ANTIANDROGENIC EFFECTS OF SPIRONOLACTONE: HORMONAL AND ULTRASTRUCTURAL STUDIES IN DOGS AND MEN

SHIRO BABA, MASARU MURAI, SEIDO JITSUKAWA, MAKOTO HATA AND HIROSHI TAZAKI From the Department of Urology, School of Medicine, Keio University, Tokyo, Japan

ABSTRACT

A decrease in the level of plasma testosterone and an increase in the level of plasma progesterone were noted after spironolactone had been administered for 20 days in 5 patients with prostatic carcinoma, as well as in 8 male dogs. Electron microscopic observation disclosed myelin- like bodies in the cytoplasm of Leydig and adrenocortical cells in dogs, contributing to a resolution of the mode of antiandrogenic action of spironolactone.

Spironolactone, an aldosterone antagonist currently being used as a potassium-conserving diuretic, has an antiandro- genic effect by potentially inhibiting some enzymatic steps in the androgen synthesis in the testes and adrenals. To clarify the site and mode of the antiandrogenic action of spironolac- tone in vivo, we have studied spironolactone-induced altera- tions in the synthesis of sex steroids in normal mature male dogs and men with prostatic carcinoma, as well as the morpho- logical changes in the adrenals, testes and prostate glands of dogs by conventional and electron microscopic observation.

MATERIALS AND METHODS

The animal study was conducted using 10 mongrel mature male dogs, including 2 controls weighing 15.1 ± 4.2 kg. In the 8 dogs 40 mg./kg. oral spironolactone were administered daily for 20 days. Blood samples were obtained between 8 and 9 a.m. from the antecubital vein. The plasma was separated immediately by centrifugation at room temperature, and plasma progesterone and testosterone were determined by radioimmunoassay. Serum electrolyte determinations were available with the use of an autoanalyzer. On the morning of the 20th day all 10 dogs were sacrificed and the prostate, testes and adrenals were removed. The absolute weight of these organs and the weight of these organs per kilogram sac- rificed body weight were recorded. All the specimens were pre- pared for conventional and electron microscopic observation. The specimens for electron microscopic study were fixed in 2.5 per cent glutaraldehyde, washed in 0.07M. borate buffer (pH 7.4), post-fixed in 2 per cent osmium tetroxide and then em- bedded in Epon 812. Ultrathin sections were stained with uranyl acetate and lead citrate.

Five patients, ranging in age from 50 to 76 years and having histologically confirmed prostatic carcinoma, were included in the study. No patient had received any previous treatment such as castration. Three of the patients were treated with full doses of 400 mg. per day spironolactone for more than 2 weeks. The remaining 2 were treated with reduced doses, 250 and 200 mg. per day, because of complications of angina pectoris and electrolyte imbalance caused by chronic renal failure, respectively (table 1). Blood samples were obtained weekly to determine serum electrolytes, progesterone and testosterone.

Plasma progesterone and testosterone were determined by radioimmunoassay using specific antibodies .* The 1,2-3H pro- gesteronet and 1,2-3H testosteronet were diluted to a concen-

Accepted for publication May 6, 1977.

Read at annual meeting of American Urological Association, Chicago, Illinois, April 24-28, 1977.

* Clinical Research Laboratory, Central Institute for Experi- mental Animals, Tokyo, Japan.

t New England Nuclear Corporation.

tration of 20 to 25 „Ci./ml. by a mixture of benzene and etha- nol (9:1) and stored at 4C until use.

RESULTS

Animal experiments. In dogs treated with spironolactone the plasma progesterone increased significantly to a maxi- mum value of 1.9 + 0.8 ng./ml. (mean + standard deviation) from 0.27 ± 0.1 ng./ml. during a period of 2 weeks (p < 0.005), while plasma testosterone decreased significantly to 54.2 + 25 ng./dl. from 125 ± 20 ng./dl. within 20 days (p < 0.025) (fig. 1). No significant changes were noted in serum electrolytes throughout the study.

Spironolactone caused a significant decrease in the absolute and relative weights of the prostate (p < 0.025) (table 2). However, no significant difference was noted in the absolute and relative weights of the adrenals and testes.

A conventional microscopic study revealed no significant changes in the adrenals and testes but remarkable atrophic epithelial changes were noted in the prostate of the animals treated with spironolactone, characterized by a change from tall to low columnar epithelium and loss of epithelial convo- lutions (fig. 2).

Electron microscopic observation and light microscopic study showed that the normal prostates were lined by tall columnar cells with typical characteristics of secretory cells. The cytoplasm contained a large number of elongated cister- nae of rough-surfaced endoplasmic reticulum and abundant mature secretory granules containing moderately dense ma- terial near the apex (fig. 3, A). In contrast, the prostates of the animals treated with spironolactone showed a marked decrease of these secretory granules and the appearance of many irregularly shaped vacuoles in the cytoplasm (fig. 3, B).

Despite the negative findings by conventional microscopy an ultrastructural study revealed characteristic laminated membranous structures identified as so-called spironolactone bodies in the glomerular cells of the adrenal cortices of animals treated with spironolactone. Smooth endoplasmic reticulum was increased quantitatively and the mitochondria were larger and more pale than those of controls. No differ- ences in nuclei or other cell membranes were observed be- tween the animals treated with spironolactone and control animals (figs. 4 and 5). Also, seemingly similar myelin-like structures were found in the cytoplasm of Leydig cells (fig. 6). These structures were comprised of smooth-surfaced, double- layered membranes, concentrically arranged about a homoge- neous lipid-like core, that were morphologically identical to the so-called spironolactone body. No distinctive alterations in the cytoplasm of the spermatogonium and Sertoli cell were observed.

Clinical study. While plasma progesterone increased from 0.25 ± 0.10 ng./ml. (mean ± standard deviation) to a maxi-

TABLE 1. Patients treated with spironolactone
Pt.Age (yrs.)Serum Creatinine (mg./dl.)Duration (days)Dosage (mg./day)Electrolyte ImbalanceOther Complications
1761.127400NoneNone
2691.020400NoneNone
3651.224250*NoneAngina pectoris
4500.920400NoneNone
5702.222200+Hyperkalemia (transient)Anorexia

* Lower dose owing to angina pectoris.

t Lower dose owing to chronic renal failure.

TABLE 2. Antiandrogenic effect of spironolactone on canine prostate and testis
No. Ani- malsBody WeightOrgan WeightOrgan Weight/Kg. Sacrificed Animal
Before Treatment (kg.)After Treatment (kg.)Adrenal (gm.)Testis (gm.)Prostate (gm.)Adrenal (mg./kg.)Testis (mg./kg.)Prostate (mg./kg.)
Control216.3 ± 6.214.5 ± 4.91.5 ± 0.810.9 ± 4.29.9 ± 4.197 ± 20750 ±32674 ± 53
Spironolactone*814.7 ± 3.911.8 ± 2.51.2 ± 0.59.7 ± 2.55.9 ± 1.2+94 ± 30731 ± 159506 ± 69+

* 40 mg./kg. orally daily for 20 days.

+ Significant decrease (p < 0.025).

mum value of 1.3 ± 0.31 ng./ml. by 20 days after the adminis- tration of spironolactone the plasma testosterone decreased to 200 ± 80.3 ng./dl. from 427 ± 74.3 ng./dl. during the same in- terval (fig. 7). These changes were statistically significant (p < 0.025).

The serum sodium, chloride and potassium levels remained within normal range during the same period of spironolactone administration in all cases except 1. In this particular patient the serum potassium level increased to 5.3 mEq./l. tran- siently. The bilateral ureteral obstruction observed in this patient had resulted in an elevation of the serum level of blood urea nitrogen and creatinine to 28.3 and 1.4 mg./dl., respectively. Therefore, this patient was treated with a re- duced dose of 200 mg. spironolactone daily. However, the patient experienced relief of pain from bony metastasis of prostatic carcinoma after 10 days. Of the 5 men receiving spironolactone 1 exhibited nodular induration of the breasts after 2 weeks.

DISCUSSION

Since spironolactone first became widely available in 195. there have been several reports of gynecomastia and loss of libido in patients being treated with it. The reports consist of individual case reports and an incidental single observation in a group of patients being treated with spironolactone.1-3 Since Greenblatt and Koch-Weser reported on 788 hospitalized patients who received spironolactone and found a relatively lower incidence of gynecomastia (1.2 per cent) it has been thought generally that gynecomastia is a rather uncommon occurrence associated with spironolactone.+ However, Clark reported on gynecomastia occurring in 4 of 7 men during treatment with spironolactone and strongly recommended more careful examination of the breasts of all patients receiv- ing spironolactone.5 Castro and associates studied the clinical use of spironolactone for benign prostatic hyperplasia and found some temporary effects.6

The mechanism of action of spironolactone causing these estrogenic effects remains unknown; however, it is thought that alteration of the normal gonadal-hypothalamic-pituitary interrelations may have an important role. Recent research indicates that spironolactone inhibits androgen formation in microsomes, characterized by a rapid loss of testicular micro- somal p-450 with a concomitant decrease in the activity of 17 «-hydroxylase, which catalyzes the conversion of progesterone to 17 a-hydroxyprogesterone, a main precursor of the testos- terone.7 Menard and associates have suggested that the de- struction of microsomal cytochrome p-450 by spironolactone may be limited to those tissues in which microsomal 17@- hydroxylase activity is high.” The adrenal glands that produce

cortisol in such species as dog and guinea pig have a high level of 17a-hydroxylase activity. Therefore, spironolactone administration in these species as well as in human beings may cause a substantial loss of testicular and adrenal micro- somal cytochrome p-450 with a concomitant decrease in 17@- hydroxylase activity.

If spironolactone significantly decreased the production of testosterone and adrenal androgens in dog and man the plasma steroid hormonal pattern also would have changed subsequently. Such drug-induced biochemical effects might account for the clinically observed endocrine effects of spiron- olactone therapy. Stripp and associates demonstrated the increase of plasma progesterone after treatment by spironolac- tone in man. However, they could not show any significant

FIG. 1. Effect of spironolactone on plasma levels of progesterone and testosterone in dogs.

CANINE PLASMA PROGESTERONE

Ng/ml

2.0

1.0

NORMAL RANGE

0

Ng/dl

CANINE PLASMA TESTOSTERONE

150

NORMAL RANGE

100

50

0

Day 4

Day 9

Day 14

Day 20

SPIRONOLACTONE (40 mg/Kg/day)

FIG. 2. A, cross-section of prostate taken from control dog that had received no spironolactone. B, cross-section of prostate taken from dogs given 40 mg./kg. spironolactone daily for 20 days. Note decreased number of convolutions and change to low columnar epithelial cells. Reduced from ×25.
FIG. 3. A, electron microscopic appearance of prostate from control dog. Much of cytoplasm is filled with elongated cisternae of rough- surfaced endoplasmic reticulum. Mature secretory granules are abundant and more common toward apex of cell. N, nucleus. RER, rough- surfaced endoplasmic reticulum. SG, secretory granules. B, electron microscopic appearance of prostate trom dog given 40 mg./kg. spironolactone daily for 20 days. Note decreased number of secretory granules and appearance of many irregularly shaped vacuoles in cytoplasm. V, vacuole. Reduced from ×3,000.

B

10 u

SG

V

RER

change of plasma testosterone, probably owing to the short duration and small dosage of spironolactone administered.9 Walsh and Siiteri found suppression of plasma testosterone, androstenedione and dehydroepiandrosterone by high doses of spironolactone in 7 castrated men with prostatic carci- noma.1º In our study plasma testosterone decreased and pro- gesterone increased significantly after spironolactone had

been given for 20 days in 8 male dogs and in 5 men with prostatic carcinoma. These studies indicate that spironolac- tone suppresses testicular and adrenal androgen production by the inhibition of the enzymatic step involved in the conver- sion of progesterone to androgen. In addition, it has been shown that spironolactone also reduces aldosterone secretion in men.11 Although the enzymes involved in the inhibition of

FIG. 4. A, portion of cells in zona glomerulosa of adrenal cortex from control dog. Mitochondria are round, oval or elongated with lamellar cristae. M, mitochondria. Reduced from x7,000. B, portion of cells in zona glomerulosa of dog given 40 mg./kg. spironolactone daily for 20 days, demonstrating laminated membranous structure (arrow). Smooth endoplasmic reticulum is abundant. Mitochondria are large and pallid. SER, smooth endoplasmic reticulum. Reduced from ×7,200.

A

B

5 1

SER

SER

FIG. 5. Cytoplasm of another zona glomerulosa cell of spironolac- tone-treated dog exhibiting characteristic laminated membranous inclusion identified as so-called spironolactone body (arrows). Re- duced from ×26,700.

N

M

M

FIG. 6. Electron microscopic appearance of testis from spironolac- tone-treated dog. Laminated myelin-like structures (arrows) in cytoplasm of Leydig cell look seemingly similar to so-called spirono- lactone bodies. Reduced from ×9,500.

5 0

DO

-2

E

53

84

6

N

FIG. 7. Effect of spironolactone on plasma levels of progesterone and testosterone in patients with prostatic carcinoma.

Ng/ml

1.50

PLASMA PROGESTERONE

1.00

0.50

0.00

Ng /dl

500

PLASMA TESTOSTERONE

400

300

200

100

0

Day 10

Day 15

Day 20

spironolactone on aldosterone production are not understood clearly it is reasonable to suggest that spironolactone acts directly on the enzyme system contributing to the progester- one metabolic pathway in the adrenal and testis.

As shown in the animal experiments in our study electron microscopic observation revealed numerous enlarged mito- chondria and smooth endoplasmic reticulum, as well as so- called spironolactone bodies, in the cytoplasm of the adreno- cortical cell in the animals treated with spironolactone. The appearance of the laminated bodies has been reported previ- ously in the glomerular zone of the adrenal glands of patients treated with spironolactone.12 The histochemical study by Jenis and Hertzog has shown that the core of the spironolac- tone bodies consists of neutral lipids whereas the rings are probably a phospholipid-protein complex.13 Electron micros- copy revealed that the bodies are composed of concentric agranular membranes surrounding a lipid-like core.14 These results strongly suggested that the bodies are derived from the smooth endoplasmic reticulum. Mitochondria may have an important role in the formation of spironolactone bodies but the mechanisms involved in the course of the formation of these bodies were not understood clearly.15 In our study similar laminated structures also were found in the cytoplasm of Leydig cells. Dietert and Scallen have suggested the mech- anisms of formation of similar myelin-like structures appear- ing in the process of inhibition of cholesterol biosynthesis in murine adrenal gland and testis.16 It is now generally agreed that the enzymes involved in the cholesterol biosynthesis and metabolic pathway of progesterone into testosterone are found mainly in the microsomal fraction derived from the endoplas- mic reticulum. The predominant cytological feature in the adrenals and testes observed in our series by electron micro- scopic study is the occurrence of myelin-like, cytoplasmic inclusions that are considered to be implicated in the same morphogenesis.

From these observations it would seem that the so-called

spironolactone bodies might appear in the process of inhibitory action of spironolactone on the enzymatic step involved in the progesterone metabolic pathway, because the site of conver- sion is predominantly located in the smooth endoplasmic reticulum of the adrenocortical cells and Leydig cells as well. From a hormonal standpoint this observation corresponds to the morphological observations by electron microscopy ob- tained from our current study.

Our conclusion, based upon this animal experiment, strongly indicates that spironolactone causes remarkable atro- phy of the canine prostate, confirmed by light and electron microscopy, and this antiandrogenic effect of spironolactone is, so to speak, an indirect action of decreasing the plasma testosterone level, although we are aware of direct antiandro- genic action of spironolactone reported by Steelman and associates.17 The principal goal of hormonal therapy in the treatment of prostatic cancer is the suppression of androgenic stimuli to the prostate. This effect can be achieved by 1 of 4 mechanisms: 1) suppression of pituitary luteinizing hormone release, 2) removal of the androgen-producing organ, 3) direct inhibition of steroidogenesis in the testes and adrenals and 4) the inhibition of androgenic action at target tissues. Even though our speculation stemmed from our current animal experiment of non-malignant canine prostates, it is our belief that this speculation should be extended further to malignant disease of the prostate and that spironolactone and its related agents could be used as antiandrogenis drugs, leading to their clinical application to the treatment of patients with cancer of the prostate.

Drs. N. Osawa and S. Yamamoto assisted in the preparation of the hormonal assay and Mr. N. Komatsu provided technical assistance for electron microscopy.

REFERENCES

1. Smith, W. G .: Spironolactone and gynecomastia. Lancet, 2: 886, 1962.

2. Williams, E .: Spironolactone and gynecomastia. Lancet, 2: 1113, 1962.

3. Mann, N. M .: Gynecomastia during therapy with spironolac- tone. J.A.M.A., 184: 778, 1963.

4. Greenblatt, D. J. and Koch-Weser, J .: Adverse reactions to spironolactone. J.A.M.A., 225: 40, 1973.

5. Clark, E .: Spironolactone therapy and gynecomastia. J.A.M.A., 193: 157, 1965.

6. Castro, J. E., Griffiths, H. J. L. and Edwards, D. E .: A double- blind, controlled, clinical trial of spironolactone for benign prostatic hypertrophy. Brit. J. Surg., 58: 485, 1971.

7. Menard, R. H., Stripp, B. and Gillette, J. R .: Spironolactone and testicular cytochrome p-450: decreased testosterone for- mation in several species and changes in hepatic drug metab- olism. Endocrinology, 94: 1628, 1974.

8. Menard, R. H., Martin, H. F., Stripp, B., Gillette, J. R. and Bartter, F. C .: Spirnolactone and cytochrome p-450: impair- ment of steroid hydroxylation in the adrenal cortex. Life Sci., 15: 1639, 1974.

9. Stripp, B., Taylor, A. A., Bartter, F. C., Gillette, J. R., Loriaux, D. L., Easley, R. and Menard, R. H .: Effect of spironolactone on sex hormones in man. J. Clin. Endocr., 41: 777, 1975.

10. Walsh, P. C. and Siiteri, P. K .: Suppression of plasma androgens by spironolactone in castrated men with carcinoma of the prostate. J. Urol., 114: 254, 1975.

11. Sundsfjord, J. A., Marton, P., Jørgensen, H. and Aakvaag, A .: Reduced aldosterone secretion during spironolactone treat- ment in primary aldosteronism: report of a case. J. Clin. Endocr., 39: 734, 1974.

12. Janigan, D. T .: Cytoplasmic bodies in the adrenal cortex of patients treated with spironolactone, Lancet, 1: 850, 1963.

13. Jenis, E. H. and Hertzog, R. W .: Effect of spironolactone on the zona glomerulosa of the adrenal gland: light and electron microscopy. Arch. Path., 88: 530, 1969.

14. Davis, D. A. and Medline, N. M .: Spironolactone (aldactone) bodies, concentric lamellar formations in the adrenal cortices

of patients treated with spironolactone. Amer. J. Clin. Path., 54: 22, 1970.

15. Fisher, E. R. and Horvat, B .: Experimental production of so- called spironolactone bodies. Arch. Path., 91: 471, 1971.

16. Dietert, S. E. and Scallen, T. J .: An ultrastructural and bio- chemical study of the effects of three inhibitors of cholesterol

biosynthesis upon murine adrenal gland and testis. Histo- chemical evidence for a lysosome response. J. Cell Biol., 40: 44, 1969.

17. Steelman, S. L., Brooks, J. R., Morgan, E. R. and Patanelli, D. J .: Anti-androgenic activity of spironolactone. Steroids, 14: 449, 1969.