Alcohol-Induced Testicular Atrophy

GASTROENTE HOLOGY 69: 326-332, 1975 Copyright© 1975 by The Willia ms & Wilkins Co.

Vol. 69, No.2 Print ed in U.S.A.

ALCOHOL-INDUCED TESTICULAR ATROPHY An experimental model for hypogonadism occurring in chronic alcoholic men DAviD H. VAN THIEL , M.D. , GOODMA N, M.D.

JuDITH

S. GAVALEH , Ror.EH

LESTER,

M.D. , AND M. DAviD

Division of Gastro enterolo{iy, Department of Medicin e, Uni versity of Pittsburfih School of Medicine, Pittsbur{ih , Pennsy lvania; and Departm ent of Pathologv, Cedars-Sinai M edical Cent er, Mt. Sinai Hospital Division, L os An{ieles , California

Elucidation of mechanisms involved in the hypogonadism and feminization observed in chronic alcoholic men requires the development of an experimental animal model system. Such an animal system should be inducible with ethanol feeding and should duplicate endocrine changes known to occur in chronic alcoholic men. We report such an animal model system . Animals fed a diet with ethanol accounting for 36% of total calories develop significant testicular, prostatic, and seminal . vesicle atrophy (P < 0.01) and greatly reduced plasma testosterone levels (P < 0.01). Animals fed a similar diet with sucrose isocalorically substituted for ethanol do not. Testicular, prostatic, and seminal vesicular mass relative to body mass and plasma testosterone levels in these isocaloric control animals do not vary significantly from those obtained for age-matched control animals fed an ad libitum rat chow diet. These findings indicate that the caloric deprivation associated with chronic ethanol ingestion is not responsible for gonadal injury and atrophy of the sex steroid-sensitive tissues in the alcohol-fed animals. This animal model provides a useful means of directly examining perturbation in gonadal function that occurs in man as a consequence of chronic ethanol ingestion and confirms our previous data which suggest that ethanol is a primary testicular toxin . It is known that chronic alcoholic men with advanced liver disease frequently have testicular atrophy. 1 • 2 Histological studies have characterized this loss of gonadal mass as a consequence of germinal cell injury, although Leydig cells appear histologically normal. a-s Consistent with these histological changes, a recent clinical study has reported a high incidence of abnormal seminal fluid characteristics and sterility in chronic alcoholic men. 2 Despite the lack of histologically apparReceived December 30, 1974 . Accepted April 4, 1975. Address reprint requests to: D . H. Van Thiel, M.D. , Division of Gastroenterology, Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15261.

326

ent Leydig cell injury in testicular biopsies obtained from chronic alcoholic men, impotence is a common complaint in these men. 2 • 6 Consistent with this symptom, several endocrine studies have shown a greater than 50% reduction in plasma testosterone levels in this patient population. 2 • 7 ' 9 The recent report 2 that impotence, sterility, and abnormal seminal fluid characteristics can be found in chronic alcoholic men without advanced liver disease raises the question: what alteration in physiology or biochemistry has produced these histological and endocrine abnormalities in these men? In addition, the question of whether alcohol per se is a testicular toxin must also be considered. 10 • 11 Preliminary data that support but do not prove the hypothesis

August 1975

ALCOHOL-INDUCED GONODAL ATROPHY

that alcohol might be a primary testicular toxin that disturbs at least the reproductive function of the testes have been reported recently. 11 This study was initiated to evaluate the effect of c-hronic alcohol ingestion on testicular growth, development , and function.

327

250 -

Methods Twenty male white Wistar rats maintained on an ad libitum rat chow diet were killed by exsanguination at 20, 30, 40, 50, 6l , 72, and 87 days of age. Liver, testes, prostate and seminal vesicles, and pituitary were removed at autopsy, weighed, and placed in Bouin's solution for histological examination. Forty-six male white Wistar rats matched for body mass and age were paired and maintained in individual cages on a liquid diet with ethanol accounting for 36% of total calories, 12 or a similar diet in which sucrose is isocaloricallv substituted for ethanol for a period of 41 days. from age 20 days through age 61 days; they were then killed. Vitamin A content of the diets was adjusted so that animals that consumed the least number of calories per day ingested at least 4 time' the minimum daily requirement for this vitamin'" as retinyl acetate. Growth was maintained hy additional supplementation of the diets with retinoic acid. The animals were autopsied, and liver, testes, prostate and seminal vesicles, and pituitaries were removed, weighed, and placed in Bouin's solution for histological section. Histological sections were prepared and studied with hematoxylin and eosin. Plasma testosterone concentration~ were determined in duplicate using a radioimmunoassay method." All samples were measured in a single assay that eliminated any interassay variation. Serum total bilirubin concentration, alkaline phosphatase, and glutamic pyruvic transaminase activities were measured utilizing standard methods in use by the Clinical Chemistry Laboratories at Presbyterian-University Hospital, Pittsburgh, Pennsylvania. All samples were measured in a single assay that eliminated any interassay variation.

Results Control animals maintained on an ad libitum standard rat chow diet gained weight steadily to age 60 days. Weights at 61, 72, and 87 days of age are not significantly different (fig . 1). Testes and prostate and seminal vesicles increase in mass steadily until age 40 days, when they

-200 E

"'

150 -

t

I /

/

l/

100

I.

p------0 /

p'

/

_....

~

......,...-·- ·.....

(

'! 10

20

30

40

so

Go

10

eo

90

AGE IOAYS)

F1c;. I . Growth curves for animals studied . e, ad libitum fed animals; 1111, ethanol-fed animals; 0 . isocaloric control animals. Bmchets mark mean 1 SEM for 20 ad libitum fed animals and 26 exper imental animals at each point. Note that both isocaloric control and ethanol-fed animals g-ain weig-ht less rapidly than ad libitum control animals.

reached adult levels of size and histological appearance. Plasma testosterone levels reach adult male rat concentrations at age 40 to 50 days. Testosterone concentrations in animals 61, 72, and 87 days old did not differ from these at 50 days of age. Although the experimental animals were not killed until age 61 days, a full 21 days after normal animals on routine rat chow diet had been shown to reach peak normal adult testicular, prostate and seminal vesicle mass as well as plasma testosterone concentrations, the alcohol-fed animals had a marked reduction in testicular mass relative to body weight (0.0094 ± 0.0006) (mean ± SEM) compared to their isocaloric controls (0.0139 ± 0.0006) (P < 0.01) (fig. 2). Absolute testicular mass of the isocaloric controls (2.091 ± 0.074 g), although smaller than that of the ad libitum normal control animals (2.451 ± 0.069) killed at

VAN THIEL ET AL.

328 14 0

~S?

~

13 0 12 0


"'>-

10 0

0

0

'"'

V> V>

8


"'

V>

UJ

2i>-20

ISO CALORIC CONTROLS

ALCOHOL

FED

FIG. 2. Ratio of testicular mass relative to body mass in isocaloric control and ethanol-fed animals (mean ± SEM, n = 26 for each group). Note that when testicular mass is normalized by comparing it to body mass , the testicular atrophy present in the ethanol·fed animals is significantly greater than that seen in the isoc a loric control animals (P < 0.01).

the same age, was nearly twice that of the alcohol-fed animals (1.191 ± 0.095) (P < 0.01). The reduction in testicular mass in the alcohol-fed animals was caused primarily by a reduction in mean seminiferous tubular diameter and in the amount of germinal epithelium contained within these tubules (fig. 3) and is histologically similiar to the gonadal pathology reported in cirrhotic men. 3 - 5 In addition to the gross histological loss of reproductive tissue, Leydig cell function is also disturbed in the alcohol-fed animals. Figure 4 shows the greater than 50% reduction in ventral prostate and seminal vesicle mass relative to body mass in the alcohol-treated animals compared to the mass of these same androgen-sensitive tissues in isocalorically fed control animals (P < 0.01). The histological appearance of the seminal vesicles and ventral prostate of the alcohol-fed animals was one of complete androgen deprivation characterized by a single uniform layer of cells without apical secretory granules, whereas those of the normal ad libitum control and isocalorically fed control animals had redun-

Vol . 69, No.2

dant tall columnar epithelial cells with large amounts of apical secretory granuleslining these tissues. In keeping with the decreased evidence of androgenicity in these tissues, plasma testosterone concentrations in the alcohol -fed animals (291.0 ± 38.1 pg per ml) were less than one-fifth the values for isocaloric control animals (1619.3 ± 283.3) and less than controls fed an ad libitum diet (1589.0 ± 290.5) (both P < 0 .01) (fig. 5). Testosterone concentrations in these latter two groups are not significantly different. Despite these marked histological differences between the alcohol-fed animal testes, seminal vesicles, and prostate compared to the same tissues in both the ad libitum fed animals and the isocaloric control animals, no difference between groups was apparent on histological examination of the pituitary glands . As shown in figure 1, animals maintained on experimental diets gained weight throughout the period of study. After 30 days of age (10 days on the diet), alcohol-fed animals weighed less (87 .1 ± 1.6 g) than their isocaloric controls (94.5 ± 1.1 g). This statistically significant (P < 0.01) difference in body mass between the two groups matched for caloric intake progressively widened as the animals were maintained on their respective diets. The liver of animals maintained on an ad libitum rat chow diet weighed 11.2 ± 0.3 gat 61 days and had a mean liver to body ratio of 0.0401 ± 0.0004. Animals maintained on the alcohol diet had smaller livers (9.7 ± 0.5 g) than control animals fed the routine rat chow diet but had significantly larger livers (P < 0.01) than their isocaloric controls (5.7 ± 0.2 g). The ratio of liver mass to body mass more clearly demonstrated the significant (P < 0.01) increase in liver size in the alcohol -fed animals (0.0775 ± 0 .024) (P < 0.01) because of the fatty metamorphosis present in these animals when compared to similar ratios of the ad libitum fed animals (0.0401 ± 0.0004) and the isocalorically fed control animals (0.0374 ± 0.0011). Despite the obvious fatty metamorphosis present in the

August 1975

ALCOHOL-INDUCED GONODAL ATROPHY

329

FIG . 3. Photomicrographs of the testes. Isocalori c cont rol animals on the left and ethanol-fed animals on the right. Note obvious reduction in mean seminiferous tubular diameter, reduced number of germinal elements, and presence of occasiona l abnormal giant cells within t he seminiferous tubules of the alcohol-fed animals testes . Leydig cells appear hi stolo gica lly normal in both groups (hematoxylin and eosi n x 220).

alcohol-fed animals, reflected in the liver to body weight ratios , no histological evidence for acute inflammation , necrosis, or portal fibrosis was seen in the histological sections examined . Moreover, although a trend toward abnormal liver function measured biochemically was observed in the alcohol-fed animals, there was no significant difference between values observed for alcohol-fed and those of the ISO· caloric fed control animals . Discussion The hypogo nadism that occurs in chronic alcoholic men initially was thought to be a consequence of the associated liver disease that occurs in these men. 1 The recent observation that hypogonadism can occur in chronic alcoholics well before the development of advanced or permanent

histological liver disease" necessitates a rethinking of the pathogenesis of this phenomenon. This observation obviously raises the question: is ethanol per se rather than ethanol acting indirectly through alcoholic liver disease responsible for the testicular atrophy seen in these men 10 • 11 ? Preliminary studies performed in vitro utilizing rat tissues have suggested one possible etiological link between the development of gonadal atrophy and alcohol abuse. Ethanol may interfere with the activation of retinal by competitive inhibition of vitamin A metabolism and the enzyme alcohol dehydrogenase . These studies show that alcohol dehydrogenase activity is present in the testes. Furthermore, they show that rat testicular homogenates are capable of converting retinol to bioactive retinal, the form ot' vitamin A

VAN THIEL ET AL.

330

IS OCAL ORIC CONTROLS

ALC OHOL

FED

FIG. 4. Ratio of mass of ventral prostate and eminal vesicles relative to body mass of isocaloric .,_mtrol animals on the left and ethanol-fed animals on · he ri{?ht. Each bar represents the mean value (n = 'G). Brack ets demonstrate ± SEM.

14 E ~ 12 0..

~ 10

0

a: loJ

>-

~ >-

8

V)

loJ

><[

""~ ..J

4

a.

ISO CAL ORIC CONTR OLS

AL COHOL

FED

Plasma concentrations of testosterone [mean ± SEM (n = 26) for each group] in nanograms per mi. Isocaloric controls are shown on the left and ethanol-fed animals are shown on the right. FIG.

5.

required for normal spermatogenesis. 11 Moreover, they show that ethanol levels in drinking males well below the minimum plasma concentrations found in legal intoxication completely inhibit retinol oxidation by these testicular homogenates. These data extend these preliminary in vitro observations. They clearly document that both germinal cell injury and Leydig cell dysfunction occur in animals chronically maintained on a diet with ethanol ac;counting for 36% of the total caloric intake,

Vol. 69, No.2

whereas animals maintained on a similar diet without ethanol do not. Although these new observations do not prove that the mechanism responsible for these gonadal effects of ethanol ingestion is a consequence of disturbed vitamin A metabolism, they are consistent with the observations that germinal cell injury and reduced plasma testosterone concentrations occur in chronically vitamin A-deficient animals. 15 Other mechanisms, however, are still possible. Ethanol might exert its primary effect on the hypothalamic-pituitary axis, secondarily reducing gonadal testosterone production. Although clinical studies of alcoholic men demonstrate predominantly normal plasma gonadotropin concentrations, 2 • s-s it has indeed been shown that the hypothalamic-pituitary axis is less responsive in these men to · provocative stimulation than in normals. 2 Whether this central defect is a primary or a secondary phenomenon, and whether it is produced by ethanol per se or by events induced by ethanol remains uncertain. These patients are grossly feminized and estrogenization can of itself suppress hypothalamic-pituitary responsiveness. It is also possible that ethanol is directly toxic to the testes by some mechanism independent of vitamin A metabolism as it is for other tissues , 15 • 3 " but all such mechanisms remain to be evaluated. The Leydig cells are a potential site for direct alcohol toxicity within the testes. An injury to these cells could adversely effect testosterone synthesis and produce the hypogonadism so commonly seen in alcoholic men and reproduced in the model we report here. In addition , the germinal cell injury observed in alcoholic men and reflected in the high frequency of seminal fluid abnormalities found in these men might also be a consequence of altered Leydig cell function, as testosterone has been shown to be necessary for completion of normal spermatogenesis. 3 1 Even in the absence of direct histological evidence of Leydig cell injury, testosterone secretion might be abnormal as a consequence of the altered redox state of the testes which occurs as a result of ethanol metabolism within this

August 1975

ALCOHOL-INDUCED GONODAL ATROPHY

tissue. The recent study in which the presence of alcohol dehydrogenase in testicular tissue was established raises the possibility of excess NADH production at sites of testicular steroidogenesis. 11 Since several steps in the biosynthesis of testosterone are NAD-dependent, alcohol may directly interfere with testicular testosterone production. In addition to being consistent with, although more severe than, the gonadal effects of chronic vitamin A deficiency in rats, our results are consistent with the recent reports that show that acute ethanol intoxication suppresses plasma testosterone concentrations acutely in normal men 3 2 and mice 33 well before any change in liver function or histology would be expected to occur. The data reported here clearly demonstrate that the acute depression of plasma testosterone observed with alcohol intoxication is not merely a transient reaction to stress but a real tissue injury effect which is associated with permanent testicular atrophy. Specifically, they show that rats chronically fed ethanol for 41 days develop significant testicular atrophy, have a significantly reduced prostate and seminal vesicles mass, and, in addition, have greatly reduced plasma testosterone levels at a time when the only hepatic abnormality present is reversible fatty metamorphosis of the liver and age matched control rats fed an identical diet with sucrose isocalorically substituted for ethanol do not have these abnormalities. These data are therefore consistent with our hypothesis 10 that irreversible liver disease is not a prerequisite for hypogonadism in chronic alcoholic men and that alcohol may be a direct gonadotoxin. REFERENCES 1. Klatskin G: Jaundice and other manifestat ions

of liver disease. Principles of Internal Medicine. Edited by TR Harrison. New York, McGraw-Hill, 1962, p 156- 157 2. Van Thiel DH, Lester R, Sherins RJ : Hypogonadism in alcoholic liver disease: evidence for a doubl e defect. Gastroenterology 67:1188- 1199, 1974 3. Morrion e T: The effect of estrogens on the testes in hepatic insuffici e ncy . Arch Pathol 37:~9-48, 1944

331

4 . Hat her L.J: Hepati c c irrhosis and testicular atrophy. Arch Intern Med 80::l97-402, 1947 .'i. Bennett HS, Baggenstoss AH, Butt HR: The testes, breast and prostate of men who di e of cirrhosis of the liver. Am ,J Clin Pathol 20:8 14 - 820, 19!i0 6. Lemere F, Smith .JW : Alcohol induced sexual impotence. Am ,J Psychiat 1:l0:212-21:l, l97:l 7. Galveo-Teles A, And e rson D C, Hurke CW, et al: Biologically active androgens and oestradiol in men with chronic li ver disease. Lancet 1: 17:l- l77, 197:l 8. Kent .JR, Scarmuzzi R.J, Lauwers, et a!: Plasma testosterone estradiol and gonadotropins in hepatic insufficiency. Gastroenterology G4: 111 - 11 fi, 197:1 9. C hopra l.J, Tulchinsky D, Greenway FL: Estrogen-androgen imbalance in hepatic cirrhosis. Ann Intern Med 79:198- 203, 1978 10. Van Thiel DH. Lester R : Sex and alcohol. N Eng! ,J Med 29 1:2!i 1-2!i:l, 1974 11. Van Thiel DH, Gavaler .JS , Leste r R: Ethanol inhibition or vitamin A metabolism in the testes: possible mechanism li1r sterilitv in alcoholics. Science 1RG:94 1- 942, 1974 12. Lieber CS, DeCarli LM : Quantitative relationship bet ween amount of dietary fat and severity or a lcoholi c ratty liver. Am ,J Clin Nutr 2:l:474 - 478, 1970 18. Biology Data Book, vol :l. Edited by PL Altman, OS Dittner. Heth es da, Federal ion or Am erican Societies for Ex peri mental Biology, 1972, p 1467 14. Goodman RL , Hotc hkiss .J, Karsc h F.J, et al: Diurnal variations in serum testosterone concentrations in the adult male Hhesus monkey. Bioi Rep rod (in press) 15. Krueger PM, Hodge n GD, Sherins R.J: New evidence for the role of the Sertoli cell and Spomatogonia in fe edback control of FSH secretion in mal e rats . Endocrinology 95:9!i!i-9G2, 1974 16. Lieber CS: Metabolic effect produced by alcohol in the li ve r and other tissues. In Advances in Internal Medicine, vol 14. Edited by GH Stollerman. Chicago, Year Book Med ical Publishers, Inc, 1968, p 151-199 17 . Lieber CS, Davi dson CS: Some metabolic effects of ethyl alcohol. Am J Med 33:319-881, 1962 18. Beard JD, Knott DH: Hematopoietic response to experimental alcoholism. Am J Med Sci 252:518-5 22, 1966 19. Lindenbaum J, Lieb er CS: Effect of ethanol on the blood and bone marrow. Am J Clin Nutr 21:533-5:14, 1968 20. Sullivan LW, Heuh ert V: Suppression of he matopoies is by ethanol. ,J Clin Invest 48:2048-2062, 1964 21. Priest RG, Binns .JK, Kitchen AH: Electrocardiogram in alcoholism and accompanying physical disease. Br Med J 1:14!1:1-1455, 19()6

332

VAN THIEL ET AL.

22. Regan TJ, Kroxenidis G, Mosc hos CB, et a l: The acute metabolic and hemodynamic responses of the left ventricle of ethanol. J Clin Invest 45:270-280, 1966 23. Ekbom K, Hed R, Kirstein L, et al: Musc ular affections in chronic alcoholism. Arch Neural 10:449-458, 1964 24. Erleborn JW , Pilz CG: Paroxysmal myoglobinuria. JAMA 181 :1111-1115, 1964 25. Hed R, Lundmark C, Fahlgren H , et al: Acute muscular syndrome in chronic alcoholism. Acta Med Scand 171:585-599, 1962 26. Valaitis J, Pilz CG, Oliver H , et al: Myoglobinuria, myoglobinuric nephrosis and alcoholism. Arch Pathol 70:195-202, 1960 27. Symposium on Neurologic and Hepatic Complications of Alcoholism: on the etiology of the alcoholic neurologic diseases with special reference to the role of nutrition . Am J Clin Nutr

Vol. 69, No.2

9:379, 1961 28. Wallgren H , Barry H: Effects of ethanol on various physiological functions. In Actions of Alcohol. New York, Elsevier Publishing Co, 1970, p 115- 208 29. Lieber CS, Jones DP, Losows ky MJ , et al: Interrelation of uric acid and ethanol metabolism in man. J Clin Invest 41:1863-1873, 1962 30. Field JB, Williams HE, Mortimore GE: Studies on the mechanism of ethanol induced hypoglycemia . J Clin Invest 42:497-506, 1963 31. Steinberger E: Hormonal control of mammalian spermatogenesis. Phys iol Rev 51:1-22, 1971 32. Ylikahri R, Huttunen M, Harkonen M, et al: Low plasma testosterone values in men during hangover. J Steroid Biochem (in press) 33. Bard F, Bartke A: Effect of ethyl alcohol on plasma testosterone level in mice . Steroids 23:921-928, 1974