Relative and combined effects of ethanol and protein deficiency on gonadal function and histology

Alcohol,Vol. 11, No. 5, pp. 355-360, 1994 Copyright©1994ElsevierScienceLtd Printed in the USA.All rightsreserved 0741-8329/94$6.00 + .00

Pergamon 0741.8329(94)E0016-6

Relative and Combined Effects of Ethanol and Protein Deficiency on Gonadal Function and Histology E. G O N Z A L E Z - R E I M E R S , *l A. M A R T I N E Z - R I E R A , * F. S A N T O L A R I A - F E R N A N D E Z , * A. C O N D E - M A R T E L , * H. A L V A R E Z - A R G O E L L E S , T C. S A N T A N A - H E R R E R A ~ / A N D F. R O D R I G U E Z - M O R E N O *

Departments o f *Medicina Interna, TAnatomia Patologica, and Y,Fisiologia, Facultad de Medicina, Hospital Universitario de Canarias, La Laguna, Tenerife, Canary Islands, Spain Received 11 December 1992; Accepted 24 F e b r u a r y 1994 GONZALEZ-REIMERS, E., A. MARTINEZ-RIERA, F. SANTOLARIA-FERNANDEZ, A. CONDE-MARTEL, H. ALVAREZ-ARGOELLES, C. SANTANA-HERRERA, AND F. RODRIGUEZ-MORENO. Relative and combined effects of ethanol and protein deficiency on gonadalfunction and histology. ALCOHOL 11(5) 355-360, 1994.-Tbe aim of the present study is to analyse the relative and combined effects of ethanol and protein deficiency on serum testosterone and LH, and on gonadal histology, in ethanol fed rats. The study was performed in 32 animals divided into four groups, fed with the Lieber & DeCarli comrol, 36% ethanol, 2% protein, and 36% ethanol 2% protein containing diets, respectively. Two months later, rats were anaesthetized with pentobarbital and sacrificed, and the fight testes and epididymus were carefully removed. Both ethanol and protein deficiency independently lead to a decrease in serum testosterone levels, and to testicular atrophy, lowest testosterone levels and highest degrees of atrophy being observed in the rats receiving the 36% ethanol, 2% protein containing diet. Both serum testosterone and testicular size and weight significantly correlated with final weight and serum albumin. Hypospermia, atrophy of the seminiferous tubules, and reduced epididymal diameter were also observed in this last group of animals. Thus, protein deficiency may contribute to hypogonadism of alcoholics. Ethanol

Malnutrition

Hypogonadism

Protein deficiency

CHRONIC ethanol consumption is associated with hypogonadism, and controversy exists regarding some aspects of the pathogenesis of this alteration. Although liver plays a key role on metabolism of sexual hormones, and serum testosterone levels decrease according to Child's categories of severity in alcoholic (23) and nonalcoholic cirrhosis (34), there is general agreement regarding the direct effect of ethanol itself on the testis (27), impairing testosterone synthesis (13,33), altering the circadian rhythm of testosterone secretion (4,6,24), and impairing spermatogenesis (25). Moreover, hypothalamicpituitary axis also becomes affected in chronic alcoholic patients, chronic ethanol consumption leading to blunted LH response to low testosterone levels or to stimulation with clomiphene or L H R H (5,10,28,30). Thus, both primary and secondary hypogonadism are observed in chronic alcoholics (26). Another important factor, which could play a role on the development of hypogonadism among alcoholics, is the fre-

quently observed accompanying protein calorie malnutrition. Today it is known that neurons controlling release of gonadotropin-releasing hormone are extremely sensitive to changes in nutritional state (18), malnutrition leading to inhibition of its release and, consequently, of LH secretion (11). However, there are experimental data that have yielded conflicting results regarding the role of malnutrition on sexual dysfunction in alcoholic rats. Van Thiel et al.(29) have shown that although a certain degree of protein calorie malnutrition (provoked by restriction in food intake) may lead to abnormal gonadotropin levels, hypogonadism does not ensue, in sharp contrast with the effect of ethanol. Moreover, Smith et al.(20) have observed elevated LH levels in severely malnourished men (marasmus type malnutrition), levels that returned to normal values after refeeding. On the other hand, studies performed in protein deprived (22), protein deficient (12) or protein and calorie deficient rats (14) have yielded low FSH and

To whom requests for reprints should be addressed. 355

GONZALEZ-REIMERS ET AL.

Testes.

LH values and variable degrees of testicular atrophy (14,21) or absence of histological changes (12). Based on these facts we have performed the present study to determine the relative and combined effects of ethanol and protein deficiency on gonadal structure and function, as well as on LH secretion, in the male albino rat. For this purpose, we have chosen the experimental liquid diet model designed by Lieber and DeCarli, following other authors who have studied relative and combined effects of ethanol and protein deficiency on other organs or systems (16,31,32).

• Weight. • Area of the section (also termed "testicular size" throughout the text). • Number of Leydig cells observed in ten fields (at 120 × ) of interstitial tissue located between three transverse sections of seminiferous tubules. • Atrophy of seminiferous tubules. Hematoxylin-eosin stained sections of the testes showed several patterns: testes with normal tubules, testes with moderately atrophic tubules, and testes with severe tubular atrophy; although in some cases tubular atrophy was widespread, in others, normal tubules coexisted with moderately and even severely atrophied tubules. A seminiferous tubule was considered as severely atrophied when there was total absence or nearly total absence of germ cells. The number of cells (besides spermatozoa) which were present in 50 atrophied tubule cross sections of 5 different animals ranged between 20 and 40 c e l l s - i n many cases, only Sertoli cells-distributed in 1 to 3 cell layers. Normal tubules, on the contrary, showed between 5 and 8 cell layers, with more than 200 ceils. A tubule was considered as moderately atrophied when it showed an intermediate number of cells (between 50 and 100) and cell layers (between 3 and 5). Whereas no doubtful cases were observed regarding severe atrophy, it was not easy to differentiate, in a few instances, between moderately atrophied and normal. In these cases, cells were counted, and if the number of cells was over 150, the tubule was considered as a normal one. In each of the 32 cases, the proportion of severely atrophied, moderately atrophied, and normal tubules were recorded by two independent observers after inspection of at least 100 tubule cross sections (range 100-436). We have defined a "score of atrophy," assigning 1 point to a normal tubule, 2 points to a moderately atrophied tubule, and 3 points to the severely atrophied, as follows:

ANIMALS AND METHODS

Animals Thirty-two 70-day-old male Wistar rats were divided into four groups of eight animals each. The control rats received the Lieber DeCarli control, 18% protein containing diet (15,16) (Dyets Inc, Bethlehem, PA) containing I kcal/ml (Diet I); a second group consisted of another eight animals fed an isocaloric, 36% ethanol containing diet (Diet II); the third one, an isocaloric, 2% protein containing diet (Diet III); and the fourth, an isocaloric, 2% protein and 36% ethanol containing diet (Diet IV) (Table 1). This relatively severe degree of protein deprivation was instituted to accelerate the development of protein deficiency, as stated elsewhere (32). The diets were prepared weekly by dissolving the nutrient mixture and the separate vitamin mix in water. Those rats receiving the alcohol, protein deficient diet were allowed dietary consumption ad lib, and the same amount consumed by these animals was then given to the other three groups. This pair-feeding process was repeated every 2 days, always adjusting the amount of liquid diet received by the other three groups to that consumed by the animals fed the protein deficient, ethanol containing diet. The mean dally amount of diet consumed was (in ml or Kcal, x + SD): 61.7 + 4.2, controls; 59.62 + 5, alcoholic animals; 58.5 + 7, low protein, ethanol fed animals; and 60.4 + 3.5, low protein fed animals. As theoretically expected, considering the pair-feeding method followed, these slight differences are not statistically significant, as described elsewhere (8). All the animals were alive at the end of the experiment, 2 months later. At this time, animals were anaesthetized by pentobarbital and sacrificed.

Score = (% of normal tubules) × 1 + (%o of moderately atrophied tubules) × 2 + (%o of severely atrophied tubules) × 3 Also, the parameter "atrophy" has been defined as the following proportion: (moderately atrophied + severely atrophied) × 100/total number of tubules.

Histological Studies Testes and epididymides were carefully removed, fixed in formaldehyde, included in paraffin, cut following the longitudinal axis, and stained with hematoxylin-eosin. The following parameters were recorded:

Global histological appearance o f testis tubules. We have classified it in five categories, according to Coward et al. (9) (Table 2).

TABLE 1 MEAN CONSUMPTION OF CALORIES, PROTEIN, AND ALCOHOL; WEIGHT AT THE BEGINNING AND AT THE END OF THE STUDY (X + SD) Group Control (Diet I) Alcohol (Diet II) Protein Poor (Diet III) Alcohol + Protein Poor (Diet IV)

Energy (Kcal/day) 61.7 59.6 60.4 58.5

+ + ± ±

4.2 5 3.5 7

Protein (g/day) 2.55 2.47 0.28 0.27

+ ± + ±

0.17 0.21 0.02 0.04

Alcohol (g/day) 0 3.79 + 0.3 0 3.72 ± 0.45

Weightat start (8) 324 316 322 310

+ ± ± ±

18 20 17 29

Weightat end-point(g) 392 313 259 204

+ ± + ±

ll 32 16 33

CTO: HYPOGONADISM, E T H A N O L , AND PROTEIN DEFICIENCY TABLE 2 CLASSIFICATIONOF HISTOLOGICALAPPEARANCE OF TESTIS TUBULES (9) Type A: Type B: Type C: Type D:

Type E:

Complete spermatogenesis in all tubules Complete spermatogenesis in most tubules: a few tubules lack spermatids and spermatocytes. Many tubules show complete spermatogenesis, some conrain early spermatid stages and in a few there is a severe loss of germinal epithelium. Severe loss of germinal epithelium. Many tubules are lined by Sertoli ceils and spermatogonia. In some tubules there are numerous spermatocytes but few spermatids are present. Some tubules do not have a lumen. Lining cells show a vacuolated cytoplasm. Most of the tubules are lined by Sertoli cells and spermatogonia. Spermatocytes are rare or absent, and spermatids are absent. There are tubules without a lumen, and the cytoplasmic lining of others is vacuolated.

correlation coefficients were calculated to determine the significance of the relation between pairs of variables. Two-way variance analysis was also performed to discern the independent a n d / o r interactive effects of ethanol and protein deprivation on the changes observed. Semiquantitative parameters were compared between the four groups using Kruskall-Wallis test. This test was also employed for analysing some quantitative parameters as testosterone, atrophy score, and percent of atrophy, which, in our study, showed marked differences in the variances of each of the groups (Tables 3 and 4). However, these parameters also underwent logarithmic transformation, to stabilize variances, before statistical tests (two-way variance analysis and calculation of Pearson's product moment correlation coefficient) were performed. Statistical tests and logarithmic transformations were all performed using the SPSS (Statistical package for Social Sciences, Chicago, IL). RESULTS

Weight Change and Serum Albumin Hypospermia in the tubules. Spermatozoa in some tubules were so numerous that the histological picture o f their lumen has a tanglelike appearance. On the contrary, spermatozoa in other tubuli are less abundant, so that there is possible to count them. Indeed, in these last, the number of spermatozoa varies from 7 to 63 in each tubule cross section, whereas in those with a tangle appearance in their lumen, more than 300 spermatozoa were counted. We have classified the tubules in two groups: those with a tangle appearance, and those without a tangle appearance, considering these last as hypospermic. We have counted the proportion of tubules with and without tangle appearance, and this proportion has been used in the present study to evaluate hypospermia: low values of this proportion correspond to hypospermic animals. Epididymis. Histological data were always recorded from the caudal portion of this duct in the following manner: • Measurement of duct diameter (as the mean of 10-12 sections of the epididymus in each animal); • Hypospermia, grading it in two categories, moderate or severe vs. slight hypospermia + normality. The presence of hypospermia was assessed, in a semiquantitative way, by two different observers, by inspection of epididymus. The presence of immature cells was invariably associated with few spermatozoids, and was considered a criterion of moderate to severe hypospermia. A perfect concordance was noted between assessment of hypospermia in epididymus and in seminiferous tubules.

Biochemical Studies Blood samples were obtained before killing for serum albumin determination (Bromcresol green, automated BM/Hitachi 717, Boehringer Mannheim, Mannheim, Germany). Serum testosterone levels were determined by RIA using a solid-phase '25I-antibody. Serum LH levels were also determined by RIA according to previously validated procedures (1), results being expressed in terms of N I A M D D rat LHRP-3.

Statistics The different parameters mentioned before were compared between the four groups using analysis of variance, and, after, the Student-Newman-Keuls test. Pearson's produet-moment

Weight changes suffered by the different experimental groups axe shown in Table 1, being fully in accordance with results of other authors using the same diet model (31,32). A significant decrease in serum albumin was observed in rats fed diets II, III, and IV (Table 3). No differences were observed between ethanol fed and protein deficiently fed animals, whereas statistically significant differences were observed between ethanol fed and protein poor ethanol fed animals.

Hormones Serum testosterone levels were significantly lower in the protein deficient + ethanol fed animals, independent effects of ethanol and protein deficiency being present ( p = 0.001 a n d p < 0.001, respectively; Table 5). Serum testosterone was significantly related to serum albumin (r = 0.68, p < 0.001, n = 32), final weight (r = 0.73, p < 0.001, n = 32), and weight difference (end-weight minus beginning-weight) (r = 0.73, p < 0.001, n = 32), and also with testicular size and weight (n = 32; r = 0.50, p < 0.005; r = 0.69, p < 0.001, respectively), atrophy of seminiferous tubules (r = - 0 . 5 1 , p < 0.005, n = 32), and epididymal diameter (r = 0.72, p < 0.001, n = 32) (Tables 4 and 6). There was a trend o f serum LH to be lower both in the protein deficient and in the alcoholic animals, interactions existing between both ethanol and protein deficiency (p = 0.024). Serum LH levels correlated with weight difference (r = 0.41, p < 0.05, n = 30) (Table 6).

Histological Changes Testicular weight and size. Weight of the testes of the protein deficient, ethanol fed animals (1034 =t: 511 rag, x + S D ) was clearly reduced when compared with that of the controls (2479 =l= 242.7 rag), the protein deficient animals (2142.6 + 261.3 rag), and the ethanol fed animals (2296.3 + 446.6 rag, p < 0.001), both ethanol and protein deficiency exerting independent effects on testes weight reduction, a positive interaction (p = 0.01) existing between both factors. Area o f testicular section was clearly reduced in the protein deficient, ethanol fed animals (Table 5), both ethanol and protein deficiency showing independent effects on this parameter (/7 = 0.019 a n d p = 0.001, respectively). Both testicular size and weight significantly correlated with serum albumin, weight difference and final weight (/7 < 0.001

G O N Z A L E Z - R E I M E R S ET AL~ TABLE 3 SERUM ALBUMIN, EPIDIDYMAL DIAMETER, HYPOSPERMIA, AND TUBULE ATROPHY IN THE FOUR GROUPS OF ANIMALS Albumin (g/L)

Epididymal Diameter~)

Group 1 (Control)

3.25 ± 0.32

333.1 + 18.4

82

Group2 (Alcohol)

2.84 + 0.21

315.1 ± 38.5

68.4 ± 15.6

Group 3 (Protein Poor)

2.70 ± 0.15

314.1 ± 34

81.6 ±

Group4 (Alcohol + Protein Poor)

2.16 ± 0.44

139

29.5 ± 24.9

± 37.1

Normospermic Tubules (%) ± 10

Atrophy Score 100.02 + 0.04 Median = 100 Range = 100-100.1 102.26 + 2.14 Median = 101.45 Range = 100-105.4

8.1

114.29 ± 27.48 Median = 100 Range = 100-173 152.49 ± 66.71 Median = 121.2 Range = 101.6-295

Albumin: F = 17.8, p < 0.001; 4 vs. 3, 4 vs. 2, 4 vs. 1, p < 0.05; 1 vs. 2, 1 vs. 3, p < 0.05; main effects = low protein, p < 0.001, and ethanol, p < 0.001. Epididymal diameter: F = 50,p < 0.001; 4vs. 3, 4vs. 2, 4 vs. 1, p < 0.05; main effects = low protein, p < 0.001, ethanol, p < 0.001, and interactions, p < 0.001. Normospermic tubules: F = 15.4, p < 0,001; 4 vs. 3, 4 vs. 2, 4 vs. 1, p < 0.05; main effects = ethanol, p < 0.001, low protein, p = 0.003 and interactions, p = 0.007. Atrophy score (after logarithmic transformation): F = 5.66,p = 0.009; 4 vs. 1, 4 vs. 2, p < 0.05; main effects = low protein, p = 0.007. Kruskail-Wallis test, p = 0.0006.

in all the cases for size [Table 6], p < 0.01 in all the cases for weight). Leydig cell number. We have failed to find any significant difference between the four groups regarding this parameter (Table 5). Hypospermia. Hypospermia in seminiferous tubules was observed in the protein deficient, ethanol fed animals (more than 7007o o f tubules with hypospermia vs. 32% in alcoholics, 19070 in protein deficiently fed animals, and 18070 in control rats, F = 15.4, p < 0.001). Both ethanol and protein deficiency exert independent effects on hypospermia (F = 26.3 and 11, respectively, p < 0.001 a n d p = 0.003, respectively), a sinergistic interaction between both factors being also observed (F = 8.84, p = 0.007). The proportion o f tubules without hypospermia correlated with serum albumin (r = 0.63, p < 0.001), final weight (r = 0.57, p < 0.001), weight difference (r = 0.60, p < 0.001), and serum testosterone (r = 0.64, p < 0.001), as well as with testicular size (r = 0.71, p < 0.001), testicular weight (r = 0.74, p < 0.001), epididymal diameter (r = 0.81, p < 0.001), and the score o f tubular atrophy (r = - 0 . 6 3 , p < 0.001).

Regarding histological appearance of the testes, severe loss o f germinal epithelium was observed only in the protein poor fed, alcoholic animals. Indeed, testes o f rats fed with a protein poor, alcoholic diet fit into Coward's E category (25070), D category (25070), C categories (25070), and B and A category (12.5070 in each case). All the control rats, and 75°70 of both alcohol fed and protein poor fed animals showed normal testes (A category), whereas 12.507o o f protein poor fed animals and 25070 o f ethanol fed animals showed testes belonging to Coward's B category and 12.5070 of the 2% protein fed animals belonged to category C. As expected, atrophy score was highly significantly related with Coward's histological classification o f the testes (Chi-square = 18.9, p = 0.0008 by Kruskall-Wallis tes0, and the same was observed when hypospermia was compared with Coward's groups (F = 23.5, p < 0.0001). Epididymal diameter. Protein deficient, ethanol fed animals showed a significantly reduced epididymal diameter, both low protein and ethanol exerting independent effects (p < 0.001 in both cases), and also a positive interaction, effect o f both factors together being much more pronounced

TABLE 4 RELATIONSHIPS BETWEEN HISTOLOGICAL PARAMETERS AND GONADAL SIZE, SERUM ALBUMIN, WEIGHT DIFFERENCE, TESTOSTERONE, AND LH Size Atrophy Score Epididymal size Normospcrmia

- 0.62~/ - 0.53 ¶ 0.74~t 0.71 :~

Albumin -0.37§ - 0.25 0.71~t 0.63:~

Weight Difference Testosterone* -0.50¶ - 0.461 0.72~/ 0.60~:

-0.51¶ - 0.50¶ 0.72~/ 0.64~t

LH't 0.20 0,20 0.11 - 0.25

*After logarithmic transformation; tn = 30; ~p < 0.001, §p < 0.05; ~o < 0.01.

CTO: H Y P O G O N A D I S M , E T H A N O L , A N D P R O T E I N D E F I C I E N C Y TABLE 5 TESTICULAR SIZE, LEYDIG CELLS' NUMBER, SERUM TESTOSTERONE AND LH IN THE FOUR GROUPS OF ANIMALS Testosterone (ng/dl)

LH (ng/ml)

94.75 + 20

166.6 + 103.1 Range = 41.7-369.2 Median =- 146.15

1.65 + 0.41

166.6 ± 31.5

109.4 ± 17.4

74.7 + 76.8 Range = 4.2-209 Median = 46.95

1.09 ± 0.67

148.9 ± 27.3

107

74.9 ± 28.3 Range = 3.4-188.5 Median = 74.4

0.98 ± 0.13

2.4 ± 2.5 Range = 0.1-5.9 Median = 1.5

1.4 ± 0.48

Size~2)

Leydig

Group 1 (Control)

174.5 + 38.9

Group 2 (Alcohol) Group 3 (Protein Poor) Group4 (Alcohol + Protein Poor)

93.4 ± 37.7

± 17.8

115.5 ± 18

Size: F = 9, p --- 0.0003; 4 vs. 3, 4 vs. 2, 4 vs. 1, p < 0.05; main effects: protein deficiency, p = 0.001, ethanol, p = 0.019. Number of Leydig cells: nonsignificant (F = 2). Testosterone (after logarithmic transformation): F = 18.4,p < 0.001; 1 vs. 2, I vs. 3, I vs. 4,p < 0.05; main effects: protein deficiency, p < 0.001, ethanol, p = 0.001. Kruskail-Wailis test, p = 0.0014. LH: F -- 3, p = 0.049; no two groups significantly different by SNK test. Interactions of ethanol protein deficiency, p = 0.024. than the sum of the effects of each of them ( p < 0.001). Close correlations were observed between epididymal diameter and testicular size (r = 0.74) and weight (r = 0.77), serum albumin (r = 0.71), weight difference (r = 0.72), and testosterone (r = 0.72) (Table 4). DISCUSSION Our results clearly show that both ethanol and protein deficiency play independent, sinergistic roles both o n gonadal endocrine function and spermatogenesis. This conclusion is in disagreement with the aforementioned study of Van Thiel et al. (29), but in accordance with those observations that point to the development of hypogonadism in undernourished individuals (7,20). However, it is necessary to take in mind that these studies deal with patients with protein-calorie malnutrition, not pure protein deficiency, as in our study; in the same way, Van Thiers weight-restricted rat model (29) is also different from the protein deficiently fed animals here described. Other authors, using diets containing 907e protein have failed to observe significant testicular histological alterations (12), and although total absence o f protein in the diet leads to a certain degree of testicular changes, atrophy was not as

marked as that observed in hypophysectomized animals (21). Indeed, in our study, independent effects of protein deficiency and ethanol were observed both regarding testosterone levels and most of the histological parameters, although markedly depressed testosterone levels and outstanding histologic alterations were observed only when both factors coexisted, suggesting an interactive effect of ethanol and protein deficiency. It is also important to point out that our results are solely due to protein deprivation, not to low intakes of specific nutrients as zinc or vitamin A, which probably play important roles on spermatogenesis and Leydig cell function (17,19). In our study, serum LH levels showed a tendency towards a reduction both in alcoholics and protein deficient animals despite low testosterone levels. The normality of L H values despite low testosterone values in alcoholic rats is in accordance with the results reported by other authors (3,29), and normal or near-normal L H values have also been obtained in animals fed low protein diets (12,14). This double defect, i.e., low testosterone secretion and an apparently blunted response of the hypothalamic-hypophyseal axis to these low levels is particularly worth noting in the protein poor + ethanol fed animals. However, an interactive effect seems to exist between ethanol and protein deficiency, so that protein deficient, etha-

TABLE 6 RELATIONSHIPS BETWEEN TESTOSTERONE, LH, TESTICULAR SIZE. AND LEYDIG CELLS' NUMBER WITH SEVERAL PARAMETERS

Testosterone* LH Testicular size Leydig cell Testicular weight

Albumin

Final Weight

0.62t 0.27 0.64t -0.35 0.50§

0.61t 0.36 0.56t - 0.36~/ 0.56t

Weight Difference 0.63t 0.41~: 0.60f -0.36~ 0.62t

Testosterone*

LH

0.50~ -0.26 0.691"

0.23 - 0.02 -0.15 - 0.10

*After logarithmic transformation; tp < 0.001; ~p < 0.05; §p < 0.01; r vaiues are given.

GONZALEZ-REIMERS ET A L nol fed animals showed a trend to higher serum LH levels than the others. Perhaps strong stimulation of hypothalamicpituitary axis due to the very low testosterone levels in these animals may account for this result. Serum testosterone levels and histological data showed significant relations with serum albumin, weight change, and final weight, reinforcing the relationship between protein deficiency and hypogonadism. Also, serum testosterone levels showed a relation with the intensity of tubular and epididymal atrophy; as with serum testosterone, protein deficient, ethanol fed rats showed the most severe degrees of epididymal and tubular atrophies, and of hypospermia, both in tubules and epididymides, and these parameters were strongly related with those derived from protein deficiency, as albumin and weight difference. Both ethanol and protein deficiency exert indepen-

dent, significant roles on testicular size and weight, epididyreal atrophy, and hypospermia, but protein deficiency seems to play a more intense role on tubular atrophy than ethanol does. It is also important to point out the absence of variation in Leydig cell number, despite variations in testosterone levels and the aforementioned histological changes. However, this fact (preservation or even increase in Leydig cell number) has been described in other conditions in which testicular atrophy ensues (2). In summary, this study shows that both a low protein diet and ethanol play independent roles on testicular dysfunction of the male Wistar rats, hypogonadism becoming aggravated when both ethanol and protein deficiency coexist in the same animal. Thus, protein deficiency may contribute to hypogonadism in alcoholic individuals.

REFERENCES 1. Alonso, R.; Prieto, L.; Herndndez, C.; Mas, M. Antiandrogenic effect of the pineal gland and meiatonin in castrated an intact prepuberal male rats. J. Endocrinol. 79:77-83; 1978. 2. Arighi, M.; Singh, A.; Bosu, W. T. K.; Homey, F. D. Histology of the normal and retained equine testis. Acta. Anat. 129:127130; 1987. 3. Bannister, P.; Handley, T.; Chapman, C.; Losowsky, M. S. Hypogonadism in chronic liver disease: impaired release of luteinising hormone. Br. Med. J. 293:1191-1193; 1986. 4. Bertello, P.; Agrimonti, F.; Gurioli, L.; et al. Circadian patterns of plasma cortisol and testosterone in chronic male alcoholics. Alcohol Clin. Exp. Res. 6:474-481; 1982. 5. Boyden, T. W.; Pamenter, R. W. Effects of ethanol on the male hypothalamic-pituitary-gonadal axis. Endocr. Rev. 4:389-395; 1983. 6. Castiila-Garcia, A.; Santolaria-Ferndmdez,F.; Gonz~ez-Reimers, E.; et al. Alcohol-induced hypogonadism: reversal after ethanol withdrawal. Drug Alcohol Depend. 20:255-260; 1987. 7. Coble, Y. D.; Bardin, C. W.; Ross, G. T.; Darby, W. T. Studies of endocrine function in boys with retarded growth, delayed sexual maturation and zinc deficiency. J. Clin. Endocrinol. Metabol. 32:361-367; 1971. 8. Conde-Martei, A.; Gonzitlez-Reimers,E.; Santolaria-Fern~ndez, F.; et al. Relative and combined roles of ethanol and protein deprivation on hepatic iron, zinc, copper, and manganese. Alcohol 9:341-348; 1992. 9. Coward, W. A.; Howell, J. McC.; Pitt, G. A. J.; Thompson, J. N. Effects of hormones on reproduction in rats fed a diet deficient in retinol (Vitamin A alcohol) but containing methyl retinoate (Vitamin Aacid methyl ester). J. Reprod. Fert. 12:309-317; 1966. 10. Faedo-Presa, I.; Rodrigo-Sdez, L.; P~rez-Alvarez,R.; et al. Valoraci6n del eje hipottllamo-hipofisario-gonadal en la cirrosis hep~itica en varones. Gastroenterologfa y Hepatologia 2:141-146; 1979. 11. Foster, D. L.; Olster, D. H. Effect of restricted nutrition on puberty in the lamb: patterns of tonic hiteinizing hormone (LH) secretion and competency of the LH surge system. Endocrinology 116:375-381; 1985. 12. Glass, A. R.; MeUitt, R.; Vigersky, R. A.; Swerdloff, R. S. Hypoandrogenism and abnormal regulation of gonadotrOpin secretion in rats fed a low protein diet. Endocrinology 104:438-442; 1979. 13. Gordon, G. G.; Airman, K.; Southren, A. L.; Rubin, E.; Lieber, C. S. Effect of alcohol administration on sex hormone metabolism in normal men. N. Eng. J. Med. 295:793-797; 1976. 14. Herbert, D. C. Growth patterns and hormonal profile of male rats with protein-calorie malnutrition. Anat. Record 197:339354; 1980. 15. Lieber, C. S.; DeCarli, L. M.; Sorrell, M. Experimental methods of ethanol administration. Hepatology 10:501-510; 1989. 16. Lieber, C. S.; DeCarli, L. M. Liquid diet technique of ethanol administration: 1989 update. Alcohol Alcohol 24:197-211; 1989.

17. McClain, C. J.; Van Thiel, D. H.; Parker, S.; Bradzin, L. K.; Gilbert, H. Alterations in zinc, vitamin A, and retinol-binding protein in chronic alcoholics: a possible mechanism for night blindness and hypogonadism. Alcohol Clin. Exp. Res. 3:135-138; 1979. 18. Robinson, J. J. Nutrition in the reproduction of farm animals. Nutr. Res. Rev. 3:253-276; 1990. 19. Russell, R. M. Vitamin A and zinc metabolism in alcoholism. Am. J. Clin. Nutr. 33:2741-2749; 1980. 20. Smith, S. R.; Chetri, M. K.; Johanson, A. J.; Radfar, N.; Migeon, C. J. The pituitary-gonadal axis in men with protein-calorie malnutrition. J. Clin. Endocrinol. Metabol. 41:60-69; 1975. 21. Srebnik, H. H. Sex differences in pituitary gonadotrophic function of protein-deficient rats. Endocrinology 75:716-720; 1964. 22. Srebnik, H. H.; Nelson, M. M. Anterior pituitary function in male rats deprived of dietary protein. Endocrinology 70:723-730; 1961. 23. Valim~iki,M.; Salaspuro, M.; H~k6nen, M.; Ylikahri, R. Liver damage and sex hormones in chronic male alcoholics. Clin. Endocrinol. 17:469-477; 1982. 24. Valimiki, M.; Pelkonen, R.; H~k6nen, M.; Ylikahri, R. Hormonal changes in noncirrhotic male alcoholics during ethanol withdrawal. Alcohol Alcohol 19:235-242; 1984. 25. Van Thiel, D. H.; Gavaler, J. S.; Lester, R. Ethanol inhibition of vitamin A metabolism in testes: possible mechanism for sterility in alcoholics. Science 186:941-942; 1974. 26. Van Thiel, D. H.; Lester, R.; Sherin, R. J. Hypogonadism in alcoholic liver disease: evidence for a double defect. Gastroenterology 67:1188-1199; 1974. 27. Van Thiei, D.; Gavaler, J. S.; Lester, R.; Goodman, D. Alcoholinduced testicular atrophy. Gastroenterology 69:326-332; 1975. 28. Van Thiel, D. H.; Lester, R. Alcoholism: its effect on hypothalamic pituitary gonadal function. Gastroenterology 71:318-327; 1976. 29. Van Thiel, D. H.; Gavaler, J. S.; Herman, G. B.; Lester, R.; Smith, W. I.; Gay, V. An evaluation of the respective roles of liver disease and malnutrition in the pathogenesis of hypogonadism seen in alcoholic rats. Gastroenterology 79:533-538; 1980. 30. Van Thiel, D. H. Ethanol: its adverse effect upon the hypothalatalc-pituitary-gonadal axis. J. Lab. Clin. Med. 101:20-23; 1983. 31. Wilson, J. S.; Korsten, M. A.; Lieber, C. S. The combined effects of protein deficiency and chronic ethanol administration on rat ethanol metabolism. Hepatology 6:823-829; 1986. 32. Wilson, J. S.; Korsten, M. A.; Leo, M. A.; Lieber, C. S. The combined effects of protein deficiency and chronic ethanol consumption on rat pancreas. Dig. Dis. SCi. 33:1250-1259; 1988. 33. Ylikahri, R.; Huttunen, M. Low plasma testosterone values in men during hangover. J. Steroid Biochem. 5:655-658; 1974. 34. Zifroni, A.; Schiavi, R. C.; Schaffner, F. Sexual function and testosterone levels in men with nonalcoholic liver disease. Hepatology 14:479-482; 1991.