Environmental Health Perspectives Vol. 57, pp. 225-231, 1984 Spermatotoxicity Associated with Acute and Subchronic Ethoxyethanol Treatment by Harold Zenick,* Deborah Oudiz* and Robert J. Niewenhuist Investigations of the male reproductive toxicity of ethoxyethanol (ethylene glycol monoethyl ether) have been restricted exclusively to histopathological assessments of the testes. The present study consisted of two experiments designed to document the spermatotoxicity of ethoxyethanol (EE) as reflected in evaluations of ejaculated rat semen. The basic strategy involved the evaluation of individual ejaculates recovered from the genital tract of a female rat prior to exposure of the males to EE. Repeated assessments of the ejaculate were made during the experimental phase according to specified protocols. Adult Long-Evans hooded male rats received 0, 936, 1872 or 2808 mg/kg (PO) of EE for five consecutive days. Semen evaluations were then conducted at weeks 0, 1, 4, 7, 10 and 14 after exposure. Males in the highest two dose groups showed declines in sperm counts by week 4 and were essentially azoospermic by week 7. At this time, males receiving the lowest dose of EE also exhibited decreases in ejaculated sperm counts. An increase in abnormal sperm shapes was also observed. Over the ensuing weeks all males exhibited varying degrees of recovery as reflected by increasing ejaculated sperm counts. In a second experiment, males were treated with 0 or 936 mg/kg of EE daily (5 days/week) for 6 weeks with semen evaluations conducted weekly. By week 5, sperm counts were significantly depressed and there was an increase in the number of abnormal shapes. Sperm motility was depressed by week 6. The temporal trends seen in these experiments suggested a differential sensitivity of the spermatocyte stage to EE toxicity. Histopathological assessments confirmed this observation while distinguishing the pachytene spermato- cytes as the most sensitive cells. Subchronic treatment did not appear to alter the spermatogonial population, suggesting the potential for recovery as witnessed following acute EE administration. The majority of work evaluating the male reproduc- tive toxicity of glycol ethers has been with methoxy- ethanol (ethylene glycol monomethyl ether). Studies on ethoxyethanol (ethylene glycol monoethyl ether, EE) have been limited and are summarized in Table 1. These studies have been restricted essentially to histopatho- logical evaluations of the testes. No published reports are available on other spermatogenic endpoints or fertility aside from the report by Lamb in these proceedings (1). Ethoxyethanol appears to produce the same pattern of testicular toxicity as that seen with the methyl derivative, although higher doses of EE are required (2). The present study consisted of two experiments. The first experiment employed acute EE treatment and was designed to determine the sensitiv- ity of the various stages of spermatogenesis to EE insult as well as to estimate recovery from such toxicity. The second investigation examined the spermatoxicity and associated testicular histopathology accompanying subchronic EE treatment using the lowest dose which produced effects in Experiment I. Both studies utilized *Department of Environmental Health, University of Cincinnati Medical Center, Cincinnati, OH 45267. tDepartment of Anatomy, University of Cincinnati Medical Center, Cincinnati, OH 45267. evaluations of ejaculated rodent semen to assess spermatoxicity. Materials and Methods. Ethoxyethanol was obtained from Aldrich Chemical (99.9% pure) and its purity confirmed by gas chroma- tography. EE was diluted with distilled water such that all animals received equivalent volumes of fluid. Treatment Protocol Male and female Long-Evans hooded rats (Charles River, Wilmington, MA) were obtained at approxi- mately 70 to 80 days of age. All animals were kept on a reverse 10:14 light-dark cycle with the lights turned off at 0900 hours. Purina Lab Chow #5001 and tap water were available ad libitum. Males were single-caged while females were housed three per cage. The same basic protocol was employed in Experi- ments I and II. The experimental strategy involved semen evaluations on ejaculates recovered from the genital tract of a female rat, 15 min after copulation. This approach allowed for the semen evaluations to be performed on each male prior to exposure (baseline) and then repeatedly throughout the experimental phase. ZENICK, OUDIZ AND NIEWENHUIS ITable 1. Summary of effects of ethoxyethanol on the male reproductive system. Species Dose Route Duration Effects Reference Mice 0.5, 1.0, 2.0% Drinking ca. 100 days Total infertility at 2.0% 1 (1) water live pups/litter at 1.0% I epididymal sperm motility l abnormal sperm T testes weight at 2.0% 1 abnormal forms at 1.0% t Rat 250-1000 mg/kg PO 11 days Testes weight and testicular atrophy 1 (2,4) Pachytene spermatocytes most sensitive Mice 500-4000 mg/kg PO 5 weeks Marked degeneration of germinal epithelium (3) Rat 1.45% Diet 2 years Testicular edema and atrophy of germinal epithelium (5) Rat 94-750 jig/kg SC 4 weeks Testicular edema, absence of more mature sperm cell (6) stages Rat 94-750 ,ug/kg PO 13 weeks Similar pattern to that seen with SC injection (6) Dog 50-20 ,uL/kg PO 13 weeks Similar pattern to that seen in rats (6) Prior to the treatment phase, the males were allowed to mate with ovariectomized, hormonally primed fe- males on several occasions over a 2-week period. Fe- males received 0.1 mg/0. 1 mL of estradiol benzoate (SC) (Sigma) 48 hr prior to mating followed by 0.1 mg/0. 1 mL (SC) of progesterone (Lilly, Indianapolis, IN) 4 hr before mating. These initial mating trials were neces- sary in order to insure a high level of mating proficiency by the time the experiment was initiated. All the males were mated on a weekly basis for the duration of the experiment in order to maintain a constant interval between ejaculations. At the beginning of the experi- mental period, the males were intubated with distilled water for 5 consecutive days. This procedure was employed in order to acclimate the animals to the procedure of intubation. On the fifth day of the water intubation, the males were mated with ovariectomized hormonally primed females in order to obtain pre- exposure baseline measures on each animal. Mating behaviors were observed and at 15 min post-ejaculation, the females were sacrificed. The semen was recovered from the uterine horns and processed for determinations of sperm count, percent motile sperm and sperm morphology (see below). The sperm count data were also used to match males for assignment to treatment groups. The males were rank-ordered by sperm counts and assigned to treatment groups such that the mean sperm count of each group was essentially equivalent. Copulatory Behaviors Evaluations of copulatory behaviors were conducted between 1300 and 1500 hr during the dark cycle. The male was placed in a Plexiglas observation cage (33 x 28 x 23 cm) and allowed to habituate for approximately 15 min. An ovariectomized, hormonally primed female was then introduced into the cage and copulatory behaviors monitored. Four behaviors were recorded: (1) mount latency, defined as the elapsed time between introduc- tion of the female and the first mount; (2) number of mounts; (3) number of intromissions, distinguished from mounts by increased vigorous thrusting, back- pedaling following desengagement, and immediate geni- tal grooming; and (4) ejaculation latency, defined as the elapsed time between the initial mount and ejaculation. The presence of a copulatory plug upon sacrifice in 100% of the females confirmed our accuracy in determining time of ejaculation. Sperm Evaluation Approximately 15 min after copulation, the female was sacrificed by CO2 asphyxiation. The uterine horns were exposed, and the contents of the uteri (semen and uterine fluid) aspirated into a syringe. Subsequently, the uteri and vagina were excised, cut open and flushed to obtain residual sperm, and the copulatory plug was removed, washed and weighed. A drop of the sample was placed on a slide and a subjective estimate was made of the degree of sperm concentration (low, moderate, high) in order to dilute the sample for motility evaluation. An aliquot of the sample was diluted with phosphate-buffered saline. The buffer and slides were maintained at 370C. A 10 ,uL portion of this diluent was pipetted onto a slide, cover-slipped and placed on the microscope stage main- tained at 37?C by an air curtain incubator. In Experi- ment I, percent motility was then determined for 100 sperm viewed in three to four areas of the slide. In Experiment II, three fields on the slide were video- taped and this tape subsequently scored for percent motile. In Experiment I, approximately 50 FiL (one drop) of the diluted sample was stained with Eosin Y and subsequently scored for sperm morphology (i.e., per- cent normal shapes seen in 200 sperm). In Experiment II, a stain was used that was developed by Bryan (7). This consisted of air drying an unstained sperm smear on a slide and staining for 40 min in an acidic Fast Green, Eosin Y and Naphthol Yellow solution. The slides were then rinsed in 1% acetic acid, cleared in xylene and mounted. This stain allowed for a more detailed inspection of the head and tail structures. The remaining undiluted sample was transferred to a graduated cylinder. The residual sperm from the uteri, vagina and copulatory plug were added to the gradua- 226 SPERMA TO TOXICITY ASSOCIATED WITH EGEE TREATMENT ted cylinder along with washings of the syringes, pipette tips, and slides. The sample was then diluted to a final volume of 50 mL. Both chambers of a hemocytometer were charged with 10 ,uL of sample and sperm counts were determined. If counts differed by more than 10% the chamber was recharged and sperm count was redetermined. Data Analyses For both experiments, a two-way repeated measures analysis of variance (ANOVA) with one between factor (treatm'ent) and one within factor (weeks) was used for analysis of body weights, sperm counts, sperm motility, plug weights, mount latency, ejaculation latency, num- ber of mounts and intromissions. The latency measures were subjected to a logarithmic transformation prior to analysis to reduce heterogeneity of variance. Percent of baseline was calculated for each animal by dividing values obtained during the experimental phases by his pre-exposure baseline values. These data were also analyzed employing a two-way repeated measures ANOVA. The morphology data were transformed using an arc sine transformation to normalize distributions and analyzed by a one-way ANOVA. Data collected at sacrifice (see below) were analyzed employing one-way ANOVAs to test for the presence of treatment-related effects. All significant trends were further tested with univariate ANOVAs to contrast the various treatment groups. Experiment I Method Two days following baseline semen evaluations, males were intubated with either 0, 936, 1972, or 2808 mg/kg of EE (four males/group) for five consecutive days. At the cessation of EE exposure (week 1, post-baseline), the animals were again mated and copulatory behaviors and semen samples evaluated. The males were subse- quently sampled at weeks 4, 7, 10 and 14. Sample weeks 1, 4, 7 and 10 corresponded to the transit time required in the rat for germ cells in the various spermatogenic stages (i.e., spermatozoa, spermatids, spermatocytes, and spermatogonia) to appear in the ejaculate subse- quent to exposure. A sample was also obtained one month later (week 14), since week 10 samples had indicated that some of the animals exposed to EE were demonstrating recovery (i.e., elevated sperm counts). Animals were mated in the interim weeks, but samples were not recovered. At week 16, all animals were sacrificed and weights of the liver, kidney, testis, epididymis (total), cauda epididymis, vas deferens, ventral prostate, and seminal vesicles were obtained. A cauda epididymis was processed to obtain a sperm count. The other cauda epididymis and one testis were fixed in Bouin's solution, dehydrated, embedded in paraffin, sectioned, stained with hematoxylin and eosin by routine procedures and examined by light microscopy. Results All groups showed significant increases in body weights during the course of the experiment (p - 0.01). However, animals in the high dose group exhibited less weight gain relative to the other groups. This trend was only significant at week 1 (p - 0.05), reflecting concur- rent EE treatment. EE produced no effects on copulatory behaviors or seminal plug weights. However, EE treatment did result in a significant decline in ejaculated sperm counts in all groups (Fig. 1). At the fourth week post exposure, males in the two highest dose groups had significantly decreased sperm counts and by the seventh week the males were essentially azoospermic. At this time males in the lowest dose group also exhibited significant decreases in ejaculated sperm relative to the controls. Sperm counts were so depressed in the middle and high dose groups that assessments of sperm motility WEEK FIGURE 1. Group mean sperm counts expressed both as (A) raw values and (B) as percent of each animal's baseline. Letters L, M and H represent low (936 mg/kg), medium (1872 mg/kg) and high (2808 mg/kg) doses of ethoxyethanol administered (PO) for 5 consecutive days. 227 ZENICK, OUDIZ AND NIEWENHUIS and morphology were negated. The low dose group showed a significant decrease in normal forms at week 7 compared to controls (mean + SD = 95.8 + 2.9 vs. 99.4 + 0.7, respectively). These abnormal forms were unlike any seen previously in the laboratory and reflected marked deformation of the sperm head (Fig. 2). Motility, although lower in this group, was not significantly different from controls. All treated males evidenced some degree of recovery by week 10 post-exposure. By week 16, only males in the middle dose group lagged behind in recovery as reflected in ejaculated sperm count. Histological evalua- tions of the testes of rats in this group suggested the presence of some residual lesions. These lesions were characterized by depleted or disorganized seminiferous tubules interspersed among normal tubules. Data collected at sacrifice are presented in Table 2. Analyses of organ weights revealed no differences between the groups except in the case of the epididymis (p S 0.01). In this instance, epididymal weights in the middle dose group were significantly lower than controls (p - 0.05). Differences between controls and high dose animals approached significance (p - 0.10). Although not significant, similar trends were observed in the cauda epididymal weights and caudal sperm counts for these two groups again indicating that recovery had not been totally achieved by this time. The data from Experiment I suggested that EE exerted its most dramatic effects at week 7 correspond- ing to a heightened sensitivity of the spermatocyte stage to this agent. Moreover, the spermatogonia appear to remain intact as witnessed by the high levels of recovery even in animals rendered azoospermic by EE treatment. FIGURE 2. Photomicrograph of sperm heads: (right) high dose animal (2808 mg/kg) at week 7; (left) normal shape seen in a non- exposed animal. x 1000 Nemarski. Table 2. Body weight, organ weight and cauda epididymal sperm counts obtained at sacrifice following acute ethoxyethanol exposure. Body and organ weight, g (mean ? SD) Treatment (PO), Epididymis Cauda Cauda sperm mg/kg Body Testis (total) epididymis count x 106 0 (control) 555 ? 37.4 1.65 ? 0.10 0.633 ? 0.07 0.263 ? 0.09 161 ? 30 936 540 ? 59.1 1.85 + 0.18 0.650 ? 0.04 0.238 ? 0.03 187 ? 25 1872 594 ? 40.6 1.57 ? 0.09 0.542 ? 0.03* 0.213 + 0.01 129 ? 27 2808 517 ? 16.0 1.64 ? 0.15 0.543 ? 0.06 0.190 ? 0.04 134 ? 36 *p < 0.05 significantly different from controls. 228 SPERMATOTOXICITY ASSOCIATED WITH EGEE TREATMENT In light of these data, Experiment II was designed to utilize subehronic exposure and weekly semen evalua- tions so as to better document the development of the spermatoxicity. Moreover, histological evaluations of the testes were done concurrently in a subpopulation of males. Experiment 11 Method The experimental strategy was identical to that employed in Experiment I with the one exception: that ejaculated semen samples were evaluated weekly. After obtaining a baseline semen evaluation on each male, treatment was initiated. Males were intubated with either 0 or 936 mg/kg EE daily (5 days/week), for 6 weeks (nine males/group). At the end of the sixth week, all animals were sacrificed and weights were obtained on all major organs. One of the cauda epididymides was processed for determination of sperm count and sperm morphology. The hematocrit and hemoglobin levels were also determined at this time. Fourteen additional males were used for histological evaluations. Twelve males received 936 mg/kg EE (PO) 5 days/week for up to 6 weeks. Two control males were intubated with equivalent volumes of water. Two treated males were sacrificed at weekly intervals through week 6. At this time, the control males were sacrificed. The males were sacrificed by penotobarbital overdose, per- fused with a 1% procaine saline solution, followed by a 5% glutaldehyde fixative. For light microscopy, testicu- lar tissue was embedded in paraffin, sliced in 5 ,um sections and stained with hematoxylin and eosin. Results Subchronic EE treatment produced a significantly slower rate of weight g4n relative to the controls (Fig. 3). This difference increased over weeks, suggesting the gradual appearance of a more pronounced systemic toxicity. Evidence of a generalized toxicity was also seen in a number of other endpoints. Data obtained at sacrifice (Table 3) revealed that EE animals had ele- vated brain and spleen weights either expressed as absolute values or as organ/body weight ratios (p - 0.05). Hemoglobins and hematocrits were significantly 500 (!) M: ck_ CD F- CD LUJ 475 1 450 1 425 1 400 - 375 F 0 1 2 WEEKS OF 3 4 5 EXPOSURE 6 FIGURE 3. Plots of body weight gain ( * ) in controls and ( * ) in animals treated with 936 mg/kg ethoxyethanol (PO) for 6 weeks. CD (n :D C-) u LU (F) 80 F 40 I 20 p 0 1 2 3 4 5 WEEKS OF EXPOSURE 6 FIGURE 4. Alterations in ejaculated sperm counts associated with subchronic ethoxyethanol treatment (936 mg/kg, PO): ( *) controls; ( * ) treated with EE; *p - 0.01. Tble 3. Body weight, organ weights and hematologic measures obtained at sacrifice following subchronic ethoxyethanol exposure. Body and organ weight, g (mean ? SD) Hemoglobin, Treatment, mg/kg Body Brain Spleen Liver Kidneys Hematocrit, % mg/dL 0 487 ? 39 1.78 ? 0.17 0.79 ? 0.21 17.0 ? 2.28 3.65 ? 0.55 46.1 + 1.62 14.2 ? 0.70 (0.365)a (0.160) (3.49) (.746) 936 470 ? 39 2.05 ? 17* 1.12 ? 0.17* 16.94 ? 2.06 3.62 ? 0.43 44.3 ? 1.80* 13.5 ? 0.70* (0.438)* (0.239)* (3.61) (0.771) aOrgan weight/100 g body weight. *p < 0.05. 229 ZENICK, OUDIZ AND NIEWENHUIS 100 CD CD I GD CD '37 1 95 90 85 80 75 0 1 2 3 4 5 6 WEEKS OF EXPOSURE FIGURE 5. Alterations in sperm morphology associated with sub- chronic ethoxyethanol treatment (936 mg/kg, PO): ( * ) controls; ( * ) treated with EE; *p < 0.01. 0 1 2 3 4 5 6 WEEKS OF EXPOSURE FIGURE 6. Alterations in sperm motility associated with subchronic ethoxyethanol treatment (936 mg/kg, PO): ( * ) controls; ( * ) treated with EE; *p S 0.01. Table 4. Reproductive organ weights and sperm parameters obtained at sacrifice following subchronic ethoxyethanol exposure. Reproductive organ weight, g (mean ? SD) Caudal sperm Normal lreatment, Seminal Vas Cauda count morphology, mg/kg Prostate vesicle deferens Testes Epididymides epididymis X 106 % 0 0.84 ? 0.15 2.21 ? 0.24 0.26 ? 0.03 3.33 ? 0.36 1.22 ? 0.11 0.15 ? 0.03 114 ? 26 95 ? 2.5 (0. 174)a (0.455) (0.054) (0.690) (0.251) (0.031) 936 0.70 ? 0.14 2.41 ? 0.31 0.24 ? 0.03 1.37 ? 0.03* 0.75 ? 0.08* 0.09 ? 0.01* 11 7* 61 ? 14* (0.150) (0.516) (0.051) (0.292)* (0.159)* (0.018)* aOrgan weight/100 g body weight. *p < 0.01. depressed (p S 0.05). Other investigators have reported similar hematopoietic effects (3). All sperm parameters were adversely affected by EE treatment. Significant declines in sperm count (Fig. 4) and percent normal forms (Fig. 5) were seen at weeks 5 and 6, and a significant decline in motility was seen at week 6 (Fig. 6). The abnormal shapes were similar to those observed in Experiment I. Data on the reproductive organs obtained at sacrifice (Thble 4) further supported these observations. While accessory organ weights were unaffected, testicular and epididymal weights were severely depressed. Cauda epididymal sperm counts were only 10% of that seen in controls and the percent of abnormal forms was mark- edly increased. The time course for appearance of spermatoxicity (as assessed in ejaculates) corresponded closely to that seen with acute exposure in Experiment I. This sup- ported the previous observation of the heightened sensitivity of the spermatocytes to EE insult. The more profound depression in sperm counts reflected the extended dosing period allowing for a more complete encompassing of cells passing through this stage. Histological evaluations as early as week 1 confirmed the sensitivity of the spermatocyte stage. At this time the testes showed scattered lesions primarily seen as vacuolization of the tubules directly above the Sertoli cell nuclei and spermatocytes with pyknotic nuclei. By the second and third weeks, the damage to the seminif- erous tubules was widespread, and a number of tubules showed exfoliation of developing germ cells. The testes showed little, if any progressive damage from week 4 to week 6. There was no apparent damage to the Leydig cells, Sertoli cells, spermatogonia, or early spermato- cytes (i.e., leptotene stage). However, there was a total absence of maturing germ cells beyond the leptotene- zygotene spermatocyte stages. These observations are essentially identical to those reported by Foster et al. (2,4). Discussion The data from these studies emphasize the remark- able sensitivity of the testes to EE toxicity. Few clinical signs were observed at the end of the 6 weeks of dosing 230 SPERMATOTOXICITY ASSOCIATED WITH EGEE TREATMENT 231 in spite of marked testicular insult. The similar patterns of insult observed across Experiments I and II would suggest the potential for recovery of testicular function within the dose and time period examined in these studies. Histological data would support this view since the spermatogonial population remained intact in the males exposed to EE for 6 weeks. Germ cell repopulation has also been noted by Foster et al. (2). Such recovery would argue against the hypothesis that EE produces its toxicity by interfering with mecha- nisms regulating cell division (3). This work was supported by NIOSH R01-OH-01272, EPA-CR808880 and NIEHS ES-07073. REFERENCES 1. Lamb, J. C., Gulati, D. K., Russell, V S., Hommel, L., and Sabharwal, P S. Reproductive toxicity of ethylene glycol monoethyl ether contrasted with the effects of ethylene glycol. Environ. Health Perspect. 57: 85-90 (1984). 2. Foster, P M. D., Creasy, D. M., Foster, J. R., Thomas, L. V, Cook, M. W, and Gangolli, S. D. Testicular toxicity of ethylene glycol monomethyl and monoethyl ethers in the rat. Toxicol. Appl. Pharmacol. 69: 385-399 (1983). 3. Nagano, K., Nakayama, E., Koydno, M., Oobayashi, H., Adachi, H., and Yamada, T. Mouse testicular atrophy induced by ethylene glycol monoalkyl ethers. Japan. J. Ind. Health. 21: 29-35 (1979). 4. Foster, P M. D., Creasy, D. M., Foster, J. R., and Gray, T. J. B. Testicular toxicity produced by ethylene glycol monomethyl and monoethyl ethers in the rat. Environ. Health Perspect. 57:207-217 (1984). 5. Morris, H. J., Nelson, A. A., and Calvery, H. 0. Observations on the chronic toxicities of propylene glycol, ethylene glycol, diethylene glycol, ethylene glycol monoethyl ether and diethylene glycol monoethyl ether. J. Pharmacol. Exptl. Therap. 74: 266-273 (1942). 6. Stenger, E. G., Aeppli, L., Mueller, D., Peheim, E., and Thomann, P. The toxicology of ethylene glycol monoethyl ether. Arzneim. -Forsch. 21: 880-885 (1971). 7. Bryan, J. D. An Eosin-Fast Green-Napthol Yellow mixture for differential staining of cytologic components in mammaliam spermatozoa. Stain Technol. 45: 231-236 (1970).