THE ANATOMICAL RECORD 247:379–387 (1997) Effects of Carbendazim (Methyl 2-Benzimidazole Carbamate; MBC) on Meiotic Spermatocytes and Subsequent Spermiogenesis in the Rat Testis MASAAKI NAKAI1* AND REX A. HESS2 of Veterinary Anatomy, Faculty of Agriculture, Miyazaki University, Miyazaki, Japan 2Department of Veterinary Biosciences, University of Illinois, Urbana, Illinois 1Department ABSTRACT Background: Benzimidazole fungicide, carbendazim, is known to adversely affect Sertoli cells by disrupting microtubules, which induces sloughing of elongate spermatids in a stage-specific manner. This study determines the direct effects on dividing germ cells and the subsequent effects on spermiogenesis. Methods: Carbendazim was administered orally to male rats (100 mg/kg), and their testes were processed for histological evaluation at various post-treatment intervals up to day 20.0. Results: The sloughing of elongate spermatids was observed as reported previously. In addition to this Sertoli cell lesion, necrosis of dividing spermatocytes in stage XIV was observed at 8 hours post-treatment. At day 1.5, empty spaces of missing step 1 spermatids were seen in stage I. At days 4.5 and 7.5, normal round spermatids were missing, but large round spermatids (megaspermatids) and binucleate spermatids were common. The megaspermatid nucleus was approximately 33% larger in diameter than normal round spermatids. At day 10.5, megasteps 10–12 spermatids, binucleate spermatids, and three to four different steps of spermatids coexisting in the same tubule section were present in stages X–XII. In addition, abnormally shaped elongating spermatids were observed having distorted heads and nuclear invagination containing microtubules. At day 20.0, empty spaces of missing diplotene spermatocytes were seen in stage XIII. Conclusions: The present observations show that carbendazim has rapid direct effects on meiotic spermatocytes and latent effects on spermatids, leading to morphological abnormalities and failure of spermiogenesis. These effects are found independent of occlusions in the efferent ductules. Anat. Rec. 247:379–387, 1997. r 1997 Wiley-Liss, Inc. Key words: carbendazim; rat; spermatocytes; spermatids; spermiogenesis; testis Studies of the effects of benzimidazole fungicides, benomyl and carbendazim, on the testis have shown that these agents cause cleavage of the apical cytoplasmic processes of Sertoli cells and sloughing of immature germ cells (Parvinen and Kormano, 1974; Hess et al., 1991; Nakai and Hess, 1994). The proposed mechanism contributing to the sloughing is deformation of Sertoli cells, mainly due to disruption of its microtubules (Nakai and Hess, 1994; Nakai et al., 1995). When the animals are exposed to relatively high doses of benomyl or carbendazim, the efferent ductules become irreversibly occluded with the sloughed materials from the seminiferous epithelium, which subsequently leads to seminiferous tubular atrophy and male infertility in the rat (Carter et al., 1987; Hess et al., 1991; Nakai et al., 1992; 1993). r 1997 WILEY-LISS, INC. The present study was designed to determine the effects, unrelated to occlusions of the efferent ductules by sloughed germ cells, of carbendazim on the seminiferous epithelium. The design incorporates the exposure of animals to a low dose of carbendazim. This experimental design is important for determining the prognosis of carbendazim-induced male infertility, as there is a possibility of recovery of the seminiferous epithelium that has undergone sloughing, except when efferent ductules are occluded. The data show that recovery Received 20 May 1996; accepted 9 September 1996. *Correspondence to: Masaaki Nakai, Department of Veterinary Anatomy, Faculty of Agriculture, Miyazaki University, Miyazaki 88921, Japan. 380 M. NAKAI AND R.A. HESS from massive sloughing is possible if the efferent ductules are intact; however, abnormal spermatids could be produced in the recovering seminiferous epithelium due to the extended effects of carbendazim. MATERIALS AND METHODS Animals and Experimental Design A total of 43 male Sprague-Dawley rats (90–100 days of age) were used. They were housed two or three per cage with a 12-hour alternating light-dark cycle and were allowed free access to diet and water. Five or six animals were assigned to individual sampling intervals. Carbendazim suspended in corn oil was administered to the animals by a single oral gavage. The known minimum effective dose of carbendazim is 50 mg/kg, but changes in the seminiferous epithelium with this dose are subtle (Nakai et al., 1992). Therefore, the dose of 100 mg/kg was used in the present study. At selected time intervals after treatment with carbendazim, the animals were deeply anesthetized with an intraperitoneal injection of pentobarbital (1 ml/ animal), and the testes and epididymides were fixed with 4% glutaraldehyde in 0.1 M cacodylate buffer, using a vascular perfusion technique (Hess and Moore, 1993), at 8 hours, and again at 1.5, 4.5, 7.5, 10.5, and 20.0 days post-treatment. As controls, three animals each were assigned to days 7.5 and 20.0, and one animal each to the remaining intervals. Corn oil alone was given to control animals, and their testes were processed using the same technique as used for the treatment group. Histological Methods Testicular tissue blocks were embedded in JB-4 plastic resin (Polyscience, Inc., PA). Sections were cut with glass knives at 2.5 µm, and stained with periodic acid-Schiff reaction (PAS) and hematoxylin. Caput epididymides, excised away from the testes, were embedded in paraffin. Caput epididymides, containing the efferent ductules and the initial segment, were serially sectioned at 4 mm. Every 10th section was mounted on a glass slide and stained with hematoxylin and eosin. Patency of the efferent ductules was confirmed by the presence of intact ductules and sperm in the epididymal ducts. Testes were used if there were no associated efferent ductal occlusions, or when only minor ductal occlusions were observed and the associated seminiferous tubules exhibited no expansion in diameter. A total of 64 out of 86 testes, including controls, were utilized to evaluate the testicular damages. When particular cell types were missing or showed abnormalities in the course of observation, we predicted the original cell type affected by treatment using a software program named STAGES (version 1.0, Vanguard Productions, Inc., IL). In the STAGES program, it is possible to backtrack in time to the original cell type that the observed affected cell would have been at the time of treatment. Testes collected at day 10.5 were further processed for transmission electron microscopy, because various abnormalities in spermatid head morphology were observed by light microscopy. Tissue blocks were rinsed in 0.1 M cacodylate buffer overnight. They were post-fixed in 1% osmium tetroxide containing 1.25% potassium ferrocyanide (Russell and Burguet, 1977), dehydrated in graded ethanol series, and embedded in Quetol 812. Thick sections for light microscopy were stained with toluidine blue. Ultrathin sections were stained with uranyl acetate and lead citrate and observed by a Hitachi H-800MU transmission electron microscope. Quantitative Analysis To determine whether to include a testis in which a few efferent ductules were occluded, the mean diameter of 50 nearly circular seminiferous tubules were measured in individual testes and compared to the mean diameter of 50 seminiferous tubules in control testes. Nuclear diameters of steps 5 and 7 were measured, and the length between the apical and caudal ends of step 11 spermatid heads was determined in the carbendazimtreated and control testes collected at days 4.5, 7.5, and 10.5. Differences in the mean values between treated and control testes were compared in individual parameters using the t-test. Stages of seminiferous tubules were determined in the treated and control testis collected at days 7.5 and 20.0, and the frequency of stage VII tubules was compared using the x2 test. A probability of less than 1% was considered significant in both tests. RESULTS Eight Hours Sloughing of immature elongate spermatids was observed in stages I, III, IV, VI, VII, and X–XIV tubules, but sloughing of steps 16 and 17 spermatids in stages III and IV was rare. Another change seen at this post-treatment interval was necrosis of meiotic spermatocytes in stage XIV tubules (Fig. 1). The necrosis occurred most frequently in metaphase primary and secondary spermatocytes, but newly formed secondary spermatocytes occasionally showed necrotic features. Back-calculation using STAGES software indicated that the cells were meiotic spermatocytes in stage XIV at the time of treatment. Day 1.5 Immature elongate spermatids were often missing in stages I–III, V–VII, and XIII–XIV, and were apparently sloughed during the first 36 hours post-exposure. Further changes characteristic of this post-treatment interval were mainly observed in stage I. These changes included simultaneous disappearance of step 1 and step 15 spermatids in the same stage I tubule, and empty spaces of missing step 1 spermatids alone (Fig. 2a). Round bodies that were PAS-positive and about the same size as step 1 spermatids were often seen in stage I tubules that also contained empty spaces of missing step 1 spermatids (Fig. 2a). Round spermatids that were larger in diameter than normal step 1 spermatids (Fig. 2b) and binucleate step 1 spermatids were occasionally seen in stage I tubules (Fig. 2a). Backtracking indicated that adversely affected step 1 spermatids had most probably been meiotic spermatocytes at the time of treatment. 381 EFFECTS OF MBC ON SPERMIOGENESIS Fig. 1. Stage XIV. a: At 8 hours post-treatment, necrosis of meiotic spermatocytes is noted by the granular appearance of the cytoplasm (arrowheads). PAS-hematoxylin. 3390. b: A control stage XIV tubule. PAS-hematoxylin. 3390. Day 4.5 Areas of missing elongate spermatids were seen in stage I–VII and XI–XIV tubules. At this post-treatment interval, additional abnormalities were observed mainly in stage V, where two spermatid steps (5 and 17) were found missing in the same tubule, and empty spaces of missing step 5 spermatids were observed (Fig. 3a). Another characteristic feature common to stage V tubules in the treated testis was the presence of large round spermatids (megaspermatids, Fig. 3b). The megaspermatids were usually observed in tubules that also showed missing either round or elongate spermatids. These megaspermatids had normal morphology, but were significantly larger in cellular and nuclear diameters (approximately 33%) than normal step 5 spermatids (Table 1). In addition, binucleate cells were often seen in tubules containing megastep 5 spermatids. In these cells, two neighboring round nuclei shared one acrosome (Fig. 3b inset). Multinucleate giant cells containing more than three nuclei were rarely observed in stage V tubules. Backtracking indicated that step 5 spermatids had most probably been meiotic spermatocytes at the time of treatment. In stage X tubules of the treated testes, nuclei of elongate spermatids (step 19 in appearance) were seen in the basal level of the seminiferous epithelium, indicating failure of sperm release (not shown). Day 7.5 Areas of missing elongate spermatids were seen in all stages except for stages VIII and IX. At this posttreatment interval, additional abnormalities were observed, mainly in mid- to late-stage VII, where both step 7 and step 19 spermatids were missing or empty spaces of missing step 7 spermatids were observed (Fig. 4a). The megastep 7 spermatids were often contained in stage VII, exhibiting missing spermatids (Fig. 4b). As observed at day 4.5, the megastep 7 spermatids showed an appearance similar to that of normal step 7 spermatids, except for an increase in the cellular and nuclear diameters (Table 1). Binucleate cells were often ob- served in stage VII tubules containing the abovementioned abnormalities (Fig. 4b, inset). The neighboring nuclei often shared the acrosome. Multinucleate giant cells with more than three nuclei were observed in stage VII (Fig. 4a, inset). Backtracking indicated that step 7 spermatids at day 7.5 had most probably been meiotic spermatocytes at the time of treatment. Failure of sperm release was also seen in stage X and XIII tubules (not shown). At day 7.5, frequency of stage VII tubules in the treated testis was 23.8%, which was not significantly different from that in the control (P , 0.01), suggesting a normal progression of spermiogenesis after treatment (Table 2). Day 10.5 Areas of missing elongate spermatids were observed in stage I–III, VII, and X–XIV tubules. Additional abnormalities at this post-treatment interval were observed mainly in stages X–XII. Megasteps 10–12 spermatids were observed, but megastep 11 spermatids were the most numerous (Fig. 6a). However the number of megaspermatids was less than previously observed at steps 5 and 7. The heads of megastep 11 spermatids were significantly longer in length than those of the control (Table 1). Another striking observation at this interval was the coexistence of abnormal steps of spermatids in stages X–XI (approximately steps 7–10 or 11; Fig. 6b), indicating retardation of spermiogenesis. In these tubules, spermatids that showed the most advanced development were used for stage identification. Abnormally shaped spermatid nuclei were seen in stages X–XI. These abnormalities included various distortions of nuclei (Fig. 6c), nuclei without chromatin condensation (Fig. 6c), nuclear invaginations (Fig. 6e, inset), and binucleate elongate spermatids that shared an acrosome (Fig. 6d). Multinucleate giant cells with round nuclei were rarely observed, but those of elongate spermatids were not. The nuclear invaginations usually occurred at the caudal surface of the nucleus, and 382 M. NAKAI AND R.A. HESS Figs. 2–5 383 EFFECTS OF MBC ON SPERMIOGENESIS TABLE 1. Size differences between normal spermatids in control and mega spermatids in carbendazim-treated testes Spermatid steps Normal spermatidsc Mega spermatidsc 5a 7a 11b 8.2 6 0.07 8.1 6 0.59 15.0 6 0.17 10.7 6 0.11d 10.8 6 0.14d 19.0 6 0.42d aNuclear diameter. bDistance between apical and caudal ends of the spermatid heads. cMean (µm) 6 SEM. N 5 23 for mega step 11 spermatids. N 5 50 for the rest. dSignificantly different from normal (p , 0.01). TABLE 2. Frequency of stage VII tubules in the control and carbendazim-treated testes Post-treatment intervals (days) Controla Carbendazim-treateda 7.5 20.0 21.7 6 0.45 22.6 6 0.79 23.8 6 0.69b 24.2 6 0.40b percent 6 SEM. N 5 2,301 for control at day 7.5, 2,432 for carbendazim-treated at day 7.5, 2,265 for control at day 20.0, and 3,130 for carbendazim-treated at day 20.0. bNot significantly different from the controls at each interval (p , 0.01). aMean the invagination was directed toward the dorsal aspect or apex of the nucleus. The invaginations contained aggregates of microtubules (Fig. 6e). The ends of these microtubules were embedded in electron-dense flocculent material that was similar in appearance to that associated with normally positioned manchette microtubules (Fig. 6e). Spermatids with invaginated nuclei often showed manchettes originating from the normally positioned nuclear ring. However, the nuclear ring was displaced somewhat caudally in a few spermatids (Fig. 6e). Backtracking indicated that steps 10–12 spermatids had been meiotic spermatocytes at the time of treatment. Failure of sperm release was seen in stage X. Day 20.0 The major change in the germ cell population at this post-treatment interval was missing elongate spermatids in stage I, II, VI, and VII tubules. In addition, diakinetic spermatocytes were occasionally missing in Fig. 2. Stage I tubules at day 1.5. a: Step 1 spermatids are missing, leaving spaces in the epithelium. PAS-positive round bodies (arrowheads) and a binucleate cell (arrow) are seen. PAS-hematoxylin. 3350. b: Large step 1 spermatids (arrowheads). Toluidine blue. 3870. Fig. 3. Stage V tubules at day 4.5. a: Many step 5 spermatids are missing. PAS-hematoxylin. 3350. b: Megastep 5 spermatids (arrowheads). Toluidine blue. 3870. Inset: Binucleate cell sharing an acrosome. Toluidine blue. 3870. Fig. 4. Stage VII tubules at day 7.5. a: Step 7 spermatids are missing, leaving spaces in the epithelium. PAS-hematoxylin. 3350. Inset: Multinucleate giant spermatid sitting near the lumen. PAShematoxylin. 3350. b: Megastep 7 spermatids (arrowheads) are approximately 33% larger than normal. Toluidine blue. 3870. Inset: Binucleate cell sharing an acrosome. Toluidine blue. 3870. Fig. 5. Control seminiferous tubules. a: Stage I. PAS-hematoxylin. 3350. b: Stage V. PAS-hematoxylin. 3350. c: Stage VII. PAShematoxylin. 3350. stage XIII, leaving behind large empty spaces in the epithelium (Fig. 7a). Megastep 19 spermatids were not identifiable in stage VII at this interval. Backtracking indicated that step 19 spermatids had most probably been meiotic spermatocytes, and that diakinetic spermatocytes had been either late type B spermatogonia or preleptotene spermatocytes in stage VI. Again, failure of sperm release was seen in stages X–XI (Fig. 7b). Frequency of stage VII tubules at day 20.0 was not significantly different from controls (Table 2). Controls Testes of the control animals showed normal histological structures (Figs. 1B; 5a–c; 7c,d), except for a rare incidence of necrotic spermatocytes, failure of sperm release, megaround spermatids, and binucleate spermatids. However, cells with these abnormalities occurred independently in the control testes, not in clusters as in the treated testes. Spermatids with nuclear invaginations, and multinucleate giant cells with more than three nuclei were not observed in the controls. DISCUSSION Benzimidazole compounds, benomyl and carbendazim, are known to cause premature sloughing of germ cells, along with cleaved cytoplasmic processes of Sertoli cells, (Hess et al., 1991; Nakai and Hess, 1994), necrosis of meiotic spermatocytes (Parvinen and Kormano, 1974), occlusion of the efferent ductules (Hess et al., 1991; Nakai et al., 1992; 1993), and seminiferous tubular atrophy (Carter et al., 1987; Hess et al., 1991; Nakai et al., 1992). It is the occlusion of efferent ductules that is a crucial factor for the atrophy of seminiferous tubules. In other words, continuation of spermatogenesis depends on the degree of injury to the efferent ductules (Hess et al., 1991; Nakai et al., 1992). In the present study, at a dosage that does not induce occlusions, we predicted that spermiogenesis would continue in a normal manner, except for a temporal absence of spermatids that were sloughed and meiotic spermatocytes that showed necrosis. However, our observations reveal that carbendazim induces not only necrosis of meiotic spermatocytes in stage XIV, but also a number of various morphological abnormalities in developing spermatids, and a partial failure of spermiogenesis. The incidence of some morphological abnormalities was greatly increased compared to the controls, and others were newly induced by carbendazim. These effects occurred in testes having intact efferent ductules. Therefore, carbendazim had direct effects on the seminiferous epithelium independent of efferent ductule dysfunction. Megaspermatids Large, round spermatids have been observed in the mouse testis after treatment with various chemotherapeutic agents, including microtubule-disrupting agents (Lu and Meistrich, 1979; Meistrich et al., 1982). Large, round spermatids are about the same size as secondary spermatocytes, and are assumed to be diploid cells or cells with two times the DNA content of haploid. In 384 M. NAKAI AND R.A. HESS Fig. 6. Spermatids in stage X–XI tubules at day 10.5. a: Megastep 11 spermatid (arrow). PAS-hematoxylin. 3370. b: Retardation of spermiogenesis. Different steps of spermatids (arabic numerals) coexist in this tubule. PAS-hematoxylin. 3370. c: Abnormally shaped nuclei without chromatin condensation (arrowheads). Toluidine blue. 3920. d: Elongating binucleate spermatid sharing an acrosome (arrow), and round spermatid (step 7) showing retardation of development (arrowhead). Toluidine blue. 3920. e: An electron micrograph of elongating spermatid with a nuclear invagination containing microtubules (asterisk) and an abnormally positioned manchette. The ends of microtubules within the nuclear invagination are embedded in a flocculent material (small arrow). Nuclear ring of the ectopic manchette is displaced caudally (open arrow). Large arrow 5 flagellum. 311,300. Inset: A light micrograph of a nucleus with a nuclear invagination (arrow). Toluidine blue. 3920. addition, it is reported that near-diploid spermatozoa, especially those with giant heads, contain near-diploid amounts of DNA, and that they are originally due to an abnormal meiotic division (Stolla and Gropp, 1974). This is supported by recent studies indicating that colchicine and vinblastine induce aneuploidy in spermatocytes (Miller and Adler, 1992; Leopardi et al., 1993). Although the number of chromosomes and the amount of DNA were not determined, megaspermatids observed in the present study are probably the same cell type as large round spermatids; and, therefore, it is possible that carbendazim induces spermatids with aneuploidy. In the mouse, carbendazim caused an increase in the number of diploid cells of the testis at 7 days post-treatment (Evenson et al., 1987). Presumably these cells could include a subpopulation of megaspermatids. The induction of aneuploidy by carbendazim or benomyl is also known in other cell types (Hummler and Hansmann, 1988; Zelesco et al., 1990; Zuelke and Perreault, 1995). Megaspermatids seemed to develop normally up to step 12. It is reported that aneuploid spermatogenic precursor cells (secondary spermatocytes) develop to be mature spermatozoa despite the genetic defect (Stolla and Gropp, 1974). The origin and fate of megaspermatids must be clarified to evaluate the potential risks of chromosomal aberrations in spermatozoa induced by carbendazim. 385 EFFECTS OF MBC ON SPERMIOGENESIS Fig. 7. Seminiferous tubules at day 20.0. a: Diakinetic spermatocytes are missing in stage XIII tubule (asterisks). PAS-hematoxylin. 3390. b: Step 19 spermatids in stage XI tubule, indicating failure of sperm release. PAS-hematoxylin. 3390. c: Control stage XIII tubule. PAS-hematoxylin. 3390. d: Control stage XI tubule. PAS-hematoxylin. 3390. Failure of Spermiogenesis ment to retard their development, and that the retardation likely becomes obvious in spermatids later than step 7 spermatids. Retardation may be attributable to abnormality of spermatids, dysfunction of Sertoli cells, or both. However, if retardation is due to dysfunction of Sertoli cells, it should be observed in other stages and post-treatment intervals besides stage X–XI at day 10.5. Therefore, it is likely that the cause of retardation mainly resides in the spermatids. The fate of retarded spermatids is unknown, but they are probably removed from the seminiferous epithelium in the subsequent stages by phagocytosis or sloughing, because retarded spermatids did not occur in stage VII at day 20.0. When specific types of germ cells are destroyed, there appear windows or spaces where cells are missing in later stages of spermatogenesis (Russell et al., 1990). In the present study, necrosis of meiotic spermatocytes resulted in windows of missing round spermatids in the expected stages at individual post-treatment intervals. Although these spaces were seen neither in stages X–XII at day 10.5 nor in stage VII at day 20.0, the elongate spermatid steps 10–12 and 19 were missing in these tubules, respectively. In addition, the frequency of stage VII tubules in the testicular sections at days 7.5 and 20.0 did not differ from that of control. These suggest that spermatids that were not in meiotic division at the time of treatment continued with normal development, which is consistent with the claim that arrest and retardation of spermatogenesis does not generally occur (Russell et al., 1990). However, three or four different steps of spermatids were sometimes observed to coexist in stage X–XI tubules at day 10.5, but not in stage VII at day 7.5, nor earlier. This suggests that carbendazim causes some spermatocytes that are in meiosis at the time of treat- Binucleate Cells Multinucleate giant cells of round spermatids are a common abnormality in the testis under a variety of experimental and pathological conditions (Smith, 1962; Russell et al., 1987; Singh and Abe, 1987; Russell et al., 1991). They are assumed to be formed by opening of intercellular bridges among spermatids (Russell et al., 1990), and this phenomenon has been attributed to the dysfunction of cytoskeletal elements, especially actin 386 M. NAKAI AND R.A. HESS filaments supporting the bridges (Russell et al., 1987; Singh and Abe, 1987). Formation of multinucleate giant cells observed in the present study could be explained by this manner. However, it is uncertain whether formation of multinucleate giant cells is a result of the direct effect of carbendazim, because they are first observed at day 4.5 (although rarely) and become more common at days 7.5 and 10.5. On the other hand, binucleate cells, one of the forms of multinucleate cells, were often observed in the present study. If these cells are formed by the same mechanism as mentioned above, each nucleus should have its own acrosome, which is not the case in the present study. It is reasonable, therefore, to assume that they are formed by failure of cytokinesis of secondary spermatocytes following nuclear division, as explained for other binucleate cells (Carter, 1967; Meistrich et al., 1982; Russell et al., 1987). This is supported by the present observation that binucleate cells occur as early as day 1.5. Whereas multinucleate giant cells other than binucleate cells always contain round nuclei, nuclei of binucleate cells show nuclear elongation, at least up to the step 10 spermatid. Although the fate of binucleate spermatids is not known, it is possible that they develop into binucleate spermatozoa. Abnormally Positioned Manchette Abnormally positioned manchette microtubules (‘‘ectopic manchette’’ by Meistrich et al., 1990), including those in the nuclear invagination, have been observed in testes treated with colchicine (Handel, 1979) and taxol (Russell et al., 1991), and in the testes of mutant mice (Bryan, 1977; Cole et al., 1988; Meistrich et al., 1990). In the testes treated with these microtubule poisons, the time lag between treatment and the incidence of abnormality is between 1 and 3 days. If carbendazim acts directly on spermatids that are just beginning their nuclear elongation and development of manchette, then ectopic manchettes should occur at earlier intervals. In the present study, however, step 10–11 spermatids with ectopic manchettes occurred in stages X–XI at day 10.5. Thus, carbendazim does not seem to act directly on the developing manchettes. Rather, it seems that the germ cells originally affected by carbendazim are the spermatocytes in meiosis at the time of treatment, and that these effects are not expressed until the spermatid steps at the start of nuclear elongation and manchette development, which is similar to the ectopic manchette in the mutant spermatozoa (Meistrich et al., 1990). The underlying mechanisms of carbendazim-induced manchette displacement are currently unknown, but they may be different from those of other microtubule poisons, although the outcomes of treatments are similar. The present study did not focus on germ cells earlier than meiotic spermatocytes. However, backtracking for missing elongate spermatids (data not shown) and diakinetic spermatocytes suggests that spermatocytes and spermatogonia could be a target of carbendazim, as reported in the mouse testis (Evenson et al., 1987). In conclusion, the present study demonstrates that despite the presence of intact efferent ductules, car- bendazim induces various morphological abnormalities in developing spermatids, which is more serious than first expected. The observed abnormalities in spermatids are delayed effects of carbendazim that cause severe disruption of spermatogenesis in the affected tubules, either due to direct effects on meiotic spermatocytes, or indirect effects through the known effects on Sertoli cells (Nakai and Hess, 1994; Nakai et al., 1995). ACKNOWLEDGMENTS The authors thank Dr. K. Toshimori, Department of Anatomy, Miyazaki Medical College, for his valuable comments on the manuscript. LITERATURE CITED Bryan, J.H.D. 1977 Spermatogenesis revisited: IV. Abnormal spermiogenesis in mice homozygous for another male-sterility-inducing mutation, hpy (hydrocephalic polydactyl). Cell Tissue Res., 18:187–201. Carter, S.B. 1967 Effects of cytochalasins on mammalian cells. Nature, 213:261–264. Carter, S.D., R.A. Hess, and J.W. Laskey 1987 The fungicide methyl 2-benzimidazole carbamate causes infertility in male SpragueDawley rats. Biol. Reprod., 37:709–717. Cole, A., M.L. Meistrich, L.M. Cherry, and P.K. Trostle-Weige 1988 Nuclear and manchette development in spermatids of normal and azh/azh mutant mice. Biol. Reprod., 38:385–401. Evenson, D.P., F.C. Janca, and L.K. Jost 1987 Effects of the fungicide methyl-benzimidazole-2-yl carbamate (MBC) on mouse germ cells as determined by flow cytometry. J. Toxicol. Environ. Health, 20:387–399. Handel, M.A. 1979 Effects of colchicine on spermiogenesis in the mouse. J. Embryol. Exp. Morph., 51:73–83. Hess, R.A. and B.J. Moore 1993 Histological methods for evaluation of the testis. In: Male Reproductive Toxicology. R.E. Chapin and J.J. Heindel, eds. Academic Press, San Diego, pp. 52–85. Hess, R.A., B.J. Moore, J. Forrer, R.E. Linder, and A.A. Abuel-Atta 1991 The fungicide benomyl (methyl l-(butylcarbamoyl)-2-benzimidazole carbamate) causes testicular dysfunction by inducing the sloughing of germ cells and occlusion of efferent ductules. Fund. Appl. Toxicol., 17:733–745. Hummler, E. and I. Hansmann 1988 Pattern and frequency of nondisjunction in oocytes from the Djungarian hamster are determined by the stage of first meiotic spindle inhibition. Chromosoma, 97:224–230. Leopardi, P., A. Zijno, B. Bassani, and F. Paccherotti 1993 In vivo studies on chemically induced aneuploidy in mouse somatic and germinal cells. Mutat. Res., 287:119–130. Lu, C.C. and M.L. Meistrich 1979 Cytotoxic effects of chemotherapeutic drugs on mouse testis cells. Cancer Res., 39:3575–3582. Nakai, M. and R.A. Hess 1994 Morphological changes in the rat Sertoli cell induced by the microtubule poison carbendazim. Tissue Cell, 26:917–927. Nakai, M., R.A. Hess, B.J. Moore, R.F. Guttroff, L.F. Strader, and R.E. Linder 1992 Acute and long-term effects of a single dose of the fungicide carbendazim (methyl 2-benzimidazole carbamate) on the male reproductive system in the rat. J. Androl., 13:507–518. Nakai, M., R.A. Hess, J. Netsu, and T. Nasu 1995 Deformation of the rat Sertoli cell by oral administration of carbendazim (methyl 2-benzimidazole carbamate). J. Androl., 16:410–416. Nakai, M., B.J. Moore, and R.A. Hess 1993 Epithelial reorganization and irregular growth following carbendazim-induced injury of the efferent ductules of the rat testis. Anat. Rec., 235:51–60. Meistrich, M.L., M. Finch, M.F. da Cunha, U. Hacker, and W.W. Au 1982 Damaging effects of fourteen chemotherapeutic drugs on mouse testis cells. Cancer Res., 42:122–131. Meistrich, M.L., P.K. Trostle-Weige, and L.D. Russell 1990 Abnormal manchette development in spermatids of azh/azh mutant mice. Am. J. Anat., 188:74–86. Miller, B.M. and I.-D. Adler 1992 Aneuploidy induction in mouse spermatocytes. Mutagenesis, 7:69–76. Parvinen, M. and M. Kormano 1974 Early effects of antispermatogenic benzimidazole derivatives U 32.422 E and U 32.104 on the seminiferous epithelium of the rat. Andrologia, 6:245–253. EFFECTS OF MBC ON SPERMIOGENESIS Russell, L.D. and S. Burguet 1977 Ultrastructure of Leydig cells as revealed by secondary tissue treatment with a ferrocyanideosmium mixture. Tissue Cell, 9:751–766. Russell, L.D., R.A. Ettlin, A.P. Sinha Hikim, and E.D. Clegg 1990 Histopathology of the testis. In: Histological and Histopathological Evaluation of the Testis. Russell, L.D., R.A. Ettlin, A.P. Sinha Hikim, and E.D. Clegg, eds. Cache River Press, Florida, pp. 210–266. Russell, L.D., J.A. Russell, G.R. MacGregor, and M.L. Meistrich 1991 Linkage of manchette microtubules to the nuclear envelope and observations of the role of the manchette in nuclear shaping during spermiogenesis in rodents. Am. J. Anat., 192:97–120. Russell, L.D., A.W. Vogl, and J.E. Weber 1987 Actin localization in male germ cell intercellular bridges in the rat and ground squirrel and disruption of bridges by cytochalasin D. Am. J. Anat., 180:25–40. 387 Singh, S.K. and K. Abe 1987 Light and electron microscopic observations of giant cells in the mouse testis after efferent duct ligation. Arch. Histol. Jpn., 50:579–585. Smith, G. 1962 The effects of ligation of the vasa efferentia and vasectomy on testicular function in the adult rat. Endocrinology, 23:385–399. Stolla, R. and A. Gropp 1974 Variation of the DNA content of morphologically normal and abnormal spermatozoa in mice susceptible to irregular meiotic segregation. J. Reprod. Fertil., 38:335–346. Zelesco, P.A., I. Barbieri, and J.A. Marshall Graves 1990 Use of a cell hybrid test system to demonstrate that benomyl induces aneuploidy and polyploidy. Mutat. Res., 242:329–335. Zuelke, K.A. and S.D. Perreault 1995 Carbendazim (MBC) disrupts oocyte spindle function and induces aneuploidy in hamsters exposed during fertilization (meiosis II). Mol. Reprod. Dev., 42:200– 209.