Abstract
Sertoli cells are the primary cellular target for a number of pharmaceutical and environmental testicular toxicants, including 2,5-hexanedione, carbendazim, and mono-(2-ethylhexyl) phthalate. Exposure to these individual compounds can result in impaired Sertoli cell function and subsequent germ cell loss. The loss of testicular function is marked by histopathological changes in seminiferous tubule diameter, seminiferous epithelial sloughing, vacuolization, spermatid head retention, germ cell apoptosis, and altered microtubule assembly. The present study investigates dose-response relationships for these classic Sertoli cell toxicants using histopathology endpoints. Understanding the relationship between the Sertoli cell toxicant dose and its histopathologic manifestations will help establish the sensitivity of these endpoints as markers of testicular injury. The results indicate that no single histopathology endpoint was sensitive on its own in identifying altered testicular morphology resulting from toxicant exposure. However, when multiple endpoints were combined dose-response relationships could be associated with incremental alterations in histopathology. The data generated from these experiments will be useful in further investigating the effects of Sertoli cell toxicant exposure in animal toxicity studies. In addition, this work is fundamental to a planned investigation of the histopathologic and gene expression changes associated with testicular toxicant co-exposures, which may occur both occupationally and environmentally.
Introduction
Sertoli cells are the primary supportive cells of the testis, facilitating the physical, hormonal, and nutritive development of germ cells. Adjacent Sertoli cells form a highly regulated microenvironment by maintaining the blood-testis barrier, separating the resident germ cells in an adluminal compartment. Contact between Sertoli and germ cells occurs through adhesion junctions supported along extensive cytoskeletal networks. Disruption of Sertoli cell function results in impaired spermatogenesis and subsequent germ cell loss.
Microtubule dynamics are an important component of the Sertoli cell cytoskeleton. Alterations in testicular microtubule assembly are a suspected mechanism of carbendazim- (CBZ) and 2,5-hexanedione- (2,5-HD) induced Sertoli cell toxicity (reviewed by Boekelheide et al., 2005). The benzimidazole fungicide, benomyl, elicits testicular toxicity through the active metabolite CBZ (Lim et al., 1997), resulting in decreased fertility, increased testis weights, and testicular atrophy (Carter et al., 1987; Gray et al., 1990; Hess, 1998). Testicular toxicity develops rapidly, with sloughing of the seminiferous epithelium present as early as 1 h postexposure (Lim et al., 1997).
CBZ inhibits microtubule polymerization in a colchicine-like manner (Correa et al., 2001; Winder et al., 2001), decreasing the rate and stability of microtubule assembly by binding the β-tubulin subunit of the αβ-tubulin heterodimer (Howard et al., 1980; Quinlan et al., 1980). In contrast, 2,5-HD exposure is characterized by rapid assembly and enhanced microtubule stability (Boekelheide, 1987a, 1987b). 2,5-HD is a metabolite of
Phthalates are another important class of Sertoli cell toxicants. Utilized commercially as plasticizers, these compounds have become persistent environmental pollutants. Mono-(2-ethylhexyl) phthalate (MEHP), the active metabolite of di-(2-ethylhexyl)phthalate (Albro et al., 1989), is one of the most studied Sertoli cell toxicants, leading to secondary germ cell apoptosis 6 and 12 hours following an acute exposure (Lee et al., 1997). Whereas the bulk of phthalate research has focused on exposures in prenatal and juvenile rats, some evidence suggests that adult rats are also susceptible to developing testicular pathology following acute phthalate exposure (Creasy et al., 1987). The short duration between MEHP exposure and the onset of pathology suggests a targeted mode of action, but the mechanism has yet to be resolved.
Multiple histopathology markers are commonly used to quantify testicular injury including seminiferous tubule diameter, sloughing, vacuolization, and apoptosis. Increased seminiferous tubule diameter is indicative of fluid retention resulting from impaired emptying through the efferent ducts, whereas decreased seminiferous diameter may indicate germ cell loss. Toxicant-induced sloughing of the seminiferous epithelium results in increased amounts of luminal cellular debris, obstruction of the efferent ducts and, subsequently, retained seminiferous tubule fluid. Vacuoles form from the fusion and swelling of cytoplasmic organelles, which may be an early indicator of Sertoli cell toxicity. Damaged germ cells undergo apoptosis, offering easy quantification using the terminal dUTP nick-end labeling (TUNEL) assay. Mounting evidence suggests that condensed elongated spermatid retention in the basement compartment (at or below the level of Sertoli cell nucliei) is a sensitive measure of altered testicular morphology (Saito et al., 2000; Yang et al., 2006), the molecular mechanisms responsible for spermatid head retention following toxicant exposure is presently unknown. Finally, the levels of α-tubulin mRNA were assessed to evaluate the effects of these Sertoli cell toxicants on microtubule assembly.
The present study investigates dose-response relationships for classical Sertoli cell toxicants using these histopathology endpoints. Understanding the relationship between the histopathologic alterations and Sertoli cell toxicant dose will establish the sensitivity of these endpoints as markers of testicular injury. Furthermore, data generated from these experiments will be used in subsequent studies to establish the dose-response and time dependence of 2,5-hexanedione-induced sensitization to carbendazim or MEHP co-exposures.
Materials and Methods
Chemicals
All chemicals were purchased from Sigma-Aldrich Corporation (St. Louis, MO) and were of reagent grade or better.
Animals
Adult male Fischer 344 rats weighing 250–275g (Charles River Laboratories, Wilmington, MA) were maintained in a temperature and humidity controlled vivarium with a 12-h alternating light-dark cycle. All rats were housed in community cages with free access to water and Purina Rodent Chow 5001 (Farmer’s Exchange, Framingham, MA). The Brown University Institutional Animal Care and Use Committee approved all experimental animal protocols in compliance with National Institute of Health guidelines.
Toxicant Exposure
Rats were treated with CBZ by a single gavage dose of 40, 67, 100, or 200 mg/kg of body weight in a corn oil vehicle (2 ml/kg body weight). Rats were euthanized by CO2 asphyxiation 24 h posttreatment. Then, 2,5-HD was administered in drinking water ad libitum for 18d at concentrations of 0.125, 0.21, 0.3125, and 0.625% 2,5-HD, then euthanized by CO2 asphyxiation. The average 2,5-HD consumption per rat for each dose group is listed in Supplemental Table 1. MEHP was dosed by a single gavage treatment of 0.2, 0.33, 0.5, or 1.0 g/kg of body weight in a corn oil vehicle (2 ml/kg body weight). Rats were euthanized by CO2 asphyxiation 12 h posttreatment. In all treatment paradigms, 10 rats were utilized for each dose.
Histopathology
Cross-sections were taken from the middle of formalin-fixed testes, embedded in glycol methacrylate (Technovit 7100, Heraeus Kulzer GmBH, Wehrheim, Germany), sectioned (3 μm), and stained with periodic acid-Schiff reagent followed by a hematoxylin counterstain (PAS-H) for measurement of tubule diameter, sloughing, vacuoles, and retained spermatid heads. Fifty seminiferous tubules were randomly counted from each testis using predetermined coordinates from a stage engraved Vernier scale for assessing seminiferous tubule diameter, sloughing, and vacuoles. To be acceptable for counting, seminiferous tubules were required to have a major:minor axis less than 1.5:1, and the diameter was identified as the minor axis length. Seminiferous tubules were randomly examined for sloughing indicated by the presence of luminal cellular material (≥24 μm in major axis length) detached from the seminiferous epithelium (Markelewicz et al., 2004). Seminiferous tubule vacuolization was defined as the presence of one or more vacuoles ≥16 μm in greatest diameter located within 1 cell layer of the seminiferous tubule basement membrane (Markelewicz et al., 2004). Spermatid head retention was conducted by counting all stage IX-XI seminiferous tubules in one testis cross section and scoring each as having 0, 1–3, or >3 spermatid heads within the basement compartment. Sloughing, vacuoles, and retained spermatid head values are all represented as the percentage of total tubules affected per total tubules counted. A Zeiss Standard microscope (Karl Zeiss, New York, NY) was utilized to view all histopathological sections.
Detection of Apoptosis
TUNEL staining was conducted on O.C.T. (Tissue-Tek, Torrance, CA; CBZ study) or paraffin embedded testes (2,5-HD and MEHP studies), sectioned (7 μm CBZ; 5 μm 2,5-HD and MEHP), and stained using an ApopTag Peroxidase In Situ Apoptosis Detection Kit (Chemicon, Temecula, CA) as directed by the manufacturer. Sections were counterstained with methyl green. Percent of seminiferous tubules with 0, 1–3, or >3 TUNEL positive nuclei were assessed by counting all seminiferous tubules with a major:minor axis less than 1.5:1 in 2 cross-sections with a Zeiss Standard microscope (Karl Zeiss, New York, NY).
Quantitative Real-Time PCR
Real-time PCR was used to measure α-tubulin mRNA levels following Sertoli cell toxicant exposure. Six random samples from each treatment group of 10 were selected for analysis. RNA was extracted from the rat testis using Tri Reagent (Sigma, St. Louis, MO) per manufacturer’s protocol and optimized for real-time PCR. Total RNA (5 μg) was treated with DNase I (Invitrogen, Carlsbad, CA) before reverse transcription with iScript Reverse Transcriptase (Bio-Rad, Hercules, CA). Real-time PCR was run on a Bio-Rad iCycler iQ thermocycler at 95° C for 4.5 min and cycled 40 times at 95° C for 10 s and 59.5° C for 45s, with a final hold at 95° C for 1 min. Samples were run in triplicate in 25 μl reactions with 25ng/well of sample (based on total RNA) using iQ SYBR green supermix (Bio-Rad, Hercules, CA). Relative α-tubulin mRNA levels were normalized to the housekeeping gene hypoxanthine guanine phosphoribosyl transferase (HPRT). The following primer concentrations and sequences were utilized: α-tubulin; 300nM, 5′: ggctgccctagagaaggatt, 3′: ctgtgaaagcagcaccttgt, HPRT; 300nM 5′: caggccagactttgttggat, 3′: taggctttgtacttggcttt. Final Mg2+ concentrations were 3 μM for α-tubulin and 4 μM for HPRT. Relative α-tubulin mRNA levels were calculated from Ct values using the following equation: (2(Ctα–tubulincontrol –Ctα–tubulinsample))/(2(Ct HPRT control–Ct HPRT sample)), where Ctα–tubulin (or Ct HPRT) is the cycle where the fluorescent signal from α-tubulin product (or HPRT) crosses the threshold in the control or sample tissue (Pfaffl, 2001).
Statistical Analysis
Results are expressed as mean ± standard error of the mean (SEM) for 6 (quantitative PCR only) or 10 rats per dose. A comprehensive table of all means, SEMs, and value ranges for each histopathological endpoint can be found in Supplemental Table 2. Statistical differences were determined by ANOVA and Bonferroni’s post hoc analysis with
Results
Body and Testis Weights
Body and testis weights were recorded immediately following CO2 asphyxiation (Table 1), and indicated dose-dependent changes in body weight with 2,5-HD, but not CBZ or MEHP treatment. The observed changes in body weight with the highest 2,5-HD dose were likely attributable to decreased water consumption in 2,5-HD treated rats (Table 1). Dose-dependent changes in testis weights were detected following CBZ, but not 2,5-HD and MEHP treatment. The increase in testis weight with CBZ treatment is consistent with fluid accumulation resulting from blocked efferent ducts (Klinefelter et al., 1998).
Dose and Time Point Selection
All toxicant doses were selected to establish a dose-response relationship using histomorphological and immunohistochemical endpoints of seminiferous epithelial alterations. Times of sacrifice were selected based on results of previous studies, allowing sufficient time to appreciate histopathological differences at low doses without developing overt pathology at high doses (Markelewicz et al., 2004). A 24 h time point was selected for CBZ given the initial onset of sloughing within hours, with extensive germ cell loss occurring by 1.5 d (Nakai et al., 1997). The long latency between exposure and development of histopathology warranted an 18-d time point for assessment following 2,5-HD administration. Extensive studies have established that 2,5-HD exposures of 1% in the drinking water do not present overt histopathology after 2 weeks of exposure, but significantly altered testicular morphology is present within 3 weeks (reviewed by Boekelheide et al., 2003). Selection of a 12-h time point for MEHP exposure is based upon previous studies indicating significant differences in germ cell apoptosis for 1000 mg/kg exposure but not 400 mg/kg MEHP, 12 h postexposure (Dalgaard et al., 2001; Rasoulpour et al., 2005). These studies indicate that a 1000 mg/kg MEHP dose, 12 h postexposure is likely to provide an effective dose and time frame to assess a dose-response.
Testicular Histopathology
Histopathologic endpoints were evaluated for their sensitivity to dose-dependent changes following Sertoli cell toxicant exposure. These endpoints included seminiferous tubule diameter, sloughing, vacuolization, retained spermatid heads, and germ cell apoptosis, as illustrated in Figure 1.
Seminiferous Tubule Diameter, Sloughing, and Vacuolization
Treatment with CBZ increased seminiferous tubule diameter at 67 and 100 mg/kg to 112% of control (Figure 2). Similarly, 200 mg/kg CBZ increased seminiferous tubule diameter to 113% of control. No dose-dependent changes in seminiferous tubule diameter were measured in the 2,5-HD or MEHP treated rats. Treatment with CBZ also induced a dose-dependent increase in the percent of tubules sloughing seminiferous epithelium per total seminiferous tubules counted (Figure 3). Doses of 67, 100, and 200 mg/kg CBZ resulted in significant sloughing to 515%, 655%, and 1015% of control, respectively. No differences in sloughing were detected at any dose of 2,5-HD or MEHP. Differences in Sertoli cell vacuolization from control were only detected following CBZ treatment (Figure 4). Although these increases were not dose-dependent, 67 mg/kg of CBZ produced significantly higher levels of vacuoles (203%) compared to control.
Basal Retained Spermatid Heads
The number of basal spermatid heads retained during stages IX through XI per seminiferous tubule was counted following Sertoli cell toxicant exposure. Both CBZ and 2,5-HD demonstrated increases in the percent of seminiferous tubules with >3 spermatid retained heads per total seminiferous tubules counted (Figure 5). Treatment with 200 mg/kg of CBZ produced a ~4-fold increase in the percent of seminiferous tubules with >3 retained spermatid heads. 2,5-HD demonstrated significantly higher levels of basal spermatid head retention at nearly every dose with nearly 100% of seminiferous tubules having retained spermatid heads at 0.625%. Doses of 0.125%, 0.3125%, and 0.625% 2,5-HD significantly increased >3 spermatid head retention to 293%, 524%, and 680% of control, respectively. Quantification of 1–3 retained spermatids per seminiferous tubule showed differing effects for CBZ than 2,5-HD. CBZ had increased levels of 1–3 retained heads at 200 mg/kg, only, whereas, 2,5-HD had a decreased percentage of 1–3 basal retained spermatid heads at 0.3125% and 0.625%. MEHP treatment did not alter the number of retained spermatid heads at any dose.
Detection of Apoptosis
Germ cell apoptosis was measured by counting TUNEL positive nuclei as an indicator of apoptosis. Both CBZ and MEHP showed threshold increases in apoptosis with >3 TUNEL positive nuclei present per seminiferous tubule (Figure 6). Every dose of CBZ evaluated had significantly higher levels of TUNEL positive nuclei compared to control. Administration of 1000 mg/kg MEHP increased apoptosis to 1184% of control. No differences in TUNEL staining were detected at any dose of 2,5-HD.
α-Tubulin mRNA Levels in Response to Toxicant Treatment
Given the purported role of microtubule assembly in the progression of CBZ- and 2,5-HD-induced Sertoli cell dysfunction, α-tubulin mRNA levels were evaluated for each toxicant using quantitative PCR (Figure 7). Although decreased α-tubulin mRNA levels were anticipated with CBZ treatment, no significant differences in mRNA levels were detected at any dose. Administration of 2,5-HD elevated α-tubulin mRNA levels for the 0.125% and 0.21% doses to 160% and 171% of control, respectively. The 330 mg/kg dose of MEHP increased α-tubulin mRNA levels to 175% of control.
Discussion
The present study examined dose-response relationships for several histopathologic endpoints following administration of Sertoli cell toxicants. No individual endpoint assessed was singularly capable of identifying testicular injury. Testis weight is a general indicator of overall testicular health, effective at indirectly reflecting changes in seminiferous tubule fluid retention or germ cell loss with the gain or loss of mass, respectively (reviewed by Creasy, 2002). Similarly, changes in seminiferous tubule diameter can be indicative of these same processes. Sloughing occurs when spermatogenic cells exfoliate into the lumen of seminiferous tubules, resulting from germ cells precociously losing adhesion to Sertoli cells. Administration of CBZ resulted in significant changes in testis weight, seminiferous tubule diameter, and sloughing. These related pathological alterations occur with CBZ treatment from disruption of the Sertoli cell cytoskeleton, propagating loss of germ cell adhesion (Nakai et al., 2002). The germ cells slough into the lumen, blocking the efferent ducts, impairing seminiferous tubule fluid passage from the testis to the epididymis, resulting in increased seminiferous tubule diameter and testis weight (Nakai et al., 1992).
Detection of altered testicular morphology was much more subtle for 2,5-HD and MEHP exposure at the chosen doses and time points. MEHP-treated rats developed increased levels of TUNEL positive nuclei at 1000 mg/kg MEHP likely from Sertoli cell dysfunction, inducing germ cell loss through apoptosis (Rasoulpour et al., 2005). Similar dose-response changes in germ cell apoptosis were observed with CBZ, but not 2,5-HD treatment. In fact, 2,5-HD administration only induced changes in one histopathology endpoint, spermatid head retention.
The presence of retained spermatids along the basement compartment of the seminiferous tubule epithelium was identified as a sensitive measure of 2,5-HD-induced testicular injury. Luminal retention of spermatids is not uncommon following testicular dysfunction. However, this marker of testicular injury can be confounded by the presence of both viable and non-viable spermatids failing to undergo spermiation, such as in the case of 2,5-HD, tri-
The dynamic interaction between toxicant and molecular target varies between toxicants, with CBZ and 2,5-HD providing extreme examples. Testicular pathology develops within hours following CBZ exposure, while 2,5-HD requires at least 2 weeks of exposure (Nakai et al., 1994; Boekelheide et al., 2003). In general, exposure to environmental toxicants occurs over long periods of time at low doses, often through the ingestion of contaminated water. Although the present study did not demonstrate many histopathological changes for the toxicants at low doses, a different outcome may be expected from long-term environmental exposures. Therefore, timing is a critical consideration when selecting pathological endpoints to compare toxicity. This variability in the time for onset of injury is likely the cause for minimal differences in α-tubulin mRNA levels following toxicant treatment.
Both CBZ and 2,5-HD are presumed to induce Sertoli cell dysfunction by inhibiting or promoting microtubule assembly, respectively (Markelewicz et al., 2004). The availability of free tubulin monomers provides feedback, regulating mRNA levels of tubulin (reviewed by Honore et al., 2005). Therefore, CBZ was expected to decrease, and 2,5-HD to increase α-tubulin mRNA levels. MEHP was not expected to modulate α-tubulin levels given that this toxicant acts by a non-microtubule-dependent mechanism to produce Sertoli cell toxicity (Richburg et al., 1996). However, no differences in α-tubulin mRNA levels were measured following CBZ treatment, and only modest changes were found at low doses of 2,5-HD treatment. This lack of change in α-tubulin mRNA levels is likely due to the occurrence of such a molecular change at an earlier time in the exposure, as a precursor event, before the histopathologically relevant time points examined in this study.
Our previous studies have indicated that 2,5-HD- and CBZ-coexposure synergistically disrupts rat spermatogenesis despite opposing molecular effects on microtubules (Markelewicz et al., 2004). It is unclear if 2,5-HD will elicit similar effects with other testicular toxicants. Therefore, the present study was conducted to establish dose-dependent changes in Sertoli cell toxicants for several parameters of testicular pathology. The data generated from this study will be used to establish the dose-response and time dependence of 2,5-HD-induced sensitization to CBZ or MEHP co-exposures. Furthermore, this study establishes quantification of basal retained spermatid heads as a sensitive marker for assessing 2,5-HD-induced Sertoli cell toxicity.
Selection of histopathological markers to assess alterations in testicular morphology is highly dependent upon the toxicant mechanism of action, dose, and duration of exposure. Identification of sensitive histopathology endpoints for evaluating potential testicular toxicants is especially relevant to preclinical toxicity testing. The present study underscores the importance of assessing multiple histopathological endpoints, as no single characterization was capable of identifying altered testicular morphology resulting from these classic Sertoli cell toxicants.
Footnotes
Acknowledgments
The project described was supported by grant number 5 P42 ES013660-02 from the National Institute of Environmental Health Sciences (NIEHS), NIH.