TOXICOLOGY
AND
APPLIED
PHARMACOLOGY
In Vitro/in
107,460-47
1 (199
1)
Vivo Effects of Ethane Dimethanesulfonate on Leydig Cells of Adult Rats’
GARY R. KLINEFELTER,
*,2 JOHN W. LASKEY,~
AND NAOMI
L. ROBERTS*
NSI, Inc.,* and HERL USEPA,? Reproductive Toxicology Branch. MD No. 72, Research Triangle Park, North Carolina 27711
Received June 1. 1990; accepted October IS, 1990 In Vitro/in Vivo Effects of Ethane Dimethanesulfonate on Leydig cells of Adult Rats. KLINEG. R., LASKEY, J. W.. AND ROBERTS, N. L. ( 1991). Toxicol. Appl. Pharmacol. 107,46047 1. Although ethane dimethanesulfonate (EDS) is well recognized as a Leydig cell toxicant, the dose responsiveness of Leydig cells to EDS, both in vitro and in vivo. is not well established. In addition, the cellular site of action of EDS during Leydig cell toxicity and the status of Leydig cell viability during the affected period remain controversial. We determined the in v&o EC50 (370 PM) and in vivo ED50 (60 mg/kg) for human chorionic gonadotropin (hCG>stimulated testosterone (T) production using both highly purified (98%) and interstitial (14%) Leydig cell preparations, respectively. Leydig cells were recovered in approximately equal numbers following all in vivo and in vitro EDS exposures. The Leydig cells in these preparations were viable and steroidogenically active (3@-HSD positive) subsequent to all exposures, both before and afier incubations to stimulate T biosynthesis. When hCG-stimulated T production was decreased 50% following in vivo or in vitro exposures, the morphological integrity of the Leydig cells appeared normal, with no discernible lesion at either the light or the electron microscope level. We used stimulants of various reactions in the pathway of T biosynthesis (ZOa-hydroxycholesterol and prenenolone) to determine the site of action impaired when T biosynthesis was decreased. Our results indicate that when Leydig cells are exposed to EDS either in vitro or in vivo,the biosynthesis of T is compromised between the cyclic adenosine monophosphate activation of protein kinase and the cholesterol side chain cleavage enzyme. 0 1991 Academic Press. Inc FELTER,
The transient infertility of male rats given a single 50 mg/kg ip injection of ethane dimethanesulfonate (EDS) was described by Jackson (1973). He also reported that the simultaneous administration of testosterone (T) prevents much of the damage which occurs to the seminiferous epithelium and the shrinkage of the prostate and seminal vesicles, suggesting Leydig cell dysfunction (Jackson, 1973). Bu’Lock and Jackson (1972, 1975) provided the first
indications that the Leydig cell was targeted by EDS when radiolabeled precursors of testosterone biosynthesis failed to be converted into testosterone in vitro by testes from animals given EDS in vivo. Subsequently, numerous studies have suggested that EDS is a specific Leydig cell toxicant which causes a reversible inhibition of Leydig cell steroidogenesis followed by cell death (Bartlett et al., 1986; Kerr et al., 1985, 1986, 1987; Morris et al., 1986; Rommerts et al., 1985, 1988; Edwards et al., 1988). This apparent Leydig cell specificity makes EDS an extremely attractive model compound for toxicological study of androgen-dependent reproductive processes. How-
’ Although the research described herein has been funded by the U.S. Environmental Protection Agency, it does not necessarily reflect the views of agency and no official endorsement should be inferred. * To whom correspondence should be addressed. 0041-008X/91
$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
460
EDS EFFECTS ON LEYDIG
ever, numerous questions concerning the cytotoxicity of EDS on the Leydig cell remain unresolved. We have addressed several of the important issues in this study. First, there is very little data on the dose response to EDS on T production by the Leydig cells exposed either in vivo or in vitro. Second, the results of in vitro studies have differed with respect to the site of EDS action, before or after pregnenolone biosynthesis, during the inhibition of T production (Bu’Lock and Jackson, 1975; Rommerts et al., 1985, 1988). Third, it remains unclear whether the inhibition in T biosynthesis caused by EDS is accompanied by alterations in Leydig cell morphology and/ or cell viability which would be indicative of cell death, or whether the steroidogenic effects of EDS can be divorced from cell death. Fourth and finally, no single study has attempted to correlate the in vitro and in vivo effects of EDS. Recently we developed a multistep isolation procedure for obtaining a highly purified preparation of Leydig cells (98%) from adult male rats (Klinefelter et al., 1987). These highly purified Leydig cells maintain maximal T production and morphological integrity for 24 hr (Klinefelter and Ewing, 1988). In the present study, we tested the dose responsiveness of such highly purified Leydig cells to EDS on T production in vitro. Once we established that Leydig cells in both highly purified and interstitial cell preparations ( 14% Leydig cells) responded similarly to an effective dose of EDS in vitro, we used interstitial cell preparations to determine the effect of EDS on Leydig cells following in vivo exposures. This allowed us to examine a range of in vivo exposures in a single experiment. Thus, we compared T production by highly purified Leydig cells following exposure to EDS in vitro, to T production in vitro by interstitial cell preparations from rats exposed in vivo to a range of doses of EDS. Leydig cells during the EC50 (in vitro) exposure and following the ED50 (in vivo) exposure were incubated with substrate-saturating levels of various stimulators of T production to identify a region(s) in the pathway of T bio-
461
CELLS
synthesis compromised by the action of EDS. The viability of Leydig cells exposed to EDS both in vitro and in vivo was assessed by morphological integrity with light and electron microscopy, 3/3-hydroxysteroid dehydrogenase (3P-HSD) staining, and trypan blue exclusion. Our results demonstrate that EDS produces a dose-dependent decrease in human chor-ionic gonadotropin (hCG)-stimulated T production during a 3-hr in vitro exposure. A dose-dependent decrease in hCG-stimulated T production was observed when interstitial cells were incubated similarly in vitro 24-hr after exposure to EDS in vivo. At the EC50 or ED50 for both in vitro and in vivo exposures, respectively, the biosynthesis of T in the Leydig cell appears to be compromised between the second messenger CAMP and the cholesterol side chain cleavage enzyme. Finally, at the EC50 or ED50 there were no discernable changes in the viability of these Leydig cells. METHODS Animals. Adult (90- to 120day-old) male SpragueDawley rats weighing 350-450 g were purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN) and housed two to three per cage (clear plastic, 20 X 25 X 47cm) with laboratory grade pine shavings as bedding. Rats were maintained under controlled temperature (22°C) humidity (40-50%). and light (14L: lOD, lights out at 7: 00 PM EST) conditions and were given Purina laboratory chow and tap water ad libitum. In vitro treatment. Individual preparations (N = 4) of highly purified Leydig cells were isolated from six rats by the multistep procedure as originally described by Klinefelter et al. (1987) with only minor changes. The Percoll centrifugation was performed at 27,OOOg max and the Hanks’ balanced salts solution used to dilute the isotonic Percoll was buffered with 2.1 g/liter Hepes, pH 7.2. The isolated Leydig cell fraction averaged 98% 3/3-HSD positive staining cells over the course of the experiments. The cells in the purified Leydig cell fraction were enumerated by averaging triplicate hemacytometer counts and 2.0 X IO5 cells were incubated in 1.O ml of Medium I99 (M- 199) in a IS-ml microfuge tube with shaking at 34°C for 3 hr. The order of additions for the Leydig cell incubations was EDS (0,100,200, and 500 P&I in IO ~1 of M-i99 containing 0.1% DMSO), incubation medium (0.8 ml), the Leydig cells (0.2 ml), and finally the stimulant (hCG, dibutyl cyclic adenosine monophosphate (db-CAMP), 20a-hydroxycholesterol (HCHOL). and pregnenolone (PREG)). EDS (99%
462
KLINEFELTER.
LASKEY.
purity as assessed by chromatography) was kindly provided by the Sterling Research Group. A stock of EDS was made fresh for each experiment by dissolving 37.5 mg of EDS in 1.0 ml ofM-199 containing 0.1% DMSO. Either a maximally stimulating concentration of hCG (100 mIU/mi). or a substrate-saturating concentration of db-CAMP (5 mM), HCHOL (5 FM), or PREG (2 pM) was added. These concentrations were previously determined to be maximally stimulating/saturating to Leydig cells and to provide linear T production over the course of a 3-hr incubation in this laboratory (Laskey et al., 1986). Prior to initiating the experiments in this study. similar EDS-free incubations were performed to confirm the linear responsiveness of these cells. Samples of media were collected at the termination of all incubations after cells were pelleted by centrifugation (250g). Samples were subsequently frozen at -70°C for radioimmunoassay of testosterone. In vivo treatment/in vitro assessment. A preparation of interstitial cells containing 14% Leydig cells was used to assess the steroidogenic capacity of Leydig cells following in vivoexposures to EDS. Since interstitial cell preparations utilize only two testes and require much less time than Leydig cell purifications. these preparations were used to examine a range of in vivo exposures in a single experiment (block). Rats (four/treatment group) were assigned randomly to either a 3- or 24-hr vehicle (DMSOwater: I .5: 3.5, v/v) treatment group, or a 3- or 24-hr EDS-treated group (25, 50, 75. or 100 mg EDS/kg body wt via a single ip injection). On each of 4 consecutive days, one rat per treatment group (exposure time and EDS dose) was euthanized by decapitation and blood was collected for serum T assay. Interstitial cells were prepared as originally described by Klinefelter et al. ( 1987) with the single exception that two testes were dissociated in a total of 10 ml in a 50-ml tube. This cell fraction averaged only 14% 3@-HSD positive staining cells. To obtain l-2 X IO’ Leydig cells. I .06 cells were incubated in 1 .O ml of M- I99 as described above for 3 hr. The cells were incubated either without stimulation or challenged by maximal stimulation with hCG. At the onset of the study. interstitial cells in control incubations were also checked for linear responsiveness to hCG stimulation. Again. samples of media were removed following centrifugation and frozen for radioimmunoassay of T. Histology The fixation and subsequent ceil/tissue processing for light and electron microscopy have been described elsewhere (Klinefelter et al., 1990) with minor modifications for preserving isolated Leydig cell morphology. The isolated Leydig cells (1-2 X 106, N = 3) incubated with or without an ED50 concentration (370 pM) of EDS to identify the time course of the effect of EDS on T production, were gently pelleted at the end of the 3-hr incubation and fixed in 5% glutaraldehyde buffered in 0.05 M collidine buffer containing 0. I M sucrose, pH 7.4. After 1 hr. the cells were rinsed twice with buffer by gentle centrifugation. To evaluate the morphology of Leydig cells in the testis 24 hr following an ED50 in vivo EDS exposure (60 mg/
AND
ROBERTS
kg) or a vehicle exposure, three rats per group were anesthetized with Nembutal (50 mg/kg body wt ip), and 2.0 cc of blood was withdrawn from the left ventricle of the heart using a 25-ga needle for T determination. A blunted 16-ga needle was reinserted into the left ventricle and blood was exchanged with DPBS during whole body perfusion. The tissues were then perfuse-fixed with 5% glutaraldehyde in 0.05 M collidine buffer containing 0.1 M sucrose, pH 7.4. Following this initial fixation, the testes were removed and three, small cubes of each testis were excised and immersed in fixative overnight at 4°C. The next day, the tissue cubes were washed in two changes of collidine buffer. Testis cubes and purified Leydig cells were postfixed in 1.O% aqueous osmium tetroxide and 2.5% potassium ferrocyanide for I hr on ice, partially dehydrated in 70% ethanol. immersed in LR White:70% ethanol (2:1, v/v) for I hr. and infiltrated with 100% LR White overnight. After final embedding in LR White, tissue blocks were polymerized overnight at 55°C. Thick sections (1 pm) for light microscopy were stained with an aqueous solution of 1% toluidine blue and 1% borax and examined with a Vanox microscope (Olympus, Japan). Sections from two tissue blocks from each animal were evaluated. Thin sections (silver and gold) of two or more tissue blocks for a given region and treatment were stained with many1 acetate and lead citrate according to the procedure of Sato ( 1967) and examined with a Phillips 300 electron microscope. Testosterone assay and viability measures. Serum collected by centrifugation of blood in serum separator tubes remained frozen until assay. The assay for T in both serum and media samples was carried out using a validated CoatA-Count kit (Diagnostic Products Corp.) procedure (Klinefelter et al., 1990). Cell viability was measured by trypan blue exclusion following a 1: 100 dilution with M199. Staining for 3&HSD enzyme activity was performed as previously described (Klinefelter et al., 1987). The percentage of positively stained cells was used to enumerate the number of steroidogenic Leydig cells in cell preparations following in vivo EDS exposure, as well as both before and after in vitro EDS exposure. The T assay data (T produced in 3 hr) obtained for all interstitial and purified Leydig cell incubations were normalized for IO6 Leydig cells (3P-HSD staining cells) present at the start of the 3hr incubation. Statistics. The data were analyzed using the General Linear Model (GLM) procedure (one-way analysis of variance) of the Statistical Analysis System (SAS) (SAS User’s Guide, 1985). The EC50 and ED50 for in vitro and in vivo exposures, respectively, were estimated by linear regression analysis. Since both the in vitro and the in vivo components of the study were conducted over several experiments (blocks) an analysis was done to determine the presence or absence of block effects (p < 0.05) and, where block effects were found. block was used as a covariable in the analysis. Where overall significance (p < 0.05) was found, the least squares means were compared for significant (p < 0.05) differences.
EDS EFFECTS ON LEYDIG
RESULTS Efect of in Vitro Exposure of Purified Leydig Cells to EDS on Testosterone Production and Morphology Figure 1 shows the dose-response effects of a 3-hr EDS exposure in vitro on T production by highly purified Leydig cells. Similar concentration-dependent decreases in hCG- and db-CAMP-stimulated T production were observed at the end of the 3-hr incubation. An estimated EC50 of 370 PM was calculated for the EDS-induced decrease in hCG- and dbCAMP-stimulated T production in vitro. Substrate-saturating amounts of both HCHOL and PREG were able to maintain T production regardless of the concentration of EDS used, suggesting that in 3 hr EDS had no effect on the activities of the steroidogenic enzymes
- hCG
+ hCG
463
CELLS
in the mitochondria or smooth endoplasmic reticulum of the Leydig cell. Figure 2 shows that hourly hCG-stimulated T production by highly purified Leydig cells increased in a linear fashion during the 3-hr incubation in the absence of EDS. When a 370 PM (EC50) concentration of EDS was added at the onset of the incubation, however, hCG-stimulated T production by highly purified Leydig cells was significantly decreased throughout each of the 3 hr of the incubation, with the greatest decline in T production occuring during the final hour. At the termination of the 3-hr incubation with 370 PM EDS, 90% of the highly purified Leydig cells excluded trypan blue dye and stained intensely for 3@-HSD activity. Moreover, the same number of trypan blue excluding, 3/3-HSD staining cells were present in the vehicle control and EDS-treated incubations (data not shown), suggesting that the effect of
+dbcAMP
+ HCHOL
+ PREG
IN VITRO ADDITION
FIG. I. Graph showing the dose and substrate response of EDS on T production by highly purified Leydig cells in vitro. Leydig cells were incubated for 3 hr with 0, 100,200, or 500 pM EDS either without stimulation or with maximally stimulating levels of hCG or db-cAMP, or with substrate-saturating amounts of HCHOL or PREG. Values represent the mean f SEM for four experiments. Values that are significantly different from vehicle-treated incubations are indicated by * (p < 0.05) and ** (p < 0.0 1). The estimated EC50 for hCG- and db-CAMP-stimulated T production was 370 PM.
464
KLINEFELTER.
LASKEY,
I - EDS/+
AND ROBERTS
hCG
L”
I
0 0
I
I
I
1
I
2 INCUBATION
TIME
I 3
(hours)
FIG. 2. Graph showing the T production by highly purified Leydig cells over time during a typical 3-hr incubation with maximal hCG stimulation, without and with the addition of 370 pM EDS (estimated EC50). Values represent the mean + SEM, N = 6 (duplicate incubations per experiment).
EDS on steroidogenic activity was not accompanied by cell death. Leydig cell viability following this in vitro EDS exposure was confirmed by electron microscopy. Figure 3 (left) is a low power electron micrograph of a purified Leydig cell at the end of the 3-hr incubation in the absence of EDS. The morphological integrity of these highly differentiated cells has been well preserved. Following a 3hr in vitro exposure to 370 PM EDS, the ultrastructural integrity of the isolated Leydig cells still appears to have been well maintained (Fig. 3, right), indicating that these cells are still viable. These EDS-exposed Leydig cells still possessa full complement of steroidogenic organelles, including a filamentous array of microtubules, suggesting that cholesterol transport to the mitochondria was not compromised. In Vivo Treatment/in
Vitro Assessment
The number of 3@-HSD staining Leydig cells recovered in the interstitial cell prepara-
tions following both 3- and 24-hr exposures was not significantly different which suggests that no loss of Leydig cells had occurred in response to EDS exposure (Fig. 4). Surprisingly, the percentage of cells excluding trypan blue was unchanged in the interstitial cell preparations regardless of EDS dose and exposure time. Together these results suggest that Leydig cell viability was not affected by in vivo EDS exposure during these time periods. Figure 4 shows that 3 hr after exposure to EDS in vivo, hCG-stimulated T production by the Leydig cells in the interstitial cell preparations was unchanged, although the values were highly variable. Likewise, serum T was unchanged following 3-hr EDS exposures (not shown). Following a 24-hr exposure to EDS in vivo, hCG-stimulated T production by the Leydig cells in interstitial cell preparations was 225 + 47, 290 + 42, 135 f 45, 33 f 48, and 62 rt 50 ng/106 Leydig cells in the 0, 25, 50, 75, and 100 mg/kg EDS treatment groups, respectively. An estimated ED50 of 60 mg/kg was calculated for the 24-hr in vivo EDS exposure. Although serum T was significantly
EDS EFFECTS ON LEYDIG
CELLS
465
FIG. 3. (Left] An electron micrograph showing the ultrastructure of a typical Leydig cell after a 3-hr incubation with maximal hCG stimulation, but without EDS, X5,554. (Right) An electron micrograph of a Leydig cell after a similar incubation with hCG plus 370 pM EDS. Note that the appearance of the nucleus and all characteristic steroidogenic organelles have been maintained. M, mitochondria; S, smooth endoplasmic reticulum; T, microtubules, X6,653:
decreased 60-70% when 24-hr exposures exceeded 25 mg/kg. no dose-dependent decreases were observed (not shown). To compare the results obtained with interstitial cells following in vivo exposure with those obtained using highly purified Leydig cells during in vitro exposure, purified Leydig cells (98%) were isolated from animals exposed to a 60 mg/kg dose of EDS in vivo for 24 hr. These Leydig cells produced 64 and 52% of the T produced by Leydig cells purified from control animals in response to hCG and dbCAMP, respectively (Fig. 5). Moreover, the Leydig cells purified from EDS-treated animals were able to maintain T production when incubated with HCHOL. These results are consistent with the results obtained when highly purified Leydig cells were incubated in vitro with EDS and various stimulators of T biosynthesis (see above).
The morphology of the Leydig cells within the testis of 24-hr vehicle- and 60 mg/kg EDStreated animals is shown at both the light (Fig. 6) and the electron microscope (Fig. 7) levels. Leydig cells are found in clusters within the interstitial space in both control animals (Fig. 6, left) and animals given a 60 mg/kg dose of EDS (Fig. 6, right). At the light microscope level both control and EDS-exposed Leydig cells appear similar. The Leydig cells contain moderately dense cytoplasm filled with profiles representative of mitochondria and a nucleus with characteristic patches of dense heterochromatin. Likewise, the ultrastructural integrity of cells in both groups of animals appears quite similar. Leydig cells within the testis of control animals (Fig. 7, left) contain an abundance of smooth endoplasmic reticulum and mitochondria, the organelles necessary for T biosynthesis. Following a 60 mg/kg dose of
466
KLINEFELTER,
WOE25
ti0
q
LASKEY,
75
q
3 HOUR
- hCG
100
AND ROBERTS
mg EDS/kg
BW - IN VIVO
TREATMENT
24 HOUR
+ hCG
- hCG
+ hCG
IN VITFIO ADDITION
FIG. 4. A graph showing the dose response of in viva-treated. in vitro-stimulated Leydig cells in interstitial cell preparations following both 3- and 24-hr EDS exposures. The number within the bars indicates the number (X 105) of Leydig cells recovered per animal per time/dose group. Bar values represent the mean T production + SEM of four experiments. Values significantly different from vehicle-treated animals are indicated by ** (p < 0.01). An estimated ED50 of 60 mg/kg EDS was calculated for hCG-stimulated T production following a 24-hr EDS exposure.
EDS, the Leydig cells in the testis still contain a full complement of these organelles (Fig. 7, right). DISCUSSION A requisite in the identification of a mechanism for a chemical which is cytotoxic to a particular cell type is to establish whether a dose-response relationship exists so that the events involved in the cytoxicity can be related to one another as a function of toxicant exposure. When we began to study the effects of EDS on the Leydig cell we found one study which presented dose-response data for the response of Leydig cells in vitro to EDS (Morris et al., 1985). These investigators reported that when Leydig cells in an interstitial cell prep-
aration were incubated in vitro, LH-stimulated T production decreased as the concentration of EDS increased. The EC50 we have estimated in the present study is approximately lo-fold higher than that which we have estimated from the data presented by Morris et al. (1985). We noticed, however, that the T production by the untreated, adult rat Leydig cells was not linear during the 36-hr cultures used in this earlier study. Second, the hourly LH-stimulated T production by the untreated Leydig cells was only one-tenth of the T production we routinely observe following hCG stimulation in our preparations (Klinefelter and Ewing, 1988). We felt that these differences indicated that the steroidogenic capacity of the untreated Leydig ceils was compromised, which in turn might influence the accuracy of dose-response data. Moreover, since
EDS EFFECTS ON LEYDIG
n
-hCG
B
+hCG
+ db CAMP
CELLS
q
467
+ HCHOL
60 mgikg BW
VEHICLE
IN VW0 TREATMENT FIG. 5. A graph showing the T production by highly purified Leydig cells during a 3-hr incubation in vitro following a 24-hr exposure to vehicle or 60 mg/kg EDS (estimated ED50) in vivo. Values significantly different from vehicle-treated animals are indicated by ** (p < 0.01). The number within the bars indicates the number (X 106) of purified Leydig cells recovered from six animals per treatment group. As with Leydig cells exposed to EDS in vitro, both hCG- and db-CAMP-stimulated T production are significantly decreased, whereas HCHOL stimulates T production - 60% over control levels.
our results indicate that EDS affects T production acutely during the first and third hours of in vitro incubation, incubation periods longer than 3 hr would be expected to yield erroneously low EC50 determinations. Finally, our estimated in vitro EC50 exposure approximates the response to EDS recently observed (Dr. William Kelce, personal communication) using a 3-hr in vitro-perfused testis model which maintains Leydig cells in their 3-d& mensional cytoarchitecture (Chubb and Ewing, 1979). In the present study we used a preparation of highly purified Leydig cells which maintains full steroidogenic capacity in culture for several days (Klinefelter et al., 1989). We were able to detect a significant decrease in T production by these cells in response to EDS in just 3 hr and maximally stimulated T production by untreated, purified Leydig cells was
linear during this time. Moreover, the Leydig cells in the interstitial cell preparations used to determine the effects of in vivo EDS exposure showed an in vitro response to EDS which was similar to the effect obtained using purified Leydig cells in vitro. Leydig cells in interstitial preparations showed a 30% decrease in hCGstimulated T production when incubated with 200 PM EDS for 3 hr, whereas highly purified Leydig cells showed a 36% decrease in T production when exposed to this EDS concentration (data not shown). Thus, the Leydig cells in the interstitial cell preparations were equally well maintained and were useful in determining the response to an in vivo exposure to EDS, and in fact, responded in a manner similar to that of highly purified Leydig cells following in vivo exposure. It might be argued that the Leydig cells obtained in the interstitial cell preparation, and also in the purified prepa-
468
KLINEFELTER.
LASKEY.
AND
ROBERTS
FIG. 6. (Left) A light micrograph showing an interstitial region of a testis from a control animal which contains many Leydig cells. X 1.067. (Right) A similar light micrograph of the interstitial region of a testis 24 hr after a 60 mg/kg dose of EDS (estimated ED50) was administered. Leydig cell clusters are still evident. These Leydig cells possess a moderately dense cytoplasm with profiles representative of mitochondria and a nucleus with patches of dense heterochromatin resembling those in vehicle-treated animals. X 1.067.
ration, are not representative of the Leydig cell population in the testis, but rather represent a select population of Leydig cells. However, Klinefelter ef al. (1987) demonstrated that these highly purified Leydig cells respond similarly to the Leydig cells in the in vitro-perfused testis when LH is withdrawn in viva Furthermore, in this study the Leydig cells in both interstitial and purified preparations responded similarly to both in vitro and in vivo EDS exposures. Within 24 hr following EDS administration in vivo, EDS has been reported to cause morphological degeneration of the Leydig cells in the rat testis (75 mg/kg: Kerr et al. 1985, 1986; 100 mg/kg: Morris et al., 1986). Our results seem to differ from these observations on several accounts. Twenty-four hours following in vivo exposures to EDS, viability as measured
by trypan blue exclusion was high among cells in the interstitial preparations regardless of the dose of EDS given. Moreover, the Leydig cells in these preparations stained intensely for 3pHSD enzyme activity and were obtained in similar numbers at all doses of EDS used in the in vivo experiments. Finally, the Leydig cells in the testes of animals 24 hr following the estimated ED50 dose of EDS were indistinguishable from Leydig cells in control animals at both the light and the electron microscope level. In accordance with the literature, we presume that the morphology of Leydig cells in the testes of rats given higher doses of EDS would have indicated some degenerative changes. However, based on the criteria for Leydig cell viability which we have chosen: 3P-HSD staining, morphological preservation, the ability to respond to KG,
EDS EFFECTS ON LEYDIG
CELLS
469
FIG. 7. (Left) An electron micrograph revealing the Leydig cell ultrastructure in testes of control animals. Numerous profiles of smooth endoplasmic reticulum (S) and mitochondria (M), organelles intimately involved with T biosynthesis, are evident, X 13,095. (Right) The ultrastructure of Leydig cells in testes 24 hr following a 60 mg/kg dose of EDS appear unchanged, X4.9 18.
and trypan blue exclusion, the viability of the Leydig cells does not appear to have been altered by an ED50 dose of EDS in vivo. Perhaps it is reasonable to assume that the cells are in a degenerative state but that none of these measures are sensitive enough to detect this change. We found that the effects of EDS on T production in vitro occurred rapidly yet cell viability, as assessed by the criteria discussed above, was unchanged during the 3-hr exposure. These results are consistent with the study by Morris et ul. (1985) but are in contrast to the study by Rommerts et al. (1988). The latter in vitro study reported that EDS (400 PM) altered the morphology of the Leydig cell. We were unable to identify any consistent morphological change in Leydig cell morphology at either the light or the electron microscope level with a 370 PM (EC50) exposure. We sus-
pect that if Leydig cell viability was diminished in culture as discussed above, the morphological integrity of all cells, control and treated, would be severely affected. Previous in vitro studies which have evaluated the possible sites of EDS action in the pathway of T biosynthesis have yielded contradictory results. For example, studies by Bu’Lock and Jackson (1972, 1975) reported that in vivo exposure to EDS reduced the ability of the Leydig cell to convert pregnenolone to testosterone in vitro, and implied that EDS exposure reduced the activity of the steroidogenic enzymes in the smooth endoplasmic reticulum of the Leydig cell. Rommerts et al. (1985) reported that when Leydig cells were exposed to EDS in vitro, LH-stimulated, but not 22R-hydroxycholesterol-stimulated, T production diminished. This suggests that the lesion induced by EDS is prior to the choles-
470
KLINEFELTER,
LASKEY.
terol side chain cleavage enzyme which converts cholesterol to pregnenolone in the mitochondria, and hence prior to the enzymes in the smooth endoplasmic reticulum. Later, Rommerts et al. (1988) reported that 22R-hydroxycholesterol did not prevent a decrease in pregnenolone production caused by EDS. These results would have confirmed the findings of Bu’Lock and Jackson ( 1972, 1975) that EDS induces a lesion in the smooth endoplasmin reticulum. Unfortunately the untreated Leydig cells cultured in the study by Rommerts et al. (1988) failed to produce steroid (pregnenolone) linearly over time (24 hr), which indicated to us that the Leydig cells were losing steroidogenic function over time in vitro. Again, since EDS affects the Leydig cell in vitro acutely, we felt that a 3 hr incubation was best to monitor the initial effects on T biosynthesis. In the present study, we have shown that the biosynthesis of T, following both in vitro and in vivo exposure to EDS, is compromised between the CAMP activation of protein kinase and the cholesterol side chain cleavage enzyme because maximal stimulation with db-CAMP failed to maintain T production by the Leydig cell in the presence of EDS, while substrate-saturating amounts of 2Oahydroxycholesterol maintained T production. However, our results do not preclude the possibility that EDS competes with the steroidogenie enzymes involved in T biosynthesis when the substrates (HCHOL, PREG) are present in physiological concentrations. We have established that the responsiveness of the Leydig cells to EDS both in vitro and in vivo is a function of the concentration or dose of EDS to which the Leydig cells are exposed. By identifying an EC50 and ED50 for in vitro and in vivo EDS exposure, respectively, we were able to correlate the in vitro and in vivo effects of EDS as a function of both the steroidogenic capacity and morphology of the Leydig cell. We conclude that a 3-hr in vitro EC50 exposure or a 24-hr in vivo ED50 exposure to EDS does not alter Leydig cell viability, but does, however, compromise the function of some aspect of testosterone bio-
AND ROBERTS
synthesis. The biochemical lesion appears to be between the production of CAMP and the action of cholesterol side chain cleavage enzyme. Now that we feel satisfied that the in vitro effects of EDS are similar to the in vivo effects, our goals are twofold. First, we will attempt to establish in vitro whether the action of EDS on Leydig cell steroidogenesis is reversible. Second, we will attempt to determine whether the final event in the cytotoxicity of EDS, namely, Leydig cell death, is mediated through other testicular factors.
REFERENCES BARTLETT, J. M. S., KERR, J. B.. AND SHARPE, R. M. (1986). The effect of selective destruction and regeneration of rat Leydig cells on the intratesticular distribution of T and morphology of the seminiferous epithelium. J. Androl. 7, 240-253. BU’LOCK, D. E., AND JACKSON,C. M. (1972). Suppression of testicular androgen synthesis in the rat by ethylene dimethanesulphonate. Gynecol. Invesf. 2, 305-308. Bu’LocK, D. E.. AND JACKSIN, C. M. ( 1975). Suppression of testicular androgen synthesis in the rat by ethane- 1,2dimethanesulphonate. J Steroid B&hem. 6, 118 l-l 185. CHUBB, C.. AND EWING, L. L. (1979). Steroid secretion by in vitro perfused testes: Testosterone biosynthetic pathways. Amer. J. Physiol. 237, 231-238. EDWARDS, G., Fox, B. W., JACKSON, H.. AND MORRIS, I. D. (1988). Lqvdig Cell Cytoto.xicity qfPutative Male Antifertility Compounds Related to Ethane-1,2-dimethanesulphonate, pp. 185- 19 1. Serono Symp. Publ., Raven Press, New York. JACKSON, H. ( 1973). Chemical methods of male contraception. Amer. Sci. 61, 188-193. KERR, J. B., BARTLETT, J. M. S., DONACHIE, K. (1986). Acute response of testicular interstitial tissue in rats to the cytotoxic drug ethane dimethanesulphonate. Cell Tissue Rex 243, 405-4 14. KERR, J. B., DONACHIE, K., ROMMERTS, F. F. G. (1985). Selective destruction and regeneration of rat Leydig cells in viva. A new method for the study of seminiferous tubular-interstitial tissue interaction. Cell Tissue Res. 242, 145-l 56. KERR, J. B., KNELL, C. M.. ABBOTT, M., AND D~NACHIE, K. ( 1987). Ultrastructural analysis of the effect of ethane dimethanesulphonate on the testis of the rat, guinea pig, hamster, and mouse. Cell Tissue Res. 249,45 l-457. KLINEFELTER. G. R.. LASKEY, J. W., ROBERTS, N. R., SLOTT, V. S., AND SUAREZ,J. D. (1990). Multiple effects of ethane dimethanesulfonate on the epididymis of adult rats. Toxicol. Appl. Pharmacol. 105, 27 l-287.
EDS EFFECTS ON LEYDIG G. R., AND EWING, L. L. (1988). Optimizing testosterone production by purified adult rat Lcydig cells in vitro. in Vitro Cell. Dev. Biol. 24, 545-549. KLINEFELTER, G. R., AND EWING, L. L. (1989). Maintenance of testosterone production by purified adult rat Leydig cells for 3 days in vitro. In Vitro Cell. Dev. Biol. 25,283-288. KLINEFELTER,
KLINEFELTER,
G.
R.,
HALL,
P. F., AND
EWING,
L.
L.
(1987). Effect of luteinizing hormone deprivation in situ on steroidogenesis of rat Leydig cells purified by a multistep procedure. Biol. Reprod. 36,769-783. LASKEY,
J. W.. PHI:LPS,
P. V., LAWS,
S. D., AND
FERRELL,
J. ( 1986). Leydig cell function in vitro following metal cation treatment. Toxicologist 6(l), 86. MORRIS,
I. D., BARDIN,
C. W., AND
MATHER,
J. P. (1985).
Inhibition of LH-stimulated but not basal testosterone secretion of isolated Leydig cells by the cytotoxic ethylene dimethanesulphonate. IRCS Med. Sci. 13, 1052- 1053.
MORRIS,
471
CELLS 1. D.,
PHILLIPS,
D.
M.,
AND
BARDIN,
C. W.
(1986). Ethylene dimethanesulphonate destroys Leydig cells in the rat testis. Endocrinology 118, 709-7 19. ROMMERTS, BRUGE,
F. F. G., J. W., AND
GROOTENHLJIS, J. W., VAN DER MOLEN, H.
HOOGERJ. (1985).
Ethane dimethanesulphonate (EDS) specifically inhibits LH-stimulated steroidogenesis in Leydig cells isolated from mature rats but not in cells from immature rats. Mol. Cell. Endocrinol. 42, 105- Ill. ROMMERTS, BRUGGE,
F. F. G.,
TEERDS,
K.
J., AND
HOOGER-
J. W. (1988). In vitro effects of ethylene-dimethane sulfonate (EDS) on Leydig cells: Inhibition of steroid production and cytotoxic effects are dependent on speciesand age of rat. Mol. Cell. Endocrinol. 55,8794. SAS Institute, Inc. (1985). SAS User’s Guide: Basics, Version 5. SAS Institute Inc., Cary, NC. SATO, T. (1967). A modified method for lead staining. J. Electron Micros. 16, 133- 135.
AND
APPLIED
PHARMACOLOGY
In Vitro/in
107,460-47
1 (199
1)
Vivo Effects of Ethane Dimethanesulfonate on Leydig Cells of Adult Rats’
GARY R. KLINEFELTER,
*,2 JOHN W. LASKEY,~
AND NAOMI
L. ROBERTS*
NSI, Inc.,* and HERL USEPA,? Reproductive Toxicology Branch. MD No. 72, Research Triangle Park, North Carolina 27711
Received June 1. 1990; accepted October IS, 1990 In Vitro/in Vivo Effects of Ethane Dimethanesulfonate on Leydig cells of Adult Rats. KLINEG. R., LASKEY, J. W.. AND ROBERTS, N. L. ( 1991). Toxicol. Appl. Pharmacol. 107,46047 1. Although ethane dimethanesulfonate (EDS) is well recognized as a Leydig cell toxicant, the dose responsiveness of Leydig cells to EDS, both in vitro and in vivo. is not well established. In addition, the cellular site of action of EDS during Leydig cell toxicity and the status of Leydig cell viability during the affected period remain controversial. We determined the in v&o EC50 (370 PM) and in vivo ED50 (60 mg/kg) for human chorionic gonadotropin (hCG>stimulated testosterone (T) production using both highly purified (98%) and interstitial (14%) Leydig cell preparations, respectively. Leydig cells were recovered in approximately equal numbers following all in vivo and in vitro EDS exposures. The Leydig cells in these preparations were viable and steroidogenically active (3@-HSD positive) subsequent to all exposures, both before and afier incubations to stimulate T biosynthesis. When hCG-stimulated T production was decreased 50% following in vivo or in vitro exposures, the morphological integrity of the Leydig cells appeared normal, with no discernible lesion at either the light or the electron microscope level. We used stimulants of various reactions in the pathway of T biosynthesis (ZOa-hydroxycholesterol and prenenolone) to determine the site of action impaired when T biosynthesis was decreased. Our results indicate that when Leydig cells are exposed to EDS either in vitro or in vivo,the biosynthesis of T is compromised between the cyclic adenosine monophosphate activation of protein kinase and the cholesterol side chain cleavage enzyme. 0 1991 Academic Press. Inc FELTER,
The transient infertility of male rats given a single 50 mg/kg ip injection of ethane dimethanesulfonate (EDS) was described by Jackson (1973). He also reported that the simultaneous administration of testosterone (T) prevents much of the damage which occurs to the seminiferous epithelium and the shrinkage of the prostate and seminal vesicles, suggesting Leydig cell dysfunction (Jackson, 1973). Bu’Lock and Jackson (1972, 1975) provided the first
indications that the Leydig cell was targeted by EDS when radiolabeled precursors of testosterone biosynthesis failed to be converted into testosterone in vitro by testes from animals given EDS in vivo. Subsequently, numerous studies have suggested that EDS is a specific Leydig cell toxicant which causes a reversible inhibition of Leydig cell steroidogenesis followed by cell death (Bartlett et al., 1986; Kerr et al., 1985, 1986, 1987; Morris et al., 1986; Rommerts et al., 1985, 1988; Edwards et al., 1988). This apparent Leydig cell specificity makes EDS an extremely attractive model compound for toxicological study of androgen-dependent reproductive processes. How-
’ Although the research described herein has been funded by the U.S. Environmental Protection Agency, it does not necessarily reflect the views of agency and no official endorsement should be inferred. * To whom correspondence should be addressed. 0041-008X/91
$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
460
EDS EFFECTS ON LEYDIG
ever, numerous questions concerning the cytotoxicity of EDS on the Leydig cell remain unresolved. We have addressed several of the important issues in this study. First, there is very little data on the dose response to EDS on T production by the Leydig cells exposed either in vivo or in vitro. Second, the results of in vitro studies have differed with respect to the site of EDS action, before or after pregnenolone biosynthesis, during the inhibition of T production (Bu’Lock and Jackson, 1975; Rommerts et al., 1985, 1988). Third, it remains unclear whether the inhibition in T biosynthesis caused by EDS is accompanied by alterations in Leydig cell morphology and/ or cell viability which would be indicative of cell death, or whether the steroidogenic effects of EDS can be divorced from cell death. Fourth and finally, no single study has attempted to correlate the in vitro and in vivo effects of EDS. Recently we developed a multistep isolation procedure for obtaining a highly purified preparation of Leydig cells (98%) from adult male rats (Klinefelter et al., 1987). These highly purified Leydig cells maintain maximal T production and morphological integrity for 24 hr (Klinefelter and Ewing, 1988). In the present study, we tested the dose responsiveness of such highly purified Leydig cells to EDS on T production in vitro. Once we established that Leydig cells in both highly purified and interstitial cell preparations ( 14% Leydig cells) responded similarly to an effective dose of EDS in vitro, we used interstitial cell preparations to determine the effect of EDS on Leydig cells following in vivo exposures. This allowed us to examine a range of in vivo exposures in a single experiment. Thus, we compared T production by highly purified Leydig cells following exposure to EDS in vitro, to T production in vitro by interstitial cell preparations from rats exposed in vivo to a range of doses of EDS. Leydig cells during the EC50 (in vitro) exposure and following the ED50 (in vivo) exposure were incubated with substrate-saturating levels of various stimulators of T production to identify a region(s) in the pathway of T bio-
461
CELLS
synthesis compromised by the action of EDS. The viability of Leydig cells exposed to EDS both in vitro and in vivo was assessed by morphological integrity with light and electron microscopy, 3/3-hydroxysteroid dehydrogenase (3P-HSD) staining, and trypan blue exclusion. Our results demonstrate that EDS produces a dose-dependent decrease in human chor-ionic gonadotropin (hCG)-stimulated T production during a 3-hr in vitro exposure. A dose-dependent decrease in hCG-stimulated T production was observed when interstitial cells were incubated similarly in vitro 24-hr after exposure to EDS in vivo. At the EC50 or ED50 for both in vitro and in vivo exposures, respectively, the biosynthesis of T in the Leydig cell appears to be compromised between the second messenger CAMP and the cholesterol side chain cleavage enzyme. Finally, at the EC50 or ED50 there were no discernable changes in the viability of these Leydig cells. METHODS Animals. Adult (90- to 120day-old) male SpragueDawley rats weighing 350-450 g were purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN) and housed two to three per cage (clear plastic, 20 X 25 X 47cm) with laboratory grade pine shavings as bedding. Rats were maintained under controlled temperature (22°C) humidity (40-50%). and light (14L: lOD, lights out at 7: 00 PM EST) conditions and were given Purina laboratory chow and tap water ad libitum. In vitro treatment. Individual preparations (N = 4) of highly purified Leydig cells were isolated from six rats by the multistep procedure as originally described by Klinefelter et al. (1987) with only minor changes. The Percoll centrifugation was performed at 27,OOOg max and the Hanks’ balanced salts solution used to dilute the isotonic Percoll was buffered with 2.1 g/liter Hepes, pH 7.2. The isolated Leydig cell fraction averaged 98% 3/3-HSD positive staining cells over the course of the experiments. The cells in the purified Leydig cell fraction were enumerated by averaging triplicate hemacytometer counts and 2.0 X IO5 cells were incubated in 1.O ml of Medium I99 (M- 199) in a IS-ml microfuge tube with shaking at 34°C for 3 hr. The order of additions for the Leydig cell incubations was EDS (0,100,200, and 500 P&I in IO ~1 of M-i99 containing 0.1% DMSO), incubation medium (0.8 ml), the Leydig cells (0.2 ml), and finally the stimulant (hCG, dibutyl cyclic adenosine monophosphate (db-CAMP), 20a-hydroxycholesterol (HCHOL). and pregnenolone (PREG)). EDS (99%
462
KLINEFELTER.
LASKEY.
purity as assessed by chromatography) was kindly provided by the Sterling Research Group. A stock of EDS was made fresh for each experiment by dissolving 37.5 mg of EDS in 1.0 ml ofM-199 containing 0.1% DMSO. Either a maximally stimulating concentration of hCG (100 mIU/mi). or a substrate-saturating concentration of db-CAMP (5 mM), HCHOL (5 FM), or PREG (2 pM) was added. These concentrations were previously determined to be maximally stimulating/saturating to Leydig cells and to provide linear T production over the course of a 3-hr incubation in this laboratory (Laskey et al., 1986). Prior to initiating the experiments in this study. similar EDS-free incubations were performed to confirm the linear responsiveness of these cells. Samples of media were collected at the termination of all incubations after cells were pelleted by centrifugation (250g). Samples were subsequently frozen at -70°C for radioimmunoassay of testosterone. In vivo treatment/in vitro assessment. A preparation of interstitial cells containing 14% Leydig cells was used to assess the steroidogenic capacity of Leydig cells following in vivoexposures to EDS. Since interstitial cell preparations utilize only two testes and require much less time than Leydig cell purifications. these preparations were used to examine a range of in vivo exposures in a single experiment (block). Rats (four/treatment group) were assigned randomly to either a 3- or 24-hr vehicle (DMSOwater: I .5: 3.5, v/v) treatment group, or a 3- or 24-hr EDS-treated group (25, 50, 75. or 100 mg EDS/kg body wt via a single ip injection). On each of 4 consecutive days, one rat per treatment group (exposure time and EDS dose) was euthanized by decapitation and blood was collected for serum T assay. Interstitial cells were prepared as originally described by Klinefelter et al. ( 1987) with the single exception that two testes were dissociated in a total of 10 ml in a 50-ml tube. This cell fraction averaged only 14% 3@-HSD positive staining cells. To obtain l-2 X IO’ Leydig cells. I .06 cells were incubated in 1 .O ml of M- I99 as described above for 3 hr. The cells were incubated either without stimulation or challenged by maximal stimulation with hCG. At the onset of the study. interstitial cells in control incubations were also checked for linear responsiveness to hCG stimulation. Again. samples of media were removed following centrifugation and frozen for radioimmunoassay of T. Histology The fixation and subsequent ceil/tissue processing for light and electron microscopy have been described elsewhere (Klinefelter et al., 1990) with minor modifications for preserving isolated Leydig cell morphology. The isolated Leydig cells (1-2 X 106, N = 3) incubated with or without an ED50 concentration (370 pM) of EDS to identify the time course of the effect of EDS on T production, were gently pelleted at the end of the 3-hr incubation and fixed in 5% glutaraldehyde buffered in 0.05 M collidine buffer containing 0. I M sucrose, pH 7.4. After 1 hr. the cells were rinsed twice with buffer by gentle centrifugation. To evaluate the morphology of Leydig cells in the testis 24 hr following an ED50 in vivo EDS exposure (60 mg/
AND
ROBERTS
kg) or a vehicle exposure, three rats per group were anesthetized with Nembutal (50 mg/kg body wt ip), and 2.0 cc of blood was withdrawn from the left ventricle of the heart using a 25-ga needle for T determination. A blunted 16-ga needle was reinserted into the left ventricle and blood was exchanged with DPBS during whole body perfusion. The tissues were then perfuse-fixed with 5% glutaraldehyde in 0.05 M collidine buffer containing 0.1 M sucrose, pH 7.4. Following this initial fixation, the testes were removed and three, small cubes of each testis were excised and immersed in fixative overnight at 4°C. The next day, the tissue cubes were washed in two changes of collidine buffer. Testis cubes and purified Leydig cells were postfixed in 1.O% aqueous osmium tetroxide and 2.5% potassium ferrocyanide for I hr on ice, partially dehydrated in 70% ethanol. immersed in LR White:70% ethanol (2:1, v/v) for I hr. and infiltrated with 100% LR White overnight. After final embedding in LR White, tissue blocks were polymerized overnight at 55°C. Thick sections (1 pm) for light microscopy were stained with an aqueous solution of 1% toluidine blue and 1% borax and examined with a Vanox microscope (Olympus, Japan). Sections from two tissue blocks from each animal were evaluated. Thin sections (silver and gold) of two or more tissue blocks for a given region and treatment were stained with many1 acetate and lead citrate according to the procedure of Sato ( 1967) and examined with a Phillips 300 electron microscope. Testosterone assay and viability measures. Serum collected by centrifugation of blood in serum separator tubes remained frozen until assay. The assay for T in both serum and media samples was carried out using a validated CoatA-Count kit (Diagnostic Products Corp.) procedure (Klinefelter et al., 1990). Cell viability was measured by trypan blue exclusion following a 1: 100 dilution with M199. Staining for 3&HSD enzyme activity was performed as previously described (Klinefelter et al., 1987). The percentage of positively stained cells was used to enumerate the number of steroidogenic Leydig cells in cell preparations following in vivo EDS exposure, as well as both before and after in vitro EDS exposure. The T assay data (T produced in 3 hr) obtained for all interstitial and purified Leydig cell incubations were normalized for IO6 Leydig cells (3P-HSD staining cells) present at the start of the 3hr incubation. Statistics. The data were analyzed using the General Linear Model (GLM) procedure (one-way analysis of variance) of the Statistical Analysis System (SAS) (SAS User’s Guide, 1985). The EC50 and ED50 for in vitro and in vivo exposures, respectively, were estimated by linear regression analysis. Since both the in vitro and the in vivo components of the study were conducted over several experiments (blocks) an analysis was done to determine the presence or absence of block effects (p < 0.05) and, where block effects were found. block was used as a covariable in the analysis. Where overall significance (p < 0.05) was found, the least squares means were compared for significant (p < 0.05) differences.
EDS EFFECTS ON LEYDIG
RESULTS Efect of in Vitro Exposure of Purified Leydig Cells to EDS on Testosterone Production and Morphology Figure 1 shows the dose-response effects of a 3-hr EDS exposure in vitro on T production by highly purified Leydig cells. Similar concentration-dependent decreases in hCG- and db-CAMP-stimulated T production were observed at the end of the 3-hr incubation. An estimated EC50 of 370 PM was calculated for the EDS-induced decrease in hCG- and dbCAMP-stimulated T production in vitro. Substrate-saturating amounts of both HCHOL and PREG were able to maintain T production regardless of the concentration of EDS used, suggesting that in 3 hr EDS had no effect on the activities of the steroidogenic enzymes
- hCG
+ hCG
463
CELLS
in the mitochondria or smooth endoplasmic reticulum of the Leydig cell. Figure 2 shows that hourly hCG-stimulated T production by highly purified Leydig cells increased in a linear fashion during the 3-hr incubation in the absence of EDS. When a 370 PM (EC50) concentration of EDS was added at the onset of the incubation, however, hCG-stimulated T production by highly purified Leydig cells was significantly decreased throughout each of the 3 hr of the incubation, with the greatest decline in T production occuring during the final hour. At the termination of the 3-hr incubation with 370 PM EDS, 90% of the highly purified Leydig cells excluded trypan blue dye and stained intensely for 3@-HSD activity. Moreover, the same number of trypan blue excluding, 3/3-HSD staining cells were present in the vehicle control and EDS-treated incubations (data not shown), suggesting that the effect of
+dbcAMP
+ HCHOL
+ PREG
IN VITRO ADDITION
FIG. I. Graph showing the dose and substrate response of EDS on T production by highly purified Leydig cells in vitro. Leydig cells were incubated for 3 hr with 0, 100,200, or 500 pM EDS either without stimulation or with maximally stimulating levels of hCG or db-cAMP, or with substrate-saturating amounts of HCHOL or PREG. Values represent the mean f SEM for four experiments. Values that are significantly different from vehicle-treated incubations are indicated by * (p < 0.05) and ** (p < 0.0 1). The estimated EC50 for hCG- and db-CAMP-stimulated T production was 370 PM.
464
KLINEFELTER.
LASKEY,
I - EDS/+
AND ROBERTS
hCG
L”
I
0 0
I
I
I
1
I
2 INCUBATION
TIME
I 3
(hours)
FIG. 2. Graph showing the T production by highly purified Leydig cells over time during a typical 3-hr incubation with maximal hCG stimulation, without and with the addition of 370 pM EDS (estimated EC50). Values represent the mean + SEM, N = 6 (duplicate incubations per experiment).
EDS on steroidogenic activity was not accompanied by cell death. Leydig cell viability following this in vitro EDS exposure was confirmed by electron microscopy. Figure 3 (left) is a low power electron micrograph of a purified Leydig cell at the end of the 3-hr incubation in the absence of EDS. The morphological integrity of these highly differentiated cells has been well preserved. Following a 3hr in vitro exposure to 370 PM EDS, the ultrastructural integrity of the isolated Leydig cells still appears to have been well maintained (Fig. 3, right), indicating that these cells are still viable. These EDS-exposed Leydig cells still possessa full complement of steroidogenic organelles, including a filamentous array of microtubules, suggesting that cholesterol transport to the mitochondria was not compromised. In Vivo Treatment/in
Vitro Assessment
The number of 3@-HSD staining Leydig cells recovered in the interstitial cell prepara-
tions following both 3- and 24-hr exposures was not significantly different which suggests that no loss of Leydig cells had occurred in response to EDS exposure (Fig. 4). Surprisingly, the percentage of cells excluding trypan blue was unchanged in the interstitial cell preparations regardless of EDS dose and exposure time. Together these results suggest that Leydig cell viability was not affected by in vivo EDS exposure during these time periods. Figure 4 shows that 3 hr after exposure to EDS in vivo, hCG-stimulated T production by the Leydig cells in the interstitial cell preparations was unchanged, although the values were highly variable. Likewise, serum T was unchanged following 3-hr EDS exposures (not shown). Following a 24-hr exposure to EDS in vivo, hCG-stimulated T production by the Leydig cells in interstitial cell preparations was 225 + 47, 290 + 42, 135 f 45, 33 f 48, and 62 rt 50 ng/106 Leydig cells in the 0, 25, 50, 75, and 100 mg/kg EDS treatment groups, respectively. An estimated ED50 of 60 mg/kg was calculated for the 24-hr in vivo EDS exposure. Although serum T was significantly
EDS EFFECTS ON LEYDIG
CELLS
465
FIG. 3. (Left] An electron micrograph showing the ultrastructure of a typical Leydig cell after a 3-hr incubation with maximal hCG stimulation, but without EDS, X5,554. (Right) An electron micrograph of a Leydig cell after a similar incubation with hCG plus 370 pM EDS. Note that the appearance of the nucleus and all characteristic steroidogenic organelles have been maintained. M, mitochondria; S, smooth endoplasmic reticulum; T, microtubules, X6,653:
decreased 60-70% when 24-hr exposures exceeded 25 mg/kg. no dose-dependent decreases were observed (not shown). To compare the results obtained with interstitial cells following in vivo exposure with those obtained using highly purified Leydig cells during in vitro exposure, purified Leydig cells (98%) were isolated from animals exposed to a 60 mg/kg dose of EDS in vivo for 24 hr. These Leydig cells produced 64 and 52% of the T produced by Leydig cells purified from control animals in response to hCG and dbCAMP, respectively (Fig. 5). Moreover, the Leydig cells purified from EDS-treated animals were able to maintain T production when incubated with HCHOL. These results are consistent with the results obtained when highly purified Leydig cells were incubated in vitro with EDS and various stimulators of T biosynthesis (see above).
The morphology of the Leydig cells within the testis of 24-hr vehicle- and 60 mg/kg EDStreated animals is shown at both the light (Fig. 6) and the electron microscope (Fig. 7) levels. Leydig cells are found in clusters within the interstitial space in both control animals (Fig. 6, left) and animals given a 60 mg/kg dose of EDS (Fig. 6, right). At the light microscope level both control and EDS-exposed Leydig cells appear similar. The Leydig cells contain moderately dense cytoplasm filled with profiles representative of mitochondria and a nucleus with characteristic patches of dense heterochromatin. Likewise, the ultrastructural integrity of cells in both groups of animals appears quite similar. Leydig cells within the testis of control animals (Fig. 7, left) contain an abundance of smooth endoplasmic reticulum and mitochondria, the organelles necessary for T biosynthesis. Following a 60 mg/kg dose of
466
KLINEFELTER,
WOE25
ti0
q
LASKEY,
75
q
3 HOUR
- hCG
100
AND ROBERTS
mg EDS/kg
BW - IN VIVO
TREATMENT
24 HOUR
+ hCG
- hCG
+ hCG
IN VITFIO ADDITION
FIG. 4. A graph showing the dose response of in viva-treated. in vitro-stimulated Leydig cells in interstitial cell preparations following both 3- and 24-hr EDS exposures. The number within the bars indicates the number (X 105) of Leydig cells recovered per animal per time/dose group. Bar values represent the mean T production + SEM of four experiments. Values significantly different from vehicle-treated animals are indicated by ** (p < 0.01). An estimated ED50 of 60 mg/kg EDS was calculated for hCG-stimulated T production following a 24-hr EDS exposure.
EDS, the Leydig cells in the testis still contain a full complement of these organelles (Fig. 7, right). DISCUSSION A requisite in the identification of a mechanism for a chemical which is cytotoxic to a particular cell type is to establish whether a dose-response relationship exists so that the events involved in the cytoxicity can be related to one another as a function of toxicant exposure. When we began to study the effects of EDS on the Leydig cell we found one study which presented dose-response data for the response of Leydig cells in vitro to EDS (Morris et al., 1985). These investigators reported that when Leydig cells in an interstitial cell prep-
aration were incubated in vitro, LH-stimulated T production decreased as the concentration of EDS increased. The EC50 we have estimated in the present study is approximately lo-fold higher than that which we have estimated from the data presented by Morris et al. (1985). We noticed, however, that the T production by the untreated, adult rat Leydig cells was not linear during the 36-hr cultures used in this earlier study. Second, the hourly LH-stimulated T production by the untreated Leydig cells was only one-tenth of the T production we routinely observe following hCG stimulation in our preparations (Klinefelter and Ewing, 1988). We felt that these differences indicated that the steroidogenic capacity of the untreated Leydig ceils was compromised, which in turn might influence the accuracy of dose-response data. Moreover, since
EDS EFFECTS ON LEYDIG
n
-hCG
B
+hCG
+ db CAMP
CELLS
q
467
+ HCHOL
60 mgikg BW
VEHICLE
IN VW0 TREATMENT FIG. 5. A graph showing the T production by highly purified Leydig cells during a 3-hr incubation in vitro following a 24-hr exposure to vehicle or 60 mg/kg EDS (estimated ED50) in vivo. Values significantly different from vehicle-treated animals are indicated by ** (p < 0.01). The number within the bars indicates the number (X 106) of purified Leydig cells recovered from six animals per treatment group. As with Leydig cells exposed to EDS in vitro, both hCG- and db-CAMP-stimulated T production are significantly decreased, whereas HCHOL stimulates T production - 60% over control levels.
our results indicate that EDS affects T production acutely during the first and third hours of in vitro incubation, incubation periods longer than 3 hr would be expected to yield erroneously low EC50 determinations. Finally, our estimated in vitro EC50 exposure approximates the response to EDS recently observed (Dr. William Kelce, personal communication) using a 3-hr in vitro-perfused testis model which maintains Leydig cells in their 3-d& mensional cytoarchitecture (Chubb and Ewing, 1979). In the present study we used a preparation of highly purified Leydig cells which maintains full steroidogenic capacity in culture for several days (Klinefelter et al., 1989). We were able to detect a significant decrease in T production by these cells in response to EDS in just 3 hr and maximally stimulated T production by untreated, purified Leydig cells was
linear during this time. Moreover, the Leydig cells in the interstitial cell preparations used to determine the effects of in vivo EDS exposure showed an in vitro response to EDS which was similar to the effect obtained using purified Leydig cells in vitro. Leydig cells in interstitial preparations showed a 30% decrease in hCGstimulated T production when incubated with 200 PM EDS for 3 hr, whereas highly purified Leydig cells showed a 36% decrease in T production when exposed to this EDS concentration (data not shown). Thus, the Leydig cells in the interstitial cell preparations were equally well maintained and were useful in determining the response to an in vivo exposure to EDS, and in fact, responded in a manner similar to that of highly purified Leydig cells following in vivo exposure. It might be argued that the Leydig cells obtained in the interstitial cell preparation, and also in the purified prepa-
468
KLINEFELTER.
LASKEY.
AND
ROBERTS
FIG. 6. (Left) A light micrograph showing an interstitial region of a testis from a control animal which contains many Leydig cells. X 1.067. (Right) A similar light micrograph of the interstitial region of a testis 24 hr after a 60 mg/kg dose of EDS (estimated ED50) was administered. Leydig cell clusters are still evident. These Leydig cells possess a moderately dense cytoplasm with profiles representative of mitochondria and a nucleus with patches of dense heterochromatin resembling those in vehicle-treated animals. X 1.067.
ration, are not representative of the Leydig cell population in the testis, but rather represent a select population of Leydig cells. However, Klinefelter ef al. (1987) demonstrated that these highly purified Leydig cells respond similarly to the Leydig cells in the in vitro-perfused testis when LH is withdrawn in viva Furthermore, in this study the Leydig cells in both interstitial and purified preparations responded similarly to both in vitro and in vivo EDS exposures. Within 24 hr following EDS administration in vivo, EDS has been reported to cause morphological degeneration of the Leydig cells in the rat testis (75 mg/kg: Kerr et al. 1985, 1986; 100 mg/kg: Morris et al., 1986). Our results seem to differ from these observations on several accounts. Twenty-four hours following in vivo exposures to EDS, viability as measured
by trypan blue exclusion was high among cells in the interstitial preparations regardless of the dose of EDS given. Moreover, the Leydig cells in these preparations stained intensely for 3pHSD enzyme activity and were obtained in similar numbers at all doses of EDS used in the in vivo experiments. Finally, the Leydig cells in the testes of animals 24 hr following the estimated ED50 dose of EDS were indistinguishable from Leydig cells in control animals at both the light and the electron microscope level. In accordance with the literature, we presume that the morphology of Leydig cells in the testes of rats given higher doses of EDS would have indicated some degenerative changes. However, based on the criteria for Leydig cell viability which we have chosen: 3P-HSD staining, morphological preservation, the ability to respond to KG,
EDS EFFECTS ON LEYDIG
CELLS
469
FIG. 7. (Left) An electron micrograph revealing the Leydig cell ultrastructure in testes of control animals. Numerous profiles of smooth endoplasmic reticulum (S) and mitochondria (M), organelles intimately involved with T biosynthesis, are evident, X 13,095. (Right) The ultrastructure of Leydig cells in testes 24 hr following a 60 mg/kg dose of EDS appear unchanged, X4.9 18.
and trypan blue exclusion, the viability of the Leydig cells does not appear to have been altered by an ED50 dose of EDS in vivo. Perhaps it is reasonable to assume that the cells are in a degenerative state but that none of these measures are sensitive enough to detect this change. We found that the effects of EDS on T production in vitro occurred rapidly yet cell viability, as assessed by the criteria discussed above, was unchanged during the 3-hr exposure. These results are consistent with the study by Morris et ul. (1985) but are in contrast to the study by Rommerts et al. (1988). The latter in vitro study reported that EDS (400 PM) altered the morphology of the Leydig cell. We were unable to identify any consistent morphological change in Leydig cell morphology at either the light or the electron microscope level with a 370 PM (EC50) exposure. We sus-
pect that if Leydig cell viability was diminished in culture as discussed above, the morphological integrity of all cells, control and treated, would be severely affected. Previous in vitro studies which have evaluated the possible sites of EDS action in the pathway of T biosynthesis have yielded contradictory results. For example, studies by Bu’Lock and Jackson (1972, 1975) reported that in vivo exposure to EDS reduced the ability of the Leydig cell to convert pregnenolone to testosterone in vitro, and implied that EDS exposure reduced the activity of the steroidogenic enzymes in the smooth endoplasmic reticulum of the Leydig cell. Rommerts et al. (1985) reported that when Leydig cells were exposed to EDS in vitro, LH-stimulated, but not 22R-hydroxycholesterol-stimulated, T production diminished. This suggests that the lesion induced by EDS is prior to the choles-
470
KLINEFELTER,
LASKEY.
terol side chain cleavage enzyme which converts cholesterol to pregnenolone in the mitochondria, and hence prior to the enzymes in the smooth endoplasmic reticulum. Later, Rommerts et al. (1988) reported that 22R-hydroxycholesterol did not prevent a decrease in pregnenolone production caused by EDS. These results would have confirmed the findings of Bu’Lock and Jackson ( 1972, 1975) that EDS induces a lesion in the smooth endoplasmin reticulum. Unfortunately the untreated Leydig cells cultured in the study by Rommerts et al. (1988) failed to produce steroid (pregnenolone) linearly over time (24 hr), which indicated to us that the Leydig cells were losing steroidogenic function over time in vitro. Again, since EDS affects the Leydig cell in vitro acutely, we felt that a 3 hr incubation was best to monitor the initial effects on T biosynthesis. In the present study, we have shown that the biosynthesis of T, following both in vitro and in vivo exposure to EDS, is compromised between the CAMP activation of protein kinase and the cholesterol side chain cleavage enzyme because maximal stimulation with db-CAMP failed to maintain T production by the Leydig cell in the presence of EDS, while substrate-saturating amounts of 2Oahydroxycholesterol maintained T production. However, our results do not preclude the possibility that EDS competes with the steroidogenie enzymes involved in T biosynthesis when the substrates (HCHOL, PREG) are present in physiological concentrations. We have established that the responsiveness of the Leydig cells to EDS both in vitro and in vivo is a function of the concentration or dose of EDS to which the Leydig cells are exposed. By identifying an EC50 and ED50 for in vitro and in vivo EDS exposure, respectively, we were able to correlate the in vitro and in vivo effects of EDS as a function of both the steroidogenic capacity and morphology of the Leydig cell. We conclude that a 3-hr in vitro EC50 exposure or a 24-hr in vivo ED50 exposure to EDS does not alter Leydig cell viability, but does, however, compromise the function of some aspect of testosterone bio-
AND ROBERTS
synthesis. The biochemical lesion appears to be between the production of CAMP and the action of cholesterol side chain cleavage enzyme. Now that we feel satisfied that the in vitro effects of EDS are similar to the in vivo effects, our goals are twofold. First, we will attempt to establish in vitro whether the action of EDS on Leydig cell steroidogenesis is reversible. Second, we will attempt to determine whether the final event in the cytotoxicity of EDS, namely, Leydig cell death, is mediated through other testicular factors.
REFERENCES BARTLETT, J. M. S., KERR, J. B.. AND SHARPE, R. M. (1986). The effect of selective destruction and regeneration of rat Leydig cells on the intratesticular distribution of T and morphology of the seminiferous epithelium. J. Androl. 7, 240-253. BU’LOCK, D. E., AND JACKSON,C. M. (1972). Suppression of testicular androgen synthesis in the rat by ethylene dimethanesulphonate. Gynecol. Invesf. 2, 305-308. Bu’LocK, D. E.. AND JACKSIN, C. M. ( 1975). Suppression of testicular androgen synthesis in the rat by ethane- 1,2dimethanesulphonate. J Steroid B&hem. 6, 118 l-l 185. CHUBB, C.. AND EWING, L. L. (1979). Steroid secretion by in vitro perfused testes: Testosterone biosynthetic pathways. Amer. J. Physiol. 237, 231-238. EDWARDS, G., Fox, B. W., JACKSON, H.. AND MORRIS, I. D. (1988). Lqvdig Cell Cytoto.xicity qfPutative Male Antifertility Compounds Related to Ethane-1,2-dimethanesulphonate, pp. 185- 19 1. Serono Symp. Publ., Raven Press, New York. JACKSON, H. ( 1973). Chemical methods of male contraception. Amer. Sci. 61, 188-193. KERR, J. B., BARTLETT, J. M. S., DONACHIE, K. (1986). Acute response of testicular interstitial tissue in rats to the cytotoxic drug ethane dimethanesulphonate. Cell Tissue Rex 243, 405-4 14. KERR, J. B., DONACHIE, K., ROMMERTS, F. F. G. (1985). Selective destruction and regeneration of rat Leydig cells in viva. A new method for the study of seminiferous tubular-interstitial tissue interaction. Cell Tissue Res. 242, 145-l 56. KERR, J. B., KNELL, C. M.. ABBOTT, M., AND D~NACHIE, K. ( 1987). Ultrastructural analysis of the effect of ethane dimethanesulphonate on the testis of the rat, guinea pig, hamster, and mouse. Cell Tissue Res. 249,45 l-457. KLINEFELTER. G. R.. LASKEY, J. W., ROBERTS, N. R., SLOTT, V. S., AND SUAREZ,J. D. (1990). Multiple effects of ethane dimethanesulfonate on the epididymis of adult rats. Toxicol. Appl. Pharmacol. 105, 27 l-287.
EDS EFFECTS ON LEYDIG G. R., AND EWING, L. L. (1988). Optimizing testosterone production by purified adult rat Lcydig cells in vitro. in Vitro Cell. Dev. Biol. 24, 545-549. KLINEFELTER, G. R., AND EWING, L. L. (1989). Maintenance of testosterone production by purified adult rat Leydig cells for 3 days in vitro. In Vitro Cell. Dev. Biol. 25,283-288. KLINEFELTER,
KLINEFELTER,
G.
R.,
HALL,
P. F., AND
EWING,
L.
L.
(1987). Effect of luteinizing hormone deprivation in situ on steroidogenesis of rat Leydig cells purified by a multistep procedure. Biol. Reprod. 36,769-783. LASKEY,
J. W.. PHI:LPS,
P. V., LAWS,
S. D., AND
FERRELL,
J. ( 1986). Leydig cell function in vitro following metal cation treatment. Toxicologist 6(l), 86. MORRIS,
I. D., BARDIN,
C. W., AND
MATHER,
J. P. (1985).
Inhibition of LH-stimulated but not basal testosterone secretion of isolated Leydig cells by the cytotoxic ethylene dimethanesulphonate. IRCS Med. Sci. 13, 1052- 1053.
MORRIS,
471
CELLS 1. D.,
PHILLIPS,
D.
M.,
AND
BARDIN,
C. W.
(1986). Ethylene dimethanesulphonate destroys Leydig cells in the rat testis. Endocrinology 118, 709-7 19. ROMMERTS, BRUGE,
F. F. G., J. W., AND
GROOTENHLJIS, J. W., VAN DER MOLEN, H.
HOOGERJ. (1985).
Ethane dimethanesulphonate (EDS) specifically inhibits LH-stimulated steroidogenesis in Leydig cells isolated from mature rats but not in cells from immature rats. Mol. Cell. Endocrinol. 42, 105- Ill. ROMMERTS, BRUGGE,
F. F. G.,
TEERDS,
K.
J., AND
HOOGER-
J. W. (1988). In vitro effects of ethylene-dimethane sulfonate (EDS) on Leydig cells: Inhibition of steroid production and cytotoxic effects are dependent on speciesand age of rat. Mol. Cell. Endocrinol. 55,8794. SAS Institute, Inc. (1985). SAS User’s Guide: Basics, Version 5. SAS Institute Inc., Cary, NC. SATO, T. (1967). A modified method for lead staining. J. Electron Micros. 16, 133- 135.
