Diametric Effects of Bacterial Endotoxin Lipopolysaccharide on ...

8 downloads 170 Views 460KB Size Report
ABSTRACT. Immune activation results in the activation of adrenal steroido- genesis and inhibition of gonadal steroidogenesis. Previous studies indicated that ...
0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society

Vol. 141, No. 11 Printed in U.S.A.

Diametric Effects of Bacterial Endotoxin Lipopolysaccharide on Adrenal and Leydig Cell Steroidogenic Acute Regulatory Protein KAREN HELD HALES, THORSTEN DIEMER, SALIL GINDE, BIRINDER K. SHANKAR, MARETHA ROBERTS, H. BRUCE BOSMANN, DALE B. HALES

AND

Department of Physiology and Biophysics (K.H.H., T.D., S.G., B.K.S., M.R., H.B.B., D.B.H.), University of Illinois at Chicago, Chicago, Illinois 60612-7342; and Department of Urology (T.D.), University Hospital of the Justus-Liebig-University, 35392 Giessen, Germany ABSTRACT Immune activation results in the activation of adrenal steroidogenesis and inhibition of gonadal steroidogenesis. Previous studies indicated that these effects were caused primarily by activation and suppression of the secretion of ACTH and LH, respectively. However, other evidence indicated a direct effect of the immune system on the gonads. In this study, serum testosterone, quantitated by RIA after lipopolysaccharide injection, showed a significant decrease within 2 h. Parallel measurement of serum LH showed no change. There were no differences in LH receptor or cAMP produced in Leydig cells between vehicle- and lipopolysaccharide-injected mice. The 30-kDa form of the steroidogenic acute regulatory (StAR) protein was quantitated, by

A

HALLMARK OF the inflammatory response is activation of the hypothalamic-pituitary-adrenal axis and marked increase in glucocorticoid secretion by the adrenal cortex. Another characteristic of the inflammatory response is a decrease in testicular steroid hormone secretion (for review, see Ref. 1). LH, acting via its intracellular secondmessenger cAMP, regulates testosterone synthesis in the Leydig cell acutely at the level of cholesterol transport to the inner membrane of the mitochondria, as well as having a more long-term influence on the levels of the steroidogenic enzymes (for reviews, see Refs. 2 and 3). Transfer of cholesterol across the inner-mitochondrial space is regulated by, and dependent on, the action of steroidogenic acute regulatory protein (StAR). LH stimulation of the Leydig cell results in the activation of StAR transcription, subsequent translation of StAR protein, and transfer of cholesterol to cholesterol side-chain cleavage cytochrome P450 (P450scc), which converts cholesterol to pregnenolone (4). Pregnenolone diffuses out of the mitochondria to the smooth endoplasmic reticulum, where it is further metabolized via the action of 3␤-hydroxysteroid dehydrogenase ⌬4-⌬5isomerase (3␤-HSD) to progesterone. Progesterone, in turn, is converted, by a two-step process, to androstenedione via the action of 17␣-hydroxylase/C17–20 lyase (P450c17). The Received March 21, 2000. Address all correspondence and requests for reprints to: Dale B. Hales, Department of Physiology and Biophysics, University of Illinois at Chicago, 835 South Wolcott Avenue, Chicago, Illinois 60612-7342. E-mail: [email protected].

Western blot, in Leydig cells and was found to decrease in a timedependent manner. No change in StAR protein messenger RNA (mRNA) was detected by Northern analysis during this time, nor were any changes found in the levels of mRNA for the steroidogenic enzymes P450scc, 3␤-hydroxysteroid dehydrogenase ⌬4-⌬5-isomerase, or P450c17. In the adrenal, StAR protein was increased, as was StAR protein mRNA. No changes were observed in the levels of mRNA for P450scc, 3␤-hydroxysteroid dehydrogenase ⌬4-⌬5-isomerase, or P450c21. Thus, although the mechanisms of regulation differ, changes in the levels of StAR protein are a sensitive indicator of the steroidogenic capacity of these two tissues. (Endocrinology 141: 4000 – 4012, 2000)

conversion of androstenedione to testosterone is catalyzed by 17␤-hydroxysteroid dehydrogenase (17␤-HSD) (for reviews see Refs. 3 and 5). The biosynthesis of corticosteroids in the mouse adrenal is regulated by ACTH and angiotensin II. Analogous to the action of LH on Leydig cells, ACTH, via the production of cAMP, stimulates StAR, P450scc, and 3␤HSD expression and activity. However, the mouse does not express P450c17 in the adrenal cortex and consequently does not generate 17-hydroxysteroids. Thus, progesterone is converted to deoxycorticosterone by 21-hydroxylase cytochrome P450 (P450c21) and to corticosterone by 11␤-hydroxylase cytochrome P450 (P45011␤) or to aldosterone by the further action of aldosterone synthase (P450aldo). Immune activation of the adrenal is thought to occur by cytokine stimulation of CRF production in the median eminence and/or cell bodies in the hypothalamus, which, in turn, stimulates the secretion of ACTH from the pituitary. In parallel, immune activation has been shown to inhibit GnRH and LH secretion, resulting in inhibition of gonadal steroidogenesis (1). However, inflammatory mediators also act directly on the gonad to inhibit steroidogenesis. We have observed direct perturbation of Leydig cell steroidogenesis in three models of immune activation. Intracerebroventricular injection of interleukin (IL)-1␤ in male rats, in addition to decreasing secretion of GnRH and LH, also causes a blunting of the Leydig cell response to human CG (hCG) (6). Experimental sepsis, induced in male rats by cecal slurry, results in a significant decrease in serum testosterone (7). After a single ip injection of the gram negative endotoxin

4000

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

lipopolysaccharide (LPS) into male mice, serum testosterone levels are decreased by 80% within 2 h and are still completely inhibited at 24 h (8). In all three of these models, we have shown that the acute inhibition of testosterone production is correlated to the inhibition of steroidogenic acute regulatory (StAR) protein expression in Leydig cells (6, 8, 9). During LPS endotoxemia, Leydig cell StAR protein, but not messenger RNA (mRNA), is decreased. In contrast, the expression of the steroidogenic enzymes P450scc, 3␤-HSD, and P450c17 did not decrease appreciably until at least 6 – 8 h after LPS (10). The purpose of the present study was to determine whether testosterone inhibition during LPS-induced endotoxemia is correlated to a decrease in serum LH, Leydig cell LH receptor number or affinity, or cAMP production. A further objective was to examine the effects of LPS injection on adrenal steroidogenesis. We find that, in contrast to the inhibitory effects observed in the Leydig cell, adrenal steroidogenesis is markedly elevated. The diametric effects of LPS on adrenal vs. testicular steroidogenesis include a 1.5- to 2-fold increase in StAR protein and mRNA levels in the adrenal within 2 h, concomitant with increased production of adrenal corticosteroids. Materials and Methods Materials [␣-32P]deoxycytidine triphosphate and Na125I were purchased from Amersham Pharmacia Biotech (Arlington Heights, IL). A Random Primed Labeling kit was purchased from Roche Molecular Biochemicals (Indianapolis, IN). Metrizamide was purchased from Accurate Chemical and Scientific Corp. (Westbury, NY); collagenase was purchased from Worthington Biochemical Corp. (Lakewood, NJ). (22R)-hydroxycholesterol, HEPES, BSA (fraction V), bovine insulin, EDTA, and sodium bicarbonate were purchased from Sigma (St. Louis, MO). Medium 199, DME/F12, Waymouth’s MB752/1, penicillin, streptomycin, and phenol were obtained from Life Technologies (Gaithersburg, MD). Mouse P450scc and P450c21 complementary DNA (cDNA) were a gift from Dr. Keith Parker (University of Texas South Western Medical Center, Dallas, TX). Mouse P450c17 and 3␤-HSD type I cDNA were a gift from Dr. Anita H. Payne (Stanford University, Palo Alto, CA). Mouse StAR cDNA was a gift from Dr. Douglas M. Stocco (Texas Tech University, Lubbock, TX).

4001

Steroid RIAs Blood was collected by cardiac puncture. After clotting at room temperature for 1 h, the clot was loosened from the edge of the tube and centrifuged at 1000 ⫻ g for 20 min at 4 C. The serum was removed to a new tube and stored at ⫺20 C before assaying. Culture media were boiled for 5 min and centrifuged at 2,000 ⫻ g for 20 min, at 4 C, before assaying. Testosterone, corticosterone, and progesterone concentrations were determined with Coat-a-Count RIA kits (Diagnostic Products, Los Angeles, CA).

LH RIA LH RIAs were performed by Brigitte G. Mann, as described (12–14).

hCG binding assay Highly purified hCG (batch CR127) was obtained from the National Hormone and Pituitary Program, NIDDK (Rockville, MD) and was radiolabeled with carrier-free Na125I using the chloramine T method, as described by Dufau et al. (15). Briefly, 25 ␮g hCG and 12.5 ␮g chloramine T were added to 1 mCi Na125I, vortexed, and incubated on ice for 30 sec. After termination of the reaction by adding 0.125 mg sodium metabisulfite containing 1.0% KI, the purified iodinated hCG was obtained by gel-filtration on a Sephadex G-50 column that had been equilibrated with 50 mm Tris buffer containing 1 mm EDTA and 0.1% BSA. The specific activity, as determined by self-displacement analysis, was 3 ␮Ci/␮g protein from mouse testis homogenate. After being thawed on ice, cells were dispersed in 0.25 ml PBS containing 0.1% BSA and filtered through three layers of organza. The resulting suspension was diluted to a final vol of 6.5 ml and used for binding studies. Incubation of 300-␮l cell suspensions with 125I-hCG (100,000 cpm in 750 ␮l), in the presence of increasing concentrations of unlabeled hCG, was performed in polystyrene tubes at 27 C. After 18 h, the suspensions were washed twice with cold PBS-BSA and centrifuged at 20,000 ⫻ g for 20 min. After aspiration of the second supernatant, radioactivity was determined by counting for 1 min in a ␥-counter. Specific binding was determined by subtracting the total amount of radioactivity bound from radioactivity present when excess unlabeled hCG was added. Binding data were analyzed by the method of Scatchard (16).

cAMP assay Leydig cells, purified from vehicle- or LPS-injected animals, were extracted immediately after isolation, for cAMP determination, or incubated at 37 C with 0.125 mm IBMX and ⫾ 2 nm hCG in 95% O2-5% CO2 for 3 h with shaking. cAMP was extracted from the cells with 65% ethanol and quantitated using the BioTRak cAMP RIA assay (Amersham Pharmacia Biotech, Arlington Heights, IL), according to the manufacturer’s instructions for the nonacetylation assay.

Animals Mice were housed for at least 1 week in groups of five per cage. They were given food and water ad libitum and maintained on a 14-h light,10-h dark schedule. Adult (60 –70 days old) male outbred pathogen-free CD-1 mice (Charles River Laboratories, Inc. Portage, MI), averaging 33 g, were injected ip with LPS [Escherichia coli strain JS (Re mutant), Sigma] or vehicle alone (PBS); and blood, testes, and adrenals were collected at the times indicated. The mice were procured, maintained, and used in accordance with the Animal Welfare Act and were killed by CO2 asphyxiation before exsanguination.

Isolation of Leydig cells Testes from each treatment group were collagenase-dispersed, and Leydig cells were isolated on Metrizamide gradients, as described previously (11). For the (22R)-hydroxycholesterol experiments, Leydig cells were cultured in serum-free DME/F12 culture medium [a 1:1 mixture of DMEM and Ham’s nutrient mixture F-12 supplemented with 2.2 g/liter sodium bicarbonate, 10 mm HEPES (pH 7.4), 500 ng/ml insulin, 100 IU/ml penicillin, 100 ␮g/ml streptomycin, and 1 mg/ml BSA] and incubated in a humidified atmosphere of 95% air-5% CO2 at 32 C.

Northern analysis RNA was extracted from isolated Leydig cells using RNeasy spin columns (QIAGEN, Chatsworth, CA). RNA was extracted from adrenals by acid guanidine phenol chloroform (17) after grinding the tissue to a powder on dry ice. Northern blots were performed as described (11) and quantitated by phosphorimage analysis (Molecular Dynamics, Inc., Sunnyvale, CA).

Western analysis of StAR Pelleted Leydig cells were resuspended in lysis buffer (0.1% SDS in PBS), and dissected adrenals were ground to a powder in a mortar and pestle on dry ice and resuspended in lysis buffer. Samples were subjected to brief sonication. Protein concentrations were determined by microBCA protein assay (Pierce Chemical Co., Rockford IL). Equal amounts of protein were analyzed as described (10). Blots were incubated with antisera raised in rabbits against a GST-StAR fusion protein. Briefly, the mouse StAR cDNA (18) was subcloned as a SmaI-BamHI fragment using a BamHI linker to convert the blunt end to a compatible site that would preserve the reading frame when inserted into the BamHI site of pGEX-2T (Pharmacia, Piscataway, NJ). The protein was

4002

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

expressed by isopropyl ␤-d-thiogalactopyranoside induction in DH5 cells and solubilized, as described (19), using 1.5% sarkosyl. The fusion protein was bound to glutathione agarose beads (Sigma) in the presence of 4% TritonX-100 and eluted by boiling in 0.75 m HEPES (pH 7.4) with 1% SDS. Before injection into rabbits, the protein was extensively dialyzed against PBS. Western blot detection was performed using ECL (Amersham Pharmacia Biotech, Piscataway, NJ), and the signal was quantitated, after densitometry, using Imagequant software (Molecular Dynamics, Inc., Sunnyvale, CA). 35

S-labeling and immunoprecipitation

Leydig cells were isolated from control or LPS-injected mice and cultured in methionine and cysteine-free medium [DMEM; ICN, Irvine, CA) supplemented with 2.2 g/liter sodium bicarbonate, 10 mm HEPES (pH 7.4), 500 ng/ml insulin, 100 IU/ml penicillin, 100 ␮g/ml streptomycin, and 1 mg/ml BSA] for 30 min. Cells were then metabolically labeled with 250 ␮Ci/ml [35S]-methionine (35S-Translabel, ICN) in methionine and cysteine-free media for 2 h. Lysates were flash-frozen on dry ice, then subjected to immunoprecipitation, as described (20). Lysates were divided, and StAR and P450scc were immunoprecipitated in parallel from equal amounts of protein, subjected to SDS-PAGE, and analyzed by PhosphorImaging. To determine the effect of LPS injection on total protein synthesis, an aliquot of lysate from 35S-labeled Leydig cells from control and LPS-injected mice, was TCA-precipitated onto GF/A glass fiber filters, washed extensively with cold TCA and ethanol, and counted. In parallel, total protein in the sample was quantitated, and results are expressed as cpm/␮g protein.

Statistical analysis Data were presented as means ⫾ sem of three or more independent experiments. For group comparison, one-way ANOVA followed by a Student-Newmann-Keuls multiple-range test were performed using the InStat, version 3.0, statistical software package (GraphPad Software, Inc., San Diego, CA). Differences were considered as significant at P ⬍ 0.05.

Results Effect of LPS injection on serum testosterone, LH and LH receptors, and cAMP

Injection of bacterial endotoxin into mice causes cessation of gonadal steroidogenesis after 2 h (Fig. 1A). Inhibition of LH release has been observed after LPS injection. To determine whether this could account for the rapid inhibition of steroidogenesis in the testes, LH levels were measured at 2 h after animals were injected with LPS. As shown in Fig. 1B, we did not observe any change in serum LH that could account for the decrease in serum testosterone. LH levels also remained unchanged between 15 min and 2 h and at 24 h post LPS injection (data not shown). Rapid loss of LH receptors on the Leydig cells might also account for the inhibition of steroidogenesis. However, there was no significant difference in LH receptor number or binding affinity in Leydig cells from vehicle or LPS-injected animals 2 h after treatment (Fig. 2). The results indicate affinity constants of 1.58 ⫻ 108 m⫺1 for vehicle and 1.2 ⫻ 108 m⫺1 for LPS; r values were 0.969 and 0.860, respectively. Furthermore, as shown in Fig. 3, no difference was observed in the amounts of cAMP in freshly isolated cells from control or LPS-injected mice, 2 h post injection (95 ⫾ 22 and 120 ⫾ 35 fmol/␮g protein, respectively). In addition, there was no difference in basal or hCGstimulated cAMP levels in Leydig cells isolated from control or LPS-injected animals after 3 h incubation with or without hCG (39 ⫾13 vs. 28 ⫾14 fmol/␮g protein for basal, and 195 ⫾

Endo • 2000 Vol. 141 • No. 11

29 vs. 208 ⫾18 fmol/␮g protein for hCG-stimulated, for control and LPS-injected, respectively). Effect of LPS on StAR

LPS causes a time-dependent decrease in the 30-kDa form of StAR. We quantitated 30-kDa StAR protein over several hours after LPS injection. A representative Western blot is shown in Fig. 4. By 2 h, we observed a 30% decrease in the level of the mature 30-kDa StAR protein, a greater than 75% decrease by 4 h, and greater than 90% by 8 h. This loss of the 30-kDa form of StAR indicates that the protein is not accumulating in the mitochondria and could account for the rapid loss of serum testosterone. One possible explanation for this decrease in StAR might be a decrease in StAR mRNA. To address this possibility, Leydig cells were purified from animals, 2 and 24 h after LPS injection; and mRNA levels for StAR, P450scc, 3␤-HSD, and P450c17 were quantitated by Northern hybridization. No changes in the amounts of mRNA for these four proteins were observed 2 h after injection (Fig. 5, A and B), indicating that transcriptional repression of neither StAR nor the three steroidogenic enzymes could account for the rapid decrease in testosterone synthesis. There was no apparent difference in any of the StAR transcripts in control vs. LPS, and all three transcripts were included in the quantitation. At 24 h, we did observe decreases in the mRNAs for the steroidogenic enzymes (Fig. 5C and Ref. 10), which likely contributes to the chronic inhibition of testicular steroidogenesis, but found no change in the level of StAR mRNA. These data suggest that StAR protein synthesis has been uncoupled from StAR transcription. To assess the effect of LPS on StAR protein synthesis, Leydig cells were isolated from mice, 2 h after LPS injection, and incubated with 35S-methionine; and newly synthesized StAR protein was immunoprecipitated. The results, shown in Fig. 6A, indicate that LPS treatment caused a marked inhibition of 35S-incorporation into newly synthesized 30-kDa StAR. However, as shown in Fig. 6B, there was an equivalent amount of label incorporated into 37-kDa StAR in both groups of Leydig cells. To determine whether the inhibition was restricted to StAR, or whether other mitochondrial proteins were affected, 35S-incorporation into newly synthesized P450scc was assessed in parallel. As shown in Fig. 6C, LPS injection had no effect on P450scc de novo synthesis. Moreover, there was no effect on overall protein synthesis, as determined by incorporation of radioactivity into TCAprecipitable protein (Fig. 6D). To confirm that the steroidogenic machinery downstream of cholesterol transport was still intact at 2 h, purified Leydig cells from vehicle- and LPS-injected mice were challenged with cAMP or (22R)-hydroxycholesterol. Testosterone production was measured after 1 h of incubation. As shown in Fig. 7A, the production of testosterone from LPS-treated animals after cAMP treatment was inhibited by more than 60%, compared with vehicle-injected animals, whereas there was no difference in testosterone synthesis when the cells were incubated with (22R)-hydroxycholesterol. However, this was not the case 24 h after LPS injection. Figure 7B shows testosterone production is inhibited in cells from LPSinjected animals after incubation with either cAMP or (22R)-

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

4003

FIG. 1. A, Serum testosterone levels in naı¨ve, vehicle-, and LPS-injected mice. Mice were injected ip with 200 ␮g/ mouse of LPS or PBS vehicle. Naı¨ve mice were not injected. Mice were killed and bled 2 h after treatment. Naı¨ve, n ⫽ 70; vehicle, n ⫽ 31; LPS, n ⫽ 24; a, P ⬍ 0.05 vs. vehicle. B, Serum LH after LPS injection. Control and LPS, 2 h after treatment, n ⫽ 8.

hydroxycholesterol, indicating a block in steroidogenesis downstream of cholesterol transport. This likely reflects a decrease in translation of the steroidogenic enzymes resulting from the decrease in mRNAs at 24 h. Effect of LPS on adrenal steroidogenesis and StAR

It has been well established that LPS causes an increase in adrenal steroidogenesis. Both the gonad and the adrenal use

many of the same proteins and enzymes in their respective steroidogenic pathways; and yet, the outcome of endotoxin administration has diametric effects on these two steroidogenic tissues. We assessed adrenal steroidogenesis after LPS administration by measuring serum corticosterone and progesterone at 2 and 24 h post injection. As shown in Fig. 8A, animals receiving LPS had serum corticosterone levels 40fold greater than naı¨ve animals and 4-fold greater than ve-

4004

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

125

FIG. 2. Scatchard analysis I-hCG binding to Leydig cells from vehicle- and LPS- injected mice. Testicular interstitial cells were isolated 2 h after injection and were incubated with increasing concentrations of 125I-hCG. The amount of cell-associated radioactivity was determined in duplicate after overnight incubation at room temperature. Data are representative of three independent experiments. B/F, Bound/ free hCG.

Endo • 2000 Vol. 141 • No. 11

FIG. 4. Time course of StAR protein after LPS injection. Mice were injected as described in Fig. 1 and killed at the times indicated. Leydig cells were purified and lysed for Western blot analysis. Leydig cells pooled from six animals were used for each time point. IOD, Integrated optical density.

tated mRNAs for StAR, P450scc, 3␤-HSD, and P450c21. Analysis of adrenal mRNAs indicated that, at 2 h after injection, StAR mRNA was increased by 50%, whereas there was no observable difference in mRNAs for the three steroidogenic enzymes (Fig. 9). We then determined whether the increase in StAR mRNA resulted in an increase in StAR protein. A representative Western blot is shown in Fig. 10A. The 30-kDa StAR protein was quantitated by Western blot at 2, 4, 8, and 24 h after LPS injection. As shown in Fig. 10B, StAR protein increased more than 2-fold during the first 8 h after injection and remained high at 24 h. It is likely that the acute activation of adrenal steroidogenesis is mediated at the level of StAR synthesis and that the stimulus is not of sufficient duration to result in the induction of the steroidogenic enzymes. Discussion

FIG. 3. Effect of LPS on cAMP levels. Leydig cells were purified from mice, 2 h after injection with either 200 ␮g LPS or vehicle. Cells were extracted immediately after isolation for cAMP determination (fresh); or cells were incubated in the presence of IBMX without (basal) or with hCG (hCG). cAMP was quantitated by RIA, as described under Materials and Methods, and expressed as fmol cAMP/␮g protein. lps, lps injected; con, vehicle injected.

hicle-injected animals. By 24 h, serum corticosterone in LPSinjected animals had fallen to a level 16-fold greater than naı¨ve animals but were 7-fold greater than vehicle-injected animals. Similarly, progesterone was increased more than 13-fold over naı¨ve animals 2 h after LPS injection (Fig. 8B). By 24 h, this level had decreased to less than 2-fold over naı¨ve and vehicle-injected animals. Thus, adrenal steroidogenesis is acutely activated after LPS injection, returning toward more normal levels by 24 h. As expected, adrenal steroidogenesis in vehicle-injected animals returned to uninduced levels more rapidly. To determine whether transcriptional activation was involved in the response to LPS, we quanti-

In this study, we examined the steroidogenic response of the mouse to LPS. Our data show that there is a rapid inhibition of testosterone synthesis but no change in serum LH levels, Leydig cell LH receptor numbers or binding affinity, or cAMP levels. Furthermore, there is no change in the steady-state levels of Leydig cell StAR mRNA at either 2 or 24 h after injection, although the steroidogenic enzyme mRNAs are markedly decreased 24 h after injection. Instead, we observed a time-dependent decrease in the 30-kDa form of StAR protein in Leydig cells after LPS injection, with a 30% decrease at 2 h, 75% decrease at 4 h, and more than 90% decrease at 8 h. We show that the inhibition of StAR protein expression is posttranscriptional and that there is marked decrease in de novo synthesis of StAR protein. In addition, we established that the inhibition of testosterone production was at the level of cholesterol transport, as demonstrated by the ability of (22R)-hydroxycholesterol to rescue steroidogenesis in Leydig cells from LPS-injected mice after 2 h. In contrast, adrenal steroidogenesis was markedly, but transiently, stimulated after LPS injection. Serum corticosterone and progesterone levels were significantly increased 2 h after LPS injection but decreased toward control levels at

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

4005

FIG. 5. Effect of LPS on Leydig cell mRNA for StAR and steroidogenic enzymes. A, Representative Northern blots of the Leydig cell mRNA for StAR, P450scc, 3␤-HSD, P450c17 isolated from Leydig cells purified from mice 2 h after LPS or vehicle (veh) injection. The Northern blot was sequentially hybridized with probes for the four genes, as well as cyclophilin. Quantitation of Leydig cell mRNA is shown at 2 h (B) or 24 h (C) after LPS injection. The results are expressed as fraction of control (vehicle) after normalizing the integrated densities for each mRNA to that for cyclophilin. Data are presented as mean ⫾ SEM of three independent experiments. SCC, P450scc.

24 h. The mRNA for StAR was markedly increased at 2 h in adrenals from LPS injected mice, but mRNAs for steroidogenic enzymes remained at control levels. Finally, the 30-kDa

form of StAR protein was increased by 2 h, peaking at 8 h, and remained elevated at 24 h. It is now well established that inflammation and activation

4006

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

Endo • 2000 Vol. 141 • No. 11

FIG. 6. Effect of LPS on StAR and P450scc de novo synthesis. To assess StAR protein synthesis, Leydig cells were isolated from mice 2 h after LPS injection, incubated with 35S-methionine, and StAR protein was immunoprecipitated from an equal amount of protein from each sample. A, Duplicate lanes from replicate culture dishes from a representative experiment are shown. The sizes of 37- and 30-kDa StAR were deduced from molecular mass markers and confirmed by comparison to 35S-lalbeing COS-7 cells that were transfected with the CMV-StAR expression vector, and immunoprecipitating overexpressed StAR protein (not shown). B, Quantitation of StAR synthesis. Radioactivity was measured by PhosphorImager and expressed as total integrated density minus local average background. Data represent a minimum of three independent experiments and are presented as mean ⫾ SEM. C, Quantitation of P450scc synthesis was performed as described for StAR. D, Effect of LPS on total protein synthesis. TCA-precipitable radioactivity per ␮g total cellular protein was determined as described under Materials and Methods and plotted as cpm/␮g protein.

of the immune system inhibit male reproductive function [in particular, Leydig cell steroidogenesis (21)]. We have shown previously, in three different experimental paradigms, that immune activation inhibits Leydig cell steroidogenesis and that inhibition of testosterone production is correlated with decreased StAR protein expression (6, 8, 9). Initially, we observed that the decrease in serum testosterone 2 h after LPS injection was accompanied by a decrease in the 30-kDa form of StAR protein. More recently, we have observed a decrease in StAR protein in rat Leydig cells as early as 30 min after ICV injection of IL1␤, which parallels a decrease in serum testosterone. In addition, StAR levels are reduced after induction of septic shock in rats. In the present study, we extend these observations and now report that LPS injection results in a rapid cessation of testosterone production, that there is a time-dependent decrease in the mature 30-kDa form of StAR protein, and that the inhibition of steroid production is at the level of cholesterol transfer, as shown by the (22R)hydroxycholesterol experiment. Finally, the decrease in StAR protein is posttranscriptional, as shown by a lack of change in StAR mRNA but a marked decrease in StAR protein synthesis and expression. Our observations indicate that the inhibition of steroidogenesis is distal to LH receptor activation. As shown here, there is no significant change in serum LH levels after LPS

injection. Because of the pulsatile nature of LH secretion, we did observe a large interanimal variation in LH levels (22, 23). Another consequence of the pulsatile release of LH is the large interanimal variation in serum testosterone levels in mice, requiring a large number of mice to obtain statistically significant data. Further evidence that a decrease in LH is not responsible for inhibition is the observation that steroidogenesis cannot be rescued by treatment with hCG (24 –26). A loss of cAMP and/or protein kinase A (PKA) signaling is also not a likely explanation, because we do not observe a change in cAMP production or a change in StAR mRNA, which has been determined to have a half-life of about 3 h and is regulated by LH (27). We examined StAR mRNA levels ex vivo after cAMP or R22 stimulation of testosterone production, and we observed that cAMP caused an increase in the amount of StAR mRNA in Leydig cells from LPS-injected mice, which indicated that LPS did not prevent the cells from responding to cAMP, suggesting that PKA was not inhibited (data not shown). This observation supports a more distal site of action in the LH signaling pathway than inhibition of PKA by LPS. Moreover, there was no change in Leydig cell LH receptor number or binding after LPS injection at any time point examined. These results are similar to those observed after ICV injection of IL-1␤ in rats. The loss of testicular responsiveness to hCG stimulation was not attribu-

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

4007

FIG. 7. Rescue of steroidogenesis with (22R)-hydroxycholesterol. Leydig cells were isolated from mice injected with LPS, 2 h (A) or 24 h (B) previously, then incubated 1 h with control medium, 1 mM 8-Br-cAMP, or 20 mM (22R)hydroxycholesterol. Testosterone was measured in the media. n ⫽ 4 for all groups; a, P ⬍ 0.05 vs. vehicle/control; b, P ⬍ 0.05 vs. vehicle/cAMP.

table to alteration of Leydig cell LH receptor number or binding affinity (6). We do observe a decrease in steroidogenic enzyme mRNA expression in Leydig cells after 24 h that we have shown is caused by transcriptional repression

by cytokines induced by LPS. That StAR mRNA remains unaffected further suggests that this repression is occurring downstream of the initial signaling pathway for LH. In models of immune-activation that involve central ad-

4008

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

Endo • 2000 Vol. 141 • No. 11

FIG. 8. Serum corticosterone and progesterone after LPS injection. Mice were injected as described in Fig. 1 and killed at 2 and 24 h post injection. A, Corticosterone: naı¨ve, n ⫽ 9; vehicle and LPS (2 h), n ⫽ 28; vehicle (24 h), n ⫽ 24; LPS (24 h), n ⫽ 23. B, Progesterone: naı¨ve, n ⫽ 7; vehicle (2 h), n ⫽ 38; LPS (2 h), n ⫽ 35; vehicle (24 h), n ⫽ 21; LPS (24 h), n ⫽ 19. For both graphs: a, P ⬍ 0.05 vs. naı¨ve; b, P ⫽ 0.05 vs. 2 h vehicle.

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

4009

FIG. 9. Effect of LPS on adrenal mRNA for StAR and steroidogenic enzymes. A, Representative Northern blots of adrenal mRNA for StAR, P450scc, 3␤-HSD, P450c21 isolated from adrenals dissected from mice 2 h after LPS or vehicle injection. The Northern blot was sequentially hybridized with probes for the four genes, as well as cyclophilin. B, Quantitation of adrenal mRNA is shown at 2 h after LPS injection. The results are expressed as fraction of control (vehicle) after normalizing the integrated densities for each mRNA to that for cyclophilin. Data are presented as mean (fraction of control) ⫾ SEM. Adrenals were dissected from individual animals: Vehicle, n ⫽ 2; LPS, n ⫽ 8.

ministration of cytokines or LPS, LH levels are decreased. ICV administration of LPS or IL-1␤ to male rats results in a measurable decrease in serum LH, by 30 min, in both intact or gonadectomized animals (25, 28). This does not hold true for systemic injection (ip) of LPS (29), which does not alter LH, as we have also shown. In contrast, iv administration of tumor necrosis factor (TNF) to rats results in the inhibition of testosterone accompanied by the expected increase in LH (30). Interestingly, we do not observe a rebound in LH levels in the absence of testosterone, suggesting that there is a secondary inhibition of LH release. In this way, our data are reminiscent of the effects of dioxin on testosterone production (31, 32). Thus, the effect of immune-activation on LH is dependent on the particular inflammatory mediator and the manner of its administration. Diametric response: increase in adrenal steroidogenesis

LPS endotoxemia does not result in a global or systemic steroidogenic failure but is restricted to the gonads [i.e. LPS also inhibits ovarian steroidogenesis (33)]. Indeed, activation of adrenal steroidogenesis modulates the deleterious effects of global immune activation. In the present study, we show that LPS causes a rapid and marked increase in adrenal steroidogenesis. Serum corticosterone levels increased by 4-fold within 2 h (compared with vehicle-injected controls) and 40-fold [compared with naı¨ve (uninjected) mice]. The elevation of corticosterone by injection of vehicle alone is

likely caused by the stress of needle penetration. Serum progesterone levels are also markedly elevated within 2 h. Elevated progesterone production by the adrenal has previously been reported (34), in particular, in association with immune activation of the hypothalamic-pituitary-adrenal axis (25). Elevated ACTH levels are also associated with increased progesterone production by the adrenal gland in the female (35). Furthermore, the rise and fall of serum progesterone parallels changes in serum corticosterone, supporting the adrenal source. In contrast, serum testosterone levels decrease rapidly and remain depressed, whereas progesterone rises and falls again. Moreover, because Leydig cell steroid production is inhibited at the level of cholesterol transfer, it stands to reason that the progesterone must be coming from the adrenal. Our data indicate that elevation of adrenal steroid production is caused by increased StAR protein expression. Both ACTH and angiotensin II transcriptionally regulate adrenal expression of StAR (36). An increase in adrenal StAR mRNA, followed by an increase in StAR protein, is observed after injection of animals with ACTH or treatment of adrenal cells in culture with cAMP analogues or angiotensin II (36 –38). We also observed a significant and proportional increase in the level of all three StAR mRNA transcripts in the adrenals within 2 h of LPS injection, supporting the conclusion that transcriptional regulation of StAR underlies the observed increase in steroid production in the adrenals.

4010

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

Endo • 2000 Vol. 141 • No. 11

FIG. 10. Effect of LPS on adrenal StAR protein. A, Representative Western blot of adrenal StAR protein. Mice were injected as described in Fig. 1 and killed at the times indicated. The adrenals were removed and lysed for Western analysis. Each lane contains adrenal protein from a single animal. B, Quantitation of adrenal StAR protein at indicated times. LPS (24 h), n ⫽ 4; all others, n ⫽ 6.

Elevated glucocorticoids are associated with decreased male reproductive function. Glucocorticoids are known to inhibit the hypothalamic-pituitary-gonadal axis at multiple sites, including GnRH and gonadotropin secretion, and to exert direct inhibitory effects on Leydig cells (1). Indeed, we have shown previously that glucocorticoids inhibit P450scc transcription and de novo synthesis in isolated mouse Leydig cells (39). Elevated serum corticosterone, therefore, may contribute to the long-term inhibition of Leydig cell function at the level of steroidogenic enzyme expression (40). It is possible that elevated glucocorticoids are causative in the lack of rebound in serum LH levels after decreased serum testosterone. In castrated males, or males whose testosterone is inhibited pharmacologically, there is a reciprocal and pronounced elevation in LH. The lack of a secondary elevation in LH in the presence of markedly reduced serum testosterone suggests that other factors, i.e. glucocorticoids, may be inhibiting LH secretion. It is unlikely that glucocorticoidmediated inhibition of Leydig cell function contributes to the

immediate cessation of testosterone production, because the effects of glucocorticoids on Leydig cells have been shown to be at the level of transcriptional repression, and our data demonstrate that the immediate inhibition of steroid production was independent of changes in mRNA levels. The present study presents data that demonstrate that serum testosterone levels decrease rapidly after LPS injection. The serum half-life of testosterone in rodents is reported to be approximately 4 –5 min (41, 42). Testicular interstitial macrophages and Leydig cells are intimately associated, suggesting a functional interaction between these cells (for review, see Ref. 43). Our studies have established that macrophage-secreted proinflammatory cytokines, such as IL-1, TNF␣, and IL-6, are potent repressors of LH/hCG or cAMPstimulated testosterone production and steroidogenic enzyme gene expression, both in vitro and in vivo, and have demonstrated that the interstitial testicular macrophages express and produce IL-1 and TNF␣ (10, 26, 44). We have demonstrated that LPS, acting via the elaboration of proin-

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

flammatory cytokines, causes a profound inhibition of steroidogenic enzyme gene expression, analogous to what we observed in vitro when we supplied exogenous recombinant cytokines to primary cultures of Leydig cells. However, testosterone inhibition after LPS injection occurs faster than can be accounted for by the cytokine-mediated inhibition of steroidogenesis (8, 10). Moreover, in vitro treatment of Leydig cells with cytokines or LPS does not reproduce the acute inhibition of StAR protein expression, cholesterol transfer activity, or immediate cessation of testosterone production. LPS is known to activate the respiratory burst and production of reactive oxygen species from testicular macrophages. Treatment of corpora luteal cells or MA-10 tumor Leydig cells, in primary culture, with exogenous reactive oxygen donors (such as hydrogen peroxide) inhibits steroid hormone production by blocking cholesterol transfer (45, 46). Together, these observations suggest that LPS may be acting via the rapid production of reactive oxygen by testicular interstitial macrophages to perturb Leydig cell steroidogenesis. It is possible that the adrenal cortex escapes oxidative stress after LPS injection because of the elevated concentrations of the antioxidants ascorbic acid and ␣-tocopherol, reported to be among the highest of all tissues (Refs. 47 and 48) and references therein). Current studies in our laboratory are examining these possibilities. StAR protein is essential for gonadal and adrenal steroidogenesis. It is a nuclear-encoded protein targeted to the mitochondria by aminoterminal signal peptides. Controversy exists about the mechanism through which StAR facilitates cholesterol transfer to the inner mitochondrial matrix. It is clear, though, that StAR is synthesized as a larger molecular mass protein (37 kDa in the mouse) that is proteolytically processed to the mature 30-kDa form in the mitochondria (for recent reviews, see Refs. 4 and 49). The innermitochondrial 30-kDa form of StAR protein represents the inactive, postfunctional form of the protein. It has a half-life of approximately 2 h (K. Held Hales and D. B. Hales, unpublished observation). The transiently expressed 37-kDa form, thought to have a half-life of minutes, is now recognized to be the active form of StAR. Therefore, changes in the 30-kDa form are an indirect measure of changes in the active form. The function of StAR in steroidogenesis depends on new protein synthesis, yet the degradation of the StAR protein is much slower than its functional inactivation (4, 50, 51). Therefore, that we see a slower time-dependent decrease in 30-kDa StAR, even though steroidogenesis has been rapidly inhibited, is to be expected. Furthermore, mitochondrial transport and processing of active StAR, as well as steroidogenesis, are dependent upon an intact mitochondrial electrochemical gradient (52). The intriguing observation, shown in Fig. 6, A and B, that there is a greater decrease in the accumulation of newly synthesized 30-kDa form of StAR, relative to the 37-kDa form after LPS injection, suggests that StAR processing and/or mitochondrial import, as well as de novo synthesis, may have been inhibited. This observation suggests a perturbation of the mitochondrial electrochemical gradient, which is known to be required for mitochondrial import of StAR (52, 53). The present study demonstrates that a single ip injection of a sublethal dose of LPS causes an almost immediate de-

4011

crease in serum testosterone levels. At 2 h after LPS, the mature 30-kDa form of StAR protein is markedly reduced and continues to decrease over the next several hours. The inhibition of testosterone production was at the level of cholesterol transfer, as shown by the ability of R22-hydroxycholesterol to restore testosterone production. The inhibition was distal to serum LH or Leydig cell LH receptor binding on Leydig cells. In contrast, there was a marked increase in adrenal steroid production, an elevation in glucocorticoid and progesterone serum levels, and concomitant increase in StAR mRNA and protein levels. Thus, our observation of decreased accumulation of 30-kDa StAR in the Leydig cell, and increased accumulation in the adrenal, are a sensitive read-out of changes in the steroidogenic potential of the cell. Acknowledgments The authors thank Dr. Anita Payne for the mouse P450c17 and 3␤HSD type I cDNAs, Dr. Keith Parker for the mouse P450scc and P450c21 cDNAs, and Dr. Douglas Stocco for the mouse StAR cDNA. In addition, we would like to thank Dr. Stocco for his insightful comments, advice, and technical assistance.

References 1. Rivier C, Rivest S 1991 Effect of stress on the activity of the hypothalamicpituitary-gonadal axis: peripheral and central mechanisms. Biol Reprod 45:523–532 2. Stocco DM, Clark BJ 1996 Role of steroidogenic acute regulatory protein (StAR) in steroidogenesis. Biochem Pharmacol 51:197–205 3. Payne AH, O’Shaughnessy PJ 1996 Structure, function, and regulation of steroidogenic enzymes in the Leydig cell. In: Payne AH, Hardy MP, Russell LD (eds) The Leydig Cell. Cache River Press, Vienna, IL, pp 259 –285 4. Stocco DM 1999 Steroidogenic acute regulatory (StAR) protein: what’s new? Bioessays 21:768 –775 5. Saez JM 1994 Leydig cells: endocrine, paracrine, and autocrine regulation. Endocr Rev 15:574 – 626 6. Ogilvie KM, Held Hales K, Roberts ME, Buchanan Hales D, Rivier C 1999 The inhibitory effect of intracerebroventricularly injected interleukin 1beta on testosterone secretion in the rat: role of steroidogenic acute regulatory protein. Biol Reprod 60:527–533 7. Sharma AC, Bosmann HB, Motew SJ, Hales KH, Hales DB, Ferguson JL 1996 Steroid hormone alterations following induction of chronic intraperitoneal sepsis in male rats. Shock 6:150 –154 8. Bosmann HB, Hales KH, Li X, Liu Z, Stocco DM, Hales DB 1996 Acute in vivo inhibition of testosterone by endotoxin parallels loss of steroidogenic acute regulatory (StAR) protein in Leydig cells. Endocrinology 137:4522– 4525 9. Sam 2nd AD, Sharma AC, Lee LY, Hales DB, Law WR, Ferguson JL, Bosmann HB 1999 Sepsis produces depression of testosterone and steroidogenic acute regulatory (StAR) protein. Shock 11:298 –301 10. Xiong Y, Hales DB 1994 Immune-endocrine interactions in the mouse testis: cytokine-mediated inhibition of Leydig cell steroidogenesis. Endocr J 2:223–228 11. Hales DB 1992 Interleukin-1 inhibits Leydig cell steroidogenesis primarily by decreasing 17 alpha-hydroxylase/C17–20 lyase cytochrome P450 expression. Endocrinology 131:2165–2172 12. Beamer WG, Murr SM, Geschwind II 1972 Radioimmunoassay of mouse luteinizing and follicle stimulating hormone. Endocrinology 90:823– 827 13. Kovacic N, Parlow AF 1972 Alterations in serum FSH/LH ratios in relation to the estrous cycle, pseudopregnancy, and gonadectomy in the mouse. Endocrinology 91:910 –915 14. Ryan KD, Schwartz NB 1980 Changes in serum hormone levels associated with male-induced ovulation in group-housed adult female mice. Endocrinology 106:959 –966 15. Dufau ML, Mendelson CR, Catt KJ 1974 A highly sensitive in vitro bioassay for luteinizing hormone and chorionic gonadotropin: testosterone production by dispersed Leydig cells. J Clin Endocrinol Metab 39:610 – 613 16. Scatchard G 1949 The interaction of proteins with small molecules and ions. Ann NY Acad Sci 51:660 – 672 17. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156 –159 18. Clark BJ, Wells J, King SR, Stocco DM 1994 The purification, cloning, and expression of a novel luteinizing hormone-induced mitochondrial protein in MA-10 mouse Leydig cells: characterization of the steroidogenic acute regulatory protein (StAR). J Biol Chem 269:28314 –28322

4012

EFFECTS OF LPS ON ADRENAL AND LEYDIG CELL StAR

19. Frangioni JV, Neel BG 1993 Solubilization and purification of enzymatically active glutathione S-transferase (pGEX) fusion proteins. Anal Biochem 210:179 –187 20. Hales DB, Sha LL, Payne AH 1987 Testosterone inhibits cAMP-induced de novo synthesis of Leydig cell cytochrome P-450(17 alpha) by an androgen receptor-mediated mechanism. J Biol Chem 262:11200 –11206 21. Hales DB, Diemer T, Held Hales K 1999 Review: the role of cytokines in testicular function. Endocrine 10:201–217 22. Coquelin A, Desjardins C 1982 Luteinizing hormone and testosterone secretion in young and old male mice. Am J Physiol 243:E257–E263 23. Sisk CL, Desjardins C 1986 Pulsatile release of luteinizing hormone and testosterone in male ferrets. Endocrinology 119:1195–1203 24. Turnbull AV, Rivier C 1997 Inhibition of gonadotropin-induced testosterone secretion by the intracerebroventricular injection of interleukin-1␤ in the male rat. Endocrinology 138:1008 –1013 25. Turnbull A, Rivier C 1995 Brain-periphery connections: do they play a role in mediating the effect of centrally injected interleukin-1␤ on gonadal function? Neuroimmunomodulation 2:224 –235 26. Hales DB, Xiong Y, Tur-Kaspa I 1992 The role of cytokines in the regulation of Leydig cell P450c17 gene expression. J Steroid Biochem Mol Biol 43:907–914 27. Clark BJ, Combs R, Hales KH, Hales DB, Stocco DM 1997 Inhibition of transcription affects synthesis of steroidogenic acute regulatory protein and steroidogenesis in MA-10 mouse Leydig tumor cells. Endocrinology 138: 4893– 4901 28. Bonavera JJ, Kalra SP, Kalra PS 1993 Mode of action of interleukin-1 suppression of pituitary LH release in castrated male rats. Brain Res 612:1– 8 29. Rivier C, Vale W 1990 Cytokines act within the brain to inhibit LH secretion and ovulation in the rat. Endocrinology 127:849 – 856 30. Mealy K, Robinson B, Millette CF, Majzoub J, Wilmore DW 1990 The testicular effects of tumor necrosis factor. Ann Surg 211:470 – 475 31. Moore RW, Potter CL, Theobald HM, Robinson JA, Peterson RE 1985 Androgenic deficiency in male rats treated with 2,3,7,8-tetrachlorodibenzo-pdioxin. Toxicol Appl Pharmacol 79:99 –111 32. Moore RW, Parsons JA, Bookstaff RC, Peterson RE 1989 Plasma concentrations of pituitary hormones in 2,3,7,8-tetrachlorodibenzo-p-dioxin-treated male rats. J Biochem Toxicol 4:165–172 33. Taylor C, Terranova PF 1996 Lipopolysaccharide inhibits in vitro luteinizing hormone-stimulated rat ovarian granulosa cell estradiol but not progesterone secretion. Biol Reprod 54:1390 –1396 34. Rivier C, Vale W 1981 Interactions of GnRH agonists, pituitary hormones and corticosteroids on progesterone secretion in the male rat. Life Sci 29:1523–1529 35. Rivier C 1994 In the female rat, the ability of centrally injected IL-1b to increase progesterone secretion and rescue the corpora lutea does not depend on the initial, ACTH-dependent rise in adrenal progesterone. Endocr J 2:477– 484 36. Clark BJ, Combs R 1999 Angiotensin II and cyclic adenosine 3⬘,5⬘-monophosphate induce human steroidogenic acute regulatory protein transcription through a common steroidogenic factor-1 element. Endocrinology 140: 4390 – 4398

Endo • 2000 Vol. 141 • No. 11

37. Ariyoshi N, Kim YC, Artemenko I, Bhattacharyya KK, Jefcoate CR 1998 Characterization of the rat Star gene that encodes the predominant 3.5- kilobase pair mRNA. ACTH stimulation of adrenal steroids in vivo precedes elevation of Star mRNA and protein. J Biol Chem 273:7610 –7619 38. Lehoux J-G, Bird IM, Briere N, Martel D, Ducharme L 1997 Influence of dietary sodium restriction on angiotensin II receptors in rat adrenals. Endocrinology 138:5238 –5247 39. Hales DB, Payne AH 1989 Glucocorticoid-mediated repression of P450scc mRNA and de novo synthesis in cultured Leydig cells. Endocrinology 124: 2099 –2104 40. Hardy MP, Ganjam VK 1997 Stress, 11-beta-HSD, and Leydig cell function [review]. J Androl 18:475– 479 41. Gibson-Berry KL, Chase DJ 1990 Continuous and pulsatile infusions of luteinizing hormone have identical effects on steroidogenic capacity and sensitivity of Leydig cells in rats passively immunized against gonadotropinreleasing hormone. Endocrinology 126:3107–3115 42. Ellis GB, Desjardins C 1982 Male rats secrete luteinizing hormone and testosterone episodically. Endocrinology 110:1618 –1627 43. Hutson JC 1994 Testicular macrophages. Int Rev Cytol 149:99 –143 44. Xiong Y, Hales DB 1993 Expression, regulation, and production of tumor necrosis factor-alpha in mouse testicular interstitial macrophages in vitro [published erratum appears in Endocrinology 1994 Mar;134(3):1597]. Endocrinology 133:2568 –2573 45. Behrman HR, Aten RF 1991 Evidence that hydrogen peroxide blocks hormone-sensitive cholesterol transport into mitochondria of rat luteal cells. Endocrinology 128:2958 –2966 46. Stocco DM, Wells J, Clark BJ 1993 The effects of hydrogen peroxide on steroidogenesis in mouse Leydig tumor cells. Endocrinology 133:2827–2832 47. Fand I, McNally WP 1981 Whole-body localization of 14C-tocopheryl acetate in the rat following oral administration. Arch Int Pharmacodyn Ther 250:4 –17 48. Odumosu A 1981 Ascorbic acid and cortisol metabolism in hypovitaminosis C guinea pigs. Int J Vitam Nutr Res 52:175–184 49. Strauss 3rd JF, Kallen CB, Christenson LK, Watari H, Devoto L, Arakane F, Kiriakidou M, Sugawara T 1999 The steroidogenic acute regulatory protein (StAR): a window into the complexities of intracellular cholesterol trafficking. Recent Prog Horm Res 54:369 –394 50. Stocco DM, Clark BJ 1996 Regulation of the acute production of steroids in steroidogenic cells. Endocr Rev 17:221–244 51. Tsujishita Y, Hurley JH 2000 Structure and lipid transport mechanism of a StAR-related domain. Nat Struct Biol 7:408 – 414 52. King SR, Liu Z, Soh J, Eimerl S, Orly J, Stocco DM 1999 Effects of disruption of the mitochondrial electrochemical gradient on steroidogenesis and the steroidogenic acute regulatory (StAR) protein. J Steroid Biochem Mol Biol 69:143–154 53. King SR, Stocco DM 1996 ATP and a mitochondrial electrochemical gradient are required for functional activity of the steroidogenic acute regulatory (StAR) protein in isolated mitochondria. Endocr Res 22:505–514