| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
a Département de pathologie et biologie cellulaire, Faculté de Médecine, Université de Montréal, Montréal, Québec, Canada H3T 1J4
b Department of Cell and Molecular Biology, Lund University, Lund, Sweden
c Department of Biological Sciences, Idaho State University, Pocatello, Idaho 83209
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sertoli cells, testis, cholesterol, hormone-sensitive lipase (HSL)
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The testis is made up of seminiferous tubules separated from each other by the interstitial tissue; cholesterol is contained in both cellular compartments. Although Sertoli cells are known to be capable of synthesizing cholesterol from acetate [7], there is no evidence that tubules make physiologically significant contributions to androgen production. Cholesterol is indispensable to all cells; however, the cholesterol concentration must be regulated to prevent excessive accumulation of the compound within the cell, which would compromise vital cellular functions. We attempt to address the hypothesis that, to preserve the integrity of the functions typical of each compartment of the testis, namely, the synthesis of testosterone in the interstitium and the production of fertile gametes in the tubules, the main enzymes implicated in the regulation of intracellular cholesterol, and specifically HSL, have to act differently in each location of the testis for maintaining adequate cholesterol concentrations. The studies described herein were designed to test this hypothesis by defining the differential HSL protein expression and activity and the localization of the enzyme in each individual compartment of the testis and by assessing the relationship of HSL in each compartment to serum pituitary hormone concentrations for the first time in the mink. Here, we took advantage of the following two facts: 1) the male mink is a seasonal breeder with an annual seasonal reproductive cycle that includes an active spermatogenic phase with and an inactive phase without production of spermatozoa [8, 9] and 2) developmental and seasonal testicular changes are accompanied by variations in serum hormone levels [10]. In addition, the use of the mink model allows the unique advantage of distinguishing factors that regulate testicular cholesterol metabolism during postnatal sexual maturation of the developing testis from those acting during the seasonal annual reproductive cycle in the adult testis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fluorescein isothiocyanate (FITC)-conjugated anti-rabbit IgG, alkaline phosphatase-conjugated anti-mouse IgG, PMSF, soybean trypsin inhibitor, BSA, 1,4 diazabiscyclo-[2.2.2] octane (DABCO), p-nitrophenylbutyrate (PNPB), and a triglycerides (TG) detection kit were obtained from Sigma-Aldrich, (Oakville, ON, Canada). Anti-h vimentin, biotinylated anti-rabbit IgG, collagenase D, leupeptin, aprotinin, antipain, pepstatin and Lumi-lightplus chemiluminescence detection kit were purchased from Boehringer-Mannheim, (Laval, PQ, Canada). Peroxidase-conjugated anti-rabbit IgG, peroxidase-conjugated anti-mouse IgG, and tetramethyl-rhodamine isothiocyanate (TRITC)-conjugated anti-mouse IgG were purchased from Jackson ImmunoRes Lab, (Mississauga, ON, Canada). The NCL-MACRO mouse monoclonal macrophage marker was obtained from Novocastra (New Castle-Upon-Tyne, UK). Peroxidase-conjugated streptavidin was obtained from Molecular Probes, (Eugene, OR). Goat anti-guinea pig IgG was from Antibodies Inc., (Davis, CA). Mouse monoclonal anti-hFSH was from Cedarlane, (Hornby, ON, Canada). The testosterone ELISA kit was purchased from American Laboratory Products, (Windham, NH). Minimum essential medium (MEM) was purchased from Gibco-BRL, (Oakville, ON, Canada). The Bradford protein measurement reagent was from Bio-Rad (Mississauga, ON, Canada). Somnotol was purchased from MCI Pharmaceutical, (Mississauga, ON, Canada).
Animals
Male mink (Mustela vison) were purchased from the R.B.R. Fur Farm (St. Marys, ON, Canada) and from the Visonnière St. Damase (St. Damase, PQ, Canada). Each animal was individually caged and fed a high-protein diet. Animals were kept under natural lighting conditions, and they were allowed food and water ad libitum. Animals were anesthetized by an intraperitoneal injection of 0.9 ml/kg body weight of Somnotol (MTC Pharmaceutical, Cambridge, ON, Canada). The protocol was approved by the Université de Montréal Animal Care Committee. Testes were dissected from neonatal (30–90-day-old), pubertal (115-, 150-, and 200-day-old), and adult (1–2-year-old) mink. Adult mink were killed monthly throughout the annual reproductive cycle. A monthly calendar had been initially proposed by Pelletier [8] to evaluate the dynamics in the germ cell population throughout the annual reproductive cycle in the adult mink. In the present report, the method of Pelletier [8] was used for the identification of the germ cells. The collection of tissues took place during the last week of each month from birth throughout adulthood. Five mink per age group were used except for the onset of puberty, where 10 animals were used. The tissues were not pooled but treated individually.
Isolation of Seminiferous Tubule and Interstitial Cell-Enriched Fractions
Freshly decapsulated testes were placed in cold MEM containing 0.25 mg/ml collagenase D and 0.1 mg/ml soybean trypsin inhibitor in a water bath equipped with a shaking mechanism set at 80 cycles/min and 37°C and processed as described elsewhere [11]. Further separation was achieved by centrifugation at 600 rpm for 20 min in a GS-6R Beckman centrifuge (Beckman, Mississauga, ON, Canada) equipped with a GH 3.8 rotor. The resulting seminiferous tubule-enriched fraction (STf), in the pellet, and the interstitial cell-enriched fraction (ITf), in the supernatant, were washed in PBS (137 mM NaCl, 3 mM KCl, 8 mM Na2HPO4, 1.5 mM KH2PO4, pH 7.4) and, except for HSL activity measurements, homogenized with a glass tissue grinder in PBS containing 2 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM EDTA, 2 µg/ml leupeptin, and 2 µg/ml aprotinine on ice. Purity of the enriched fractions was evaluated by light microscopy.
Epididymal Spermatozoa
Epididymal spermatozoa were obtained by dicing adult mink caudae of epididymides in cold PBS. The resulting cell suspension was filtered through a 74-µm mesh and centrifuged at 2000 rpm for 15 min in a GS-6R Beckman centrifuge to recover the spermatozoa. The gametes were resuspended in 10 mM Tris-HCl, pH 8, containing 1 mM EDTA for 5 min to lyse contaminating epithelial and blood cells [12]. For HSL immunoblot analyses, spermatozoa were diluted 1:1 in cold PBS containing 2 mM PMSF, 1 mM EDTA, 2 µg/ml leupeptin, and 2 µg/ml aprotinine and were sonicated in a Fisher Sonic dismembrator (Fisher, model 300; Fisher, Farmington, NY) at maximum setting during three consecutive intervals of 30 sec each. For HSL immunofluorescence localization, spermatozoa were diluted in PBS, placed on poly-L-lysine-coated coverslips at a density of 1 x 106 cells/ml, and air dried.
Isolation of Testicular Macrophages
Macrophages were isolated from ITf obtained from normal adult mink testes killed in February using the protocol of Hutson [13]. Briefly, the ITf were washed in PBS and resuspended in MEM containing 5% FBS. The ITf suspension was plated on cell culture Petri dishes and incubated in a 95% air/5% CO2 atmosphere at 37°C for 3 min. Next, nonadherent cells were discarded. Adherent cells (predominantly macrophages) were scraped off and cultured in MEM containing 5% FBS for 24 h on poly-L-lysine-coated coverslips.
Antibodies Against HSL
Antibodies against human (h) HSL were generated in rabbit [14]. Rabbit antiserum was affinity purified against the recombinant hHSL coupled to a CNBr-activated Sepharose 4B column (Amersham, Pharmacia Biotech, Solna, Sweden). The affinity-purified antibodies are specific for HSL and do not cross-react with other proteins upon Western blot analysis of tissue extracts. The specificity of HSL antibody was tested using mink adipose tissue as positive control. In addition, controls were made with anti-HSL preadsorbed with human adipose tissue obtained from autopsies. Human tissue was obtained following written consent of the close family. For negative controls, preimmune serum as well as the primary or secondary antibody alone were used.
Electrophoresis and Western Blots
Twenty micrograms of total proteins of each STf and/or ITf sample were denatured in electrophoresis sample buffer and loaded on a 10% polyacrylamide minigel, subjected to electrophoresis, and electrotransferred onto nitrocellulose membranes. Membranes were blocked with 5% skim milk in Tris-buffered saline (TBS: 140 mM NaCl, 50 mM Tris-HCl, pH 7.4) for 60 min at 37°C and then incubated with anti-HSL (1:150 dilution) or monoclonal anti-h vimentin (1:1000 dilution) for 2 h at 37°C. Membranes were thoroughly rinsed and incubated with peroxidase-conjugated anti-rabbit IgG (1:2000 dilution) or peroxidase-conjugated anti-mouse IgG (1:2000 dilution). Antigen-antibody complexes were revealed by enhanced chemiluminescence.
Immunolabeling
Testes were perfusion fixed with PBS, pH 7.4, followed by Bouin fixative, and then were embedded in paraffin. Endogenous peroxidase activity was inhibited with 0.3% hydrogen peroxide (H2O2) as previously described [11]. Labeling was done following the protocol described earlier [11]. Blocking of nonspecific binding was achieved with 0.5% skim milk. Sections were incubated with anti-HSL (1:35 dilution) and next with biotinylated anti-rabbit IgG (1:1000 dilution) followed by peroxidase-conjugated streptavidin (1:200 dilution) [11]. Controls included the use of preimmune or preadsorbed serum and of the second antibody alone.
Fluorescence Microscopy
Spermatozoa were fixed-permeabilized with -20°C methanol for 10 min and processed as described elsewhere [11]. Preparations were incubated with 3% skim milk in PBS to block nonspecific binding. Next, cells were incubated with anti-hHSL (1:6 dilution) for 60 min at 37°C, rinsed with PBS, and incubated with FITC-conjugated anti-rabbit IgG (1:400 dilution).
Testicular macrophage-enriched fractions were fixed with formaldehyde in PBS and permeabilized with acetone as described elsewhere [15]. Blocking of nonspecific labeling was achieved with 3% skim milk in PBS. To test for the presence of HSL in testicular macrophages, the cells were incubated with a mixture of antibodies consisting of rabbit anti-hHSL (1:30 dilution) and the mouse monoclonal macrophage marker NCL-MACRO (1:20 dilution) for 60 min at 37°C. Next, cells were incubated with goat FITC-conjugated anti-rabbit IgG (1:400 dilution) and donkey TRITC-conjugated anti-mouse IgG (1:150 dilution).
Coverslips were mounted in PBS-glycerol (1:1) containing 5% DABCO.
HSL Enzymatic Activity
To estimate HSL activity, a spectrophotometric esterase assay based on the hydrolysis of PNPB was used as described earlier [11, 16]. STf, ITf, spermatozoa, and adipose tissue were homogenized on ice in 0.25 M sucrose, 1 mM EDTA, pH 7.0, 1 mM dithioerythreitol, 20 µg/ml leupeptin, 2 µg/ml antipain, 1 µg/ml pepstatin. Aliquots of 10–20 µg of protein of each sample were incubated with PNPB (diluted in acetonitril) and buffer (0.1 M NaH2PO4, pH 7.25, 0.9% NaCl, 1 mM dithioerythreitol) at 37°C for 10 min. The reaction was stopped by addition of 3.25 ml of methanol:chloroform:heptane (10:9:7). After centrifugation at 800 g for 20 min, solutions were incubated for 3 min at 42°C and the absorbance of the supernatant was measured at 400 nm in an ultraviolet spectrophotometer (UV 160; Shimatzu, Corporation, Japan). The enzymatic activity was expressed in units, one unit being equivalent to the release of 1 µmol of p-nitrophenol/min. All samples were analyzed in triplicate and the HSL activity was related to the total protein concentration of the sample.
Free and Esterified Cholesterol Measurements in STf and ITf
Measurements were performed as described earlier [11]. Briefly, lipids in the samples were extracted in hexane:isopropanol (3:2 v/v) containing the internal standards, stigmasterol and stigmasteryl oleate, at room temperature (RT). Next, lipid extracts were dried under N2, resuspended in chloroform, and separated by thin-layer chromatography (TLC) on silica G250 µm plates with hexane:diethyl ether:NH4OH(c) (80:20:1) as the solvent. The separated lipids were visualized by placing the TLC plates in a chamber of iodine vapors. Areas corresponding to the separated cholesterol and cholesteryl esters were marked and the iodine was allowed to sublime. The areas containing cholesterol, which also contained the stigmasterol internal standard, were scraped into a glass tube containing chloroform:methanol (2:1). The mixture was vortexed, centrifuged at low speed, and the supernatant used for analysis. The cholesteryl ester bands, which also contained the stigmasteryl oleate internal standard, were scraped into a glass tube and hydrolyzed in KOH by heating at 80°C for 30 min. After cooling, 0.5 ml of H2O and 1 ml of hexane were added, the mixture was vortexed, and the phases were separated by low-speed centrifugation. The upper hexane layer, containing the cholesterol and the internal standard stigmasterol, was used for analysis [17]. The cholesterol samples were analyzed by gas-liquid chromatography (Gas-Liquid Chromatographer 6890 series; Hewlett-Packard, Boise, ID) using a Supelco SPB 1701 column (Sigma-Aldrich) as described by Brown et al. [18]. The average elution times for cholesterol and stigmasterol were 17 and 22.5 min, respectively. Quantification was performed by reference to the internal standard stigmasterol. The sensitivity of the method was 4 ng cholesterol.
Triglyceride Determination in STf and ITf
Quantification of TG in STf and ITf was carried out using a kit according to the manufacturer's protocol. TG were hydrolyzed to glycerol and free fatty acids by incubation with a lipase. The glycerol produced was measured by coupled enzyme reactions catalyzed by glycerol kinase, glycerol phosphate oxidase, and peroxidase. The sensitivity of the method was 0.01 g of glycerol/L.
Determination of Serum Levels of LH, FSH, Prolactin, and Testosterone
Bovine ßLH (NIH AFP-8102), rabbit anti-rat LH (NIH IC-3 AFP571292393), hFSH (AFP-7298A), rabbit anti-h ßFSH (NIH IC-3 AFP891891), c-prolactin (AFP2451B), and guinea pig anti-c-prolactin (AFP 1062091GP) were generously donated by Dr. A.F. Parlow, Scientific Director of the National Hormone and Pituitary Program of the National Institute of Diabetes and Digestive and Kidney Diseases.
LH concentrations were measured by double antibody-sandwich ELISA. Bovine ßLH diluted in 0.1% BSA in PBS was used as the standard. Wells of immulon 2 high-binding microtiter plates (Dynex Technology, Chantilly, VA) were coated with 200 µl rabbit anti-rat ßLH (1:1000 dilution in PBS) for 6 h. Blocking of nonspecific binding was performed with a blocking buffer (42 mM Tris base, pH 7.0, 50.65 mM NaCl, 14.61 mM sucrose, 0.19 mM MgCl2, 0.1 mM ZnCl2, 0.25 mM thimerosal, 0.1% BSA, and 1.32 ml/L Tween 20). Two hundred microliters of the standard dilution or serum sample were added. After an overnight incubation at room temperature (RT), anti-bovine ßLH monoclonal mouse antibody (1:200 dilution purchased from Dr. F. Roser, Department of Animal Science, University of California, Davis, CA) was added. Plates were incubated for 2 h at RT and, next, incubated with alkaline phosphatase-conjugated anti-mouse IgG (1:3000 dilution). After a 2-h incubation period, 200 µl of 3 mM p-nitrophenyl phosphate in 0.05 M Na2CO3 and 0.05 mM MgCl2 were added to each well. Plates were incubated for two additional hours at RT. The enzymatic reaction was read at 405 nm using a microtiter plate reader (Beckman, Mississauga, ON, Canada). Dilutions of mink serum were parallel to the ßLH standard curve. All samples and standards were assayed in triplicate within the same experiment. The results are expressed in terms of NIH bovine ßLH standard. The sensitivity of the assay was 1 ng bovine LH/ml. All samples were measured in one assay. The intraassay coefficient of variation was 11% (n = 25).
The FSH concentrations were measured by applying a protocol similar to the one used for LH determination. Rabbit anti-hßFSH (1:1000 dilution) was used for coating the plates and a mouse monoclonal anti-hFSH as the second antibody. The hFSH was used as standard. The results were expressed in terms of NIH hFSH standard. The sensitivity of the assay was 0.5 ng/ml. The intraassay coefficient of variation was 9.5% (n = 30). All samples and standards were assayed in triplicate within the same experiment.
Serum prolactin concentrations were measured by heterologous double-antibody RIA in a single run, as previously described [19]. In brief, serum samples (200 µl) were incubated with 100 µl of a 1:70 000 dilution of the first Ab (guinea-pig anti-c-prolactin) and 50 µl of 125I-labeled c-prolactin for 16–18 h at 4°C. Subsequently, 200 µl of assay buffer containing goat anti-guinea pig IgG at a 1:12 dilution and were added to the tubes, vortexed, and incubated for 3–4 h. The reaction was stopped by addition of 2.0 ml of cold distilled water, followed by centrifugation at 1500 x g for 30 min at 4^def;C. The intraassay coefficient of variation was 4.35%. The sensitivity of the assay was 1.0 ng/ml. All serum samples were analyzed in triplicate.
Testosterone concentration in mink sera was measured by ELISA. The sensitivity of the method was 0.07 ng/ml. All samples were measured in one assay. The intraassay coefficient of variation was 5.2%.
Data and Statistical Analyses
The relative HSL protein content of scanned immunoreactive bands in Western blots was estimated by densitometry (Scion Image Software; Scion Corporation, Frederick, MD). The bands of three independent immunoblots were scanned using a laser scanner (Astra 1200S; Umax Data System, Inc., Hainchu, Taiwan, ROC). Descriptive statistics are presented in the form of figures with means ± SEM. Pearson correlation coefficients were calculated for subsets of variables that were observed in the same animal. For each variable, a multivariate one-way ANOVA was used with age as the factor and the variable as the component of the response vector. Post hoc comparisons (t-tests) between age groups were performed with the Bonferroni (Dunn) method. Where necessary, logarithmic transformations were applied. The significance level of 0.05 was used. All analyses were carried out with SAS software (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Western Blot Analyses
Western blot analyses performed with anti-hHSL revealed two immunoreactive bands of 104 and 108 kDa in the STf (Fig. 1A, STf) and a 90-kDa band in mink adipose tissue (Fig. 1A, Fat). Adsorption of anti-hHSL with human adipose tissue caused a major decrease in the intensity of the immunoreactive bands in mink tissues (Fig. 1B), demonstrating that the anti-hHSL antibody recognizes the enzyme in mink tissues. Western blots performed with preimmune serum were negative (not shown).
|
Both the 104- and 108-kDa HSL immunoreactive bands were present in mink STf and ITf; however, the intensity of the two bands varied during development and the annual reproductive cycle. In addition, the changes detected were different depending on the band and on the testicular compartment under study. In the STf, the 104-kDa band was stronger during adulthood and the active spermatogenic phase, whereas the 108-kDa band was more intense during periods when spermatogenesis was not completed (Fig. 2A). A 104-kDa immunoreactive band was detected in epididymal spermatozoa when membranes were overexposed (Fig. 2B). Because we did not detect expression of the 108-kDa band in spermatozoa, we reasoned that this isoform could be expressed only by Sertoli cells within the seminiferous tubules. We therefore analyzed the levels of the intermediate filament protein vimentin, which is normally expressed by Sertoli cells but not spermatozoa. Vimentin levels in the same STf samples showed a profile of variation that was similar to that of the 108-kDa HSL band during development and during the annual reproductive cycle in the adult (Fig. 2C). In the interstitial compartment, the expression of the 104-kDa protein also increased coincidentally with adulthood and the active spermatogenic phase and in May–June, while the 108-kDa protein was mainly expressed during development and testicular regression (Fig. 2D).
|
HSL Concentration and HSL (Esterase) Activity in STF and ITf
Densitometry analyses of the 104- and 108-kDa HSL immunoreactive bands in the STf and ITf were carried out on six independent immunoblots similar to those shown in Figure 2 (Fig. 3). The 104- and 108-kDa bands were always detected in the STf, whereas only the 104-kDa HSL was detected in the ITf of all age groups studied. The 108-kDa HSL expression in STf and ITf was inversely related to the number of germ cells present within the seminiferous tubules: high during development and testicular regression but low during the active spermatogenic phase (Fig. 3, A and B). In the STf and ITf, the expression of the 108- and the 104-kDa HSL immunoreactive bands together as well as the expression of the 104-kDa alone increased with development, whereas these were lowest by the onset of the next seasonal spermatogenic phase in the adult (Fig. 3).
|
HSL activity measured at 34 and 37°C gave similar values. HSL-specific activity was generally lower in the STf (Fig. 3A) than in the ITf (Fig. 3B). HSL-specific activity increased significantly in both STf and ITf with the completion of spermatogenesis in the adult during the active phase of the annual reproductive cycle (Fig. 3, A and B). HSL-specific activity increased with the levels of the 108- and 104-kDa HSL proteins together and with the levels of the 104-kDa HSL protein alone in STf and ITf. The enzymatic activity decreased more rapidly than the protein levels (Fig. 3, A and B). In addition, HSL was active in epididymal spermatozoa (42 mU/mg protein).
Correlation Between HSL Activity and FC, EC, and Triglyceride Levels in the STf and ITf
To search for a potential correlation between the cholesterol metabolism and HSL activity in the two testicular compartments, FC and EC, were measured in the STf and ITf. In the STf, FC levels were high throughout puberty. In the adult, during the annual reproductive cycle, FC levels decreased from the end of the active spermatogenic phase to the first half of testicular regression (Fig. 4A). During the second half of testicular regression, FC levels increased with the loss of germ cells (Fig. 4A). EC concentrations decreased significantly in STf during development (Fig. 4A). In the adult, EC levels were low during the active spermatogenic phase up to the first half of testicular regression, whereas they increased during the second half (Fig. 4A).
|
In the ITf, FC and EC levels showed similar profiles: high during puberty and low in the adult in February (Fig. 4B). The FC and EC levels were low in ITf during the active spermatogenic phase up to the first half of testicular regression, but they increased during the second half (Fig. 4B).
The FC/EC ratios followed a profile similar to that of HSL activity in STf: both parameters increased in the adult in February (Fig. 4C). The FC/EC ratios were highest in STf during the active spermatogenic phase, but they decreased with the loss of germ cells during testicular regression and increased again with the onset of the next active spermatogenic phase (Fig. 4C). The peak in FC/EC ratios appeared 1 mo later in the ITf than it did in the STf (Fig. 4D). There was a strong positive correlation of FC/EC ratios and HSL activity in the STf (r = 0.85, P < 0.05) but not in the ITf (r = 0.35).
Triglyceride Levels in STf and ITf
The TG content showed little variation in the STf from puberty to adulthood (Fig. 5A). The TG levels reached peak values by the second half of testicular regression (Fig. 5A). The TG profiles were similar in STf and ITf (Fig. 5, A and B). There was a small increase in TG levels concomitantly with a rise in HSL activity in STf and ITf in February, whereas a more dramatic increase in TG levels was accompanied with a decrease in HSL activity in June (Fig. 5, A and B). Correlation of TG levels with HSL activity was found in neither STf (r = 0.18 P < 0.05) nor ITf (r = -0.10).
|
Immunohistochemical Studies
No reaction product was observed in controls performed with either the preimmune serum or the second antibody alone (not shown). Immunohistochemistry studies on pubertal mink testes showed that HSL immunoreactivity was localized to Sertoli cell cytoplasmic processes surrounding germ cells throughout puberty (Fig. 6a). HSL was also detected in elongating spermatids (from step 8 to step 19 spermatid) as well as within the Sertoli cell cytoplasmic processes that surround germ cells in the adult (Fig. 6, b–f). This enzyme was detected in all stages of the cycle of the seminiferous epithelium. HSL was abundant near the base of the Sertoli cells, particularly during stage X, i.e., two stages of the cycle following the release of mature spermatids into the lumen of the tubules in this species (Fig. 6f). Within the interstitial tissue, HSL was localized to the endothelial cells of the capillaries and to some interstitial cells resembling macrophages (Fig. 6d). The presence of the enzyme was confirmed in cultured macrophages obtained from adult mink testes harvested in February using immunofluorescence labeling (Fig. 7, a and b).
|
|
Because the histological changes taking place in the mink testis during puberty have been reported to mimic those that take place in the adult during testicular regression but in reverse sequence [8] and because the changes in HSL immunolocalization taking place during puberty were similar to those encountered in the adult during testicular regression, to avoid redundancy, only the data gleaned during testicular regression will be presented here (Fig. 8, a–d). HSL was localized to Sertoli cells throughout testicular regression (Fig. 8, a–c) from the time of the collapse of the lumen (Fig. 8b) to the progressive disappearance of spermatids (Fig. 8c) up to the onset of the next spermatogenic phase in August (Fig. 8d). The presence of HSL in germ cells was observed for as long as there were elongated spermatids still remaining within the regressed tubules.
|
Epididymal spermatozoa incubated with preimmune serum showed occasional nonspecific labeling in the mitochondrial sheath of the tail (Fig. 9a). HSL was localized principally to the equatorial segment of the head, where it showed a typical pattern (Fig. 9b), and occasionally in the postacrosomal segment.
|
Gonadotrophin, Prolactin, and Testosterone Serum Levels During Puberty and the Annual Reproductive Cycle
Serum LH concentrations were highest by the onset of puberty and decreased thereafter. In the adult, serum LH levels were high during the active spermatogenic phase and they decreased by the onset of the next seasonal reproductive cycle (Fig. 10). Serum FSH levels were high throughout puberty (Fig. 10). During the annual seasonal reproductive cycle, they were high from the active spermatogenic phase to the first half of testicular regression, but they decreased coincidentally with the disappearance of most germ cells during the second half of testicular regression (Fig. 10). Serum prolactin levels were low during puberty (Fig. 10). In the adult, prolactin secretion was low during the active spermatogenic phase; it rose significantly during the first half of testicular regression to decrease sharply from the second half of testicular regression to the beginning of the next recrudescence period (Fig. 10). Serum testosterone levels were low during puberty. During the annual reproductive cycle in the adult, serum testosterone levels were high during the active spermatogenic phase and decreased to basal levels during the rest of the annual reproductive cycle (Fig. 10).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This is the first study of the testicular HSL in a seasonal breeder. Because the measurements were made here every 30 days rather than only sporadically as in most studies that used seasonal breeders, this study provides a more comprehensive and accurate account of the impact of HSL on testicular cholesterol metabolism. In addition, because the present measurements were made in seminiferous tubule- and interstitial tissue-enriched fractions rather than in whole-testis extracts, this study provides new and more precise insights into the role and the dynamics of the enzyme in individual cellular compartments of the testis. Two HSL immunoreactive bands of 104- and 108-kDa molecular masses were detected in mink STf and ITf. However, in the mink adipose tissue, the HSL molecular mass was 90 kDa. It has been shown in other species that both HSL mRNA and protein are larger in testis than in adipose tissue and other HSL-expressing tissues [11, 22]. HSL expression increased during development, reaching maximal levels during adulthood in the STf and ITf in the mink. HSL protein levels [11] and HSL mRNA [23] were also reported to increase in steroidogenic tissues, namely, the adrenal and the testis, while remaining constant in the adipose tissue during development in continual breeders.
In addition, this study shows that the expression of both HSL isoforms varied differently in each testicular compartment during the annual reproductive cycle.
Tubular Compartment
In the STf, the 108-kDa HSL isoform levels were higher during the periods when spermatogonia and spermatocytes were present, whereas the 104-kDa HSL protein levels were maximal when spermatids and spermatozoa were plentiful. The studies with vimentin, a marker present only in Sertoli cells, showed that the concentration of vimentin-containing Sertoli cells in STf samples was lowest during periods when spermatogenesis was complete, although the total number of Sertoli cells per testis remained constant through the annual reproductive cycle regardless of the changes in spermatogenic activity. In addition, the profile of vimentin concentrations recorded during development and the annual reproductive cycle were similar to the variations detected in the expression of the 108-kDa HSL isoform. Therefore, the variations recorded herein in 108-kDa HSL protein levels are concomitant with the variations in the proportion of Sertoli cells relative to the germ cell population, suggesting that the 108-kDa HSL isoform resides chiefly within Sertoli cells. Western blot analyses showed the presence of a 104-kDa HSL immunoreactive band in mink epididymal spermatozoa. Moreover, the present immunohistochemical studies show the localization of HSL to mink spermatids and confirm other immunohistochemical [11, 24] and in situ hybridization studies [25], suggesting together that haploid germ cells contribute most to the content of the 104-kDa isoform in the STf.
The finding that HSL accumulated near the base of the mink Sertoli cells following the release of mature spermatids and the accumulation of residual bodies and the observation that HSL labeling decreased in the tubules coincidentally with the appearance of diplotene spermatocytes (present paper and [11]) suggest an active participation of Sertoli cells in the removal of residual body-borne lipids and a particular chronology in the action of the enzyme during spermatogenesis. Distinct HSL isoforms could be synthesized within the tubules in response to specific phases of the germ cells' development, favoring meiosis at one time and, at other times, the shaping of the acrosome or the acquisition by the spermatozoa of cholesterol levels that are consistent with fertility. The observation that HSL localizes to the equatorial segment of mink epididymal spermatozoa and the finding that the enzyme is active in the male gamete add support to the notion that the enzyme plays a role in spermatozoa-oocyte interaction. The acquisition of motility and fertility has been associated with changes in cholesterol content of the male gametes during the epididymal transit [26] and capacitation [27], and inadequate levels of cholesterol have been reported in infertile human spermatozoa [28].
The activity of HSL in the STf varied during development and the annual reproductive cycle. The finding here that, in the STf, FC and EC levels were inversely related to the number of germ cells hints to a close relationship between the metabolism of cholesterol and the spermatogenic activity. In addition, the strong positive correlation of HSL activity with FC/EC ratios but not with TG in STf suggests a distinct substrate preference for testicular HSL and, furthermore, that HSL may be the only cholesterol esterase in the testis. The results are consistent with the report that HSL -/- knockout mice possess no cholesterol esterase activity but showed residual triglyceride lipase activity in the testis [2]. The present finding that FC/EC ratios in tubules were roughly twice those in the interstitial tissue together with the observation that HSL-specific activity was about 2-fold higher in the latter than in the former implies a regulation of the enzyme expression and activity that is typical in each compartment of the testis. The observation that pathological, experimental, and seasonal arrest of spermatogenesis were all reported to be generally accompanied with an increase in lipid droplets [29–31] reinforces the notion of a close relationship between lipid metabolism and germ cell development.
Interstitial Compartment
The present immunohistochemical and fluorescence microscopy studies showed that, within the interstitial tissue, HSL localized to macrophages and endothelial cells of the blood vessels but not to Leydig cells. The lack of staining for HSL in Leydig cells is in agreement with the report that Leydig cells do not contain HSL mRNA [32]. The observation that HSL was immunolocalized to mink testicular macrophages within the interstitial tissue is in agreement with the reports of HSL protein and mRNA in macrophages from murine origin and in macrophage-like cell lines [33–35] and with the observation of HSL mRNA in macrophages from human origin [36]. However, these findings contrast with those that showed undetectable HSL mRNA in mature human monocyte-derived macrophages [37, 38]. The number of macrophages/unit area of testicular interstitium has been shown to increase significantly during puberty [39, 40], but this expansion is due largely to local proliferation rather than originating from monocytes in the peripheral circulation [41]. Thus, the variance in the source of macrophages and in the differentiation state may account for the discrepancies mentioned above in the detection of HSL in macrophages. Furthermore, that the highest total cholesterol levels corresponded to low HSL expression and activity in the mink testicular interstitial tissue is in agreement with the report that HSL protein expression and enzymatic activity are down-regulated by accumulation of cholesterol esters in the mouse macrophage cell line J774.2 [42]. Macrophages constitute a relatively important proportion of the cells in the interstitial tissue in the adult testis [43, 44]. There are abundant direct contacts of macrophages with Leydig cells [45]. Experimental or genetic elimination of macrophages from the testis causes a decrease in testosterone production and a reduction of fertility [46, 47]. Testicular macrophages possess 25-hydroxylase mRNA and convert 14C-cholesterol to 14C-25-hydroxycholesterol [48]. 25-Hydroxycholesterol is used for testosterone synthesis when administered to the testis of macrophage-depleted animals [49]. Testicular macrophages synthesize the 25-hydroxycholesterol that is converted to testosterone by Leydig cells [50]. The present work shows that serum testosterone concentrations correlate with increased HSL activity and protein levels in the ITf. Taken together, these observations provide strong support to the notion of close cooperation between Leydig cells and macrophages in the synthesis of steroid hormones in the testis. Macrophage HSL will free cholesterol from esterified cholesterol to be converted into 25-hydroxycholesterol taken up by Leydig cells to synthesize testosterone.
In addition, testicular macrophages have been suggested to help remove lipofuscin from Leydig cells in pinnipeds [51] and bats [52] during testicular regression in preparation for a recrudescence of steroidogenic activity, and this could explain the increase in interstitial tissue HSL protein and enzymatic activity during May–June. Interstitial macrophages may be part of a system that would be involved in the recycling of sterols in the interstitial tissue, whereas in the tubules, recycling could be ensured by Sertoli cells. In the interstitium, cholesterol has been localized in Leydig cells and macrophages [53] and the contribution of 25-hydroxycholesterol to Leydig cells by macrophages has been shown [48]. In the tubules, cholesterol has been localized to germ cells [54] and to lysosomes in Sertoli cells [11]; moreover, transfer of sterols from germ cells to Sertoli cells has been documented [53].
Relationship of Testicular HSL to Serum Pituitary Hormone Concentrations
HSL and cholesterol ester hydrolase have been shown to be identical proteins hydrolyzing cholesterol esters and triglycerides, depending on the tissue [55]. In the adipose tissue, lipolytic and antilipolytic hormones reportedly regulate HSL activity via reversible HSL protein phosphorylation mediated by the cAMP-dependent protein kinase (PKA) [56]. Activation of HSL by cAMP-dependent phosphorylation leads to the translocation of the enzyme from the cytoplasm to the surface of the lipid storage droplets in adipocytes [57, 58]. FSH regulates steroid synthesis in Sertoli cells via cAMP and PKA [59]. FSH could contribute to the regulation of cholesterol levels within the tubules via cAMP-mediated HSL phosphorylation and thus to the activation of the enzyme in Sertoli cells. The serum FSH profiles provided herein, which are consistent with those reported elsewhere [10, 60, 61], show that high serum FSH levels were concomitant with high HSL activity in the STf, suggesting a possible relationship between the two parameters. The report that cholesterol hydrolase can be restored to normal levels following FSH injection in hypophysectomized rats [62] provides additional support to this notion. The cAMP-dependent transduction pathway may also be involved in the modulation of HSL protein expression in the tubules. The expression of the cAMP-responsive element-binding protein and the cAMP-responsive element modulator in germ cells and Sertoli cells is correlated with the fluctuation in cAMP signaling induced by FSH [63]. Increased expression of the transcription factor may boost HSL expression in Sertoli cells. This hypothesis is consistent with our finding of increased HSL protein levels in the STf during the active spermatogenic phase, which is characterized by high serum FSH concentrations. LH regulates steroid synthesis in Leydig cells also via cAMP-PKA [59]. However, the present data show a relationship between LH serum levels and HSL activity in the ITf that is not clear-cut. This could be due to the fact that LH acts on Leydig cells whereas interstitial HSL is localized to macrophages. In the present study, HSL activity in ITf was inversely correlated with serum prolactin levels. Macrophages possess prolactin receptors [64]. Prolactin has been shown to modulate proliferation and phagocytic activity in macrophages [65, 66]. The increase in prolactin secretion in spring may cause macrophages to reduce HSL protein expression and activity in the interstitium, and this in turn may be one of the causes that would decrease testosterone synthesis by Leydig cells. The report that prolactin induces accumulation of cholesterol esters in response to LH stimulation [67] and the observation that prolactin inhibition of testosterone production involves interstitial cells other than Leydig cells in the rat [68] are in agreement with this hypothesis.
Summary and Conclusion
The finding of a differential HSL protein expression and enzymatic activity in the interstitial tissue and seminiferous tubules entails distinct cholesterol regulatory mechanisms in individual compartments of the testis. This assumption is further supported by the finding that the peak in FC/EC ratios appeared 1 mo later in the interstitial tissue than it did in the tubules. The finding of a strong correlation of the FC/EC ratios with HSL activity in the tubules adds support to the notion of a distinctive substrate preference for testicular HSL and to the concept that HSL may be the only cholesterol esterase in this location.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: R.-Marc Pelletier, Département de pathologie et biologie cellulaire, Faculté de Médecine, Université de Montréal, 2900 Édouard-Montpetit, Montréal, QC, Canada H3T 1J4. FAX: 514 485 7932; e-mail: marc.pelletier{at}umontreal.ca
3 M.L.V. and R.M.P. made equal contributions to these studies, and the order of their names is arbitrary
Received: 12 June 2002.
First decision: 13 July 2002.
Accepted: 7 August 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
on testis cholesterol and on plasma testosterone in mice. Prostaglandins 1973 3:97-104[CrossRef][Medline]
—). The values shown are mean ± SEM of six independent experiments for the assessment of HSL protein levels and of three independent experiments for the HSL activity. A) The variations recorded in the 104- and 108-kDa HSL isoform levels were not significantly different from each other in the STf. B) In the ITf, the levels of the 108-kDa isoform were significantly higher in the 200-day-olds than in the 115-day-olds and in the adult mink of August (**P < 0.05). In addition, the levels of 104 + 108 kDa together and of the 104 kDa were significantly higher in the ITf of the 200 day olds and in the adult mink killed in February, March, May, and June than in ITf of the 115 day olds and of the adult mink of August (*P < 0.05). The panels A and B show the HSL activity assayed at 37°C. The HSL activity was significantly higher in February than in the 115-, 150-, and 200-day-olds and in the adults in April, May, and June in the STf (+P < 0.05). In the ITf, HSL activity was significantly higher in February than in the 115-day-olds and in the adult mink in April, May, June, and August (+P < 0.05)
–) and EC (—
—) are expressed as mg cholesterol/g protein. The values shown are mean ± SEM of three independent experiments. A) FC levels were high in STf in the 115-, 150-, and 200-day-olds and in the adults during February, June, and August (*P < 0.05 adults of March, April, and May versus the 115-day-olds and the adult mink of June and August). EC levels were high in the 115 day olds and in the adults of June and August but were significantly decreased during adulthood in February, March, April, and May (+P < 0.05 in the adult mink of February, March, and April versus the 115-day-olds and the adults in June and August). B) In the ITf, FC and EC levels were high in the 115-, 150-, and 200-day-olds and in the adults in June and August. Only the EC was significantly decreased in the ITf (+P < 0.05, adult mink in February, March, and April versus the 115-day-olds and the adult mink in June and August). C) FC/EC ratios (––) and HSL activity (—
–) recorded in STf (A) and ITf (B) in the 115-, 150-, and 200-day-old mink were not significantly different from those recorded. The values shown are the mean ± SEM of three independent experiments
—), FSH (—
—), prolactin (—