| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institut de Neurobiologie de la Méditerranée (J.F., S.K.), Institut National de la Santé et de la Recherche Médicale Unité 29, Parc Scientifique de Luminy, Marseille, F-13273, France; Laboratoire de Physiologie Neurovégétative (C.C., E.M.), Unité Mixte de Recherche 6153 Centre National de la Recherche Scientifique/1147 Institut National de la Recherche Agronomique/Université Paul Cézanne-Aix-Marseille III, Faculté Saint-Jérôme, 13397 Marseille, France; and Institut Fédératif de Recherche Jean Roche (I.G.), Faculté de Médecine Secteur Nord, F-13916 Marseille, France
Address all correspondence and requests for reprints to: Dr. S. Krantic, Institut de Neurobiologie de la Méditerranée (INMED), Institut National de la Santé et de la Recherche Médicale Unité 29, Parc Scientifique de Luminy–BP13, F-13273 Marseille, Cedex 09, France. E-mail: krantic{at}inmed.univ-mrs.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Three well-defined stages of Leydig cell development are observed during the prepubertal period in rat testis. At birth, mesenchymal stem cells proliferate and differentiate into progenitor Leydig cells (PLCs). The differentiating PLCs are recognizable by the expression of Leydig cell markers such as 3ß-hydroxysteroid dehydrogenase (3ß-HSD) (2). Proliferative capacity of mesenchymal stem cells is high until d 14 after birth and then drops sharply between d 14–21 (3). The newly differentiated PLCs repopulate the intertubular space (4) and further differentiate into immature Leydig cells (ILCs) characterized by a preferential secretion of androstane-3
,17ß-diol (5). ILCs undergo an additional round of proliferation and then progressively lose their proliferative capacity and begin the terminal differentiation into nondividing adult Leydig cells (ALCs). The relaxin-like factor (RLF) is considered a good marker for fully differentiated ALC because it is exclusively expressed by this cell type in the adult testis, but it is only weakly expressed in immature testis (6, 7). The terminal differentiation of ALC is achieved between d 55–90 after birth and is associated with a shift from predominant androstane-3,17ß-diol to testosterone secretion (2). Different hormones (e.g. LH, androgens, estrogens, FSH, anti-Müllerian hormone, and thyroid hormone) and growth factors (platelet-derived growth factor, TGF- and -ß, and IGF-I) have been involved in the regulation of the balance between Leydig cell proliferation and differentiation (8, 9).
Mammalian cell proliferation is controlled by the nuclear cell cycle engine. This engine is composed of cyclins and associated cyclin-dependent kinases, and its principal role is to regulate progression and transition through different cell cycle phases. The cyclin A family is involved in both G1/S and G2/M transitions, whereas different types of cyclin B are required for initiation and progression through the M phase of the cell cycle (10). Four mammalian G1-phase cyclins have been identified to date: cyclins D1, D2, and D3 (D-type cyclins) and cyclin E. Extracellular signals impinge on the cell cycle engine mainly in the G1 phase of the cell cycle during a limited time window. As a consequence, growth-stimulatory (mitogen) and growth-inhibitory signals received from the environment are integrated at the level of cyclin D expression. Indeed, induction of cyclin D expression is highly dependent on the presence of growth factors in the extracellular environment and declines rapidly after their withdrawal. D-type cyclins are therefore considered as growth factor sensors. This is in contrast to the expression of E-, A-, and B-type cyclins, which is believed to be growth factor independent (11). The underlying mechanisms of such independence involve, at least in the case of cyclin A and cyclin E, the induction of their promoters by a transcription factor (E2F1) that is de-repressed by cyclin D-dependent kinase activities (12).
Among the known G1/S cyclins, type A2 (13), G1 (13), D1 (14), D3 (15, 16), and E (14) have been identified in rodent testis and implicated in the control of Leydig cell proliferation. More specifically, it has been proposed that all of them (except cyclin E, which was studied only in 1-wk-old and adult rat testis) (14) play a role in the regulation of the proliferative capacity of prepubertal Leydig cells at the transition from PLC to more differentiated stages (13, 14, 15, 16). However, the precise nature of the extracellular signals responsible for the observed alteration in cyclin expression in prepubertal Leydig cells remains largely unknown.
Leptin is a peripheral hormone principally produced by adipose tissue and involved in the regulation of body weight, food ingestion, body temperature, and metabolic rate (17, 18, 19). At the cellular level, leptin controls the proliferation of numerous cell types. For example, leptin inhibits proliferation of pituitary cells (20) and stimulates that of lymphocytes (21), epithelial cells (22), and hematopoietic cells (23).
In addition to these actions, leptin has a crucial role in both female and male reproductive functions (24, 25, 26). Leptin-deficient male mice are sterile, and their fertility can be restored by exogenous leptin (24, 27). The presence of the full-length, signaling-competent form of leptin receptor, Ob/Rb (28) in rodent testis (29, 30) is consistent with leptin involvement in reproduction. Indeed, it has been shown that leptin inhibits testosterone secretion from adult rat testis (31, 32), thus pointing to a direct targeting of adult Leydig cells by this hormone. By contrast, there are currently no data to suggest the possible involvement of leptin in the control of mammalian Leydig cell development. However, although there is still a controversy on the relevance of a transient leptin peak in plasma for the onset of puberty in different mammalian species (33), the occurrence of such a peak has been demonstrated in plasma of prepubertal rats (34). These latter studies suggest that leptin might play a role in the regulation of prepubertal Leydig cell proliferation at the onset of their differentiation from PLC to ILC.
In the present work, we tested this hypothesis by analyzing leptin actions on G1/S-phase cyclin expression in prepubertal Leydig cells from 21-d-old rat testis. Leptin actions on the proliferative capacity of Leydig cells were monitored by parallel assessment of 1) Leydig cell number; 2) expression of the proliferation marker Ki67; 3) bromodeoxyuridine (BrdU) incorporation. We show that leptin decreases cyclin A2 and increases cyclin D1 expression and inhibits prepubertal Leydig cell proliferation in a timely ordered manner. These alterations were further correlated with the expression of Leydig-cell-specific differentiation markers such as RLF and 3ß-HSD.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For immunohistochemistry, we used rats of various ages (10-, 21-, and 35-d-old animals; n = 3 for each age group) corresponding to infantile, limit between infantile/prepubertal, and limit between pre-/peripubertal developmental stages, respectively. The animals were transcardially perfused with ice-cold 0.1 M sodium phosphate-buffered solution of 4% paraformaldehyde (pH 7.4), through a gauge-appropriate needle connected to a peristaltic pump, for 5 min at 50 ml/min. After overnight postfixation in the same solution at 4 C, cryoprotection by immersion 48 h in 30% sucrose solution at 4 C, and snap-freezing in liquid isopentane at –40 C, testes were sectioned in a cryostat into serial 20-µm-thick sections that were mounted on positively charged glass slides (SuperFrost Plus; Menzel-Glaser, Braunschweig, Germany), dried overnight at room temperature, and kept frozen at –80 C until use.
Chemicals
DMEM-Ham’s F-12 medium and Maloney murine leukemia virus reverse transcriptase were obtained from Invitrogen (Merelbeke, Belgium). The kit RNeasy mini was from QIAGEN (Courtaboeuf, France). Recombinant mouse leptin was obtained from France Biochem (Meudon, France). All real-time PCR supplies were purchased from PE Applied Biosystems (Courtaboeuf, France). All other chemicals were purchased from Sigma (L’Isle d’Abeau, France).
Antibodies
Goat polyclonal anti-Ob/Rb and rabbit polyclonal anti-cyclin D2, D3, and A2 antibodies were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Anti-cyclin D1 antibody used is SP4 clone corresponding to the C terminus of the human cyclin D1, which cross-reacts with rat cyclin D1 (Neo Marker Inc., Fremont, CA). Polyclonal ß-actin antibody was from Sigma, rat polyclonal anti-RLF from Phoenix Pharmaceuticals Inc. (Belmont, CA), rat monoclonal anti-BrdU antibody from Harlan Sera Laboratories (Loughborough, UK), mouse monoclonal anti-Ki67 antibody from BD Biosciences PharMingen (San Diego, CA), and rabbit polyclonal anti-HSD antibody from Abnova Corp. (Taipei City, Taiwan).
Cell isolation and culture
Leydig cells were isolated from testes of 21-d-old Sprague Dawley rats (Iffa-Credo) by collagenase dispersion according to the previously published procedures to obtain the cell populations containing more than 90% of Leydig cells as assessed by staining for 3ß-HSD (35, 36). Briefly, decapsulated testes were incubated for 20 min in culture medium containing 0.5 mg collagenase/ml. Seminiferous tubules were separated from Leydig cells by gravity sedimentation. Supernatants containing Leydig cells were collected, and collagenase was removed by centrifugation (200 x g for 10 min at 4 C). The resulting pellets were resuspended, plated in 24-well plates at densities of 0.25 x 106 cells per well, 5 x 106 cells per dish, or 15 x 106 cells per dish for cell counting/immunocytochemistry, RT-PCR, and Western blot experiments, respectively. They were cultured in DMEM-Ham’s F-12 (1:1 vol/vol) medium containing 1.2 mg/ml sodium bicarbonate, 15 mM HEPES, 20 µg/ml gentamicin, 5 µg/ml transferrin, 10 µg/ml -tocopherol, 100 IU penicillin, and 0.05 mg/ml streptomycin at 37 C in a humidified atmosphere of 5% CO2 and 95% air for 1 h. After that time, Leydig cells were well attached to the surface of the dish, whereas contaminating Sertoli and germ cells were loosely attached. The contaminating cells were removed by washing (three times), and the purified Leydig cells were cultured for 2 d before starting the experiments.
For immunocytochemistry of BrdU or Ki67, Leydig cells were grown on glass coverslips (inserted at the bottom of the 24-well dishes) for different time periods in the presence of 100 nM leptin or left untreated to serve as a control. For BrdU incorporation assay, 30 µM BrdU was added for the last 24 h of culture (37); at the end of the treatment period, cells were rinsed in PBS, fixed in 4% paraformaldehyde in PBS for 15 min, rinsed, and stored in PBS at 4 C until labeling.
RT-PCR analysis of Ob/Rb and G1/S-phase cyclin expression
Leydig cells were incubated with 100 nM leptin or left untreated for the indicated time periods. This leptin concentration was chosen based on previous work using an in vitro experimental paradigm to study leptin actions on rat testicular slices (38). Moreover, this concentration corresponds to the leptin concentration that has been demonstrated as efficient in functional regulation of hormone secretion in another endocrine gland (i.e. pituitary) of male rats (39). At the end of the different treatments, cells were washed with PBS, and total RNAs were extracted using the kit RNeasy mini according to the manufacturer’s instructions.
For RT-PCR analysis, total RNAs (3 µg Ob/Rb and 1 µg cyclin D1, D2, D3, and A2 and RFL) were reverse-transcribed into cDNA. The sequences of the chosen primers for the analysis of Ob/Rb receptor RNA expression are given in Table 1
, whereas the PCR conditions used have been reported previously (32).
|
) selected for the analysis of either a given gene of interest or ß-actin (used as an internal standard) were used at 300 pM concentration. The analysis of the fragments amplified in the relevant real-time PCR by electrophoresis on 2% agarose and ethidium bromide staining demonstrated that in the chosen experimental conditions, a single amplification product of the expected size (80, 88, 84, 109, 85, and 148 bp) was obtained for type D1, D2, D3, and A2 cyclins, RLF, and ß-actin, respectively (data not shown). The two initial steps of real-time PCR performed in an ABI Prism 7700 sequence analyzer (PE Applied Biosystems) consisted of heating to 50 C for 2 min and then to 95 C for 10 min to activate the AmpliTaq Gold polymerase. They were followed by 40 iterations of two-step PCR: denaturation (95 C for 20 sec) and annealing/extension (60 C for 30 sec).
To perform a relative quantitation between control and leptin-treated samples, we normalized the amount of the amplification product corresponding to the gene of interest over an internal standard (ß-actin) and then expressed this ratio as a function of calibrator by using calculating 2–
Ct, where Ct = threshold cycle (Ct) (gene of interest) – Ct (ß-actin) and Ct = Ct (control or leptin-treated samples) – Ct (time point zero) for each time point, knowing that calibrator refers to the ratio of the amount of the fragment corresponding to the gene of interest to the amount of the fragment corresponding to ß-actin obtained for the time point zero (i.e. beginning of the experiment). This ratio was arbitrarily fixed to 1.
The above equations were applied only after the identical amplification efficiencies had been demonstrated for ß-actin primers and each of the chosen pairs of primers for cyclin detection. The efficiencies of amplification for each couple of the chosen target gene primers and ß-actin were determined by plotting the amount of cDNA input as a function of Ct (corresponding to the ratio of a given target gene to the ß-actin amplification product amounts). The absolute slope value of the resulting plots were less than 0.1 in all cases (data not shown), thus demonstrating the equal efficiencies of amplification from cDNA templates.
Western blot analysis of G1/S-phase cyclins
For small interference (si)RNA experiments, Leydig cells were transfected with either 80 pmol/well (six-well dishes) of cyclin D1 siRNA or equivalent amount of control siRNA (both from Santa Cruz Biotechnology) by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Leptin was added 4 h after transfection (leptin-treated group) or not (controls) to both control and cyclin D1 siRNA-transfected groups, and cells were incubated (37 C, 5% CO2) for a total of 48 h. At the end of transfections, control and leptin-treated Leydig cells were washed twice in PBS and then scraped off petri dishes in 50 mM Tris-HCl buffer (pH 7.5) containing 140 mM NaCl, 1 mM EDTA, 0.1 mg/ml soybean trypsin inhibitor, and 0.1 mM phenylmethylsulfonyl fluoride in the presence of 1.5% CHAPS and 0.5 mM sodium orthovanadate. The mixture was incubated for 30 min at 4 C and then centrifuged at 13,000 x g for 20 min to obtain the whole-cell protein extracts. Aliquots of the cleared supernatant were stored at –20 C until use.
For Western blot analysis, solubilized proteins (50 µg for cyclin D1, D3, and A2; 80 µg for D2 cyclins; and 30 µg for HSD) were resolved through 10% SDS-PAGE and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA). The membranes were saturated with 5% dried milk in PBS solution containing 0.05% Tween 20 (pH 7.6) and then coimmunoblotted overnight (4 C) with each of anticyclin antibodies (1:500 dilution for D1, D3, and A2 types and 1:100 dilution for the D2 type), anti-3ß-HSD (1:300), and anti-ß-actin (1:6000 dilution) antibodies. After washing three times with PBS/Tween 20 at room temperature, the detection of bound antisera was performed with horseradish-peroxidase-labeled goat-antirabbit IgG at a 1:2000 dilution (2 h, room temperature). The membranes were then washed and the immune complexes visualized by using CovaLight chemiluminescence detection system (CovalAb, Lyon, France).
The intensities of the bands obtained after light impression of autoradiographic films (Biomax; Eastman, Kodak Rochester, NY) were determined by densitometric scanning using the National Institutes of Health software (Bethesda, MD). The ODs corresponding to cyclin and ß-actin bands, respectively, in each individual sample were expressed in arbitrary units as a cyclin over ß-actin ratio.
Immunohisto- and immunocytochemical assays
For immunohistochemistry, frozen slide-mounted sections were brought back to room temperature, hydrated 5 min in 0.1 M PBS (pH 7.4), and permeabilized 30 min in PBS containing 0.5% Triton X-100.
For immunocytochemistry of BrdU and Ki67, an initial pretreatment consisted in antigen unmasking by incubation in 10 mM sodium citrate buffer (pH 5.8) for 10 min at 95 C, followed by treatment with pepsin (0.125 mg/ml in 0.1 N HCl) for 2 min at room temperature.
For all immunocyto- and immunohistochemical assays, the endogenous peroxidase activity was quenched with a 15-min incubation in 3% H2O2. Samples were then incubated in blocking buffer (PBS containing 2–5% BSA, 0.2% Triton X-100, and 5–10% normal goat serum) for 1 h at room temperature. Samples were incubated overnight at 4 C with primary antibody diluted in fresh blocking buffer (1:1000 for anti-BrdU and anti-cyclin D1, 1:200 for anti-Ki-67, and 1:500 for anti-RLF), rinsed three times for 10 min each in PBS, and incubated 2 h at room temperature with corresponding biotinylated anti-IgG (diluted 1:200 in blocking buffer). Labeling was revealed by 30-min incubation with avidin-peroxidase (Elite-ABC kit; Vector Laboratories, Burlingame, CA) and staining with diaminobenzidine (DAB kit; Vector Laboratories). After immunocytochemical labeling, cells were lightly counterstained with Mayer’s hematoxylin (Dako, Glostrup, Denmark). All samples were dehydrated through 70/95/100% ethanol and xylene and either mounted on glass slides (for cells) or coverslipped (for tissue sections) with Permount. Nonspecific labeling was assessed on samples processed without the primary antibody.
Cell counting
Two different protocols of cell counting were used. In the first one, dead cell number was assessed by trypan blue exclusion. At the end of the treatment periods, floating cells were collected and added to the cells detached from the well bottom by trypsinization. After centrifugation, cell pellets were resuspended in PBS to which trypan blue solution was added to a final concentration of 0.04%. Both total and dead cell numbers were determined by four independent hemocytometer counts in each experiment. The number of dead cells is expressed as a percentage of total cell number at the time of the treatment arrest.
In the second protocol, the number of viable cells was determined after the supernatant containing dead cells had been discarded. For adherent Leydig cells, this number corresponds to the attached cells that were removed from the well bottom by trypsinization.
For quantification of BrdU incorporation and Ki67 labeling, the positive dark-stained nuclei were manually counted on coverslips at five randomly chosen areas of identical surface (423 x 341 µm). For each experimental condition, duplicate coverslips were systematically analyzed and assessed in two distinct series of experiments (total n = 20).
Data analysis
Statistical significance of the differences observed between experimental groups was determined by one-way ANOVA using the PRISM software computer program. Post hoc comparisons between treatment group means were made with the Bonferroni test for multiple comparisons. Differences are accepted as significant if P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, lane 1) and from RNAs extracted from cultured Leydig cells (Fig. 1A, lane 2). The specificity of the amplified fragment was confirmed by a parallel amplification of the fragment with the same size (410 bp) from cDNA corresponding to the whole adult rat brain RNA extracts used as a template (Fig. 1A, lane 5). DNA sequencing of both testicular fragments (i.e. whole testis and Leydig cell derived) and the fragment obtained with brain RNA extracts showed that, as expected, they are identical to the sequence from 211–621 bp of the cloned rat Ob/Rb gene (data not shown).
|
) corresponding to ILCs rather than to PLCs based on these morphological criteria (3, 8). Accordingly, spindle-shaped cells with peritubular location corresponding to PLCs were not Ob/Rb positive (Fig. 1C). Moreover, no immunostaining was observed in peritubular space, thus indicating that peritubular cells in 21-d-old rat testis could not be targeted by leptin administration.
Leptin mediates the alteration of cyclin expression
To assess leptin actions on cyclin expression at the protein level, we performed Western blot analysis of soluble protein extracts obtained from prepubertal Leydig cells cultured with 100 nM leptin for different time periods. These experiments indicated that leptin triggers an increase of cyclin D1 protein expression that could be evidenced after 24 h of treatment (Fig. 2A). This increase in cyclin D1 expression was preceded by a decrease in cyclin A2 expression seen already after 16 h of treatment (Fig. 2D). By contrast, leptin treatment could not trigger any significant alteration in cyclin D2 and cyclin D3 expressions (Fig. 2, B and C).
|
, A and D, respectively). In accordance with Western blot data, no significant fluctuations in type D2 and D3 cyclin mRNAs were observed (Fig. 3, B and C).
|
). The addition of leptin prevented this increase, and the inhibitory action of leptin was detectable after 48 h of treatment.
|
). Quantification of BrdU-labeled nuclei showed that in the absence of leptin and at the beginning of the culture (24-h time point), a small fraction (about 4%) of the viable cells is actively synthesizing DNA. The addition of leptin (100 nM for 24 h) induced a slight but not significant decrease in BrdU incorporation by Leydig cells (Fig. 4C). At the later time points, leptin triggered a significant decrease in BrdU incorporation. However, the difference between the two experimental conditions was the most pronounced at 48 h after the beginning of treatment (Fig. 4C).
Similar results were obtained by quantification of the proliferation marker Ki67, which is expressed specifically in G1, S, and G2 phases of the cell cycle (40). These experiments showed that in the absence of leptin and at the beginning of culture, about 5% of cells are in the cell division cycle (Fig. 4E). This number increases with increasing time and reaches a maximum at 48h when more than 20% of cells are found in the cell cycle (Fig. 4E). The addition of leptin (100 nM) induced a significant decrease in the number of Ki67-positive nuclei that was more pronounced at 48 and 72 h than at 24 h after the beginning of treatment (Fig. 4, D and E).
Leptin-mediated induction of RLF expression displays a developmental pattern that correlates with the increased cyclin D1 expression
Given that in the course of development, a decrease in the proliferative capacity of Leydig cells is associated with their differentiation, we searched for a possible leptin effect on the expression of a differentiation marker, RLF. Our data show that there is a significant induction of RLF mRNA expression in prepubertal Leydig cells treated with leptin (100 nM, 72 h) (Fig. 5). This increase could not be assessed at the protein level on the extracts obtained from cultured prepubertal Leydig cells because the antibody used in our experiments (which is the only one commercially available) turned out to be unsuitable for Western blotting (data not shown).
|
, B and F, respectively). This was in contrast with the cyclin D1 labeling detected in both interstitial space and seminiferous tubules in infantile rat testis used as a positive control (Fig. 6A). The observed cyclin D1 labeling appeared to be cell type specific, and according to the morphological criteria, it could be attributed to Leydig cells in intertubular space and to both Sertoli cells and spermatogonia in the seminiferous tubule (Fig. 6A). Testicular sections of infantile rats used as a positive control displayed RLF-positive diffuse labeling in seminiferous tubules (Fig. 6D). This RLF labeling was specific because it was not detected when the anti-RLF antibody was omitted (Fig. 6E).
|
). Cyclin D1 immunoreactivity, as assessed on the adjacent testicular sections from the same animals, was found in some round-shaped Leydig cells but never in seminiferous tubules (Fig. 6C).
Leptin-mediated increase in 3ß-HSD is prevented by knocking down cyclin D1 expression
To demonstrate the involvement of cyclin D1 in pro-differentiating actions of leptin, we searched for a Leydig cell differentiation marker that can be directly assessed by Western blot. Expression of 3ß-HSD was chosen because preliminary experiments showed that the increase of 3ß-HSD expression could be readily detected after leptin treatment of cultured Leydig cells (Fig. 7A). The observed increase was slightly more prominent after 48 h than after 72 h of treatment (3-fold vs. less than 2-fold increase in leptin-treated and control cells, respectively; Fig. 7A). Knocking down cyclin D1 in vitro with siRNA, as attested by a decreased cyclin D1 protein expression (control siRNA vs. cyclin D1 siRNA in the absence of leptin treatment, Fig. 7B), prevented the leptin-mediated 3ß-HSD induction after 48 h of treatment (Fig. 7C). Importantly, knocking down cyclin D1 expression completely abolished the leptin capacity to induce cyclin D1 expression (Fig. 7B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our experimental approach was validated by RT-PCR amplification of fragments of the identical length and sequence as those reported in previous studies in which the same pair of primers has been used to study leptin receptor Ob/Rb (32). Moreover, the expression of leptin receptors in interstitial cells and their absence in seminiferous tubules have been shown in 20-d-old mouse testis (45), which is in agreement with our present results using the same receptor antibody. However, our data on Ob/Rb receptor expression in 21-d-old rat testis are not in agreement with the previously reported absence of this receptor expression in 22-d-old rat testis (32). Given that these authors used the same antibody for immunohistochemical detection, the most plausible explanation for the discrepancy between our two studies is the difference in tissue fixation procedures (Bouin in the previous study vs. paraformaldehyde in ours). Our immunohistochemical data showing the absence and presence of RLF-positive cells in testes from 21- and 35-d-old rats, respectively, are consistent with the previously published data obtained on testes of the same postnatal ages (32). Most importantly, the presently documented absence of RLF labeling at the developmental stage used in our study (21-d-old rats) allows us to exclude the presence of residual mature fetal Leydig cells (known to abundantly express this receptor) (32) and thus to exclude this cell type as a substrate for the observed Ob/Rb labeling.
The observed expression of different cyclins at both RNA and protein levels shown here in extracts obtained from cultured prepubertal Leydig cells is in agreement with previous reports of their expressions in whole testicular extracts obtained from the same developmental period in rat (13, 46). More specifically, the observed leptin-mediated actions on the increase of cyclin D1 protein occur in parallel with an increase in corresponding mRNA level. This is in agreement with both the time range of cyclin D1 mRNA induction in other cell types (47) as well as in testicular somatic cells (48).
The analysis of leptin actions on Leydig cell proliferation points to a decrease in the viable cell number that correlates with a decrease in the number of cells passing through S phase of the cell cycle as shown by the quantification of BrdU-positive nuclei. In agreement, the number of actively dividing cells is decreased in the presence of leptin as attested by a decreased number of Ki67-positive cells. Leptin actions on both proliferation indexes were the most pronounced 48 h after the beginning of the treatment, precisely at the time of the highest proliferative activity that could be reached in the chosen experimental conditions. However, at least at the 24-h time point, the percentage of the labeled nuclei was slightly higher for Ki67 immunoreactivity than for BrdU incorporation, likely because Ki67 is expressed in dividing cells along the cell cycle (40), whereas a given cell nucleus incorporates BrdU only after the cell has passed through the S phase.
The observed leptin-induced inhibition of Leydig cell proliferation was not associated with the inhibition of different types of cyclin D expression. Indeed, cyclin D1 expression was increased, whereas cyclins D2 and D3 were not altered in leptin-treated cells. These results were unexpected in light of the largely accepted assumption on the causal link existing between cyclin D expression and cell proliferation (reviewed in Ref. 49). It is indeed believed that depending on the cell type, a specific D-type cyclin is necessary and sufficient to trigger G1/S transition and entry into the cell division cycle (11). According to the generally accepted view, the induction of any of the three cyclin D forms is directly coupled to the mitogen challenge and leads to an irrevocable entry into the cell cycle provided that the mitogens remain present in the cell environment during a critical time period, i.e. until the R point in the late G1 phase (50, 51). As a consequence, leptin-mediated inhibition of Leydig cell division appears uncoupled from cyclin D regulation.
However, the present study shows that leptin actions on cyclin D1 expression are preceded by an inhibition of cyclin A2 expression at both protein and mRNA levels. The observed leptin-induced decrease in cyclin A2 protein level is in agreement with an abrupt decrease in cyclin A2 mRNA expression coincident with a loss of Leydig cell proliferative capacity that has been observed when these parameters were compared between PLCs obtained from the testes of 21-d-old rats and ILCs obtained from the testes of 35-d-old rats (13). Because this time lapse corresponds to the switch from PLC proliferation to their differentiation to ILC (3, 4), the demonstrated leptin-induced inhibition of prepubertal Leydig cell proliferation might be mediated through the leptin-induced inhibition of cyclin A2 expression by PLCs.
Our data on the independence of proliferation control from cyclin D1 are consistent with the recent data obtained from studies on cyclin D1–/– D2–/– D3–/– mice. These studies strongly suggested that D-type cyclins might not be the exclusive triggers of the cell division cycle because homozygous inactivation of all three cyclin D genes is not sufficient to generally compromise the cell proliferation (52, 53). If the regulation of cell division in prepubertal Leydig cells is independent of cyclin D, what would then be the physiological meaning of the observed leptin-mediated increase in cyclin D1 expression? To answer this question we assessed the leptin action on the expression of Leydig cell differentiation-specific marker RLF (6, 54). Our data show that leptin increases RLF transcript expression, but these effects require a longer treatment period than that necessary for detection of leptin effects on cyclin D1 expression (72 vs. 16–24 h, respectively). Similar results were seen concerning leptin effects on another marker of Leydig cell differentiation such as 3ß-HSD. The fact that leptin-mediated induction of cyclin D1 expression precedes those of RLF and 3ß-HSD is consistent with the putative role of leptin in the control of Leydig cell differentiation via cyclin D1. The observed timing of leptin effects on the inhibition of cell proliferation (48–72 h) is also compatible with the postulated pro-differentiating actions of leptin. Indeed, in the course of Leydig cell differentiation, the loss of precursor proliferative capacity occurs concomitantly with the acquisition of the characteristics of differentiated Leydig cells (2, 5). However, it should be stressed that a large delay in leptin effects on the expression of differentiation markers (in particular for RLF) might indicate that they are only a bystander consequence of Leydig cell differentiation. Indeed, such a possibility cannot be formally ruled out, although our siRNA data show that blockage of leptin-dependent cyclin D1 induction inhibits leptin-mediated 3ß-HSD expression.
To assess the physiological significance of cyclin D1 involvement in the mediation of leptin actions on Leydig cell differentiation, we compared the expression of cyclin D1 and of differentiation marker RLF in sections of testis sampled at the critical developmental stages. Our immunohistochemical data point to the correlation between expressions of cyclin D1 and RLF by Leydig cells; neither expression is detectable at the onset of puberty (21-d-old testis), whereas both are expressed at the limit between pre- and peripubertal stages (35-d-old testis). This is in line with previously published data on the induction of RLF expression in peripubertal rat testis (7, 54). In addition, our data on the down-regulation of cyclin D1 expression by Leydig cells between the postnatal d 10 and 21 are in agreement with the cluster analysis related to the PLC to ILC transition (55).
The round-shaped morphology and central location in the interstitial space of cyclin D1-positive cells are thus rather reminiscent of the differentiation-committed ILC (3, 8). Very similar morphology and location of cells immunoreactive for RLF, which is considered a differentiation-specific marker of committed Leydig cells (6, 54), point to the fact that cyclin D1 and RLF are likely coexpressed in the same subpopulation of ILC, thus further suggesting the involvement of cyclin D1 in differentiation of Leydig cells. Moreover, our data showing that siRNA-mediated cyclin D1 knockdown inhibits leptin-dependent induction of 3ß-HSD differentiation marker demonstrate the functional relevance of the cyclin D1 increase in pro-differentiating actions of leptin.
Altogether, our data are consistent with the emerging hypothesis that different types of cyclin D might be involved in the control of the switch from proliferation to differentiation (56). Indeed, the role of cyclin D1 in the induction of neuronal differentiation (57) as well as its involvement in retinoic acid-induced inhibition of proliferation and induction of differentiation in epithelial cells (58) have been reported. The expression of cyclin D3 in nondividing pachytene spermatocytes and round spermatogonia led to their proposed involvement in differentiation of male germ cells (15, 46). However, extracellular signals able to trigger cyclin D3 expression and their link to proliferation have not yet been identified.
Our data show that leptin has a functional capacity to decrease the rate of proliferation and increase cyclin D1, RLF, and 3ß-HSD expressions in cultured rat Leydig cells in a time-dependent manner. Similarly, it has been suggested that leptin plays a role in the control of proliferation/differentiation balance in mouse male germ cells, although the involvement of cyclin D1 in this process has not been previously assessed (45). We also show here that leptin exerts these actions on Leydig cells obtained from rat testes precisely at the developmental age at which these cells begin to differentiate toward adult-type, testosterone-secreting cells. Given that in rodents leptin can cross the blood-testis barrier (59) and that Leydig cells from prepubertal rat testis express the functional Ob/Rb receptors (present study and Refs. 32 and 45), it seems reasonable to assume that leptin might be involved in the physiological regulation of the progressive switch from proliferation to differentiation of Leydig cells in the prepubertal stage of testis development. In the light of the literature data and results presented here, it can be proposed that fluctuations in circulating leptin levels before sexual maturation may have an impact on the development of the adult-type Leydig cell population. Indeed, pathologically elevated leptin levels, associated for example with juvenile obesity, might be deleterious for male fertility through impairment of Leydig cell dynamics. In agreement, it has been recently reported that rats neonatally treated with an obesity-inducing agent display hyperleptinemia and hyperadiposity that are associated (among other pathological alterations) with a significant reduction in the number of Leydig cell in prepubertal age (60). Although this reduction in the size of the Leydig cell population provoked by neonatally induced hyperleptinemia could not be correlated with a reduced testosterone secretion in adulthood, these data combined with the present work point to the urgent need to clarify these issues in the context of growing health problems such as increasing infant obesity in developed countries.
In conclusion, our study adds to the evidence of leptin actions in the control of male reproductive functions. It also brings new insights on the cyclin D-independent cell cycle control and involvement of cell cycle effectors such as cyclin D in the control of cell processes other than proliferation. Better understanding of these phenomena might have important implications not only in fundamental but also in applied fields of reproductive biology, development, and carcinogenesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online February 15, 2007
Abbreviations: ALC, Adult Leydig cells; BrdU, bromodeoxyuridine; 3ß-HSD, 3ß-hydroxysteroid dehydrogenase; ILC, immature Leydig cells; PLC, progenitor Leydig cells; PVDF, polyvinylidene difluoride; si, small interference; RLF, relaxin-like factor.
Received September 8, 2006.
Accepted for publication February 6, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
