|Home | About | Journals | Submit | Contact Us | Français|
Estradiol is known to increase lactotropic cell proliferation, but estradiol susceptibility varies among human populations and among various strains of rats. We had reported that folliculostellate (FS) cells regulate estradiol’s mitogenic action on lactotropes; therefore, we studied their role in determining the susceptibility to estradiol in a high estradiol-responsive rat strain, Fischer 344 (F344), and in a low-responsive strain, Sprague Dawley (SD). Determination of total S-100-positive FS cells in the pituitary revealed that F344 rats have significantly more FS cells than do SD rats. Estradiol treatment did not change the number of FS cells in both F344 and SD rats. When cotransplanted with F344 pituitaries under the kidney capsule or cocultured with F344-derived lactotropes in vitro, FS cells derived from F344 rats increased estradiol’s mitogenic action. They also increased estradiol’s mitogenic action on SD-derived lactotropes in primary cultures. However, SD-derived FS cells failed to increase estrogen’s action on F344-or SD-derived lactotropes. The levels of basic fibroblast growth factor production and secretion by TGF-β3 and estradiol were much higher in F344-derived FS cells than in SD-derived FS cells. However, the lactotropes’ growth response to basic fibroblast growth factor was similar in both strains. These data suggest that cell-cell interaction between FS cells and lactotropes regulates estradiol’s mitogenic action on lactotropes and also determines lactotrope susceptibility to the steroid.
The ovarian steroid estradiol is known to increase lactotropic cell proliferation in humans as well as in laboratory animals (1–7). Certain populations of humans are more susceptible to estradiol’s mitogenic action on lactotropes (8). A similar difference in lactotropic cell susceptibility to estradiol is also observed among different strains of laboratory rats. For example, Fischer-344 (F344) (3, 9) and AxC-Irish strains (10, 11) are sensitive to estrogen’s growth-promoting and tumor-inducing actions on the pituitary. The F344 strain is most sensitive to estrogen, and chronic estradiol treatment in this strain induces lactotropic proliferation that results in lactotropic tumors within a few months (3, 12). Unlike F344 rats, Sprague Dawley (SD), Brown Norway, and Holzman strains show low lactotropic cell proliferation upon chronic estrogen treatment (9, 11, 13). Pituitary tumor formation in genetic crosses between estrogen-sensitive and estrogen-insensitive rats has indicated that multiple genetic pathways exist in the rat species (13–15). However, it is not well understood what mechanisms account for these differences in estradiol susceptibility.
In several tissues, it appears that the growth of cells depends not only on the mitogenic stimulus but also on cell-to-cell communications. For example, in uterine, ovarian, and mammary tissues, communication with mesenchymal cells facilitates the growth of epithelial cells (16–18). Our recent findings show that folliculostellate (FS) cells regulate estradiol cell proliferating action on lactotropes (19). FS cells are characterized by their stellate shape and long cytoplasmic processes, and they are largely devoid of secretory granules. S-100 immunoreactivity has proven to be a reliable marker of FS cells (20). It has been suggested that FS cells perform several supportive functions, including regulation of phagocytosis (21) and secretion of angiogenic factors (22), growth factors (19, 23), and cytokines (24, 25) as well as tropic and stem cell functions (26). FS cells seem to be targeted by estrogen because they have both estrogen receptor α and estrogen receptor β (27), and rapidly increase c-fos expression (28) in response to estrogen.
There are no morphological differences in the ultrastructures of FS cells of normal F344 and SD rats. However, after estrogen treatment, FS cells of F344 rats dramatically alter their morphology, are activated as phagocytes, and affect new vessel formation in the adjacent meninges. On the other hand, FS cells of SD rats do not show a comparable response (29). This report suggests that there are some functional differences between FS cells of F344 rats and SD rats. We have recently demonstrated that FS cells participate in estrogen’s mitogenic action on lactotropes (19). Estradiol induces TGF-β3 release from lactotropes. TGF-β3 then acts on FS cells to release basic fibroblast growth factor (bFGF), which acts on lactotropes to increase cell proliferation. Thus, FS cells mediate estradiol’s mitogenic action on lactotropes by releasing bFGF. We present evidence here that FS cells determine lactotrope susceptibility to estradiol by producing different amounts of bFGF.
Female F344 and SD rats with a body weight (BW) of 160–200 g, obtained from Simonsen Laboratories (Gilroy, CA), were housed in a controlled environment (temperature, 22 C; lights on, 0500–1900 h) and provided rodent chow meal and water ad libitum. Animals were ovariectomized bilaterally and sc implanted with a 1-cm SILASTIC brand capsule (Dow Corning Corp., Midland, MI) filled with estradiol-17β (Sigma Chemical Co., St. Louis, MO) or an empty SILASTIC brand capsule using sodium pentobarbital anesthesia (40 mg/kg BW, ip). Animal surgery and care were performed in accordance with institutional guidelines and complied with the National Institutes of Health policy.
After pituitaries were obtained, they were immediately fixed with 4% formalin, and paraffin blocks were made. Pituitary tissue sections were prepared at 3-μm thickness and processed for immunostaining using S-100 antibody (1:200; Zymed Laboratories, San Francisco, CA) and an ABC kit (Vector Laboratories, Gilroy, CA) as described by us previously (30). A negative control slide, incubated with normal serum from the host species, was included. No specific staining was observed in negative control slides. The sections were counterstained with Gill’s hematoxylin to detect nuclei. The distribution of S-100 immunoreactivity in the presence and absence of estrogen was compared. Routine counts of cells exhibiting S-100 immunoreactivity were conducted in five areas of each of five sections of the pituitary from ovariectomized and estradiol-treated ovariectomized rats.
Pituitary glands were obtained from female F344 rats. Pituitaries were cleaned of membranes, and the intermediate and posterior lobes were peeled off and discarded. The anterior lobes of the pituitaries were implanted under the kidney capsules of ovariectomized female F344 rats using sodium pentobarbital anesthesia (40 mg/kg, ip). Both kidneys were transplanted with one pituitary in each animal; one was cotransplanted with FS cells (1 million cells/100 μl media), and the other was transplanted alone without FS cells but received 100 μl media. These rats were additionally implanted with a 1-cm estradiol-filled SILASTIC brand capsule. Three weeks after the pituitary transplant, bromodeoxyuridine (BrdU; 100 mg/kg BW, ip) was injected, and 4 h later, animals were killed. Pituitary transplants were obtained and immediately fixed with 4% formalin for lactotropic cell proliferation assay.
Lactotropic cell proliferation was determined by identifying cells that displayed both BrdU and prolactin (PRL) immunoreactivities, as described by us previously (12, 30). Because BrdU is a marker of DNA synthesis, double-stained cells were considered to be dividing lactotropes. Rats were treated with BrdU (100 mg/kg BW) 4 h before fixation with 4% formaldehyde. Tissue sections were treated with HCl and then incubated at 4 C overnight with BrdU monoclonal mouse IgG (Becton Dickinson, San Jose, CA) and stained using the Vectastain ABC kit (Vector Laboratories, Inc., Burlingame, CA) with diaminobenzidine as the chromagen. Cells were then incubated with PRL antibody (IC-5; National Institute of Diabetes and Digestive and Kidney Diseases) at 4 C overnight and stained using the Vectastain ABC-AP kit. Negative controls were conducted by exposing cells to 3% normal serum from the host species, rather than primary antibody, and also by preabsorbing the antibody with 100-fold excess antigen. No specific staining was observed in negative control slides. Cells were counted in five separate areas in each coverslip with approximately 500 cells per area.
For some experiments using purified cell populations, only BrdU immunoreactivity was determined, and the results are presented as a percentage of BrdU-incorporating cells out of the total cell population (which is determined by counting the number of Harris hematoxylin-positive cells).
Anterior pituitaries from estradiol-treated ovariectomized rats were collected and enzymatically dissociated as described by us previously (12). Lactotropes were enriched from dissociated anterior pituitary cell suspensions using a discontinuous Percoll gradient as described before (31). Briefly, cells were dissociated, layered, using a 1-ml plastic pipette, on top of the Percoll (Sigma) gradient consisting of 60, 50, and 35% Percoll layers. The gradient was centrifuged at 450 × g for 20 min. The cells at the 35–50% interface were collected with a glass pipette and seeded on a poly-L-lysine-coated 24-well dish (Becton Dickinson) as enriched lactotropes. Enrichment by Percoll gradient separation yielded 81 ± 6.3% lactotropes. These were considered to be free of FS cells because no S-100 immunoreactivity was detected in cultures that were Percoll separated (30). Cells were plated in 24-well plates with approximately 200,000 cells per well. The cells were maintained in DMEM/Ham’s F-12 medium (DMEM/F-12, containing 1.2 mg/ml sodium bicarbonate, 100 U/ml penicillin, and 100 mg/ml streptomycin; all from Sigma) with 10% fetal calf serum (FCS; HyClone Laboratories, Inc., Logan, UT) for 1 d, then in medium containing 2.5% FCS and 10% horse serum (HyClone) for another 2 d. Cultures were then maintained in serum-free DMEM/F-12 containing serum supplement (100 μM human transferrin, 5 μM insulin, 1 μM putrescine, and 30 nM sodium selenite) during experimentation.
We have previously established a FS cell line (F344-FS) from anterior pituitary cells from F344 cyclic female rats (30). In this study, we established a FS cell line (SD-FS) from anterior pituitary cells from SD cyclic female rats. The methods for the preparation of the SD-FS cell line were identical with those described for the preparation of the F344-FS cell line. After 10 passes, immunohistochemical procedures for S-100 detection (described above) revealed that the cultures contained only S-100-positive cells. It has been shown that S-100 is an effective marker of the FS cell phenotype (30). The FS cell line was maintained in DMEM/F-12 with 10% FCS, and the cells used were between generations 20 and 30. During experimentation, FS cells were maintained in serum-free DMEM/F-12 with serum supplement.
Primary cultures of lactotropes (200,000 cells/well) were maintained in serum-containing medium for 3 d as described above. These cells were then cocultured with F344-FS or SD-FS cells (50,000 cells per well) in the presence of serum-free DMEM/F-12-containing serum supplement. After 1 d, these cocultures were used for experimentation. The estrogen used was estradiol-17β(Sigma, water-soluble) at a 10 nM concentration, which has been shown to produce the maximal lactotropic cell growth response (32). Recombinant human TGF-β3 was obtained from R&D Systems (Minneapolis, MN) and reconstituted in 0.1% BSA and 4 mM HCl. The dose used was 1 ng/ml; this dose of the growth factor has been shown to maximally increase the secretion of bFGF from the FS cells and the proliferation of lactotropes (19, 30). Recombinant human bFGF (R&D Systems) was dissolved in PBS containing 0.1% BSA and 1 mM dithiothreitol and used at concentrations of 0 –10 ng/ml. Control cultures received vehicle. In all of the cell proliferation experiments, total treatment time was 96 h with the media and treatment changed at 48-h intervals. Cultures were treated with 0.1 mM BrdU 4 h before fixation with 99% ethanol. Lactotropic cell proliferation was determined by identifying cells that displayed both BrdU and PRL immunoreactivities.
FS cells (250,000 per well) were grown in 24-well plates in serum-containing medium. After 2 d of plating, the medium was changed with defined media containing serum supplement. On the following day, cells were treated with vehicle (control), TGF-β3 (1 ng/ml), estradiol (10 nM), or TGF-β3 (1 ng/ml) and estradiol (10 nM) in serum supplement media for 24 h. Media samples were collected and bFGF levels measured using a Quantikine immunoassay kit (R&D Systems). Cells were lysed in 100 μl of lysis buffer, and 10 μl of cell lysate from each group was used to assess bFGF levels Total protein concentration in each cell lysate was determined using the Bio-Rad assay (Bio-Rad, Hercules, CA) to calculate expression of total protein per microgram.
The data shown in the text and the figures are mean ± SEM. Multiple groups were statistically analyzed using one-way or two-way ANOVA as appropriate, whereas differences between two groups were evaluated using Student’s t test. Post hoc analyses after ANOVA used the Student-Newman-Keuls test. A value of P < 0.05 was considered significant.
We determined whether the number of FS cells in the pituitary varies between F344 rats and SD rats. The FS cell number per pituitary of ovariectomized F344 rats was significantly higher than that in ovariectomized SD rats (Fig. 1 and Table 1). Estradiol treatment increased the size of the pituitary (as they develop prolactinomas) and the total number of cells in F344 rats but not in SD rats. Because of the increase in the pituitary size, the FS cell number per square millimeter of the pituitary decreased without any significant change in the cell number per total area of the pituitary after estradiol treatment in F344 rats. FS cell number per square millimeter or in the total area of the pituitary was similar in SD rats with or without estradiol treatment.
We have previously shown that FS cells regulate estradiol’s mitogenic action on lactotropes in primary cultures (19). In this study, we determined the effect of FS cells on estradiol-induced lactotropic cell proliferation in vivo. When F344 rat pituitaries were cotransplanted with F344-FS cells in donor F344 rat kidneys for a period of 3 wk, the transplanted pituitaries showed a higher number of mitotic lactotropes than those in the pituitaries transplanted without FS cells (Fig. 2). Our efforts to cotrans-plant SD pituitary with F344-derived FS cells in the kidney capsules of F344 rats were not successful; the cells were rejected.
To further characterize the strain differences in estrogen’s function on FS cells, we determined the effect of coculturing lactotropes with FS cells from the same strain or from a different strain. Estradiol increased lactotrope proliferation when F344 lactotropes were cocultured with F344-FS cells (Fig. 3). Interestingly, the estradiol-induced proliferation of SD lactotropes was also observed when SD lactotropes were cocultured with F344-FS cells. However, lactotropes derived from SD or F344 rats failed to respond to estradiol when cocultured with SD-derived FS cells (Fig. 4). These results demonstrated that both F344-derived and SD-derived lactotropes had the ability to respond to estradiol if F344-derived FS cells were present and suggested that the sensitivity of lactotropes to estradiol is determined by the function of FS cells.
As described above, we have previously shown that lactotrope-derived TGF-β3 and FS cell-derived bFGF play important mediatory roles in estradiol’s mitogenic action on lactotropes. Therefore, we compared the changes in the release and cellular levels of bFGF after estradiol and TGF-β3 treatment in F344- and SD-derived FS cells. Both estradiol and TGF-β3 alone increased bFGF production and release from F344-derived FS cells. However, bFGF production and release in SD-derived FS cells was not significantly changed by estradiol or TGF-β3 (Fig. 5). Estradiol and TGF-β3 together produced a marked stimulation of bFGF from F344-derived FS cells. They produced only a moderate increase in cellular levels and in release of bFGF in SD-derived FS cells.
Previously we have shown that estradiol’s mitogenic action on lactotropes involves TGF-β3 secretion from lactotropes. The growth factor then acts on FS cells to produce bFGF, which then acts on lactotropes to increase cell proliferation (19). We determined whether a difference in the responsiveness to bFGF also exists between SD- and F344-derived lactotropes. In both F344-derived and SD-derived lactotropes, bFGF dose-dependently stimulated cell proliferation (Fig. 6). There was no difference in the cell proliferation response to bFGF between F344 lactotropes and SD lactotropes (Fig. 6), suggesting that the major difference between F344-FS and SD-FS cells is the lower bFGF response to TGF-β3 and estradiol.
We demonstrated that FS cells derived from high estrogen-responsive F344 rats, but not from low estrogen-responsive SD rats, regulated estrogen-induced proliferation of lacto-tropes in both F344 and SD rats. The bFGF response to estradiol and TGF-β3 was high in F344-derived FS cells, but it was low in SD-derived FS cells. In addition, we showed that the cell-growth response of bFGF was similar in the lactotropes of both rat strains. These data suggest that susceptibility to estrogen depends on the ability of FS cells to produce sufficient bFGF to increase cell proliferation.
The data presented here show that the differential lactotropic cell response to estradiol in SD and F344 rats is related to the difference in the estradiol-regulated FS cell communication with lactotropes via the production of bFGF. The different responses of SD-FS cells and F344-FS cells may not be related to estradiol receptor functions, because estradiol activation of its receptor α- and β-levels in the F344 and SD pituitaries was similar (27). Interestingly, the truncated estrogen receptor products (TERP) 1 and 2, which are known to function as inactivators of estrogen receptors (32), have been shown to be expressed differently in the pituitary of F344 and SD rats (27). However, the physiological significance of these findings in relation to differential pituitary’s estrogen sensitivity of F344 and SD rats is not apparent at present.
Expression of the oncogene c-Fos in FS cells was activated by estradiol in the F344 pituitary (28). The product of c-Fos is a nuclear protein that forms heterodimers with products of the c-Jun gene family generating the transcription factor, activating protein-1 (33). Activating protein-1 and Smad protein complexes synergize in the transcriptional activation of the c-Jun promoter (34). Smad proteins mediate TGF-β signaling (35). TGF-β-induced translocation of Smad proteins has been demonstrated in FS cells (36). Recently, it has been shown that TGF-β-activated Smad proteins regulate estrogen action (37, 38). The FS cells’ bFGF response to TGF-β3 in the presence of estradiol was higher in F344 rats than in SD rats. TGF-β1 also has been shown to stimulate FS cell proliferation (39). Hence, it could be postulated that higher FS cell numbers in the F344 pituitary is partly related to the FS cells’ higher response to TGF-β. Additional work determining the F344- and SD-derived FS cell-growth responses to various isoforms of TGF-β is needed to address this issue.
TGF-β3 and estradiol treatment increased the bFGF production in F344-FS cells 6-fold but only moderately increased bFGF production in SD-FS cells. The small bFGF response of SD-FS cells to TGF-β and estradiol is consistent with the finding that estradiol did not stimulate the lactotropic cell proliferation in SD rat-derived pituitary cells. This may be because the total amount of bFGF produced by estradiol treatment is not enough to stimulate lactotropic cell proliferation in this strain. It is also possible that factors other than the low bFGF response of SD-derived FS cells are responsible for estrogen’s inability to increase lactotropic cell proliferation in SD rats. FS cells produce growth factors or cytokines, such as vascular endothelial growth factor (21) and IL-6 (24). Studies determining the role of these factors in estrogen’s differential regulation of lactotropic cell proliferation in F344 and SD rats should provide further understanding of the steroid’s mitogenic action in the pituitary.
We thank the National Hormone and Pituitary Program, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (Bethesda, MD) for their kind supply of antibodies to prolactin.
This work was supported by NIH Grants AA00220 and CA775500.