| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-Subunit in Neonatal Rat and Chicken: Rapid Decline with Age and Reappearance in Vitro under Regulatory Pressure of Corticotropin-Releasing Hormone in the Rat
Laboratory of Cell Pharmacology, University of Leuven, Medical School, B-3000, Leuven, Belgium
Address all correspondence and requests for reprints to: Prof. Carl Denef, Laboratory of Cell Pharmacology, University of Leuven, Medical School, Campus Gasthuisberg (O & N), B-3000 Leuven, Belgium. E-mail: Carl.Denef{at}med.kuleuven.be
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit (GSU) could be coexpressed with proopiomelanocortin (POMC) in vivo and under pressure of CRH in vitro. Cells coexpressing GSU and POMC mRNA amounted to 2.6% of the cells in ex vivo rat pituitary at birth [postnatal d 1 (P1)], fell to much lower level at P14, and were undetectable in adulthood. In cultured pituitary aggregates of P14 rats, GSU/POMC cells remained scarce but represented up to 6.6% after chronic treatment with CRH but not leukemia inhibitory factor. CRH was less effective in aggregates from P1 and adult rats. The total GSU population ex vivo at P1 was two times smaller than at P14, but in culture it expanded 2.5 times, concomitantly with a reciprocal change in POMC cell abundance. Tpit transcripts were detected in POMC-only and GSU/POMC cells but not in GSU-only cells. Cells coexpressing GSU and POMC mRNA were relatively abundant in P14 chicken pituitary and aggregate cultures, but occurrence was not affected by CRH. Immunostaining showed GSU and POMC colocalization in sporadic cells in intact rat pituitary and CRH-treated cultures at P1 but not at P14 and adult age. The data demonstrate the occurrence of cells coexpressing GSU and POMC in rat and chicken pituitary. The developmental dynamics of this cell population and its response to CRH in vitro in the rat suggest a relationship of these cells with the embryonic branching of the POMC and GSU cell lineages and their mutually opposite developmental course during early postnatal life. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit (GSU) but not ßGSU (23).
The function of cells displaying promiscuous gene expression in the pituitary remains unknown. It has been proposed that these cells are pluripotential progenitor cells or ambiguous cells that might differentiate into one cell type or another depending on the endocrine needs of the body (24, 25, 26). Others have suggested that bihormonal cells might be transitional cells in transdifferentiation processes (15, 27, 28) that are presumed to be important when hormonal needs increase. Bihormonal cells may also be viewed as providers of hormones with a coordinated function (27). That bihormonal cells are biologically relevant is indicated by the finding that their abundance and coexpression phenotype is influenced by developmental, physiological, and pathological factors. Hypothyroidism considerably increases the proportion of cells expressing GH and TSH (14, 15). Exposure to cold stress and vasopressin rapidly expands a population of cells coexpressing POMC and TSH (16). The number of GH cells expressing LH significantly increases during proestrus (27). During pregnancy, somatotrophs start expressing PRL mRNA (29). We have shown that the small population of cells coexpressing mRNA of GSU and PRL is increased several-fold (up to 20% of all cells) by chronic treatment of pituitary aggregate cell cultures from immature rats [postnatal d 14 (P14)] with GnRH or estrogen (23). Importantly, after induction of mRNA colocalization by GnRH, protein colocalization was not detectable by immunocytochemistry, suggesting the existence of cells that coexpress hormone genes at the mRNA level but do not store the hormone proteins or translate the mRNA as classical terminally differentiated hormonal cells do (23). Cells expressing more than one hormone mRNA are also detectable in embryonic and pubertal mouse pituitary in specific age-dependent combinations (22).
The present investigation was intended to further document promiscuous coexpression of pituitary hormone phenotypes. Inspired by our previous finding of GSU mRNA expression in cells that express PRL mRNA under the influence of GnRH, we explored whether the hypothalamic hormone CRH is capable of inducing coexpression of GSU with other pituitary hormone genes, particularly with POMC and, if so, whether this coexpression also occurs in the pituitary in vivo and alters during development. We reported some preliminary evidence (in abstract form) that CRH can indeed induce coexpression of GSU and POMC mRNA in vitro (30). We here report the occurrence of GSU/POMC cells in relative high abundance in the female ex vivo rat pituitary at birth but not in adults and the remarkable expansion of this population in vitro under the influence of CRH, particularly in the P14 female rat pituitary. Furthermore, we could demonstrate that this cell type is also present, in an even greater abundance, in the chicken pituitary, although it was not affected by CRH in this species.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The RT-PCR products AmpliTaq Gold DNA polymerase, 10x PCR Gold buffer, MgCl2, dNTPs, and random hexamers were obtained from Applied Biosystems (Applera Belgium, Lennik, Belgium), RNase inhibitor (RNasin) from Promega (Leiden, The Netherlands), dithiothreitol (DTT), Moloney murine leukemia virus reverse transcriptase and 5x first-strand buffer (FSB) from Invitrogen (Merelbeke, Belgium), and IGEPAL CA-630 from Sigma Aldrich (Steinheim, Germany). Ex Taq Hot Start DNA polymerase and 10x Ex Taq buffer were purchased from Cambrex Bioscience (Verviers, Belgium). The GenElute gel extraction kit was obtained from Sigma Aldrich, the restriction enzyme DraIII from New England Biolabs (Beverly, MA), the 100-bp marker from Eurogentec (Seraing, Belgium), and the 2.1-TOPO vector from Invitrogen. Agarose was from ICN Biomedicals (Aurora, OH)
The primary antibodies guinea pig antirat GSU and rabbit antirat GH, PRL, LHß, TSHß, and human ACTH were obtained from Dr. A. F. Parlow through the National Hormone and Pituitary Program (National Institute of Diabetes and Digestive and Kidney Diseases, Harbor-UCLA Medical Center, Torrance, CA). Monoclonal mouse antihuman N-POMC was purchased from Biogenesis (Poole, UK). Normal goat and normal donkey serum were obtained from DakoCytomation (Glostrup, Denmark), fluorescent antibodies fluorescein isothiocyanate (FITC)-labeled goat antimouse, Texas Red (TR)-labeled donkey antiguinea pig, FITC-labeled goat antirabbit, and aminomethylcoumarin acetic acid-labeled goat antimouse IgGs from Jackson ImmunoResearch Laboratories (West Grove, PA). Monoclonal mouse antibodies against chicken GSU, GH, and N-POMC were kindly provided by Prof. L. Berghman from the Texas A&M University (31, 32). Vectashield was bought from Vector Laboratories (Burlingame, CA), and T4',6-diamidine-2'-phenylindole dihydrochloride from Roche Molecular Biochemicals (Mannheim, Germany). Goat antimouse Alexa Fluor 488-conjugated IgG (GAM488), goat anti-guinea pig Alexa Fluor 555-conjugated IgG (GAGp555), and TOPRO-3 were purchased from Molecular Probes (Eugene, OR).
Bovine pancreatic trypsin type III was acquired from Sigma Aldrich, paraformaldehyde from Riedel-deHaën (Seelze, Germany), picric acid from UCB (Braine-l’Alleud, Belgium), and Ca2+- and Mg2+-free PBS from Invitrogen. Paraffin was bought from Klinipath (Turnhout, Belgium), and saponin was from Sigma Aldrich.
Animals
Newborn (P1), 14-d-old (P14), and adult female Wistar rats were obtained from Elevage Janvier (Schaijk, The Netherlands). They were kept in an environment of constant temperature, humidity, and day-night cycle with free access to water and food within the animal house facilities (University of Leuven, Belgium). P14 male Ross chicks were obtained from Vervaeke (Tielt, Belgium). Immature rats and chickens were killed by decapitation, adult rats by decapitation after carbon dioxide narcosis. The experiments were conducted in accord with the National Institutes of Health Guidelines for Care and Use of Experimental Animals and were approved by the University Ethical Committee.
Pituitary cell dispersion and reaggregate cell cultures
Pituitaries were isolated, dispersed into single cells, and cultured as reaggregates as previously described (23, 33). In a typical experiment, the total cell dispersion included pituitaries from at least 10 rats or chicks, depending on the number of treatment groups and age. For most experiments, anterior and neurointermediate lobes were not separated and whole pituitaries were dispersed. In some experiments, cultures were prepared from the neurointermediate lobe only. Briefly, the tissues were cut into small pieces and treated sequentially with a 2.5% porcine pancreas trypsin solution (ICN Biomedicals or Roche Diagnostics, Basel, Switzerland), DNase, soybean trypsin inhibitor (Sigma Aldrich), and EDTA in Ca2+- and Mg2+-free medium as previously reported (23, 33). Subsequently, tissue fragments were mechanically dispersed with a Pasteur pipette. Cells were seeded in 35-mm untreated polystyrene culture dishes in 2 ml serum-free defined culture medium at 106 cells/ml. Cells were allowed to reaggregate on a gyratory shaker at 63 rpm in a 1.5% CO2 water-saturated incubator at 37 C (33). In some experiments, T3 was added to the culture medium at a final concentration of 0.05 nM. Culture medium was serum-free defined medium as previously described (23, 33). It consists of HEPES/TES-buffered DMEM/F12 1/1 with selenite and ethanolamine (prepared as powder by Invitrogen), supplemented with 0.5% BSA, 5 mg/liter insulin-Zn, 5 mg/liter transferrin, 50 mg/liter streptomycin, 35 mg/liter penicillin, 10 mM ethanol, 1 mg/liter catalase, and 1 g/liter NaHCO3. Final concentration of iron (Fe2+ and Fe3+) was 0.6 µM and of glucose was 1.4 g/liter (7.8 mM).
Treatment of reaggregate pituitary cell cultures with CRH.
On d 2 of culture, aggregates (from 2 x 106 cells) were transferred to 55-mm untreated polystyrene culture dishes in 6 ml fresh defined culture medium. In all experiments, at least three dishes were used for each treatment condition. No substance was added to another three dishes (control condition). The medium (with or without CRH) was renewed three times a week. Because chicken and rat CRH share the same amino acid sequence (34), rat CRH was used in both rat and chicken pituitary cell cultures.
Aggregates were kept in culture for at least 2 wk. Because the availability of the fluorescence-activated cell sorting (FACS) sorter was not fully predictable, the length of treatment could not be exactly fixed. The exact culture time for each treatment is mentioned in Results.
Redispersion of aggregates and isolation of single cells by FACS
After 2 wk of culture, aggregates were pooled per group and redispersed into single cells by treating them sequentially with 2.5% porcine pancreas trypsin, trypsin inhibitor, and EDTA in Ca2+- and Mg2+-free medium as described previously (23). Finally, the cells were mechanically dispersed with a Pasteur pipette, resuspended in serum-free defined culture medium at a concentration of 106 cells/ml, and kept on ice. Cell sorting was done using a FACSVantage (BD Biosciences, Erembodegem, Belgium). Just before loading the cells into the FACS sorter, the cell suspension was diluted 1:50 in FACSFlow sheath fluid (BD Biosciences) or in PBS. Cells were deposited at one cell per well in a 96-well plate with each well containing 5 µl of a lysis solution (0.05% IGEPAL CA-630, 4 U RNasin, 2 mM DTT, 2x FSB), as described in detail previously (23). FACS parameters and gates were set to guarantee random cell selection, with exclusion of cell debris and cell clusters. After filling with single cells, the 96-well plate was centrifuged for 5 min at 3000 rpm (4 C) to bring cell and lysis buffer to the bottom of the well. Cells were further lysed for 30 min on ice and lysates subsequently snap-frozen. Cell lysates were stored at –70 C until further analysis.
Single-cell RT-PCR
The procedures have been described in detail previously (23). Both control and treatment groups were tested in the same RT and PCR run.
RT.
RT was performed on the cell lysates in the 96-well plate. To disrupt secondary structures of the mRNA, samples were first heated for 10 min at 70 C and quickly cooled to 4 C. Then, 7 µl RT reaction mixture containing 4.286 µM random hexamers, 10 U Moloney murine leukemia virus reverse transcriptase, 0.429 mM of each dNTP, 0.4 µl 5x FSB, and 0.29 mM DTT was added to the 5-µl lysate. Negative (no cell) and positive RT controls (10 and 100 pg total RNA extracted from pituitary aggregates) were included.
PCR.
PCR for detection of rat GH, PRL, POMC, GSU, LHß, TSHß, FSHß, and Tpit cDNA was performed using the GeneAmp PCR system 9700 from Applied Biosystems in ThermoFast 96, 0.2-ml 96-tube plates from Westburg (Leusden, The Netherlands). As an internal control for the presence of a cell in the well, PCR was carried out for the ribosomal protein L19. GH/PRL and LHß/TSHß cDNAs were analyzed together in duplex PCRs. Two microliters of the 12-µl RT reaction mixture were added to 8 µl of a PCR mixture composed of 0.75 U AmpliTaq Gold polymerase, 0.188 mM of each dNTP, 3.125 mM MgCl2, 0.875x PCR Gold buffer, and 1.25 µM of sense (S) and antisense (AS) oligonucleotide primers (see Table 1
). For PCR of Tpit, the PCR mixture was composed of 0.25 U Ex Taq Hot Start polymerase (a polymerase giving less ghost amplification or empty box syndrome in negative controls), 0.188 mM of each dNTP, 0.8x Ex Taq buffer, and 1.25 µM of Tpit S and AS primers (see Table 1). All amplification reactions started with a hot start (7 min at 95 C for the hormone cDNAs, 5 min for L19 cDNA, and 1 min for Tpit). The samples were then submitted to 45 cycles of three different temperatures (40 cycles for POMC), i.e. 10 sec at 95 C, 20 sec at the annealing temperature of the primers (60 C for FSHß, LHß, TSHß, GH, and PRL; 55 C for GSU and L19; 59.5 C for POMC; and 54.5 C for Tpit), and 30 sec at 72 C (20 sec for GSU). Finally, samples were kept for 7 min at 72 C. The sensitivity of each PCR analysis was verified by analyzing a dilution series (100, 20, 10, and 5 copies/µl) of full-length purified cDNA copies (for hormones and L19) or of purified 2.1-TOPO vector with the amplified cDNA fragment inserted (for Tpit). As few as 10 cDNA copies per well of GH, PRL, LHß, TSHß, POMC, Tpit, and L19 cDNA, 20 copies of the GSU cDNA, and 100 copies of the FSHß cDNA could be detected. As negative controls for the PCR, 2 µl of the PCR mixture instead of sample was added to the 8-µl PCR mixture. The amplified DNA fragment was visualized, and its length checked, by means of electrophoresis in a 2% agarose gel containing 0.5 µg/ml ethidium bromide. Amplified products from genomic DNA and from cDNA were distinguished by using primers flanking at least one intron. The authenticity of the amplified products was confirmed by restriction enzyme analysis and sequence analysis after cloning the fragment in the 2.1 TOPO vector (see Results). Amplified Tpit fragments were purified with the GenElute gel extraction kit and digested with DraIII for 3 h at 37 C for restriction analysis.
|
GSU and POMC mRNA, using the conditions described above and primers shown in Table 1
. Annealing temperatures were 57.5 C for POMC and 43 C for GSU cDNA amplification. Again, sensitivity was verified during each PCR run by testing a dilution series (100, 20, 10, and 5 copies/µl) of purified 2.1-TOPO vector in which the cDNA of the amplified fragment had been cloned. Ten copies of POMC cDNA and 20 copies of GSU cDNA could be detected.
Specificity and reproducibility of the single-cell RT-PCR procedure has been reported before (23). The negative RT controls included with each PCR run were used to measure the incidence of false-positive results. For rat, the percentages of false positives were 1.3% for GH PCR, 0.8% for GSU PCR, and 0% for all other hormone PCRs. In chicken POMC and GSU PCRs, no false positives were detected. To exclude the possibility that mRNA in the cell suspension, originating from broken cells, contaminated sorted cell samples, cells of each experimental condition were spun down (10 min at 190 x g at 4 C) and the supernatant was diluted and sorted (using electronic test pulses) in the same way as cells had been sorted. RT-PCR was performed for GH, PRL, GSU, and POMC mRNA because these were the most relevant mRNAs in the present study. The incidence of positive PCR bands was 1.03% for GH, 0.76% for PRL, 0% for GSU, and 0% for POMC mRNA.
Fluorescent immunostaining of paraffin-embedded sections
Methods were as previously described (23). Briefly, aggregates were fixed for 2 h in 4% paraformaldehyde, embedded in paraffin, and sectioned in 2-µm slices. Sections were permeabilized with 0.1% bovine pancreatic trypsin for 5 min, preincubated for 1 h in 10% normal goat or donkey serum in Tris-buffered saline (pH 7.6), and incubated overnight with the primary antibody at the final dilutions as mentioned below. Subsequently, sections were incubated with secondary antibodies for 1 h at room temperature and mounted in Vectashield containing 0.5 µg/ml 4',6-diamidino-2-phenylindole (DAPI). For double-immunofluorescent staining, sections were incubated with primary antisera for the hormones of interest simultaneously and then also with the respective secondary antibodies together. As negative control, primary antibodies were omitted. No staining occurred in these controls.
To detect cells containing both POMC and GSU, aggregate sections were incubated simultaneously with the primary antibodies guinea pig antirat GSU (1:1000) and monoclonal mouse antihuman N-POMC (POMC(1–50)) (1:100) and subsequently with TR-labeled donkey anti-guinea pig and FITC-labeled donkey antimouse (1:800) IgGs. The primary antibodies rabbit antirat LHß (1:100,000) and guinea pig antirat GSU (1:1000) were used together to detect cells containing GSU and LHß (gonadotrophs); secondary antibodies were FITC-labeled goat antirabbit (1:400) and TR-labeled donkey anti-guinea pig (1:800) IgGs. For triple immunolabeling of POMC, GSU, and GH plus PRL (simultaneous staining), the above-mentioned primary mouse anti-POMC and guinea pig antirat GSU antibodies were used, whereas GH and PRL were detected using rabbit antirat PRL (1:5000) and antirat GH (1:2000), respectively. As second antibodies, FITC-labeled goat antirabbit IgG (1:400), TR-labeled donkey antiguinea pig IgG (1:800), and goat antimouse aminomethylcoumarin acetic acid-labeled IgG (1:800) were used.
For chicken hormone immunostaining, monoclonal mouse antibodies against chicken GH (1:2000 of ascites fluid), GSU (1:5000), and N-POMC (1:2000) were used, and TR-labeled goat antimouse IgG (1:100) was used as secondary antibody. To detect cells coproducing GSU and POMC, the primary antibodies mouse antichicken GSU (1:5000) and rabbit antihuman ACTH (1:15000) were added simultaneously. Secondary antibodies were goat antimouse TR-labeled (1:100) and goat antirabbit FITC-labeled (1:400) IgGs. Cell nuclei were revealed with DAPI.
The sections were examined on a Leica DMRB microscope (Leica Microsystems, Rijswijk, The Netherlands) equipped with appropriate filter sets: FITC/RSGFP/Bodipy/Fluo3/DiO, TR, and DAPI (Chroma Technology, Rockingham, VT) and coupled to a Nikon DXM 1200 digital camera. Images were analyzed with LUCIA-G 4.71 image capturing software (Laboratory Imaging, Prague, Czech Republic). Section areas and the number of immunoreactive (ir) cells per measured area were determined for 30 captured images per condition at a magnification of 20 x 10, using public domain image analysis software (ImageJ) distributed by National Institutes of Health (Bethesda, MD) (http://rsb.info.nih.gov/ij/). Only cells of which the cytoplasm surrounded a nucleus were counted.
Double-fluorescent immunostaining of vibratome sections
Rats were perfused with PBS and 4% paraformaldehyde in PBS. Postfixation was also in 4% paraformaldehyde in PBS. Fixation time was 1 h at 4 C for intact P14 pituitary and 3 h for adult pituitary, whereas aggregates and P1/P3 pituitary were fixed for 45 min at room temperature. Embedding was done in 2% agarose in PBS. Embedding of cell aggregates was achieved by centrifugation at low speed through the hardening 2% agarose. Sections of 50 µm (sagittal on whole pituitary) were cut on a vibratome (Leica VT 1000S, Leica Microsystems, Wetzlar, Germany) by careful positioning of the agarose blocks. Afterward, sections were stored in PBS at 4 C. Sections were permeabilized by incubating twice for 10 min in 0.5% saponin in PBS. All subsequent solutions were made in PBS complemented with 0.5% saponin, and incubation was on a shaker at room temperature. To block nonspecific binding, sections were incubated in 20% normal goat serum before overnight incubation with primary antisera (monoclonal antihuman-POMC1–50 and guinea pig antirat-GSU). Negative control staining consisted of replacement of the primary antibodies with nonimmune (normal) serum at comparable dilutions. After washing twice in PBS/saponin for 10 min, sections were incubated for 90 min in the dark with goat antimouse Alexa Fluor 488-conjugated IgG and goat antiguinea pig Alexa Fluor 555-conjugated IgG, added at a dilution of 1:1000, and nuclei were counterstained with TOPRO-3 (1:100) for 10 min. After washing, sections were mounted on positively charged glass slides using Vectashield, carefully covered with coverslips and kept at –70 C until confocal laser-scanning microscopy. The microscope used was a LSM510 (Zeiss, Zaventem, Belgium) in multitrack mode. Alexa Fluor 488-conjugated antibody was excited at 488 nm, whereas emission was measured between 505 nm and 550 nm. Alexa Fluor 555-conjugated antibody was excited at 543 nm and measured between 560 and 605 nm (through a NFT545). TOPRO-3 was excited at 633 nm and emission captured above 650 nm.
5-Bromo-2'-deoxyuridine (BrdU) labeling
Entrance of POMC cells in the S-phase of the cell cycle was estimated by incorporation of the thymidine analog BrdU into replicating DNA, identified in the tissue sections by immunofluorescent staining with a specific monoclonal anti-BrdU antibody, using the Cell Proliferation Kit of Amersham Biosciences (Roosendaal, The Netherlands). On the fifth day of culture, the BrdU solution was added to the aggregates for 16 h (1:1000 in the culture medium, as recommended by the provider). Subsequently, aggregates were fixed with 4% paraformaldehyde for 1 h and embedded in paraffin. The 2-µm sections were permeabilized with 0.1% bovine pancreatic trypsin III for 5 min and treated for 1 h with the nuclease/anti-BrdU antibody solution from the kit at room temperature. Then, sections were preincubated with 20% normal goat serum and stained for ACTH as described above. ACTH antiserum was used instead of the N-POMC antibody because anti-BrdU and anti-POMC antibodies were both mouse monoclonal antibodies. Secondary antibodies used to visualize ACTH and BrdU were FITC-labeled goat antirabbit IgG (1:400) and TR-labeled goat antimouse IgG (1:100), respectively. Sections were mounted in Vectashield containing 0.5 µg/ml DAPI and kept at –70 C. All cells staining for ACTH (with FITC-labeled goat antirabbit IgG) also stained for N-POMC (with TR-labeled goat antimouse IgG).
Statistical analysis
Binary logistic regression analysis was performed to explore the influence of CRH on the incidence of the presence of the hormone mRNA of interest in a well (1 = hormone mRNA present; 0 = hormone mRNA absent). A second categorical variable was experiment, added as a fixed covariate to correct for variations due to possible differences between experiments. Statistical significance of the coefficients in the regression equation was contrasted with the Wald test. A difference between conditions was considered significant with a P value of 
0.05. ANOVA was used to compare results of immunocytochemical analyses. Statistical calculations were done using NCSS statistical software.
Presentation of data
A cell type X identified at the hormone mRNA level is indicated as XmRNA cell, whereas a cell type X identified at the immunoreactive (ir) level is designated as X-ir cell.
The number of cells of each type, as identified at the hormone mRNA level, is expressed as percentage of total cells analyzed and ordered in monohormonal and plurihormonal cells. Monohormonal cells are cells expressing only one hormone mRNA (including cells expressing GSU mRNA only). Cells expressing GSU mRNA in combination with TSHß mRNA (thyrotrophs) or GSU in combination with LHß or FSHß mRNA (gonadotrophs) were also considered as monohormonal cells. For data presentation, however, the percentage of LHß and FSHß cells and of cells containing both these hormone gene transcripts were summed and presented together as LHß/FSHß cells. Nonhormonal cells were also included in the figures. Among the bihormonal cells, those expressing GSU mRNA together with GH, PRL, or POMC mRNA and those coexpressing GH and PRL mRNA were the most abundant and are shown. The GSU and POMC mRNA-coexpressing cells included cells that contained these two hormone mRNAs only plus cells (only a very few) that in addition had another hormone mRNA as well. GH/GSU cells and PRL/GSU cells also included a few cells that also contained another hormone mRNA (usually LHß).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GSU mRNA in P14 rat pituitary cell aggregates) but caused a clear-cut rise in the number of cells coexpressing GSU and POMC mRNA (GSU/POMCmRNA cells) (16 cells, 3.6% of total cells), a combination barely present in untreated cells (two cells, 0.5%; control vs. CRH, P < 0.01) (Fig. 1B). After CRH treatment, the GSU/POMCmRNA cells represented a large fraction (29.1%) of the total POMCmRNA cell population. The proportions of cells expressing GSU mRNA together with GH or PRL mRNA did not significantly change (Fig. 1B). In the CRH-treated aggregates, no GH/PRLmRNA cells were found (but not significantly different from control, possibly because of a low number of these cells).
|
GSU and POMC PCRs (CRH-treated cells), including those present together in the same cell, were again subjected to nucleotide sequence analysis. All sequences exactly matched the primer-defined cDNA sequences of both hormone mRNAs (data not shown).
Because T3 has a widespread influence on the development and/or gene expression of all pituitary cell types (37, 38, 39, 40, 41), the effect of CRH was also tested in cell aggregates cultured in medium containing a physiological concentration of T3 (0.05 nM). The study included three independent experiments (15, 18, and 18 d in culture, respectively). In total, 482 cells from the control condition and 447 cells from the CRH condition were analyzed. T3 by itself had no appreciable effect on cell type distribution (compare Fig. 1A with 1C). The lower values for GHmRNA cells in the T3-free condition is because one experiment within this group was performed on aggregates that were kept considerably longer in culture (see above), which lowers the proportion of GHmRNA cells (36). In the presence of T3, CRH did not affect the distribution of the monohormonal cell types (Fig. 1C) but increased the proportion of GSU/POMCmRNA cells from 0.2% (one cell) in control to 6.6% (29 cells, 25 of which did not contain any other hormone mRNA) in the CRH-treated aggregates (P < 0.000001) (Fig. 1D), which is of a considerably higher magnitude than in the absence of T3. When examined within the POMC mRNA cell population, the percentage of GSU/POMCmRNA cells increased from 1.9 to 40.3%, which is also more pronounced than in the absence of T3. The total number of POMC cells (POMCmRNA + GSU/POMCmRNA cells) increased from 9.3 to 15.1% (P < 0.05). The other bihormonal cell types tended to decrease in number in the presence of CRH, although the decrease was statistically significant only when all cell phenotypes were combined (P < 0.05).
Next we investigated whether or not the up-regulation of GSU/POMC cell abundance by CRH is shared by LIF, a pleiotropic cytokine also stimulating POMC expression and POMC cell development (42, 43) but acting through a receptor mechanism different from that of CRH. LIF was used in aggregates cultured for 2 wk at a concentration of 0.5 nM, a maximal dose on growth or differentiation in LIF-sensitive cell lines (42, 44, 45). For economical reasons, these experiments were performed on male P14 rats, littermates of the females used in other experiments. In two independent experiments, no significant effect of LIF could be observed. Only one of 342 cells in control aggregates and one cell of 308 in aggregates treated with LIF contained GSU and POMC mRNA.
CRH has no obvious effect on GSU/POMC coexpression in intermediate lobe cell aggregates from P14 rats
The effect of CRH was also investigated in aggregates prepared from the neurointermediate lobe of P14 rats. These aggregates contain intermediate-lobe POMC cells as well as neural lobe pituicytes (46). Aggregates were cultured in the absence of T3. The effect was tested in three independent experiments (aggregates 13, 14, and 14 d in culture), but the control condition of one experiment was accidentally lost. In total, 243 cells from the control condition and 403 cells from the CRH condition were analyzed. The mRNA detections were limited to POMC, GSU, and L19. As shown in Fig. 2, more than 50% of the cells in the control neurointermediate lobe aggregates were POMCmRNA cells and only 0.8% (2 cells) were GSU/POMCmRNA cells. Surprisingly, this percentage is not lower than in aggregates from the entire pituitary. In the presence of CRH, the proportion of POMCmRNA cells and GSU/POMCmRNA cells was slightly, but not significantly, increased to 61.4% (249 of 403 cells) and 2.1% (nine of 403 cells), respectively. The small change in the GSU/POMCmRNA cell proportion represents an increase from 1.5 to 3.4% within the total population of POMC mRNA-expressing cells, the latter percentage being almost 10 times lower than that seen in aggregates from the entire pituitary (see Fig. 1), suggesting that coexpression of POMC and GSU is mainly induced in anterior lobe cells. In one experiment, cells were also analyzed for GH and PRL mRNA. No PRL mRNA was detected, and only two cells of 137 in the control (one with POMC mRNA) and one cell of 136 in the CRH-treated condition contained GH mRNA.
|
GSU/POMCmRNA cellsGSU mRNA (47). The presence of GSU mRNA and POMC mRNA in the same cell seems contradictory to these observations. We therefore examined the presence of Tpit mRNA in GSU/POMCmRNA cells. A number of cells sorted in the experiments shown in Fig. 1C were analyzed for the presence of mRNA of POMC, GSU, and Tpit. As shown in Fig. 3, cDNA fragments with a length corresponding to that of the Tpit amplicon were clearly detectable in POMCmRNA cells, although not in all of them (detectable in seven of 17 cells). In contrast, Tpit mRNA was not detectable in any of the 40 cells in which only GSU mRNA was present. Tpit mRNA was also detectable in four of 13 cells coexpressing GSU and POMC mRNA.
|
GSU, and Tpit. All sequences exactly matched the sequence of the primer-defined cDNA fragments of the hormone mRNAs and Tpit.
We also attempted to detect expression of other transcription factors essential for differentiation of GSU cells [steroidogenic factor (SF1) and GATA2] (48, 49). SF1 and GATA2 cDNA were detectable in a pituitary RNA extract at a sensitivity of 10 pg and in a cDNA dilution series at a sensitivity of 10 and 20 copies per well, respectively, but in the cell lysate samples, only nonspecific PCR bands were obtained despite several attempts for optimization and using different primer sets (including two different nested PCRs for SF1). We also failed to detect Prop1 mRNA in the single cells despite its being detectable in pituitary RNA extract.
GSU/POMC mRNA-expressing cells in P1 and adult rat pituitary and aggregates
In view of the recently proposed hypothesis that gonadotrophs and corticotrophs share a common precursor (47) and the assumption that the abundance of this precursor may vary during pituitary development, we investigated GSU/POMC coexpression in pituitaries and aggregates from rats of different ages. The mRNA detections were limited to POMC, GSU, and L19. In ex vivo pituitary from adult diestrus females, no cell of 471 cells analyzed (three independent experiments) contained both GSU and POMC mRNA. In P14 pituitary, GSU/POMCmRNA cells were scarce (two of 532 cells; data from three experiments). In contrast, as shown in Fig. 4, in the pituitary ex vivo from newborn female rats, 11 cells of 425 (2.6%) coexpressed GSU and POMC mRNA (two independent experiments, each with a higher number of cells sorted) (P1 vs. P14, P = 0.051), suggesting that GSU/POMCmRNA cells are related to early pituitary development and rapidly regress with advancing age. No false-positive signals were found in control samples of PCR and RT and in cell suspension medium controls, again supporting full confidence of the data. Interestingly, POMCmRNA cells in P1 pituitary accounted for 18.4% of the population ex vivo, which is double the value in the P14 pituitary (P = 0.0013). The opposite was found for GSU cells; their abundance was two times higher in the P14 than in the newborn rat pituitary (P = 0.002).
|
, in cultured aggregates of adult rat pituitary (random cycle females, three independent experiments, 14 d in culture), none of 527 cells tested were found to coexpress GSU and POMC. When cultured in the presence of CRH (0.05 nM T3 in the culture medium), however, GSU/POMCmRNA cells reappeared, reaching 1.5% (eight cells of 523). The latter proportion is four times lower than that found in the P14 female rats (P < 0.001) (Fig. 4). CRH did not raise the proportion of POMCmRNA cells but did increase the total number of POMC cells (POMCmRNA plus GSU/POMCmRNA) (P < 0.05). In control aggregates from newborn rat pituitary, GSU/POMCmRNA cells remained present but in a lower abundance than in the ex vivo pituitary (P = 0.062). CRH restored their number to the ex vivo number, although the effect was not statistically significant (P = 0.093). Importantly, the efficacy of CRH in P1 aggregates was more than three times lower than that in P14 aggregates.
Another remarkable finding in control aggregates compared with the ex vivo pituitary at P1 was that the abundance of GSUmRNA cells was increased by a factor of 2.5, together with a reciprocal change of POMCmRNA cell abundance (P < 0.000001 and P = 0.001 for effect on GSUmRNA and POMCmRNA cells, respectively). CRH counteracted both these changes, although the effect was not statistically significant (P = 0.12 and P = 0.15, respectively). At all ages, CRH tended to lower the abundance of the GSUmRNA, cells but statistical significance was not reached.
GSU/POMC mRNA-expressing cells in chicken pituitary and aggregates
To further substantiate the relevance of coexpression of POMC and GSU mRNA, we tested its occurrence in an evolutionarily distant species, the chicken. Cells were dispersed from chicken pituitary at 14 d after hatching. The mRNA detection was limited to POMC, GSU, and L19. As shown in Fig. 5, cDNA fragments were found with a base pair length expected for the chicken POMC and GSU PCRs. Numerous cells showed colocalization of both mRNAs. Several bands were subjected to nucleotide sequence analysis, and all sequences exactly matched the sequence of the primer-defined cDNA fragments of both hormone mRNAs (data not shown).
|
GSU) signal. About 35% of the wells were hormone positive (either POMC or GSU mRNA or both). As shown in Fig. 6, POMCmRNA cells accounted for 17.6% of the hormone-positive cells and GSUmRNA cells for 73.0%. In 9.4% of the hormone-positive cells, POMC and GSU mRNAs were found together, representing 35% of the total number of POMC cells. This value is considerably higher than that in ex vivo pituitary of the P14 rat.
|
GSU) in the control and of 397 hormone-positive cells in the CRH-treated condition were examined. In control aggregates, 17.8% of the hormone-positive cells were GSU/POMCmRNA cells, which represents more than 40% of the total POMC cell population (Fig 6). The abundance of the GSU/POMCmRNA cells in culture is a clear-cut contrast to that found in the P14 rat. Another difference with the rat is that in culture, 1 nM CRH did not increase the abundance of GSU/POMCmRNA cells (Fig. 6). In contrast, CRH increased the proportion of POMCmRNA cells (P < 0.001) and decreased that of GSUmRNA cells (P < 0.05) (Fig. 6).
Immunofluorescent analysis of expressed hormones in intact pituitary and pituitary cell aggregates
Paraffin sections of intact pituitary from P1 female rats (Fig. 7 A) and of their cultured aggregates (CRH treated in the presence of T3 for 2 wk) (Fig. 7 B) were immunostained for POMC and GSU. Immunoreactive material (ir) was mainly in separate cells, whereas paraffin sections double stained for GSU and LHß showed both antigens in the same cells (yellow staining; Fig. 7 C), as expected. As examined in a confocal laser scanning microscope in vibratome sections, some cells of P1 aggregates (Fig. 7, D and E) and intact P1 pituitary (Fig. 8) showed colocalization of POMC- and GSU-ir (yellow or orange staining). Figure 8 clearly shows the separate images for both POMC-ir (green) and GSU-ir (red) and the overlays. Orange to yellow staining is present in the cells positive for both antigens. The same figure (lower panels) also shows several GSU/POMC-ir cells located in a row close to the pituitary cleft (cl) (remnant of Rathke’s pouch) in a P3 intact pituitary. The GSU/POMC-ir cells were scarce and often less intensely stained than the monohormonal cells, and the relative intensity of green and red stain was variable. Also in paraffin sections of rat P14 aggregates treated with CRH, but not in the control aggregates, some cells containing both POMC- and GSU-ir (yellow staining) (Fig. 7, F and F') were seen. As examined under 20 x 10 magnification on 30 fields in each condition (fluorescence microscopy), about four double-staining cells per field were counted on the average, which is obviously less than the abundance of GSU/POMCmRNA cells, suggesting that most GSU/POMC cells store little hormone. In the P14 aggregates, POMC-ir cells were predominantly present in the periphery, whereas GSU-ir cells were located more centrally, as previously observed (36). In the P1 aggregates, this was less obvious. POMC/GSU double-stained cells were also located in the periphery of the aggregates. In intact P14 and adult female pituitary, no GSU/POMC-ir cells were found.
|
|
), we also analyzed the aggregates by immunofluorescent staining and counted the number of POMC-ir cells per unit section area (in 30 fields of each preparation with 20 x 10 magnification). CRH increased the number of POMC-ir cells with 33 and 60% (in experiments 2 and 3), respectively. Note that in these same cultures, CRH did not affect the proportion of POMCmRNA cells (see Fig. 1C) but raised the proportion of GSU/POMCmRNA cells to 3.7 and 7.1% of all cells, respectively, suggesting that the rise in POMC-ir cells mainly reflects the change in GSU/POMC cells.
Figure 7A shows a section of the intact anterior pituitary from P1 female rats, triple stained for GSU (red), GH and PRL (green), and POMC (blue). Numerous clusters of POMC-ir cells are seen intermingled with smaller clusters of GH-ir, PRL-ir, and GSU-ir cells. After 2 wk in aggregate culture, GSU-ir cell abundance was markedly increased (Fig. 7B, compare with 7A). Many of these cells also stained for LHß (Fig. 7C, yellow cells).
Immunostaining of sections of aggregates (2 wk in culture) from chicks for GH, POMC, and GSU showed staining for each of these hormones in separate cells. Single and clustered cells were seen without a particular regional distribution according to cell type within the aggregates. For double staining of POMC and GSU, rabbit antihuman ACTH and mouse antichicken GSU monoclonal antibody were used simultaneously. Both control and CRH-treated aggregates contained sporadic cells immunoreactive for both ACTH and GSU together (Fig. 7G, yellow staining). Figure 7G (inset) also shows double staining with the monoclonal antichicken POMC antibody and antihuman ACTH antibody; all positive cells were yellow stained, consistent with the presence of both antigens in the same cells and demonstrating the reliability of the antihuman ACTH antiserum to identify chicken POMC cells. The abundance of ACTH-ir cells in chicken aggregates was visibly increased in the CRH-treated aggregates, and counting revealed a rise of 105% above control (P < 0.000001). The proportion of GSU-ir cells slightly (15%) decreased after treatment with CRH (data not shown).
CRH increases BrdU labeling of POMC cells in P14 rat aggregates
Several studies have shown that CRH is mitogenic for POMC cells in adult rats (50, 51, 52). In the P14 rat aggregates cultured in the presence of T3, CRH indeed increased the total number of POMC cells at the end of culture (both at mRNA and immunoreactive level). However, under T3-free conditions, the total number of POMC cells did not increase. We therefore performed a BrdU-labeling assay in the T3-free aggregates from P14 rats. Aggregates were treated from d 2–5 in culture with 1 nM CRH (two independent experiments), with BrdU added on d 4, and BrdU incorporation in POMC cells was examined on d 5 (after 16 h of labeling) using a mouse anti-BrdU monoclonal antibody simultaneously with rabbit antihuman ACTH (see Materials and Methods for rationale). Thirty images in sections of control aggregates and 30 images of CRH-treated aggregates were analyzed in each experiment. CRH increased the percentage of BrdU-labeled cells among ACTH-ir cells more than 3-fold (from 4.0 ± 0.9 to 14.3 ± 1.4%; P < 0.0001). The total number of ACTH-ir cells and of BrdU-labeled cells per unit of section area in the CRH-treated condition was slightly but not significantly increased compared with control (26.5 ± 9.2 and 22.5 ± 6.4% above control).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GSU mRNA with POMC mRNA in rat pituitary ex vivo and in reaggregate cell culture and the occurrence of these cells also in chicken. We show that in the intact rat pituitary, GSU/POMC cells are relatively prominent (2.6%) at birth, rapidly decline within the first 2 wk of life, and can reappear under regulatory pressure of CRH in culture, e.g. in P14 pituitary from approximately 0.5 to 3.5% of total cells in hormone-free culture medium and from approximately 0.2 to 6.6% in T3-supplemented medium. This represents one third or more of all POMC cells. In P14 chicken, the cell type was present in a relevant proportion ex vivo as well as in culture and represented more than one third of all POMC cells, but CRH did not affect its steady-state abundance. Importantly, the large majority of the GSU/POMC cells did not express a ßGSU. Previously, colocalization of GSU- with ACTH-ir has been reported in certain corticotroph adenomas in humans (15, 53, 54, 55) and of ACTH with GSU/TSHß-ir (56) and with GSU/LHß/FSHß-ir (13, 57)in the rat pituitary, particularly during postnatal development, although the latter was not seen by others (58, 59). However, to our knowledge, coexpression of POMC and GSU mRNA without ßGSU has not been described yet in the normal mammalian pituitary, and this coexpression has not been reported for chicken pituitary. The effect of CRH is specific because GnRH (23) and TRH (36) do not show any effect on GSU/POMC coexpression in the same model as used here.
A remarkable aspect of POMC/GSU mRNA coexpression is that it is not paralleled by clearly detectable storage of both GSU and POMC polypeptides except in sporadic cells, suggesting that most of these cells release one or both hormone products constitutively and do not store it or that they do not translate one or both transcript(s). In any event, unlike cells with GSU and LHß, FSHß, or TSHß, GSU/POMC cells do not appear to be cells involved in the acute secretory responses of the pituitary, because that would necessitate hormone storage, clearly visible by immunostaining in fluorescence microscopy.
The present observations are relevant with respect to the development and maintenance of the GSU and POMC lineages. During embryonic development, GSU and POMC are the two earliest pituitary hormone phenotypes in rat (60), mouse (61, 62), and chicken (together with LH) (63, 64, 65). It has been shown that early committed prospective POMC cells are initially capable of reverting to cells destined to express GSU and TSHß (66). Furthermore, it has been well established that expression of Tpit drives Pit-1-independent progenitor cells to terminal differentiation into cortico-/melanotrophs, whereas SF1 and GATA2 expression in these cells dictates differentiation to gonadotrophs (49, 68). More recent research revealed that Tpit gene deletion results in the development of gonadotrophs and Pit1-independent thyrotrophs in areas of both the anterior and intermediate lobe where normally POMC cells develop (47). On the other hand, overgrowth of the POMC cell lineage under pressure of LIF occurs in part at the cost of Pit1-dependent and GSU-expressing cell populations (43, 69). Interestingly, when LIF is overexpressed under the control of the GH promoter, there is a deficiency in Pit1-dependent cells (somatotrophs and lactotrophs) (69), whereas when LIF is overexpressed under control of the GSU promoter, there is hypoplasia of gonadotrophs and thyrotrophs (43). Taking these data together with the present finding of GSU and POMC coexpression, we propose that GSU/POMC cells may be an ambiguous cell type in the branching process of the Pit1-independent progenitors into the GSU and POMC lineages and that some of these ambiguous cells are retained in a reserve cell pool capable of expanding under regulatory pressure by CRH in later life, in this way providing the tissue with a mechanism of postnatal growth, plasticity, and repair. Alternatively or in addition, GSU/POMC cells may arise from already existing GSU cells by starting coexpression of POMC, as already suggested by Childs (56) to explain the presence of ACTH with either TSH or LH in the same cells under certain physiological changes.
To further evaluate the above proposals, we explored in the rat model whether the abundance of GSU/POMCmRNA cells and their expansion in response to CRH is age dependent, reasoning that neonatal animals would have a larger population of these ambiguous cells and/or be more prone to expand it than older animals. In direct support of this reasoning, it was found that the P14 and adult pituitary ex vivo contained only sporadic GSU/POMC cells but that the neonatal pituitary contained 2.6% of these cells, an approximately 10 times higher number. Furthermore, in intact pituitary and aggregates from adult rats, GSU/POMC cells were virtually absent. They reappeared after CRH treatment, but the proportion of the cells amounted to only 1.5%, which is four times less than the proportion found in CRH-treated P14 aggregates or only 19% of the total POMC cell population instead of 44% in the P14 aggregates. However, an unexpected finding was that in P1 aggregates, CRH was also less effective on GSU/POMC cells than in aggregates from P14 rats (2.6% of all cells at P1 vs. 6.6% at P14). A possible explanation for this finding may be found when looking at the developmental course of POMC-only and GSU cells between P1 and P14 in vivo. In P14 pituitary, GSU cell abundance was double as high as in P1 pituitary, whereas the opposite was found for POMC-only cells (see Fig. 4). It is known that during the first 2 wk of life, there is an impressive growth rate of GSU cells in the direction of gonadotrophs, particularly in female rats (70, 71, 72, 73), whereas the CRH-POMC-adrenal axis is highly active in late fetal life and at birth (74, 75) to become hyporesponsive between P2 and P12, the stress-hyporesponsive period (76, 77). The relatively low responsiveness to CRH at birth in terms of GSU/POMC cell production may be the consequence of an intrinsic drive within the pituitary at birth to preferentially develop gonadotrophs (70, 71, 72, 73). This intrinsic drive was clearly manifested by P1 and to some extent P14 pituitary cells in culture. Culturing of P1 cells resulted in a clear-cut hyperplasia of GSU cells and gonadotrophs, but culturing of adult pituitary cells did not. Thus, it is plausible that the intrinsic developmental drive toward the GSU/LHß lineage at P1 may counteract the action of CRH in generating GSU/POMC cells. Interestingly, although the abundance of GSU cells increased 2.5-fold in culture of P1 aggregates, that of POMC cells decreased with a comparable extent, and GSU/POMC cell abundance also lowered, suggesting the idea that expansion of the GSU population may occur at the expense of both the POMC and GSU/POMC cells. Obviously, this process would counteract the positive action of CRH on the generation of GSU/POMC cells.
Another interesting observation was that up-regulation of GSU/POMC cells by CRH was an effect not shared by LIF, a cytokine well known to strongly stimulate POMC expression (78), to promote POMC cell differentiation and development (43, 69), and to potentiate the action of CRH on the pituitary-adrenal axis (42, 79). The failure of LIF to affect the abundance of GSU/POMC cells may be related to its specific mode of action, which can widely differ according to the cell type involved. For example, LIF inhibits proliferation of the POMC cell line AtT20 (80), whereas it stimulates proliferation of the lactosomatotroph cell line GH3 (81).
The GSU/POMC cells were additionally characterized by exploring the expression of the cell-specific transcription factor Tpit (TBX19). In mice and humans, Tpit is essential for terminal corticotroph and melanotroph cell differentiation during embryogenesis and for POMC gene expression throughout life (47). We therefore looked in the rat model to determine in which cell type Tpit was expressed. As expected, Tpit mRNA was not detected in GSUmRNA cells and was found in POMCmRNA cells although not in all of them. Unexpectedly, we also found Tpit coexpressed in GSU/POMCmRNA cells. This finding raises an important issue because it is currently believed that Tpit represses the GSU gene (47). Our data indicate that Tpit is not always preventing GSU mRNA expression. A possible explanation is that Tpit is expressed in the GSU/POMC cells at a level too low to repress GSU expression or that Tpit cooperates with another factor in repressing the GSU gene and that this factor is absent in GSU/POMC cells. Alternatively, GSU/POMC cells may contain a factor disabling the repressive action of Tpit on GSU mRNA expression. The question whether GSU/POMC cells also express the GSU cell-specific transcription factors SF1 or GATA2 could not be addressed because of technical problems. Interestingly, Tpit expression has recently been detected not only in POMC-expressing carcinoid tumors but also in many POMC-negative carcinoid tumors (82), and almost half of the carcinoid tumors also express GSU (83), again emphasizing that Tpit is not an absolute POMC cell-specific transcription factor and that these genes are not strictly mutually exclusive in expression.
We also addressed the mechanism underlying the expansion of the GSU/POMC cells by CRH. It has been reported that prolonged exposure to CRH by chronic treatment with the peptide or in CRH-overexpressing transgenic mice results in POMC cell hyperplasia (50, 84). The peptide is well known to stimulate corticotroph proliferation (51, 52) in more acute experiments. In agreement with a proliferative action, we found in the P14 rat aggregates that CRH provokes a more than 3-fold rise in the BrdU labeling index of POMC-ir cells during the first week of culture, even under conditions where no net growth of the total POMCmRNA population was seen (culture without T3). Because in the T3-free condition the GSU/POMCmRNA cells and not the POMCmRNA cells increased in number, CRH may be mitogenic for the GSU/POMCmRNA cells (or their progenitors provided there is postmitotic differentiation), although it remains impossible to conclude that with certainty, because the BrdU assay is done on cells identified on the basis of POMC-ir material. In view of the rather strong CRH-induced DNA replication in POMC-ir cells compared with the net increase in GSU/POMCmRNA cells, we presume that CRH might also increase POMC cell loss or that the effect on cell proliferation is only temporary. In the chicken aggregates, CRH did increase the proportion of POMCmRNA cells and decreased the proportion of GSUmRNA cells, but it did not raise the proportion of GSU/POMCmRNA cells. These data suggest that in the chicken, CRH may increase the number of POMC cells at the expense of the GSU cells, possibly through transdifferentiation from GSU cells to POMC-only cells.
Some GSU/POMCmRNA cells were also found in the rat neurointermediate lobe cell aggregates from P14 rats. Aggregates from this lobe consist of melanotroph POMC cells and neural lobe pituicytes (46). It is unlikely that these GSU/POMCmRNA cells are contaminating cells from the anterior lobe because the intermediate lobe can easily be dissected from the anterior lobe because of the presence of a cleft between these lobes. Furthermore, the proportion of GSU/POMCmRNA cells found in the neurointermediate lobe aggregates was not lower than in the entire pituitary aggregates, and GH or PRL cells, which are more numerous, were virtually absent. CRH did not convincingly increase the proportion of the GSU/POMCmRNA cells in the neurointermediate lobe aggregates. Taken together, these data suggest that GSU/POMC cells also exist in the intermediate lobe but are not or are less sensitive to CRH despite the presence of CRH receptors in the intermediate lobe (67).
In conclusion, we describe a remarkable cell type in the rat anterior pituitary coexpressing GSU and POMC mRNA, which is relatively abundant at birth, rapidly declines in occurrence postnatally, and reappears in vitro under regulatory pressure of CRH. This cell phenotype is also present in chicken pituitary but does not seem to be dependent on CRH in this species. Most of these cells do not store hormone products, suggesting they are not involved in acute secretory responses of the pituitary. The development and regulatory dynamics of these cells suggest a transition relationship between the POMC and GSU lineage not only during embryonic development but also in postnatal life. Whether GSU/POMC cells have a relationship to (the sexually dimorphic) development of gonadotrophs during the first 2 wk of life in the rat is an intriguing hypothesis for future research.
|
Acknowledgments |
|---|
GSU, GH, PRL, LHß, and TSHß and antihuman ACTH antisera and Prof. L. Berghman (Texas A&M University, Department of Poultry Science and Veterinary Pathobiology) for kindly providing antichicken GSU, POMC, and GH antisera. Koen Van Acker, Erik Nys, and Victor Van Duppen are kindly acknowledged for expert assistance with single-cell sorting by FACS. Kristine Rillaerts is greatly acknowledged for skillful technical assistance. We thank Dr. Pieter Van As from the laboratory for Physiology and Immunology of Domestic Animals at this university for providing chicken pituitaries. |
Footnotes |
|---|
Disclosure statement: The authors have nothing to disclose.
First Published Online July 13, 2006
Abbreviations: AS, Antisense primer; BrdU, 5-bromo-2'-deoxyuridine; DAPI, 4',6-diamidino-2-phenylindole; DTT, dithiothreitol; FACS, fluorescence-activated cell sorting; FITC, fluorescein isothiocyanate; FSB, first-strand buffer; GSU, glycoprotein hormone -subunit; ir, immunoreactive; LIF, leukemia inhibitory factor; P14, postnatal d 14; POMC, proopiomelanocortin; PRL, prolactin; S, sense primer; SF1, steroidogenic factor 1; TR, Texas Red.
Received May 22, 2006.
Accepted for publication July 6, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit: clinicopathological, immunohistochemical, ultrastructural, and immunoelectron microscopic studies in nine cases. Neurosurgery 26:397–403[Medline]
3-melanocyte-stimulating hormone detected through intracellular Ca2+ responses in immature rat pituitary constitute a fraction of all main pituitary cell types, but mostly express multiple hormone phenotypes at the messenger ribonucleic acid level. Refractoriness to melanocortin-3 receptor blockade in the lacto-somatotroph lineage. Endocrinology 140:4874–4885-subunit genes by gonadotropin-releasing hormone and estradiol-17ß in single rat pituitary cells during aggregate cell culture. Endocrinology 144:388–399 inhibits stimulated adrenocorticotropin, prolactin, and growth hormone secretion in normal rat anterior pituitary cell cultures. Endocrinology 126:2919–2926-subunit gene in pituitary gonadotropes. Mol Endocrinol 8:878–885-subunit gene in the placenta and pituitary gland. Mol Cell Biol 14:5592–5602-subunit production and plurihormonality in human corticotroph tumours: an in vitro and immunohistochemical study. Eur J Endocrinol 133:25–32-subunit in the pars tuberalis of the rat pituitary gland during ontogenesis. Neuroendocrinology 58:616–624[Medline]-subunit mRNA of glycoprotein hormones during the chick pituitary organogenesis, with special reference to the pars tuberalis. Cell Tissue Res 299:71–80[Medline]-dihydrotestosterone and 3-androstanediol formation during postnatal development and puberty in the rat. Endocrinology 94:1265–1274-subunit of glycoprotein hormones. J Clin Endocrinol Metab 82:2622–2628This article has been cited by other articles:
 |
|
 |
|
  K. Janssens, O. Krylyshkina, N. Hersmus, H. Vankelecom, and C. Denef {beta}1-Adrenoceptor Expression in Rat Anterior Pituitary Gonadotrophs and in Mouse {alpha}T3-1 and L{beta}T2 Gonadotrophic Cell Lines Endocrinology, May 1, 2008; 149(5): 2313 - 2324. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.001 vs. ex vivo; 
