| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Discipline of Physiology School of Molecular and Biomedical Sciences (K.F., I.C.M., J.S.), University of Adelaide, Adelaide, Australia; and Department of Biology (S.T.), Faculty of Science, Shizuoka University, Ohya 836, Shizuoka 422-8529, Japan
Address all correspondence and requests for reprints to: Jeffrey Schwartz, Discipline of Physiology, School of Molecular and Biomedical Sciences, University of Adelaide, North Terrace, Adelaide 5005, Australia. E-mail: jeff.schwartz{at}adelaide.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The increases in plasma levels of ACTH1–39 and precursors after CRH stimulation undergo different age-related changes during the last third of gestation (2), and there is in vitro evidence that individual corticotrophs of the late gestation fetal pituitary are differently responsive to CRH (5). It is not known, however, if the different ontogenic profiles are linked to changes in the processing of POMC to ACTH in individual cells that are differently responsive to CRH.
In addition, placental insufficiency imposes developmental burdens on the fetus that are also associated with altered HPA activity during the prepartum period. For example, fetal cortisol levels are elevated, whereas POMC transcript levels are lowered without changes to the fetal plasma levels of ACTH1–39 or total immunoreactive ACTH (6).
Despite these characterizations of net changes in vivo and in pituitaries as a whole, virtually nothing is known of changes to corticotrophs at the level of individual cells, as a function of either gestational age or placental insufficiency.
At the level of individual cells, corticotrophs display heterogeneity. In situ hybridization has identified fetal pituitary cells that express the transcript for POMC but not the transcript for the cleavage enzyme prohormone convertase (PC) 1 (7). Furthermore, the proportion of POMC mRNA-expressing cells that also express PC1 mRNA increases across late gestation (7). In addition, several in vitro studies have now identified subpopulations of corticotrophs that differentially respond to CRH and other signals in the rat (8), adult sheep (9), and fetal sheep (5, 10). This has allowed investigators to identify a differential response to chronic placental insufficiency in separate subpopulations of corticotrophs responsive and nonresponsive to CRH (10).
This study therefore aimed to identify whether all corticotrophs express the peptides for both full-length POMC and the ACTH fragment, as well as whether each corticotroph contained the molecular machinery for response to CRH, namely the CRH type 1 receptor (CRHR1), in the late gestation fetal sheep pituitary. We also identified the effects of gestational age and placental insufficiency on the expression of POMC, ACTH, and CRHR1 in individual pituitary cells in the late gestational fetal sheep. To investigate one of the most HPA-active windows of gestation, tissues were collected from fetuses during the fortnight prepartum. It was reasoned that a broad gestational age window at a time when there were exponential changes in physiological parameters of HPA activity might mask rapid changes in pituitary architecture. Therefore, fetuses were divided into groups at 139–141 and 144–145 d gestation (where term is 
150 d gestation in this breed). Specifically, we hypothesized that changes in pituitary function associated with normal prepartum development and placental insufficiency, such as processing of POMC to ACTH, and the capability to respond to CRH, would be reflected in differences among individual corticotrophs in immunohistochemical labeling for high- and low-molecular-weight fragments of the POMC peptide and CRHR1.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Uterine carunclectomy
In a subset of nonpregnant ewes [placental restriction (PR) group], the majority of endometrial caruncles were surgically removed from the uterus as described previously (11). This procedure restricts the number of placental cotyledons formed and subsequently limits placental and, hence, fetal, growth. The carunclectomy procedure was performed under aseptic conditions with general anesthesia induced by an iv injection of sodium thiopentone (1.25 g/ml; Boehringer Ingelheim, New South Wales, Australia) and maintained with 3–4% halothane in oxygen. Ewes were kept under observation for 4–7 d postsurgery. After a minimum of 10 wk recovery, these ewes, plus additional control ewes, entered a mating program, and twin pregnancies were confirmed by ultrasound at approximately 50 d gestation.
Pituitary collection
Ewes carrying PR or control fetuses were killed by an iv overdose of sodium pentobarbitone (200 mg/kg; Lethobarb; Virbac Pty Ltd., New South Wales, Australia). Both twin fetuses were delivered via laparotomy, weighed, and killed by decapitation. Pituitaries and adrenal glands were collected and weighed. In this flock, parturition occurs at 150 ± 3 d gestation
For immunohistochemistry, pituitaries from four PR fetuses and four control fetuses at 140 d and from four PR and six control fetuses at 144 d gestation were collected intact and immediately fixed in 4% formaldehyde in 0.1 M PBS at 4 C for 24 h. Tissues were washed twice in PBS, then dehydrated in 70% ethanol. Whole pituitaries were bisected in the coronal plane with the neurointermediate lobe kept intact so that the pars nervosa and pars intermedia could be used as negative and positive control tissues, respectively. Pituitaries were then processed into paraffin wax blocks, and sections (5 µm) were cut at 100-µm intervals and collected onto poly-L-ornithine-coated (Sigma, St. Louis, MO) glass slides. Slides were photobleached under a 50-W halogen light bulb for 24 h to reduce autofluorescence.
For Western blots, pituitaries from two control fetal sheep at 140 d gestation tissue were collected separately into liquid nitrogen and extracted independently for the Western blot analysis. Samples of the tissues (50 mg) were homogenized in 250 µl radioimmunoprecipitation assay lysis buffer [50 mM Tris-HCl, 150 mM NaCl (pH 8.0), 0.1% Nonidet P-40, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 1 mM disodium EDTA, and protease inhibitor cocktail] on ice. After homogenization, samples were centrifuged at 10,000 x g for 5 min at 4 C, and the clear supernatant was collected. Protein concentrations were estimated by Bio-Rad protein assay (Bio-Rad, New South Wales, Australia).
Western analysis
Protein (20 µg/well) samples from the extracts of the two fetal pituitaries were resolved in separate lanes using 4–15% gradient-ready polyacrylamide gel (Bio-Rad) and then transferred to polyvinylidene difluoride membranes (Bio-Rad). Membranes were blocked with Tris-buffered saline containing 5% (wt/vol) nonfat dried milk and incubated with POMC or ACTH antisera (1:250). Membranes were subsequently washed and incubated with either horseradish peroxidase-coupled sheep antirabbit (1:3000, Silenus Laboratories, Victoria, Australia) or goat antimouse (1:2000, Sigma) antisera. Bound peroxidase activity was detected by the enhanced chemiluminescence system (GE Healthcare, Buckinghamshire, UK) and visualized by exposure to radiographic film (Hyperfilm ECL, GE Healthcare) for 1 min for bands above 10-kDa molecular mass and for 5 min to improve the visualization of the smallest peptide bound by anti-ACTH. Precision Plus dual color protein standards (Bio-Rad) were used as molecular weight markers.
Specificity of primary antisera binding to membranes was confirmed by preincubation of each antisera with its corresponding blocking peptide (as detailed for the immunohistochemical protocol below) and nonspecific binding of the secondary antisera to membranes determined by omission of the primary antisera.
Immunohistochemistry
Pituitary sections were rehydrated in histolene (Fronine, New South Wales, Australia) and a series of 100, 90, and 70% ethanol before washing in PBS three times for 5 min each. Antigen retrieval was performed in Tris-HCl buffer [0.1 M (pH 6.6)] for 10 min at 121 C. Sections were incubated for 30 min in blocking solution (PBS, 0.01% azide, 10% normal donkey serum) at room temperature. Three-color immunofluorescence was performed by incubating the sections for 24 h at 4 C with an antibody cocktail containing: mouse anti-ACTH IgG (1:50, Dako, Glostrup, Denmark), goat anti-CRHR1 IgG (1:50, Santa Cruz Biotechnology, Santa Cruz, CA), and rabbit anti-POMC IgG (ST-1, 1:100). The POMC antibody is specific for the full-length protein being raised against the ST-1 fragment, a nonapeptide KLEFKRELE spanning the cleavage site between murine ACTH and ß-LPH (12). The ACTH antibody was raised against residues 24–39 of ACTH1–39 and therefore recognizes epitopes at the carboxyl terminus of ACTH1–39. After washing in PBS, slides were incubated for 2 h at room temperature with secondary antisera conjugated to distinct fluorophores: Cy3-conjugated donkey antimouse IgG, Cy5-conjugated donkey antigoat IgG, and Cy2-conjugated donkey antirabbit IgG (1:200, 1:100, and 1:100, respectively; Jackson ImmunoResearch, West Grove, PA). The slides were washed in PBS and incubated with 3 µM 4,6-diamidino-2-phenylindole (DAPI; Molecular Probes Inc., Eugene, OR) before a final wash in PBS. Cover slips were attached with antifade fluorescent mounting medium (Dako).
In addition to tissue controls, the specificity of the staining for the antisera raised against POMC, ACTH, and CRHR1 was confirmed by preabsorption with the ST-1 (the same POMC fragment as used to raise the POMC antibody), ACTH1–39 (AUSPEP, Victoria, Australia), and CRHR1 (Santa Cruz Biotechnology) peptides, respectively. Substitution of the monoclonal ACTH with another monoclonal antibody against a protein not expressed by the pituitary was also performed to confirm the specificity of the antibody raised against ACTH. To confirm that there was no binding of secondary antibodies to inappropriate primary antibodies and that the microscope filters could accurately discern all three fluorophores, each primary antibody was incubated separately with a tissue section before incubation with the secondary antibody cocktail.
Epifluorescent imaging of corticotrophs
Sections of pituitary were examined under an epifluorescence microscope (AX70, Olympus, Tokyo, Japan) attached to a digital camera (Photometrics Cool Snap Fx, Roper Scientific, Tucson, AZ). Eight-bit grayscale images were captured for each of the fluorescent labels resulting in four images per field. Ten to 15 nonoverlapping fields were captured randomly from each pituitary section at 400x magnification using V2+ (Total Turnkey Solutions, New South Wales, Australia). Images were saved in tagged image file format.
Confocal imaging of corticotrophs
Slides from each gestational age and treatment group were also visualized on a Bio-Rad MRC1000uv laser scanning microscope built around a Nikon DIAPHOT 300 inverted microscope (Nikon, Tokyo, Japan). DAPI signals were imaged using excitation of 351-/363-nm laser and emission with 460 long pass filter. Cy2 was detected using a 488-nm laser for excitation and imaged using a 522/35 band pass filter. Cy3 was detected using the 514-nm laser for excitation and imaged through a 585 long pass filter. Cy5 was detected using the 647-nm laser for excitation and imaged through a 680/32 band pass filter. All images were collected using a 40x/1.15 numerical aperture water immersion objective lens, and individual Z series of images were collected through the depth of cells using Z step of 0.5 µm.
Quantification
The grayscale threshold representing the positive area for each field was determined using the image analysis software, analySIS (Soft Imaging Systems, Münster, Germany). Numerous corticotrophs were found within contiguous clusters; therefore, the clusters were categorized by size, and the relative proportion of clusters in each category was determined for each gestational age and treatment group. To determine the number of individual corticotrophs within clusters of cells, the cross-sectional area of 50 randomly chosen corticotrophs at each gestational age in each treatment group was measured.
The total number of cells in each field was identified from the DAPI-stained nuclei. The number of cells positively stained for each antigen in each field was counted. Then, the cells were categorized into subpopulations depending on the combination of POMC, ACTH, and CRHR1 expressed by each cell, and total cell counts were determined for each phenotypic subpopulation. The accuracy of the automated quantification method was validated against manual cell counts for the total number of cells in each of the phenotypic subpopulations in five nonoverlapping fields. Overall, an average of 10 fields from three sections of each pituitary from four to six animals per group was counted, totaling over 100,000 cells counted per group.
Statistical analysis
A corticotroph was defined as any pituitary cell that stained positively for POMC or ACTH, and this total corticotroph population is presented as a proportion of the total number of pituitary cells. Each subpopulation of corticotrophs is presented as a proportion of pituitary cells to identify the effects of gestational age and placental restriction on each subpopulation. Because the corticotrophs make up only a small proportion of the total pituitary cells (10%) and each subpopulation of corticotrophs is subsequently smaller, we expressed the effect of changes in corticotroph subpopulations within the approximately 100,000 cells that make up the corticotroph population. Thus, the effects of gestational age and PR on the relative contribution each subpopulation made to the whole corticotroph population was identified by investigating the changes in the proportion of corticotrophs in each phenotypic subpopulation. In all cases, results are expressed as the mean ± SEM.
A two-way ANOVA with gestational age (140 vs. 144 d gestation) and treatment (PR vs. control) as the specified factors was used to determine whether there were differences between different groups in the average cross-sectional areas of the corticotrophs, the frequency of clusters of different sizes, and the proportion of cells in the different subpopulations of corticotrophs. Differences between gestational age groups or treatment groups are considered significant at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). As expected, PR fetuses were significantly smaller in weight than control fetuses at both gestation ages (Table 1, P = 0.0001, F = 162.947). Pituitary and adrenal weights were significantly lower in the PR fetuses than the control fetuses at both gestational ages (Table 1, pituitary weight, P = 0.0001, F = 22.099; adrenal weight, P = 0.030, F = 6.065), although as a proportion of body weight, pituitary and adrenal weights were significantly higher in the PR fetuses (Table 1, pituitary/fetal weight, P = 0.001, F = 15.603; adrenal/fetal weight, P = 0.001, F = 19.387).
|
, lanes 1 and 3, extract of one fetal pituitary; lanes 2 and 4, extract of a second fetal pituitary). The antibodies produced very discrete and unambiguous bands with no smearing. The results indicate that anti-POMC (lanes 1 and 2) strongly recognized an antigen of the appropriate molecular masses (31 kDa). In addition, the POMC antibody produced a band at approximately twice the molecular mass of POMC and a much weaker staining of an unidentified band at approximately 19 kDa but no staining of any bands less than 10 kDa even under extended exposure of the film to the Western membrane. In contrast, anti-ACTH (lanes 3 and 4) produced staining of bands at entirely different molecular masses, with no binding whatsoever to peptides above 25-kDa molecular mass, two distinct bands between 10 and 25 kDa, and a single band less than 10 kDa that became obvious after extending the exposure time to 5 min. All visible binding in Fig. 1 was eliminated by the preabsorption of primary antisera with the relevant peptide and omission of the primary antisera (data not shown).
|
), ACTH (Fig. 2C), and CRHR1 (Fig. 2D) were found to be localized to the cytoplasm of anterior pituitary cells. There was no staining for any of these peptides in the neural lobe of the pituitary (Fig. 2, B–D). The cells of the intermediate lobe stained positively for POMC, ACTH, and CRHR1, (Fig. 2, B–D). No specific staining of anterior pituitary tissue was observed for the preabsorption controls for CRHR1 (Fig. 3), ACTH, or POMC, nor for the replacement control for ACTH (data not shown). None of the primary antisera were found to cause fluorescent signals that indicated inappropriate cross-reactions with secondary antisera or bleed-through of signals through the microscope filters. Confocal microscopy revealed all three antigens were present throughout the entire cytoplasm. Localization of CRHR1 to the cell membrane could not be confirmed in any cells at any gestational age, in controls or PR fetuses, because the interface between the cytoplasm and cell membrane could not be identified. There was no alteration to the cellular localization of any antisera staining when the primary antisera were incubated with tissue sections separately instead of incubation with the primary antisera cocktail.
|
|
). There was no independent effect of treatment on the proportion of clusters in any size category (Fig. 4).
|
, white arrow). There were also substantial numbers of cells that stained positive for ACTH and CRHR1 but not POMC (ACTH + CRHR1; fuchsia arrow in Fig. 5) and a further set of cells that were positive for POMC but not ACTH or CRHR1 (POMC only; green arrow in Fig. 5). We have therefore identified three major phenotypic subpopulations of corticotrophs: POMC + ACTH + CRHR1, ACTH + CRHR1, and POMC only. The proportions of pituitary cells exhibiting these phenotypes are summarized in Table 2 at 140 and 144 d gestation for the PR and control groups.
|
|
).
Effect of placental restriction on the proportion of corticotrophs in each phenotypic subpopulation
There was no effect of placental restriction on the proportion of corticotrophs present in the fetal pituitary at either 140 or 144 d gestation. PR fetuses underwent the same decline in proportion of pituitary cells of the POMC + ACTH + CRHR1 phenotype (Table 2) as the control animals between 140 and 144 d gestation. Within the corticotroph population however, PR caused a significant shift in the distribution of subpopulations, resulting in a significant conservation (P < 0.05, F = 6.014) of the corticotrophs expressing POMC + ACTH + CRHR1, independent of the effects of gestation age (Fig. 6).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Differential expression of POMC, ACTH, and CRHR1 in individual cells
We have identified a substantial population of cells expressing ACTH, but expressing no POMC or too little POMC to be detected by immunohistochemistry (i.e. the ACTH + CRHR1 subpopulation). These cells are likely to be processing the POMC to ACTH at a greater rate than those cells stained positively for both POMC and ACTH (i.e. the POMC + ACTH + CRHR1 subpopulation). Such differential rates of processing of POMC to ACTH could be due to the differential activation of PC1, which is produced as an 87 kDa peptide and activated to a 66 kDa form in acidic secretory vesicles (13). We have also identified cells expressing POMC, but too little ACTH to be detected by immunohistochemistry (i.e. the POMC-only subpopulation). These cells obviously contain the full-length POMC peptide but may not cleave the POMC into ACTH, consistent with the lack of PC1 transcript in some POMC mRNA-expressing cells of the fetal sheep pituitary (7). Thus the three major subpopulations of corticotrophs: POMC-only-, POMC + ACTH + CRHR1-, and ACTH + CRHR1-expressing cells may differentially express and activate PC1 leading to the differential expression of POMC and ACTH.
Alternatively, it is also possible that individual cells observed to differentially express POMC and ACTH may have different storage capacities for POMC and ACTH (14). In either case, this represents a capacity within the fetal pituitary to separately regulate the secretion of the two peptides by stimulating individual cells. The precursor forms of ACTH are secreted by fetal pituitaries in vivo, and the present results suggest that plasma POMC levels may be maintained by secretion of POMC from cells that express POMC only and/or POMC + ACTH + CRHR1, whereas ACTH levels may be regulated by ACTH secretion from ACTH + CRHR1 and/or POMC + ACTH + CRHR1. Interestingly, a previous study using the same POMC antibody in pituitary cultures from adult sheep demonstrated that nearly half of the cells that stain positive for POMC also secreted POMC in unstimulated conditions (15).
The association of CRHR1 with corticotrophs that potentially process POMC to ACTH at a faster rate suggests a further association between CRHR1 and the activity of PC1. CRH stimulation of cultures of the murine cell line, At-20, has been shown to cause secretion of the 66-kDa active form of PC1, whereas unstimulated cultures secrete only the 87-kDa precursor of PC1 (13). Therefore, CRH may stimulate POMC processing via activation of PC1.
Although we found that 12% of pituitary cells express the CRHR1 peptide at 140 d gestation in control animals, fewer than 2.5% of pituitary cells bind CRH in cultures from fetal sheep at the same age (16). It must be noted that in the current study, the CRHR1 peptide was unexpectedly localized to the cytoplasm in all CRHR1-positive cells. Confocal microscopy was used to confirm the cytoplasmic localization of CRHR1; however, this cytoplasmic localization made it impossible to determine whether CRHR1 was localized to the cell membrane as well in all of the CRHR1-expressing cells. Therefore, it is possible that some of the cells that stained positive for CRHR1 contain only cytoplasmic CRHR1 and no membrane bound CRHR1, in which case they are likely to be unresponsive to CRH. Differential intracellular localization of the CRHR1 protein between individual cells may underlie the known differential regulation of CRHR1 protein levels, and CRH binding (17).
There are other explanations for cytoplasmic localization of CRHR1. Some of the cytoplasmic CRHR1 identified in this study may represent receptors that were recently bound by CRH and are undergoing internalization and degradation since binding of CRH to CRHR1 has been shown to cause the complex to be internalized via endocytosis (18, 19, 20) and processed in lysosomes (20). Some of the cytoplasmic CRHR1 identified in this study may be a mobilizable source of CRHR1 sequestered to the cell membrane when CRH stimulation is excessive. This may explain why 48 h of CRH pretreatment of pituitary cell cultures from 142–144 d gestation fetal sheep reduces subsequent CRH binding by only 40–50% (21). In addition, the internalized CRHR1 identified in this study may be regulated by the high plasma cortisol levels know to exist at this stage of gestation (1) because glucocorticoid pretreatment of rat pituitary cell cultures reduces the proportion of cells that bind CRH by 70% (22).
The differential staining of individual cells with the POMC and ACTH antibodies indicates that the antibody used to identify the POMC and ACTH peptides specifically reacted with different molecules. The antibody used in this study to identify POMC recognizes a sequence of POMC spanning the cleavage point between pro-ACTH and ß-LPH. The specificity of this antisera to detect POMC and not pro-ACTH or ACTH has been validated previously in AtT-20 cells (23). The Western blot results clearly demonstrate that the POMC antisera reacts intensely with peptides at approximately 70 and approximately 31 kDa and reacts to a far lesser extent with a smaller peptide. This indicates that the POMC antibody reacted predominantly with full-length POMC and possibly a POMC homodimer, which has been reported previously to be present in pituitaries (24). Interestingly, the antibody produced a faint band at approximately 19 kDa, which was completely eliminated by preincubation with the antigenic peptide, suggesting the possible existence of a POMC fragment formed by cleavage at site(s) other than that between pro-ACTH and ß-LPH. The notable cross-reaction of the POMC antisera with an approximately 19-kDa peptide was unexpected given the results of studies in AtT-20 cells (23) but may indicate a different sequence of processing steps between the cell line and fetal sheep pituitary cells; elimination of this band in preabsorption and primary omission controls suggests it was not a nonspecific reaction. Our monoclonal ACTH antibody was raised against residues 24–39 of ACTH1–39 and is therefore specific for a single epitope at or near the carboxyl terminus of ACTH1–39 (25). Previous investigation suggests this ACTH antibody is capable of binding to ACTH precursors such as POMC and pro-ACTH at high concentrations in fractions of human serum (26). In the present study, Western blot analysis demonstrated that the ACTH antibody reacted with peptides of molecular weights consistent with pro-ACTH and unglycosylated pro-ACTH. It is possible that the free carboxyl terminus present in these peptides may be required for epitope recognition, resulting in a high degree of specificity of this antibody for these POMC fragments. Critically, these two antisera recognize different peptides, suggesting that the differential immunohistochemical staining of individual cells by the POMC and ACTH antisera is indicative of individual cells expressing different POMC-related peptides.
The negative staining for POMC, ACTH, and CRHR1 in the neural lobe of the pituitary is consistent with previous reports (27, 28). In addition, negative results for the preabsorption and replacement controls indicate that the staining seen in the anterior lobe was specific for all three peptides.
In the intermediate lobe, intense POMC and lesser ACTH staining was present in most cells. Previous in situ hybridization has indicated intense POMC mRNA expression in nearly all of the cells of the intermediate lobe in the near-term fetal sheep and intense immunohistochemical staining using a promiscuous ACTH antibody (27) indicates that all cells of the intermediate lobe express an ACTH-like peptide. The intermediate lobe was stained intensely for CRHR1, consistent with previous reports of immunohistochemical staining for CRHR1 (28) and in vitro evidence of secretion of ACTH (not 
MSH or corticotropin-like intermediate lobe peptide) in response to CRH (29).
Impact of gestational age on corticotroph subpopulations
Between 140 and 144 d gestation, a short gestational window near term, there was a marked decrease in the proportion of pituitary cells expressing POMC + ACTH + CRHR1, which may refine our understanding of the maturational reduction in: 1) the proportion of pituitary cells identified as corticotrophs (30, 31), 2) the proportion of corticotrophs secreting ACTH in response to CRH (30), and 3) the proportion of pituitary cells that bind CRH (16) over the last third of gestation.
Due to the interest in the accelerated activation of the HPA axis during parturition, the changes in parameters of HPA axis activity between 10 and 7 d before term have been studied extensively, providing evidence that these two gestational windows represent entirely different fetal environments. During this period of gestation, there is an increase in the CRH binding sites in pituitary membrane preparations (32) and an increase in the POMC mRNA per cell but no change in the proportion of pituitary cells expressing POMC mRNA (27). Furthermore, Western blot has revealed a developmental increase in the ratio of ACTH to precursors in the fetal pituitary (33), and in vitro evidence indicates that slices of pituitary secrete an increased ratio of ACTH to precursors in unstimulated conditions (4). The loss of the POMC + ACTH + CRHR1 cells identified in this study is not overtly consistent with the increase in CRH binding, the lack of decrease in POMC mRNA-positive cells, or the selective increase in ACTH1–39 production and secretion without concomitant changes in precursor production and secretion. Therefore, it is likely that there is an increase in the synthetic, processing, and secretory capacity of the remaining corticotrophs as term approaches.
A decrease in the proportion of corticotrophs expressing POMC + ACTH + CRHR1 with an increase in hypothalamic CRH content as gestation progresses (34) might result in a relative increase in CRH stimulation of the ACTH + CRHR1-expressing cells. This would result in increase in pituitary secretion of ACTH, if, as suggested earlier, these cells process POMC to ACTH at a faster rate than the other subpopulations of corticotrophs. Between 135 d gestation and adulthood, there is an increase in the average amount of ACTH secreted by individual cultured pituitary cells in response to an equal dose of CRH (30).
The loss of POMC + ACTH + CRHR1 cells can be logically accounted for by: 1) apoptosis of those cells (35), 2) out-proliferation by other cells of the pituitary (36), 3) loss of one or two antigens, or 4) simultaneous loss of the POMC, ACTH, and CRHR1 peptides. We believe none of these mechanisms alone sufficiently explain the specific reduction in the POMC + ACTH + CRHR1 subpopulation. The lack of pituitary weight change suggests that neither apoptosis nor out-proliferation alone is likely; however, we did find evidence of dynamic architectural changes in the pituitary over a short gestational window. There was a marked increase in the proportion of corticotroph clusters expressing POMC + ACTH + CRHR1 that contained single cells between 140 and 144 d gestation, with significant decreases in the proportion of clusters in many of the larger cluster categories, suggesting there may be a generalized splitting of larger clusters into single cells, possibly via the proliferation of noncorticotrophs between corticotrophs. The clusters of corticotrophs observed in this study may correspond with the palisades of columnar corticotrophs described previously, which are known to decline over late gestation, whereas individual stellate corticotrophs increase (31, 37). It is interesting to note that the increased specificity of antisera used in the present study revealed that many large corticotrophic clusters contained a heterogeneous mix of subpopulations of corticotrophs that differentially express POMC, ACTH, and CRHR1. We did not find evidence for the loss of only one or two antigens because the specific reduction in the POMC + ACTH + CRHR1 subpopulation occurred with no change to the proportion of pituitary cells in any other subpopulation. Although it is possible that we have identified a transient loss of three antigens in particular pituitary cells, adult sheep have an even lower proportion of pituitary cells identified as corticotrophs (30), so the decrease in POMC + ACTH + CRHR1-expressing cells seen in this study may be a permanent loss rather than a transient down-regulation. In any event, this suggests that the mechanism for the loss of the POMC + ACTH + CRHR1 subpopulation is complex and requires further investigation.
The decrease in POMC + ACTH + CRHR1-expressing cells may be regulated by the increasing cortisol levels over the last weeks of gestation (1, 38) because adrenalectomy stops the decline in the proportion of pituitary cells expressing all ACTH-like peptides over the last third of gestation (31). It is unclear why a specific subpopulation of corticotrophs might be targeted by cortisol, although it may relate to differential expression of the glucocorticoid receptor in individual cells.
The hypothalamus may also be involved in the maturational disappearance of the POMC + ACTH + CRHR1-expressing cells. Hypothalamic-pituitary disconnection at 120 d gestational increases the proportion of pituitary cells expressing ACTH at 140 d gestation compared with control animals and stops the maturational loss of CRHR1 in late gestation (39). Regulation by CRH alone, however, cannot totally explain the specific decrease in POMC + ACTH + CRHR1-expressing cells and lack of change in ACTH + CRHR1-expressing cells.
Interestingly, hypothalamic-pituitary disconnection stops the late gestational rise in fetal plasma cortisol levels and therefore represents a model in which both the hypothalamic and adrenal input into the pituitary are removed in late gestation. It may therefore indicate that an interaction between CRH and cortisol controls the maturational changes in specific cells within a particular subpopulation of corticotrophs as term approaches. Such an interaction has been observed in rat pituitary cell cultures where pretreatment with glucocorticoids reduces the proportion of pituitary cells that bind CRH by approximately half (20, 40), which suggests that only a subpopulation of CRH target cells respond to negative feedback by cortisol.
Impact of placental restriction on corticotroph subpopulations
Fetuses subjected to restriction of placental nutrient supply underwent the same maturational reduction in the proportion of pituitary cells expressing POMC + ACTH + CRHR1 as the controls. Independently, we found a higher proportion of corticotrophs expressed POMC + ACTH + CRHR1 in pituitaries from PR fetuses compared with controls. This suggests that the maturational changes in corticotroph subpopulations are likely to occur via a different mechanism to the changes that occur as a result of placental insufficiency. In addition, previous investigation of this model of growth restriction has shown that there is an increase in the prepartum rise in the fetal plasma concentrations of cortisol (6). Therefore, it is unlikely that the growth-restricted fetus is simply displaying slower maturation than the control fetus, and it is much more likely that the pituitaries of PR fetuses have undergone a slightly different pattern of development, and may be operating at a different set-point to those of the control fetus.
Placental insufficiency is known to cause differential adaptation of several levels of the HPA axis including decreasing the pituitary POMC mRNA content, not altering the ratio of fetal plasma ACTH1–39 to its precursors but increasing the fetal plasma cortisol levels from approximately 20 d before parturition (6). Although the conservation of corticotrophs expressing POMC + ACTH + CRHR1 could be consistent with these findings, it would be inappropriate to speculate on these links without further investigation of the function of the subpopulations of corticotrophs in terms of POMC mRNA production and ACTH1–39 and precursor secretion rates. Similarly, evidence from a similar model of fetal stress, long-term hypoxia, which identifies a decrease in the pituitary protein content of POMC, pro-ACTH, ACTH1–39, and CRHR1, might directly conflict with our finding of a conservation of corticotrophs expressing these proteins. It is not known, however, if these differences are due to changes in protein content within individual corticotrophs or unique adaptations of the pituitary to different perturbations of the fetal environment. Therefore, it is important to investigate the functional characteristics of the corticotroph subpopulations identified in this study to understand the associations these findings have with the biological changes known to occur during development and perturbation of the fetal environment.
Some evidence of the functional differences between individual corticotrophs has been gained from in vitro studies on cultures of fetal pituitary cells in which the CRH-target cells have been eliminated. This experimental paradigm has been used to demonstrate that carunclectomy is associated with an increase in population(s) of non-CRH-target cells that secrete high levels of ACTH in the absence of external stimulation in vitro (10). It is impossible to make direct correlations between populations of cells identified by function in vitro and populations identified by the antigens they expressed in vivo. On the other hand, it is noteworthy that placental restriction in the present study was associated with a conservation of the predominant type of cell. Although these cells expressed CRHR1 protein, the staining appeared to be more cytoplasmic than membrane-associated, which might render them functionally non-CRH targets. Indeed, as we have stated previously, the fraction of fetal sheep pituitary cells that actually bind CRH is apparently less that the fraction that express CRHR1 by immunohistochemistry.
PR produced fetuses that were growth restricted as shown by the low body weight of these fetuses compared with control animals. Interestingly, there was sparing of both the pituitary and adrenal glands in fetuses subjected to restriction of placental function, indicated by the higher weights of these organs as a proportion of fetal body weight. Although the absolute weight of the pituitary was reduced in PR fetuses, this was not due to hypotrophy of the corticotrophs. In addition, the proportion of pituitary cells identified as corticotrophs was not altered by PR, suggesting that there is not a selective hypoplasia or hyperplasia of corticotrophs.
In conclusion, this is the first report of the differential colocalization of POMC, ACTH, and CRHR1 in individual pituitary cells. We have reported three major subpopulations of corticotrophs and have shown that between 140 and 144 d gestation, there is a swift drop in the overall proportion of pituitary cells expressing POMC + ACTH + CRHR1. In contrast, adaptations of the fetal pituitary to chronic placental restriction involve subtle shifts within the corticotroph population to produce a significant increase in the proportion of corticotrophs expressing POMC + ACTH + CRHR1. Therefore, this study has added a visual dimension to our understanding of the adaptations the HPA axis undergoes during parturition and chronic intrauterine nutrient deprivation and raises interesting questions about the differential regulation of key corticotrophic peptides in individual cells and the functional roles different phenotypic subpopulations play during physiological challenges.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Present address for I.C.M.: Chancellery, University of South Australia, Adelaide 5000, Australia.
The authors have nothing to disclose.
First Published Online July 6, 2006
Abbreviations: CRHR1, CRH type 1 receptor; DAPI, 4,6-diamidino-2-phenylindole; HPA, hypothalamic-pituitary-adrenal; PC, prohormone convertase; POMC, proopiomelanocortin; PR, placental restriction.
Received December 1, 2005.
Accepted for publication June 23, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-melanocyte-stimulating hormone derived from POMC in pituitary and brains of PC2-deficient mice. J Neurochem 86:556–563[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  C. W. H. Rumball, M. H. Oliver, E. B. Thorstensen, A. L. Jaquiery, S. M. Husted, J. E. Harding, and F. H. Bloomfield Effects of Twinning and Periconceptional Undernutrition on Late-Gestation Hypothalamic-Pituitary-Adrenal Axis Function in Ovine Pregnancy Endocrinology, March 1, 2008; 149(3): 1163 - 1172. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Szarek, K. Farrand, I. C. McMillen, I. R. Young, D. Houghton, and J. Schwartz Hypothalamic input is required for development of normal numbers of thyrotrophs and gonadotrophs, but not other anterior pituitary cells in late gestation sheep J. Physiol., February 15, 2008; 586(4): 1185 - 1194. [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 |
