Abstract
Ubiquitous overactivation of Hedgehog signaling in adult pituitaries results in increased expression of proopiomelanocortin (Pomc), growth hormone (Gh) and prolactin (Prl), elevated adrenocorticotropic hormone (Acth) production and proliferation of Sox2+ cells. Moreover, ACTH, GH and PRL-expressing human pituitary adenomas strongly express the Hedgehog target GLI1. Accordingly, Hedgehog signaling seems to play an important role in pathology and probably also in homeostasis of the adult hypophysis. However, the specific Hedgehog-responsive pituitary cell type has not yet been identified. We here investigated the Hedgehog pathway activation status and the effects of deregulated Hedgehog signaling cell-specifically in endocrine and non-endocrine pituitary cells. We demonstrate that Hedgehog signaling is unimportant for the homeostasis of corticotrophs, whereas it is active in subpopulations of somatotrophs and folliculo-stellate cells in vivo. Reinforcement of Hedgehog signaling activity in folliculo-stellate cells stimulates growth hormone production/release from somatotrophs in a paracrine manner, which most likely is mediated by the neuropeptide vasoactive intestinal peptide. Overall, our data show that Hedgehog signaling affects the homeostasis of pituitary hormone production via folliculo-stellate cell-mediated regulation of growth hormone production/secretion.
Introduction
The pituitary gland is a key regulator of body homeostasis and responsible for signal exchanges between the hypothalamus and peripheral organs. Besides of the six different endocrine cell types (e.g. corticotrophs/adrenocorticotropic hormone- (Acth), somatotrophs/growth hormone- (Gh), prolactin- (Prl), thyroid-stimulating hormone-, luteinizing hormone-, follicle-stimulating hormone-secreting cells), the anterior lobe (AL) of the pituitary consists of Sox2+ (stem) cells and a meshwork of non-endocrine Sox2+ folliculo-stellate cells (FSC). The latter ones are implicated in regulation and maintenance of the endocrine cells by delivering paracrine factors (reviewed in Cox et al. 2017).
Hedgehog (Hh) signaling plays a major role in the development of the pituitary. However, its function in homoeostasis and disease of the adult gland is far from clear. Under normal physiological conditions Hh signaling is inactive in most cells of adult tissues. Activation occurs upon binding of Hh ligands (e.g. mammalian Sonic, Indian or Desert Hh) to the receptor protein Patched1 (Ptch). This releases the inhibition of Smoothened (Smo), which results in translocation of Smo into the primary cilium and nuclear translocation of transcription factors of the Gli family to induce target gene expression (e.g. Gli1, Gli2 or Ptch) (reviewed in Bangs & Anderson 2017). Inactivation or overactivation of the pathway during pituitary organogenesis can lead to agenesis of the gland (Roessler et al. 2003), hypopituitarism and pituitary malformations (França et al. 2010, Flemming et al. 2013) or hyperplasia of the pituitary (Treier et al. 2001), respectively. Several lines of evidence additionally point toward a regulative function of the pathway in stem cell maintenance and regenerative processes in the adult pituitary. Thus, our group described the enhanced proliferation of Sox2+ cells in the AL after ubiquitous Hh signaling activation (Pyczek et al. 2016). Furthermore, other groups demonstrated that stem cells of the pituitary side population express the Hh signaling regulators Ptch and Smo (Chen et al. 2009, Vankelecom 2010) and that regenerative processes induce the expression of the Hh signaling target genes Gli1 and Gli2 in these cells (Gremeaux et al. 2012, Willems et al. 2016). Additionally, a regulatory function of Hh signaling in hormone-producing cells (e.g. corticotrophs) was proposed. Thus, Hh signaling regulates Acth expression in AtT-20 cells (Vila et al. 2005a,b, Pyczek et al. 2016) and ex vivo activation of the pathway in the whole pituitary leads to elevated Acth, Gh and Prl expression (Pyczek et al. 2016).
Additionally, there is evidence that Hh signaling is involved in hormone secretion or formation of pituitary tumors. For example patients and mice with heterozygous PTCH/Ptch germline mutation occasionally develop acromegaly-like symptoms (Kahn & Gordon 1967, Codish et al. 1973, Marcos et al. 1982, Cramer & Niederdellmann 1983, Bale et al. 1991, 1994, Kimonis et al. 1997, Wicking et al. 1997, Hahn et al. 1998, Lo Muzio et al. 1999). However, although human ACTH, GH or PRL-expressing pituitary adenoma show very high expression of the HH signaling inducer SHH and the HH target gene GLI1 (Pyczek et al. 2016), a direct link between Hh signaling (e.g. mutations or pathway overactivation) and tumor formation in the AL has not been confirmed.
Altogether, our data and those from other labs strongly suggest that Hh signaling plays a role in pathology and probably in function of the adult pituitary gland, especially in corticotrophs, somatotrophs, lactotrophs and/or Sox2+ cells. However, it never has been analyzed whether pituitary endocrine cells and/or other cell types are Hh responders under physiological conditions. Moreover, the fact that Hh signaling is a key player in tumorigenesis and obviously also plays a role in pituitary adenoma substantiates the efforts to unravel the Hh-responsive cell type/s in the normal adult pituitary gland.
Here, we investigated the Hh signaling activation status of the adult pituitary gland on cellular level and studied the impact of a deregulated pathway in endocrine and non-endocrine pituitary cells using in vivo and in vitro approaches. By investigating mouse models for lineage tracing and for conditional cell-specific deregulation of Hh signaling we demonstrate that the Hh pathway does not play a role in corticotrophs in the adult pituitary gland. However, subpopulations of somatotrophs and FSC of the adult pituitary gland express the surrogate marker for active Hh signaling Gli1 and descend from Gli1-expressing cells. Remarkably, we show here for the first time that activation of Hh signaling in FSC induces Gh release from somatotrophs in a paracrine manner, which most likely is mediated by the neuropeptide vasoactive intestinal peptide (Vip).
Materials and methods
Mice
All experiments using animals were performed in compliance with all German legal and ethical requirements and have been approved by the Lower Saxony State Office for Consumer Protection and Food Safety (file number 33.9-42502-04-15/1787). The following mouse strains were used in the study: Ptch1tm1Hahn (Ptchflox/flox (Uhmann et al. 2007), JAX stock # 012457), Smotm2Amc (Smoflox/flox (Long et al. 2001), JAX stock # 004526), Tg(Pomc-cre/ERT2)#Jke (PomcCreERT2 (Berglund et al. 2013) a kind gift from J K Elmquist), Gli1tm3(cre/ERT2)Alj (Gli1CreERT2 (Ahn & Joyner 2004), JAX stock #007913), Gt(ROSA)26Sortm9(CAG-tdTomato)Hze (tdT (Madisen et al. 2010), JAX stock #007905) and Tg(S100b-EGFP)11Lgrv (S100b-EGFP (Vives et al. 2003) a kind gift from C Legraverend and P Mollard).
Ptchflox/flox, PomcCreERT2, tdT and S100b-EGFP strains were maintained on C57BL/6 and Smoflox/flox and Gli1CreERT2 mice on a 129/Sv background. Both genders of transgenic mice were used. No sex-specific differences were observed. Genotyping of the mice was conducted by PCR on genomic DNA isolated from tail or ear biopsies using primer pairs recommended by the donating investigators (Vives et al. 2003, Berglund et al. 2013) or by The Jackson Laboratory (https://www.jax.org/jax-mice-and-services). For CreERT2-mediated homozygous deletion of Ptch or Smo Ptchflox/flox or Smoflox/flox mice, respectively, were bred to the respective CreERT2-deleter mouse strain. For lineage-tracing experiments the CreERT2-deleter strains were crossed to tdT mice that in some experiments additionally carry the S100b-EGFP transgene. The CreERT2-activity of the transgenic mice was induced by five single intraperitoneal injection (i.p.) of 1 mg tamoxifen on 5 consecutive days at an animal age of 8 weeks (Uhmann et al. 2007). Untreated mice without the respective CreERT2-recombinase gene and solvent-treated mice carrying the floxed alleles and the respective CreERT2-recombinase genes served as controls. For lineage-tracing experiments mice were analyzed after the first tamoxifen application as indicated in the respective figure legends. Body weight and blood samples of PomcCreERT2 Ptchflox/flox, PomcCreERT2 Smoflox/flox and the respective control mice were taken weekly or every second week, respectively, up to 250 days after the first tamoxifen/solvent application when the mice were sacrificed (Fig. 1B). Measurements of blood glucose and serum hormone levels are described in the Supplementary methods (see section on supplementary materials given at the end of this article). The number of analyzed animals is given in Supplementary Table 1 or in the respective figure legends.
In vivo deregulation of Hh signaling in Pomc-expressing cells of the adult pituitary gland. (A) Percentual birth rates, (B) experimental setup, (C and D) body weight (top), blood glucose level (bottom), (E) Acth serum levels, (F) pituitary weight, (G) pituitary gene expression levels, (H) genomic loci before and after recombination (top), recombination analysis on genomic pituitary DNA (bottom) of tamoxifen- or solvent-treated Pomc/Ptchf/f and Ptchf/f or Pomc/Smof/f and Smof/f mice. Analyzed animal numbers are given in Supplementary Table 1, no gender-specific differences in serum Acth levels, pituitary weight or gene expression levels were observed (data not shown). Gene expression levels were normalized to 18S rRNA expression and to the respective gene expression levels of solvent-treated controls. Each circle in F and G indicates one biological replicate (bars: mean ± s.e.m.), circles in C–E represent mean ± s.e.m. No significant differences were detected by using non-parametric Holm–Sidak method or Mann–Whitney tests.
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
Compounds
If not otherwise stated all compounds were obtained from Sigma-Aldrich, Darmstadt, Germany. Beta-Ala-Lys-N(epsilon)-aminomethylcoumarin acetate (β-Ala-Lys-N(ε)-AMCA) was obtained from Carbosynth (Berkshire, UK) and Smoothened Agonist (SAG) from Cayman Chemical (Ann Arbour, USA). β-Ala-Lys-N(ε)-AMCA was dissolved in HBSS (0.952 mM CaCl2·2H2O, 5.36 mM KCl, 0.411 mM KH2PO4, 0.812 mM MgSO4·7H2O, 136.7 mM NaCl, 0.385 mM Na2HPO4, 25 mM d-glucose·H2O, 10 mM HEPES). SAG was dissolved in dimethyl sulfoxide (DMSO). The preparation of tamoxifen/ethanol/sunflower oil for in vivo application has been previously described (Uhmann et al. 2007).
Cell culture
GH3 (CCL-82.1, January 2016) and AtT-20 cells (CCL-89, July 2014) were obtained from ATCC and grown in Ham’s F12-K Medium (Gibco, Life Technologies) supplemented with 15% Horse serum and 2.5% heat-inactivated FBS or in Ham’s F12-K Medium (Gibco) supplemented with 15 horse serum and 2.5% FBS, respectively. TtT/GF cells were obtained from RIKEN BRC (RCB1279, September 2019) and cultured in DMEM/HamF12 (Gibco) supplemented with 10% Horse serum and 2.5% FBS. Starvation medium resembles the growth medium but contains 0.5% Horse serum and 0.125% FBS (heat inactivated for GH3 cells). Routinely, all cell lines were tested for mycoplasma contamination by using Mycoplasma Detection Kit (minerva biolabs, Berlin, Germany). Identity of the cells was analyzed by marker gene expression analyses and immunofluorescent stainings against marker proteins as shown in Supplementary Fig. 3. Passage numbers between 15 and 30 of cell lines were used for the respective experiments.
Detailed information about SAG treatment, preparation of conditioned medium, medium transfer experiments, measurements of supernatant hormone/neuropeptide levels and BrdU incorporation analysis are given in Supplementary methods.
Detection of recombination of the Ptchflox and the Smoflox loci
Isolation of genomic DNA from pituitary glands was performed as previously described (Pyczek et al. 2016). For PCR-based detection of the CreERT2-mediated recombination at the Ptchflox or Smoflox locus the primer pairs indicated in Fig. 1H and G were used. The sequences of the primers are given in Supplementary Table 2.
RNA isolation and quantitative real-time PCR analyses
Gene expression analyses of murine tissue samples and in vitro cultured cells, RNA-isolation, cDNA synthesis and quantitative real-time PCR (qRT-PCR) analyses were conducted as previously described (Pyczek et al. 2016). All primer pairs, except those for amplification of 18S rRNA serving for normalization of the amount of sample cDNA, were intron-flanking and are summarized in Supplementary Table 3. Each cDNA was measured in triplicates.
Transcriptome analyses
For transcriptome analyses of three biological replicates of RNA from TtT/GF cells treated with either 100 nM SAG or solvent (see previous description) were analyzed. RNA quality control (Fragment Analyzer, Agilent Technologies), cDNA library preparation (TruSeq® RNA Sample Preparation v2; Illumina, San Diego, USA) and RNA sequencing (HiSeq 4000; Illumina) were performed at the NGS Service Facility for Integrative Genomics, Institute of Human Genetics, University Medical Center Göttingen, Germany. For detailed description see Supplementary methods. RNAseq data were deposited in the gene expression omnibus, accession: GSE153550.
Western blot and histological analyses
Immunohistological and immunofluorescent antibody stainings of paraffin and cryosections have been described previously (Pyczek et al. 2016). For detailed description of protein isolation, Western blot analysis, paraffin, cryotome and vibratome sections, immunofluorescent stainings of adherent or non-adherent cells, combined RNAScope/immunofluorescent staining and β-Ala-Lys-N(ε)-AMCA incubation see Supplementary methods. Used antibodies, antibody dilutions and antigen retrieval procedures are summarized in Supplementary Table 4.
Statistics
Statistical analyses were performed using the GraphPadPrism 6 software (GraphPad Software Inc., San Diego, USA). The used statistical tests are given in the respective figure legends.
Results
Deregulation of Hh signaling in Pomc-expressing cells has no impact on homoeostasis of adult pituitary glands
Constitutive activation of Hh signaling by a Rosa26-CreERT2-driven homozygous deletion of Ptch in ex vivo cultured adult pituitaries lead to an increase in Pomc, Gh and Prl expression and enhanced BrdU-incorporation of Sox2+ pituitary cells (Pyczek et al. 2016). Since the Rosa26-CreERT2-deleter recombines the Ptchflox locus in virtually every pituitary cell, these experiments did not allow for the determination of the specific/individual phenotype-triggering cell type. Therefore, we first tested whether Hh signaling directly regulates Acth expression in corticotrophs in vivo. For this purpose, we bred Ptchflox/flox or Smoflox/flox to PomcCreERT2 mice, which express the tamoxifen-inducible CreERT2-recombinase under the control of the murine proopiomelanocortin (Pomc, encodes for the Acth precursor polypeptide) promoter (Berglund et al. 2013). To verify inducibility, specificity and potential leakiness of the deleter strain we furthermore generated PomcCreERT2 R26-tdTomato (Pomc/tdT) mice. Each mouse cohort was subdivided into two groups that received tamoxifen or solvent at an age of 8 weeks.
As judged by the amount of tdT+ cells isolated from adult pituitaries of solvent-treated Pomc/tdT mice (experimental setup see Supplementary Fig. 1A), the Pomc/tdT reporter was highly or mildly leaky in the intermediate lobe or the AL, respectively (Supplementary Fig. 1B). However, tamoxifen-application strongly increased the number of tdT+ cells in the AL within 7 days after CreERT2-induction and the cells were trackable until 250 days after tamoxifen injection without any reduction of labeled cell numbers (Supplementary Fig. 1B). Double immunofluorescence analyses furthermore verified tdT expression in Pomc- and Acth- but not in Gh- or Prl-expressing cells in both tamoxifen- and solvent-treated Pomc/tdT mice (Supplementary Fig. 1B) indicating a cell-specific expression of the PomcCreERT2-transgene in corticotrophs. Thus, we expected that under normal physiological conditions the PomcCreERT2-deleter allows for a long-term observation of genetically modified Pomc-expressing cells and crossed the PomcCreERT2-transgene with Ptchflox/flox or Smoflox/flox mice. Irrespective of the leakiness of the deleter, PomcCreERT2 Ptchflox/flox (Pomc/Ptchf/f) and PomcCreERT2 Smoflox/flox (Pomc/Smof/f) mice were born at a Mendelian ratio (Fig. 1A) and did not show any obvious developmental abnormities without tamoxifen application. Similarly, (experimental setup see Fig. 2B) none of the Pomc/Ptchf/f and Pomc/Smof/f mice showed signs of a deregulated hormone status (e.g. alopecia, weight loss/gain (Fig. 1C and D), abnormal blood glucose levels (Fig. 1C and D), abnormal serum Acth levels (Fig. 1E) or increased pituitary weight (Fig. 1F)) 250 days after CreERT2-induction. Neither Hh signaling activity nor Gh, Prl or Pomc expression levels were altered (Fig. 1G) albeit Ptchflox or Smoflox loci were efficiently recombined in Pomc/Ptchf/f and Pomc/Smof/f mice (Fig. 1H). In addition, no histological abnormalities were observed in Pomc/Ptchf/f and Pomc/Smof/f pituitaries (Fig. 2A and B, respectively). Thus, the distribution of hormone-releasing cells was normal and the Pomc/Ptchf/f or Pomc/Smof/f pituitaries did not show any signs of hyperplasia (Fig. 2A) or hypoplasia (Fig. 2B), respectively, compared to the controls. Moreover, the percentage of Acth+ cells was not altered in tamoxifen-treated Pomc/Ptchf/f (Fig. 2C) or Pomc/Smof/f mice (Fig. 2D). Finally, combined RNAScope/immunofluorescent analyses revealed that Gli1 transcripts are not expressed in Acth-expressing cells, neither in tamoxifen-treated Pomc/Ptchf/f mice nor in the controls (Fig. 2E).
In vivo deregulation of Hh signaling in Pomc-expressing cells has no impact on the hormone expression pattern or cellular proliferation of the adult pituitary gland. (A and B) Representative (immune) histological analyses and (C and D) percentage of Acth+ cells of pituitaries of tamoxifen- or solvent-treated (A and C) Pomc/Ptchf/f and Ptchf/f and (B and D) Pomc/Smof/f and Smof/f mice 265 days post tamoxifen or solvent treatment. (E) Representative Gli1 RNAScope/anti-Acth antibody stainings of pituitaries of tamoxifen- or solvent-treated of Pomc/Ptchf/f and Ptchf/f mice 265 days post-tamoxifen or solvent treatment (for RNAScope control stainings please see Fig. 4A). Analyzed animal numbers are given in Supplementary Table 1. Boxes: zoomed areas. Black arrows: Ki67+ cells. White arrows: Gli1+ cells. Scale bars: 50 µm (A and B), 10 µm (A insets, B insets, E). Circles in C and D represent mean ± s.e.m. No significant differences were detected by using non-parametric Holm-Sidak method or Mann-Whitney tests.
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
Together, Ptch or Smo depletion in Pomc/Ptchf/f and Pomc/Smof/f pituitaries do not result in changes of Hh signaling activity or in the development of pathological phenotypes. These data show that a homozygous deletion of Ptch or Smo in Pomc-expressing cells has no impact on homeostasis of corticotrophs or other pituitary cells in vivo. These results are surprising, because ex vivo depletion of Ptch in whole pituitaries results in upregulation of Pomc (Pyczek et al. 2016).
Somatotrophs and folliculo-stellate cells but not corticotrophs of the adult pituitary gland express Gli1
Because the above-mentioned in vivo experiments clearly excluded a direct impact of Hh signaling at least on corticotrophs, we hypothesized that Hh signaling might regulate hormone release in an indirect manner. To shed light on this, we performed Gli1 lineage tracing experiments by generating Gli1CreERT2 R26-tdTomato (Gli1/tdT) mice and visualized the pituitary progeny of Gli1+ cells under normal physiological conditions (same experimental setup as for Pomc/tdT mice, Supplementary Fig. 1A). Leakiness of the Gli1CreERT2-deleter strain was excluded by simultaneously investigated pituitary glands of solvent-treated Gli1/tdT mice (Fig. 3A). Remarkably, analyses of tamoxifen-treated Gli1/tdT adult pituitaries showed that two morphologically different pituitary cell types in the AL were marked by tdT reporter expression and thus developed from Gli1-expressing cells: one cell population with a round (Fig. 3A, arrow heads) and another with a stellate-shaped morphology (Fig. 3A, double arrows). Double immunofluorescence analyses demonstrated that the round cell type was positive for Gh and negative for Prl and Acth/Pomc, thus representing somatotrophs (Fig. 3A, arrow heads). The stellate-shaped cell type did neither express Gh, Prl, Acth (Fig. 3A) nor Pdgfra (Fig. 3B), but was positive for Sox2 (Fig. 3B) and beta-Ala-Lys-N(epsilon)-aminomethylcoumarin acetate (β-Ala-Lys-N(ε)-AMCA) (Fauquier et al. 2002) uptake, resembling the phenotype of FSC (Fig. 3B). To further reinforce this assumption, we additionally examined pituitaries of tamoxifen-induced Gli1/tdT/S100b-EGFP mice, in which the progeny of Gli1+ cells and cells that express the FSC marker S100b are marked simultaneously. Indeed, this approach revealed that tdT+ stellate-shaped pituitary cells express EGFP (Fig. 3C) indicating that FSC represent progenies of Gli1+ pituitary cells. Moreover, combined RNAScope/immunofluorescent analyses (Fig. 4A, B, C, D, E and F) and subsequent quantification of Gli1+ and Gli2+ cells (Fig. 4G, H, I and J) verified that 33% (s.e.m. 1.5%) or 38% (s.e.m. 3.9%) of all somatotrophs (Fig. 4G and H) and 31% (s.e.m. 3.2%) or 34% (s.e.m. 7.2%) of S100b-EGFP+ FSC (Fig. 4I and J) express Gli1 or Gli2 transcripts, respectively, and thus show active Hh signaling. The distribution and number of Gli1+ somatotrophs and Gli1+ FSC and their offspring did not grossly vary between different age-matched animals.
Somatotrophs and FSC of the adult pituitary gland descent from Gli1+ cells. Representative fluorescence analyses of adult (A and B) Gli1/tdT and (C) Gli1/tdT/S100b-EGFP pituitaries 7 or 50 days after in vivo tamoxifen or solvent application (A), 14 days (B bottom, C), 15 days (B top) or 22 days (B middle) days after tamoxifen injection. Similar experimental setup as shown in Fig. 1. Analyses were conducted on pituitaries of at least three animals per cohort. Arrow heads: somatotrophs; double arrows: FSC. Scale bars: 10 µm.
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
Somatotrophs and FSC of the adult pituitary gland express Gli1 and Gli2. (A, B, C, D, E and F) Representative fluorescence analyses and (G, H, I and J) quantification of Gli1+ (G) and Gli2+ somatotrophs (H) and of Gli1+ (I) and Gli2+ FSC (J) of combined Gli1 or Gli2 transcripts detection using RNAScope technique and immunofluorescent stainings of murine basal cell carcinomas (A) (used as positive control for high Gli1 and Gli2 expression levels) and adult pituitary glands of tamoxifen-injected Gli1/tdT (B), C57BL/6N wildtype (C, D, G, H) and S100b-EGFP mice (E, F, I, J). Analyses were conducted on pituitaries of at least three animals per cohort. Arrow heads: somatotrops; double arrows: FSC. secAB, secondary antibody only; negative, negative control probe. Scale bars: 33 µm (A), 10 µm (C top, D top), 3.3 µm (B, C bottom, D bottom, E, F). Each open circle indicates one pituitary. Mean ± s.e.m.
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
Taken together these data demonstrate that Hh signaling is active in a constant subpopulation of somatotrophs and FSC in the adult pituitary gland and thus most likely has a function in these two pituitary cell populations.
Hh signaling is active in the folliculo-stellate cell line TtT/GF, but not in the somatotroph cell line GH3 or in the corticotroph cell line AtT-20
Next, we studied whether the aforementioned in vivo data also apply to pituitary cell lines. For this purpose, we used the folliculo-stellate cell line TtT/GF, the somatotroph cell line GH3 and the corticotroph cell line AtT-20 and studied the expression of cell-specific marker genes, the basal Hh signaling activity as well as the responsiveness to Hh signaling activation. The results revealed that TtT/GF cells grow with a stellate-shaped morphology (Supplementary Fig. 2A) and express high levels of the FSC markers Sox2 (Supplementary Fig. 2A), S100b, Vegfa, Mif and Fst (Supplementary Fig. 2B), whereas GH3 or AtT-20 cells express Gh and Ghrhr (Supplementary Fig. 2C, D and E) or Pomc and Acth, respectively (Supplementary Fig. 2F and G). In contrast to GH3 and AtT-20 cells, TtT/GF cells furthermore express robust Gli1 levels (Supplementary Fig. 3A) and show unambiguous Smo localization to primary cilia (Supplementary Fig. 3B, C and D), indicating basal Hh signaling activity. In addition, Smoothened Agonist (SAG)-treatment elevates the basal Gli1 and Gli2 transcription in TtT/GF (Supplementary Fig. 4A) but not in GH3 (Supplementary Fig. 4B) or AtT-20 cells (Supplementary Fig. 4C). This indicates that TtT/GF, but not GH3 or AtT-20 cells, are responsive to Hh signaling stimulation.
Supernatant of Hh-stimulated TtT/GF folliculo-stellate cells induces Gh production in somatotroph GH3 cells, but has no impact on the corticotroph AtT-20 cell line
Since FSC were responsive to Hh signaling activation, we hypothesized that active Hh signaling might indirectly influence hormone release in Gh- or Acth-expressing cells, potentially by secreted factors (reviewed in Morris & Christian 2011). To test this hypothesis, we treated GH3 or AtT-20 cells with conditioned medium from SAG-stimulated (CM-TtT/GFSAG) or solvent-treated TtT/GF cells (CoM-TtT/GF) (for confirmation of Hh signaling activity in TtT/GF cells after SAG treatment see Fig. 5A and Supplementary Fig. 6A), and analyzed the expression levels of Gli1, Gli2, Ptch and Gh or Pomc. CM-TtT/GFSAG-treatment neither alters Hh signaling activity in GH3 (Fig. 5B) and AtT-20 cells (Supplementary Fig. 5B), the proliferative activity of GH3 cells (Fig. 5C) nor the Pomc expression (Supplementary Fig. 5B) or Acth secretion level of AtT-20 cells (Supplementary Fig. 5C). However, CM-TtT/GFSAG-incubation significantly increases Gh expression levels (Fig. 5B) and Gh secretion of GH3 cells compared to the respective CoM-TtT/GF-treated controls (Fig. 5D).
Upregulated Gh production of somatotophs via paracrine signal transduction of Hh-activated FSC. (A and B) Gene expression levels of Gli1, Gli2, Ptch (A and B) and Gh (B) of (A) TtT/GF cells after serum starvation followed by 48 h, 100 nM Smoothened Agonist (SAG) or solvent treatment (dotted line) in GH3 serum starvation medium and (B) of GH cells after serum starvation followed by 48 h, 100 nM SAG (gray circles, same data as shown in Supplementary Fig. 5B) or solvent treatment (dotted line) or by 48 h incubation with conditioned media from TtT/GF cells (shown in A) treated with SAG (CM-TtT/GFSAG, red circles) or solvent (dotted line). Please note: Gli2 expression most likely is not induced by general Hh signaling activation by remaining SAG in the CM-TtT/GFSAG medium since SAG-treatment of GH3 cells rather leads to reduced Gli2 expression levels (Supplementary Fig. 5). (C) Relative BrdU incorporation of GH3 cells after serum starvation followed by 48 h incubation with GH3 starvation medium (dotted line), conditioned media from TtT/GF cells treated with SAG (CM-TtT/GFSAG, red circles) or solvent (CoM-TtT/GF). (D) Gh protein concentration in supernatant of GH3 cells after serum starvation followed by 48 h incubation with conditioned media from TtT/GF cells (shown in A) treated with SAG (CM-TtT/GFSAG, red circles) or solvent (dotted line). Gene expression levels were normalized to 18S rRNA expression and to the respective gene expression levels of solvent-treated control cells (dotted lines). Gh concentration was normalized to the Gh concentration of solvent-treated control cells (dotted lines). Each open circle indicates one biological replicate measured in technical triplicates (A, B and D) or sextuplets (C). Mean ± s.e.m. Significant differences were tested using the non-parametric Holm-Sidak method. Significant differences to the respective base line (dotted lines) are indicated by asterisks above the data. **P = 0.01; ***P = 0.001; ****P = 0.0001; n.s., not significant.
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
These data demonstrate that Hh activation in the FSC cell line TtT/GF apparently induces the release of paracrine factors that initiate Gh production/release from GH3 cells. The factors, however, do not initiate Acth production/release from AtT-20 cells.
Vasoactive intestinal peptide is a candidate molecule for mediating Gh production/secretion upon Hh signaling activation in folliculo-stellate cells
The current knowledge about the functional regulation of endocrine cells by FSC and the involved signal transduction molecules is sparse. However, growth factors and peptides may play a role in this process (Allaerts & Vankelecom 2005, Morris & Christian 2011). To identify potential candidate molecules that are upregulated upon Hh signaling activation and potentially mediate Gh production in GH3 cells in a paracrine manner, we conducted comparative transcriptome analyses of SAG- vs solvent-treated TtT/GF cells. This approach revealed that SAG-treatment leads to an up- and downregulation of 108 or 63 genes, respectively (Fig. 6A). Significantly upregulated genes included 8 genes associated with Hh signaling activation (Ptch, Gli1, Hhip, Psmb9, Adcy5, Pcdhga2, Gpr161, Rasl11b) (Fig. 6A, B and C) and 8 genes associated with G protein-coupled receptors (Gpcr) signaling (Reep6, Qrfp, Gna15, Fgd2, Vip, Olfr1250, Cxcr4, Hrh1) (Fig. 6A, B and D) whereas the expression of four genes associated with Gpcr signaling were downregulated (Ucn2, C5ar1, Gpr3, Gpr171) (Fig. 6B and D). Additionally, SAG-treatment increased the expression levels of insulin-like growth factor-binding protein 2 (Igfbp2), angiotensin-converting enzyme (Ace), glutamate ionotropic receptor kainate type subunit 4 (Grik4) and the putative pituitary stem/progenitor marker coxsackie virus and adenovirus receptor (Cxadr) (Supplementary Fig. 6A). In contrast, SAG-treatment merely altered FSC marker gene expression (e.g. Sox2, S100b, Mif, Anxa1) (Supplementary Fig. 6B) albeit it significantly increased the transcript levels of Cxcr4 (Fig. 6D) and Cxadr (Supplementary Fig. 6A) that are also known to be expressed in FSC (Horiguchi et al. 2012, Chen et al. 2013).
Comparative transcriptome analysis of Hh signaling activated TtT/GF cells. (A, B, C and D) Comparative transcriptome analysis of Smoothened Agonist (SAG)- vs solvent-treated (DMSO) TtT/GF cells. (A) Volcano plot of all expressed gene transcripts (cut off log2 -0.8- or log2 1.2-fold change, red vertical lines), (B) heat map and (C and D) expression profile of significant differentially expressed genes (cut off 0.8- or 1.2-fold change, grey lines) associated with Hh and Gpcr signaling. Transcriptome analyses were conducted in biological triplicates (open circles). Gene expression of SAG-treated cells were normalized to solvent-treated controls (dotted lines in C and D). Differential expression with adjusted P values (non-parametric Holm–Sidak method) below 0.05 were considered to be significant (red horizontal line in A, asterisks in C and D). Significant differences to the respective baseline (dotted lines) are indicated by asterisks above/below the data. *P = 0.05; **P = 0.01; ***P = 0.001; ****P = 0.0001.
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
To this end, we focused on the two neuropeptides RF(Arg-Phe)amide family 26 amino acid peptide (Qrfp) and vasoactive intestinal peptide (Vip), which are known to regulate pituitary hormone release (Matsushita et al. 1981, Chihara et al. 1982, Abe et al. 1985, Denef et al. 1985, Bluet-Pajot et al. 1987, Bjoro et al. 1990, Alexander & Sander 1994, Mazzocchi et al. 1998, Vleck & Patrick 1999, Fazekas et al. 2000, Christian et al. 2007, Leprince et al. 2017) and whose expression levels were significantly elevated in SAG-stimulated TtT/GF cells compared to the controls (Fig. 6A, B and D). qRT-PCR-based expression analyses verified the significant increase of Vip expression in SAG-treated TtT/GF cells (Fig. 7A), whereas the absolute Qrfp reads remained under qRT-PCR detection level. Importantly, measurement of Vip protein concentration revealed a significant increase of Vip protein in CM-TtT/GFSAG compared to CoM-TtT/GF (Fig. 7B and Supplementary Fig. 7). Expression analyses of Vip and its receptors Vipr1 (vasoactive intestinal peptide receptor), Vipr2 and pituitary adenylate cyclase-activating peptide (Pacap) type 1 receptor (Adcyap1r1) in TtT/GF, GH3, AtT-20 and NIH/3T3 (used as negative control) cells revealed that Vip transcripts were only detectable in TtT/GF cells (Fig. 7C) that also showed Vip protein expression (Fig. 7D). In addition, both TtT/GF and GH3 cells showed robust Vipr2 mRNA levels (Fig. 7E). None of the cell lines expressed Vipr1 or Adcyap1r1 (data not shown). Finally, we analyzed whether GH3 cells respond to Vip. Strikingly, treatment of GH3 cells with the hybrid Vip antagonist KPRRPYTDNYTRLRKQMAVKKYLNSILN-NH2 efficiently inhibited the CM-TtT/GFSAG-mediated Gh production (Fig. 7F).
Paracrine signal transduction of Hh-activated FSC to somatotrophs is mediated by Vip/Vipr2 signaling. (A and B) qRT-PCR- (E, same samples as shown in Supplementary Fig. 5) and EIA-based (F) verification of increased Vip/Vip expression/secretion of SAG-treated TtT/GF cells (E) or supernatant (F, for total Vip concentration see Supplementary Fig. 8). (C, D and E) Vip mRNA (C), Vip protein (D) and Vipr2 mRNA expression of TtT/GF (C, D and E), GH3 and AtT-20 cells compared to NIH/3T3 cells (C and E). (F) qRT-PCR-based analysis of Gh expression level of GH3 cells treated with CM-TtT/GFSAG or CoM-TtT/GF supplemented with 1 µM Vip antagonist (VipAntag) or solvent, respectively. (G) Combined visualization of Gli1 transcripts and Vip protein expression in an adult murine C57Bl6/N wildtype pituitary gland. Gene expression levels in A, C, E and F were normalized to 18S rRNA expression and in A and F additionally to the Vip or Gh expression level of solvent-treated control cells, respectively (dotted lines in A and F). Vip protein concentration was normalized to the Vip protein concentration of solvent-treated control cells (dotted line in B). Each open circle indicates one biological replicate measured in technical triplicates. Mean ± s.e.m. Significant differences were tested using the non-parametric Holm–Sidak method. Significant differences to the respective base line (dotted lines) are indicated by asterisks above the data. *P = 0.05; **P = 0.01; ***P = 0.001. White arrows: Vip+ TtT/GF cells. White double arrows: Gli1+ Vip+ stellate-shaped pituitary cells. Scale bars: 10 µm (D), 3.3 µm (G).
Citation: Journal of Endocrinology 248, 3; 10.1530/JOE-20-0388
These data show that Hh signaling activation in the FSC cell line TtT/GF stimulates the production and release of the neuropeptide Vip, which induces Gh production/secretion in the GH3 cells most likely via Vipr2 signaling. Moreover, the fact that Gli1+ stellate-shaped pituitary cells of the adult pituitary gland also express Vip (Fig. 7G) strongly points to a similar circuit in the pituitary in vivo.
Discussion
The Hh signaling pathway plays a prominent role in the development of the pituitary (Treier et al. 2001, Roessler et al. 2003, França et al. 2010, Flemming et al. 2013). However, its function in the adult gland is far from clear. Recently we demonstrated that Hh signaling activation in the adult pituitary gland leads to Acth, Gh and Prl production and proliferation of Sox2+ cells. Unfortunately, these experiments were not conclusive with respect to the Hh-responsive pituitary cell type in the normal gland (Pyczek et al. 2016). However, this information is of great importance because GLI1 and SHH are highly expressed by GH-, PRL- and ACTH-expressing human pituitary adenoma, which suggests that HH signaling has an impact on pituitary tumor formation (Pyczek et al. 2016).
Here we demonstrate that a cell-specific deregulation of Hh signaling in Pomc-expressing cells does not affect the homeostasis of corticotrophs in vivo. This conclusion is based on our findings that homozygous depletion of Ptch or Smo in Pomc-expressing cells neither leads to defective development of the gland nor to disturbed Hh signaling activity or defective homeostasis of the adult pituitary. At the first glance, these results are contrary to our previous ex vivo studies on Rosa26-CreERT2/Ptchf/f pituitaries that revealed a higher Acth release upon Hh signaling activation (Pyczek et al. 2016). However, Rosa26-CreERT2-driven recombination targets every pituitary cell, whereas in Pomc/Ptchf/f and Pomc/Smof/f mice Hh signaling is activated/inactivated cell-specifically in Pomc-expressing cells. Moreover, the fact that murine Pomc-expressing cells never stained positive for tdT in Gli1 lineage tracing experiments or for Gli1 transcripts in RNAScope stainings supports the conclusion that cell-intrinsic Hh signaling is not important for corticotrophs. Currently, we cannot be completely sure whether this also applies to the human pituitary since some ACTH-expressing cells of the human pituitary are immunopositive for SHH (Vila et al. 2005a, Pyczek et al. 2016). Nevertheless, our new data demonstrate that Pomc/Acth production in corticotrophs must also involve an indirect (e.g. paracrine) effect of Hh signaling.
Beyond that, our RNAScope and Gli1 lineage tracing approaches revealed that subpopulations of somatotrophs and FSC show active Hh signaling in vivo. These findings are remarkable because they suggest that Hh signaling is important for homeostasis of both pituitary cell types. However, our analyses of the Hh signaling status and responsiveness toward SAG-treatment in well-accepted pituitary cell lines revealed that GH3 cells express extremely low Gli1 levels, show very rarely ciliary Smo localization and are unresponsive to Hh signaling activation upon SAG-treatment. These facts impaired further in vitro analyses using GH3 cells to investigate the cell-intrinsic impact of Hh signaling in somatotrophs and the most elegant way to do so would be in vivo approaches. Unfortunately, until now no somatotroph-specific CreERT2-deleter mouse strains exist.
In addition, our data strongly suggest that Hh signaling influences the functionality of FSC, which activates hormone production in somatotrophs in a paracrine way. FSC represents a small (5–10%) non-hormone secreting cell population in the adult AL and are implicated in the regulation and maintenance of hormone-secreting cells by delivering paracrine factors (e.g. interleukin-6, vascular endothelial growth factor, annexin-1) (reviewed in Allaerts & Vankelecom 2005). However, the exact mechanisms of how FSC regulates endocrine cells are not well understood. Our in vitro approaches now demonstrate for the first time that activation of Hh signaling in the FSC cell line TtT/GF induces Gh production/secretion in GH3 cells via a paracrine mechanism. Since Vip expression and concentration are significantly increased in TtT/GF cells and in the respective supernatant after SAG-treatment, and since Vip antagonist treatment can block CM-TtT/GFSAG-induced Gh production from GH3 cells, this paracrine mechanism most likely encompass the neuropeptide Vip. In addition, this peptide is well known for its specific capacity to stimulate Gh production/secretion in GH3 and adenoma cells and in in vivo approaches (Matsushita et al. 1981, Chihara et al. 1982, Denef et al. 1985, Bluet-Pajot et al. 1987, Murakami et al. 1995, Fazekas et al. 2000). Apart from that Vip also induces Prl (Abe et al. 1985, Bjoro et al. 1990, Vleck & Patrick 1999, Fazekas et al. 2000, Christian et al. 2007) and Acth release (Alexander & Sander 1994, Mazzocchi et al. 1998) from the respective cell lines and endocrine and/or pituitary adenoma cells. In the normal pituitary gland Vip is expressed throughout the organ (Arnaout et al. 1986, Hsu et al. 1989) including in a so far unidentified pituitary cell type with FSC-like morphology (Hagen et al. 1986). Vip signal transmission into the target cells is mediated by binding to the G protein-coupled membrane-bound receptors Vipr1 or Vipr2 (type 2 receptors), but not via the Pacap-specific Pac1 receptor (type 1 receptor, encoded by Adcyap1r1 gene) (reviewed in Hirabayashi et al. 2018). Interestingly, TtT/GF cells express neither Pacap, Adcyap1r1 nor Vipr1. However, they express Vip and Virp2 and the expression and concentration of Vip increases upon Hh signaling activation in TtT/GF cells. Most strikingly, GH3 cells express Vipr2 but not Vip, Adcyap1r1 or Vipr1. Thus, the increased Gh production/release of GH3 after incubation with CM-TtT/GFSAG is most likely transmitted via Vip/Vipr2 signaling. Similar findings have been reported for the AtT-20 substrains AtT-20/D16-16 (Cellosaurus CVCL_GZ35) and AtT20/D16v (Cellosaurus CVCL_4W08), in which Vip-binding to the Vipr2 receptor induces Acth-release (Reisine et al. 1982, Aoki et al. 1997). Paternal AtT-20 cells (Cellosaurus CVCL_2300) used in our study do not express Vipr2 (Fig. 7H). This may explain the unresponsiveness of AtT-20 cells toward Vip-enriched CM-TtT/GFSAG in our setting.
Together, our data demonstrate for the first time that Hh signaling is involved in FSC-mediated regulation of Gh production/release at least in vitro. Moreover, our results strongly hint toward a similar role of Hh signaling in vivo. Nevertheless, additional studies are needed to show whether this concept is indeed transferrable to the in vivo situation and potentially also to Acth-expressing cells. For this purpose, in vivo depletion of Gli1, Ptch or Smo in FSC would be advantageous which is so far hampered by missing availability of an FSC-specific Cre- or CreERT2-deleter mouse strain. However, our findings could be of importance for several pituitary adenoma subtypes, in which HH signaling is activated (Pyczek et al. 2016). It is possible that Hh signaling activation in tumor-associated FSC, which are found in large numbers at the periphery of adenomas and other pituitary lesions (Nishioka et al. 1991, Voit et al. 1999, Horvath & Kovacs 2002, Cimpean et al. 2017) support hormone production from tumor cells. This opens the intriguing possibility that hormone production of tumor cells depends on Hh signaling activity in adjacent FSC, which thus might represent a target for future therapeutic intervention.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/JOE-20-0388.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
Research was supported by the Deutsche Forschungsgemeinschaft to A U and R B (UH-221/6-1).
Author contribution statement
D S B designed and performed research, collected and analyzed data, prepared the figures and wrote the manuscript. N B, A F and I H performed research and collected data. A W analyzed data, A Z analyzed data, H H contributed vital reagents and analytical tools and reviewed the paper. R B contributed vital reagents and analytical tools and reviewed the paper. A U designed research, collected and analyzed data, prepared the figures and wrote the manuscript. All authors reviewed the manuscript.
Acknowledgements
We are grateful to the members of the animal facility of the University Medical Center Göttingen for excellent animal care. We thank A F Parlow (National Hormone & Peptide Program; Torrance, California) for providing anti-Acth, anti-Gh and anti-Prl antibodies, J K Elmquist (Division of Hypothalamic Research, UT Southwestern Medical Center, Dallas, Texas) for providing PomcCreERT2 mice, C Legraverend and P Mollard (Institut national de la santé et de la recherche médicale, Montpellier, France) for providing S100b-EGFP mice, W Brück, C Stadelmann-Nessler and O Kowatsch (Department of Neuropathology, University Medical Center, Göttingen, Germany) for providing help by preparing vibratome sections.
References
Abe H, Engler D, Molitch ME, Bollinger-Gruber J & Reichlin S 1985 Vasoactive intestinal peptide is a physiological mediator of prolactin release in the rat. Endocrinology 116 1383–13 90. (https://doi.org/10.1210/endo-116-4-1383)
Ahn S & Joyner AL 2004 Dynamic changes in the response of cells to positive hedgehog signaling during mouse limb patterning. Cell 118 505–5 16. (https://doi.org/10.1016/j.cell.2004.07.023)
Alexander LD & Sander LD 1994 Vasoactive intestinal peptide stimulates ACTH and corticosterone release after injection into the PVN. Regulatory Peptides 51 221–22 7. (https://doi.org/10.1016/0167-0115(9490068-x)
Allaerts W & Vankelecom H 2005 History and perspectives of pituitary folliculo-stellate cell research. European Journal of Endocrinology 153 1–12. (https://doi.org/10.1530/eje.1.01949)
Aoki Y, Iwasaki Y, Katahira M, Oiso Y & Saito H 1997 Regulation of the rat proopiomelanocortin gene expression in AtT-20 cells. II: effects of the pituitary adenylate cyclase-activating polypeptide and vasoactive intestinal polypeptide. Endocrinology 138 1930–193 4. (https://doi.org/10.1210/endo.138.5.5116)
Arnaout MA, Garthwaite TL, Martinson DR & Hagen TC 1986 Vasoactive intestinal polypeptide is synthesized in anterior pituitary tissue. Endocrinology 119 2052–205 7. (https://doi.org/10.1210/endo-119-5-2052)
Bale SJ, Amos CI, Parry DM & Bale AE 1991 Relationship between head circumference and height in normal adults and in the nevoid basal cell carcinoma syndrome and neurofibromatosis type I. American Journal of Medical Genetics 40 206–2 10. (https://doi.org/10.1002/ajmg.1320400217)
Bale AE, Gailani MR & Leffell DJ 1994 Nevoid basal cell carcinoma syndrome. Journal of Investigative Dermatology 103 (Supplement) 126S–130S. (https://doi.org/10.1111/1523-1747.ep12399438)
Bangs F & Anderson KV 2017 Primary cilia and mammalian hedgehog signaling. Cold Spring Harbor Perspectives in Biology 9 a028175. (https://doi.org/10.1101/cshperspect.a028175)
Berglund ED, Liu C, Sohn JW, Liu T, Kim MH, Lee CE, Vianna CR, Williams KW, Xu Y & Elmquist JK 2013 Serotonin 2C receptors in pro-opiomelanocortin neurons regulate energy and glucose homeostasis. Journal of Clinical Investigation 123 5061–50 70. (https://doi.org/10.1172/JCI70338)
Bjoro T, Sand O, Ostberg BC, Gordeladze JO, Torjesen P, Gautvik KM & Haug E 1990 The mechanisms by which vasoactive intestinal peptide (VIP) and thyrotropin releasing hormone (TRH) stimulate prolactin release from pituitary cells. Bioscience Reports 10 189–1 99. (https://doi.org/10.1007/BF01116578)
Bluet-Pajot MT, Mounier F, Leonard JF, Kordon C & Durand D 1987 Vasoactive intestinal peptide induces a transient release of growth hormone in the rat. Peptides 8 35–38. (https://doi.org/10.1016/0196-9781(8790161-6)
Chen J, Gremeaux L, Fu Q, Liekens D, Van Laere S & Vankelecom H 2009 Pituitary progenitor cells tracked down by side population dissection. Stem Cells 27 1182–11 95. (https://doi.org/10.1002/stem.51)
Chen M, Kato T, Higuchi M, Yoshida S, Yako H, Kanno N & Kato Y 2013 Coxsackievirus and adenovirus receptor-positive cells compose the putative stem/progenitor cell niches in the marginal cell layer and parenchyma of the rat anterior pituitary. Cell and Tissue Research 354 823–8 36. (https://doi.org/10.1007/s00441-013-1713-8)
Chihara K, Iwasaki J, Minamitani N, Kaji H, Matsukura S, Tamaki N, Matsumoto S & Fujita T 1982 Effect of vasoactive intestinal polypeptide on growth hormone secretion in perifused acromegalic pituitary adenoma tissues. Journal of Clinical Endocrinology and Metabolism 54 773–77 9. (https://doi.org/10.1210/jcem-54-4-773)
Christian HC, Chapman LP & Morris JF 2007 Thyrotrophin-releasing hormone, vasoactive intestinal peptide, prolactin-releasing peptide and dopamine regulation of prolactin secretion by different lactotroph morphological subtypes in the rat. Journal of Neuroendocrinology 19 605–6 13. (https://doi.org/10.1111/j.1365-2826.2007.01567.x)
Cimpean AM, Ceausu AR, Corlan A, Melnic E, Jitariu AA & Raica M 2017 The ‘game’ of glial fibrillary acidic and S100 proteins in pituitary adenomas: two players or several? Endokrynologia Polska 68 380–389. (https://doi.org/10.5603/EP.a2017.0029)
Codish SD, Kraszeski J & Pratt K 1973 CNS developmental anomaly in the basal cell nevus syndrome: another congenital neurocutaneous syndrome? Neuropadiatrie 4 338–3 43. (https://doi.org/10.1055/s-0028-1091750)
Cox B, Roose H, Vennekens A & Vankelecom H 2017 Pituitary stem cell regulation: who is pulling the strings? Journal of Endocrinology 234 R135–R158. (https://doi.org/10.1530/JOE-17-0083)
Cramer H & Niederdellmann H 1983 Cerebral gigantism associated with jaw cyst basal cell naevoid syndrome in two families. Archiv für Psychiatrie und Nervenkrankheiten 233 111–1 24. (https://doi.org/10.1007/BF00343432)
Denef C, Schramme C & Baes M 1985 Stimulation of growth hormone release by vasoactive intestinal peptide and peptide PHI in rat anterior pituitary reaggregates. Permissive action of a glucocorticoid and inhibition by thyrotropin-releasing hormone. Neuroendocrinology 40 88–91. (https://doi.org/10.1159/000124057)
Fauquier T, Lacampagne A, Travo P, Bauer K & Mollard P 2002 Hidden face of the anterior pituitary. Trends in Endocrinology and Metabolism 13 304–30 9. (https://doi.org/10.1016/s1043-2760(0200616-1)
Fazekas I, Bacsy E, Varga I, Slowik F, Balint K, Pasztor E, Czirjak S & Glaz E 2000 Effect of vasoactive intestinal polypeptide (VIP) on growth hormone (GH) and prolactin (PRL) release and cell morphology in human pituitary adenoma cell cultures. Folia Histochemica and Cytobiologica 38 119–1 27.
Flemming GM, Klammt J, Ambler G, Bao Y, Blum WF, Cowell C, Donaghue K, Howard N, Kumar A & Sanchez J et al.2013 Functional characterization of a heterozygous GLI2 missense mutation in patients with multiple pituitary hormone deficiency. Journal of Clinical Endocrinology and Metabolism 98 E567–E5 75. (https://doi.org/10.1210/jc.2012-3224)
França MM, Jorge AA, Carvalho LR, Costalonga EF, Vasques GA, Leite CC, Mendonca BB & Arnhold IJ 2010 Novel heterozygous nonsense GLI2 mutations in patients with hypopituitarism and ectopic posterior pituitary lobe without holoprosencephaly. Journal of Clinical Endocrinology and Metabolism 95 E384–E3 91. (https://doi.org/10.1210/jc.2010-1050)
Gremeaux L, Fu Q, Chen J & Vankelecom H 2012 Activated phenotype of the pituitary stem/progenitor cell compartment during the early-postnatal maturation phase of the gland. Stem Cells and Development 21 801–8 13. (https://doi.org/10.1089/scd.2011.0496)
Hagen TC, Arnaout MA, Scherzer WJ, Martinson DR & Garthwaite TL 1986 Antisera to vasoactive intestinal polypeptide inhibit basal prolactin release from dispersed anterior pituitary cells. Neuroendocrinology 43 641–64 5. (https://doi.org/10.1159/000124594)
Hahn H, Wojnowski L, Zimmer AM, Hall J, Miller G & Zimmer A 1998 Rhabdomyosarcomas and radiation hypersensitivity in a mouse model of Gorlin syndrome. Nature Medicine 4 619–6 22. (https://doi.org/10.1038/nm0598-619)
Hirabayashi T, Nakamachi T & Shioda S 2018 Discovery of PACAP and its receptors in the brain. Journal of Headache and Pain 19 28. (https://doi.org/10.1186/s10194-018-0855-1)
Horiguchi K, Ilmiawati C, Fujiwara K, Tsukada T, Kikuchi M & Yashiro T 2012 Expression of chemokine CXCL12 and its receptor CXCR4 in folliculostellate (FS) cells of the rat anterior pituitary gland: the CXCL12/CXCR4 axis induces interconnection of FS cells. Endocrinology 153 1717–17 24. (https://doi.org/10.1210/en.2011-1937)
Horvath E & Kovacs K 2002 Folliculo-stellate cells of the human pituitary: a type of adult stem cell? Ultrastructural Pathology 26 219–2 2 8. (https://doi.org/10.1080/01913120290104476)
Hsu DW, Riskind PN & Hedley-Whyte ET 1989 Vasoactive intestinal peptide in the human pituitary gland and adenomas. An immunocytochemical study. American Journal of Pathology 135 329–3 3 8.
Kahn LB & Gordon W 1967 The basal cell naevus syndrome – report of a case. South African Medical Journal 41 832–83 5.
Kimonis VE, Goldstein AM, Pastakia B, Yang ML, Kase R, Digiovanna JJ, Bale AE & Bale SJ 1997 Clinical manifestations in 105 persons with nevoid basal cell carcinoma syndrome. American Journal of Medical Genetics 69 299–308. (https://doi.org/10.1002/(SICI)1096-8628(19970331)69:3<299::AID-AJMG16>3.0.CO;2-M)
Leprince J, Bagnol D, Bureau R, Fukusumi S, Granata R, Hinuma S, Larhammar D, Primeaux S, Sopkova-De Oliveiras Santos J & Tsutsui K et al.2017 The Arg-Phe-amide peptide 26RFa/glutamine RF-amide peptide and its receptor: IUPHAR review 24. British Journal of Pharmacology 174 3573–3607. (https://doi.org/10.1111/bph.13907)
Lo Muzio L, Nocini PF, Savoia A, Consolo U, Procaccini M, Zelante L, Pannone G, Bucci P, Dolci M & Bambini F et al.1999 Nevoid basal cell carcinoma syndrome. Clinical findings in 37 Italian affected individuals. Clinical Genetics 55 34–40. (https://doi.org/10.1034/j.1399-0004.1999.550106.x)
Long F, Zhang XM, Karp S, Yang Y & Mcmahon AP 2001 Genetic manipulation of hedgehog signaling in the endochondral skeleton reveals a direct role in the regulation of chondrocyte proliferation. Development 128 5099–5 108.
Madisen L, Zwingman TA, Sunkin SM, Oh SW, Zariwala HA, Gu H, Ng LL, Palmiter RD, Hawrylycz MJ & Jones AR et al.2010 A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nature Neuroscience 13 133–1 40. (https://doi.org/10.1038/nn.2467)
Marcos MV, Querol X, Armengol A, Hierro FR & Cruz M 1982 Basal cell nevus syndrome and gigantism. Anales Españoles de Pediatria 16 513–51 9.
Matsushita N, Kato Y, Katakami H, Shimatsu A, Yanaihara N & Imura H 1981 Stimulation of growth hormone release by vasoactive intestinal polypeptide from human pituitary adenomas in vitro. Journal of Clinical Endocrinology and Metabolism 53 1297–1 300. (https://doi.org/10.1210/jcem-53-6-1297)
Mazzocchi G, Rebuffat P, Gottardo L & Nussdorfer GG 1998 Vasoactive intestinal peptide stimulates rat adrenal glucocorticoid secretion, through an ACTH receptor-dependent activation of the adenylate cyclase signaling pathway. Hormone and Metabolic Research 30 241–24 3. (https://doi.org/10.1055/s-2007-978874)
Morris J & Christian H 2011 Folliculo-stellate cells: paracrine communicators in the anterior pituitary. Open Neuroendocrinology Journal 4 77–89. (https://doi.org/10.2174/1876528901104010077)
Murakami Y, Koshimura K, Yamauchi K, Nishiki M, Tanaka J, Furuya H, Miyake T & Kato Y 1995 Pituitary adenylate cyclase activating polypeptide (PACAP) stimulates growth hormone release from GH3 cells through type II PACAP receptor. Regulatory Peptides 56 35–40. (https://doi.org/10.1016/0167-0115(9500003-t)
Nishioka H, Llena JF & Hirano A 1991 Immunohistochemical study of folliculostellate cells in pituitary lesions. Endocrine Pathology 2 155–160. (https://doi.org/10.1007/BF02915456)
Pyczek J, Buslei R, Schult D, Holsken A, Buchfelder M, Hess I, Hahn H & Uhmann A 2016 Hedgehog signaling activation induces stem cell proliferation and hormone release in the adult pituitary gland. Scientific Reports 6 24928. (https://doi.org/10.1038/srep24928)
Reisine T, Heisler S, Hook VY & Axelrod J 1982 Multireceptor-induced release of adrenocorticotropin from anterior pituitary tumor cells. Biochemical and Biophysical Research Communications 108 1251–125 7. (https://doi.org/10.1016/0006-291x(8292134-9)
Roessler E, Du YZ, Mullor JL, Casas E, Allen WP, Gillessen-Kaesbach G, Roeder ER, Ming JE, Ruiz i Altaba A & Muenke M 2003 Loss-of-function mutations in the human GLI2 gene are associated with pituitary anomalies and holoprosencephaly-like features. PNAS 100 13424–1342 9. (https://doi.org/10.1073/pnas.2235734100)
Treier M, O’Connell S, Gleiberman A, Price J, Szeto DP, Burgess R, Chuang PT, Mcmahon AP & Rosenfeld MG 2001 Hedgehog signaling is required for pituitary gland development. Development 128 377–3 86.
Uhmann A, Dittmann K, Nitzki F, Dressel R, Koleva M, Frommhold A, Zibat A, Binder C, Adham I & Nitsche M et al.2007 The hedgehog receptor patched controls lymphoid lineage commitment. Blood 110 1814–18 23. (https://doi.org/10.1182/blood-2007-02-075648)
Vankelecom H 2010 Pituitary stem/progenitor cells: embryonic players in the adult gland? European Journal of Neuroscience 32 2063–20 81. (https://doi.org/10.1111/j.1460-9568.2010.07523.x)
Vila G, Papazoglou M, Stalla J, Theodoropoulou M, Stalla GK, Holsboer F & Paez-Pereda M 2005a Sonic hedgehog regulates CRH signal transduction in the adult pituitary. FASEB Journal 19 281–28 3. (https://doi.org/10.1096/fj.04-2138fje)
Vila G, Theodoropoulou M, Stalla J, Tonn JC, Losa M, Renner U, Stalla GK & Paez-Pereda M 2005b Expression and function of sonic hedgehog pathway components in pituitary adenomas: evidence for a direct role in hormone secretion and cell proliferation. Journal of Clinical Endocrinology and Metabolism 90 6687–66 94. (https://doi.org/10.1210/jc.2005-1014)
Vives V, Alonso G, Solal AC, Joubert D & Legraverend C 2003 Visualization of S100B-positive neurons and glia in the central nervous system of EGFP transgenic mice. Journal of Comparative Neurology 457 404–4 19. (https://doi.org/10.1002/cne.10552)
Vleck CM & Patrick DJ 1999 Effects of vasoactive intestinal peptide on prolactin secretion in three species of passerine birds. General and Comparative Endocrinology 113 146–1 54. (https://doi.org/10.1006/gcen.1998.7191)
Voit D, Saeger W & Ludecke DK 1999 Folliculo-stellate cells in pituitary adenomas of patients with acromegaly. Pathology, Research and Practice 195 143–14 7. (https://doi.org/10.1016/S0344-0338(9980026-0)
Wicking C, Shanley S, Smyth I, Gillies S, Negus K, Graham S, Suthers G, Haites N, Edwards M & Wainwright B et al.1997 Most germ-line mutations in the nevoid basal cell carcinoma syndrome lead to a premature termination of the PATCHED protein, and no genotype-phenotype correlations are evident. American Journal of Human Genetics 60 21–2 6.
Willems C, Fu Q, Roose H, Mertens F, Cox B, Chen J & Vankelecom H 2016 Regeneration in the pituitary after cell-ablation injury: time-related aspects and molecular analysis. Endocrinology 157 705–7 21. (https://doi.org/10.1210/en.2015-1741)
