Regulation and Function of the Sonic Hedgehog Signal Transduction Pathway in Isolated Gastric Parietal Cells*

Shh (Sonic hedgehog) regulates gastric epithelial cell differentiation. We reported that incubation of purified canine parietal cells with epidermal growth factor (EGF) for 6–16 h, stimulates H+/K+-ATPase α-subunit gene expression through the activation of Akt. We explored if Shh mediates some of the actions of EGF in the parietal cells. EGF induced a 6-fold increase in Shh expression, measured by Western blots, after 5 h of incubation. This effect was inhibited by both the phosphatidylinositol 3-kinase inhibitor LY294002 and by transduction of the cells with an adenoviral vector expressing dominant negative Akt. EGF stimulated the release of Shh-like immunoreactivity from the parietal cells, after 16 h of incubation. Shh induced H+/K+-ATPase α-subunit gene expression, assessed by Northern blots, it stimulated a luciferase reporter plasmid containing the EGF-responsive sequence (ERE) of the canine H+/K+-ATPase α-subunit gene promoter, and it induced parietal cell nuclear protein binding to the ERE. Gli transcription factors mediate the intracellular actions of Shh. Co-transfection of the parietal cells with the H+/K+-luc plasmid together with one expressing Gli2, induced H+/K+-luciferase activity 5-fold, whereas co-transfection of the cells with the H+/K+-luc plasmid together with one expressing dominant negative Gli2, inhibited EGF induction of H+/K+-luciferase activity. Identical results were observed in the presence of the Shh signal transduction pathway inhibitor, cyclopamine. Transfection of the cells with dominant negative Akt inhibited EGF, but not Shh stimulation of H+/K+-ATPase-luciferase activity. Thus, EGF but not Shh signals through Akt. Preincubation of the cells for 16 h with either Shh or EGF enhanced histamine-stimulated [14C]aminopyrine uptake by 50%. In conclusions, some of the actions of EGF in the parietal cells are mediated by the sequential activation of the Akt and the Shh signal transduction pathways. These effects might represent novel mechanisms mediating the actions of growth factors on gastric epithelial cell differentiation.

EGF 1 is a member of a large family of polypeptide growth factors that has been shown to exert numerous physiological actions in the gut such as regulation of growth, differentiation, restitution, and secretion (1)(2)(3)(4). Several studies have indicated that stimulation of gastric canine parietal cells in primary culture for 7-16 h with EGF exerts a direct stimulatory action on both the expression and the transcription of the H ϩ ,K ϩ -ATPase ␣-subunit, a gene considered to be a marker of parietal cell differentiation and a key element in the process of gastric acid production (5)(6)(7)(8)(9)(10)(11)(12).
The intracellular signal transduction pathways that mediate the multiple, complex actions of growth factors in the stomach have been only partially elucidated. We have recently reported that, in the canine gastric parietal cells, EGF stimulates H ϩ ,K ϩ -ATPase ␣-subunit gene expression and it enhances secretagogue-stimulated gastric acid secretion, through a signal transduction pathway that requires, at least in part, the activation of Akt (9). Akt has been linked to the induction of cellular growth and survival and to the expression and the maintenance of highly differentiated cellular phenotypes (13)(14)(15)(16). The specific intracellular targets of the Akt signal transduction pathway in the stomach are currently not known.
Sonic, Indian, and Desert hedgehog (Shh) are members of the family of the Hedgehog proteins, peptides known to exert important regulatory functions in patterning and growth in a large number of tissues during embryogenesis (17)(18)(19). In the mammalian stomach, in particular, Shh has been shown to be an important factor for the regulation of gastric epithelial cell maturation and differentiation (20 -22). Recent studies have shown that Shh null mice fail to develop a normal gastric epithelium (20) and that inhibition of Shh signaling in the gastric mucosa leads to enhanced cellular proliferation and to diminished expression of factors, such as bone morphogenic protein 4 and hepatocyte nuclear factor 3␤, which promote gastric epithelial cell differentiation (21).
The intracellular mechanisms that mediate the actions of Shh on target cells have been only partially understood. According to recent investigations, Shh appears to bind to a transmembrane receptor protein, known as patched (Ptc) which, in the absence of Shh, exerts an inhibitory effect on the seven transmembrane receptor smoothened (Smo). Binding of Shh to Ptc blocks the inhibitory effect of Ptc on Smo. Once activated, Smo induces a complex series of intracellular reactions that targets the Gli family of transcription factors (17,18,23,24). At least three members of this family of nuclear proteins have been identified in mammalian tissues (17,24). Al-though Gli1 and Gli2 are transcriptional activators, Gli3 seems to function primarily as a transcriptional repressor (17,24). Gli2 appears to be the principal effector of Shh signaling, because disruption of the Gli2 gene leads to developmental defects involving several Shh target tissues, while Gli1 null mice are born without detectable abnormalities (17,(25)(26)(27)(28)(29)(30).
Immunohistochemical studies conducted in the gastric mucosa of both mice and humans have demonstrated that Shh is expressed in the parietal cells (21), highly specialized gastric epithelial cells, known to produce and secrete growth factors, and regulatory peptides in the gastric mucosa (21,31,32). Indeed, loss of mature parietal cells, achieved by genetic, pharmacological, and immunological methods, appears to be associated with profound abnormalities in the differentiation and development of multiple cell lineages in the stomach (33)(34)(35)(36)(37). Taken together, these observations underscore the importance of the parietal cells in the regulation of complex programs of cellular growth and differentiation in the gastric mucosa.
Accordingly, we took advantage of a well established system based on primary cultures of highly purified canine gastric parietal cells to study the function and regulation of Shh in the stomach. In addition, we investigated if the stimulatory actions of EGF on parietal cell maturation and differentiation are mediated by the activation of the Shh signal transduction pathway.

Adenoviral Vectors and Plasmids-
The replication defective adenoviral vector expressing hemagglutinin-tagged, dominant negative Akt 1(with serine 473 and threonine 308 mutated to alanine, Ad.dn-Akt) under the control of the CMV promoter, was a gift of K. Walsh (Tufts University, Boston, MA) (38). The adenoviral vector expressing the ␤-galactosidase enzyme under the control of the CMV promoter (Ad-.CMV-␤-gal) was previously described (39). The plasmid 8X3ЈGli-BS-Luc (40) was a gift of H. Sasaki (Osaka University, Osaka, Japan). The vectors expressing Gli2⌬C4p and Gli2p, gifts of Andrzej Dlugosz (University of Michigan, Ann Arbor, MI), were generated by subcloning Gli2⌬C4 (24) and Gli2 (24)  Primary Parietal Cell Preparation and Culture-For preparation of primary parietal cells we utilized a modification of the method of Soll et al. (43)(44)(45)(46). The mucosal layer of freshly obtained canine gastric fundus was bluntly separated from the submucosa and rinsed in Hank's balanced salt solution containing 0.1% bovine serum albumin. The cells were dispersed by sequential exposure to collagenase (0.35 mg/ml) and 1 mM EDTA and parietal cells were enriched by centrifugal elutriation using a Beckman JE-6B elutriation rotor. Elutriator fractions 8 and 9, which contain up to 70% parietal cells as determined by hematoxylin and eosin and periodic acid-Shiff reagent staining, were further purified by centrifugation through density gradients generated by 50% Percoll (Amersham Biosciences) at 30,000 ϫ g for 20 min. The cell fraction at ϭ 1.05 consisted of virtually all parietal cells as determined by staining with an anti-H ϩ ,K ϩ -ATPase ␣-subunit monoclonal antibody (MBL, Nagoya, Japan). The isolated parietal cells (2 ϫ 10 6 cells/well) were cultured according to the method of Chew et al. with some modifications (44,45). Briefly, the cells were cultured in Ham's F-12/Dulbecco's modified Eagle's medium (1:1) containing 0.1 mg/ml gentamycin, 50 units/ml penicillin G, 0.01 mg/ml ciprofloxacin, and 2% Me 2 SO (Sigma) on 6-well culture dishes (Corning Inc., Corning, NY) coated with 150 l of H 2 O-diluted (1:5) growth factor reduced Matrigel (BD Biosciences). For our studies, the parietal cells were incubated, according to the experiments, with EGF (10 nM, BD Biosciences), histamine (100 M, Sigma), and recombinant Shh amino-terminal peptide (0.1-0.5 g/ml, R&D Systems, Minneapolis, MN) for various time periods. In some experiments, LY294002 (10 M, Calbiochem) and cyclopamine (1 M, Toronto Research Chemicals, North York, Ontario, Canada) were added 30 min prior to the addition of the stimulants. LY294002 and cyclopamine were dissolved in Me 2 SO; all other test substances were dissolved in water. Control experiments with untreated cells were performed by incubating the cells in either vehicle (0.1% Me 2 SO) or incubation buffer without the test sub-stances. The parietal cells were transduced, when indicated, with a multiplicity of infection of 100 of adenoviral vectors expressing either dominant negative Akt or ␤-galactosidase.
Northern Blot Analysis-The parietal cells were lysed with TRIzol (Invitrogen) according to the manufacturer's instructions. Northern blot hybridization assays were performed as previously described (47). Equal amounts of each RNA sample, with ethidium bromide (10 mg/ml) in a final volume of 20 l, were electrophoresed on a 1.25% agarose gel containing formaldehyde, and the RNA was transferred from the gel to nitrocellulose filters. The ethidium-stained ribosomal RNA bands in the gel were photographed before and after transfer to ensure that equivalent amounts of RNA were loaded onto each lane and that no residual RNA was left on the gel. The canine H ϩ ,K ϩ -ATPase ␣-subunit cDNA probe was a gift of Il Song (University of Michigan). The human glyceraldehyde-3-phosphate dehydrogenase cDNA probe was obtained from Clontech (Palo Alto, CA). The cDNAs were labeled with [ 32 P]dCTP by the random priming procedure, and the nitrocellulose filters were hybridized to the 32 P-labeled cDNA probes as previously described (47).
Transfection of Primary Cultured Parietal Cells-The parietal cells were transfected according to previously reported techniques (7,45,46). Before transfection the cells were washed once with 1 ml of Opti-MEM I serum-reduced media (Invitrogen) and fed with 400 l of Opti-MEM I media supplemented with 2% Me 2 SO. The cells were transfected with the luciferase reporter plasmids (3 g of HK(Ϫ619 ϩ 35)-Luc, 2 g of 8X3ЈGli-BS-Luc, 5 g of ERE-HK(Ϫ54 ϩ 34)-Luc and 2 g of SRE-Luc) and, where indicated, with different amounts of the expression vectors (1 g of dnAkt, 1 g of Gli2⌬C4, 0.75 g of Gli2, 0.25 g of pCMV-␤Gal). Control experiments were conducted in the presence of the empty vectors. Transfections were carried out using Lipofectin (Invitrogen) as previously described (7,45,46). The day after transfection the cells were fed with fresh media and stimulated with the test substances. At the end of the incubation period (16 h), the cells were washed twice with cold calcium-free PBS. After addition of 100 l of cell lysis buffer (25 mM Tris-HCl, pH 7.8, 2 mM EDTA, 2 mM 1,2-diaminocyclohexane-N,N,NЈ,NЈ-tetraacetic acid, 10% glycerol, 1% Triton X-100), the cells were incubated at room temperature for 15 min. The cells were then scraped and transferred to Eppendorf tubes. After quick centrifugation to pellet large debris, the supernatant was transferred to a new tube. , and luminescent intensity was measured for 10 s utilizing a Lumat LB9501 luminometer (Berthold, Germany). Luciferase activity was expressed as RLU (relative light units) and normalized for ␤-galactosidase activity. ␤-Galactosidase activity was measured by the luminescent light derived from 10 l of each sample incubated in 100 l of Lumi-Gal 530 (Lumigen, Southfield, MI) and used to correct the luciferase assay data for transfection efficiency.
Electrophoretic Mobility Shift Assays-For gel mobility shift assays, nuclear extracts from parietal cells were prepared using previously described techniques (7,45). Extracts were stored at Ϫ80°C, and protein concentrations were measured by the Bradford method using bovine serum albumin standards (48). The oligonucleotide probe used to quantify the ERE-binding proteins present in the nuclear extracts was 5Ј-CTAGCAGACATGGCAGATC-3Ј (7). Labeling of the probe was performed using [ 32 P]dATP and the T4 kinase reaction (45). The specificity of nuclear protein binding to the ERE probe was confirmed by competition with an unlabeled ERE probe. Gel shift mobility assays were performed as previously described (7, 45) using 5 g of parietal cells nuclear extracts and the 32 P-labeled ERE oligonucleotide probe (Ϸ10,000 cpm/0.5 ng) in a 20-l solution containing 10 mM Tris-HCl (pH 7.9), 100 mM KCl, 5 mM MgCl 2 , 1 mM EDTA, 1 mM dithiothreitol, 10% glycerol, and 300 ng of double-stranded poly(dI-dC) as nonspecific competitor. The mixture was preincubated, when indicated, at 25°C for 10 min with unlabeled competitors, for 10 min with labeled probe and then resolved on a 4% nondenaturing polyacrylamide gel (acrylamide/bisacrylamide ratio, 30:1). After pre-electrophoresis at 10 V/cm for 2 h, gels were run at 10 V/cm in 45 mM Tris base/45 mM boric acid/1 mM EDTA. The gels were dried prior to autoradiography at Ϫ80°C.
Amplification and Purification of Adenoviral Vectors-Briefly, the recombinant adenoviruses were amplified as previously described using HEK293 cells (39). The viruses were subsequently concentrated and purified on a cesium chloride gradient (39). The concentration of the recombinant adenoviruses was assessed on the basis of the absorbance at 260 nm and on a limiting dilution plaque assay (39).
Histochemistry-These studies were carried out according to previously reported methods with minor modifications (49,50). Briefly, the parietal cells were cultured on glass slides and fixed in 4% paraformaldehyde. The cells were permeabilized in 0.2% Triton X-100 for 15 min at room temperature. After three rinses with PBS the slides were blocked for 1 h in 5% milk, PBS. At the end of the incubation period, the slides were rinsed with PBS and incubated, for 1 h at room temperature, with rabbit polyclonal antibodies directed against either Shh (sc-9024, 1:50) or ptc (sc-9016, 1:50) and with mouse monoclonal antibodies directed against the H ϩ ,K ϩ -ATPase ␣-subunit 1:200, MBL). The primary antibodies were diluted in PBS, 1% milk. Immunostaining for either Shh or ptc preceded staining of the parietal cells with the H ϩ ,K ϩ -ATPase ␣-subunit antibodies. After three rinses with PBS, the slides were incubated with specific donkey secondary antibodies (1:100, Jackson ImmunoResearch Laboratories, West Grove, PA) diluted in PBS, 1% milk, for 1 h at room temperature. The secondary antibodies were labeled with either fluorescein isothiocyanate (FITC) or Cy-3. At the end of the incubation periods with the secondary antibodies, the slides were rinsed with PBS and mounted in Vectashield (Vector Laboratories, Burlingame, CA). In control experiments, the parietal cells were incubated with the secondary antibodies without the primary antibodies. Visualization of slides was performed with a Zeiss LSM 510 version 3.2 confocal microscope. The microscope settings, such as lasers, excitation wavelengths, emission filters, and gain settings, were initially defined for proper image acquisition using empirical image quality. The settings were stored in the microscope data base and recalled through the re-use function for analysis of all subsequent experiments. With this system, the microscope settings were strictly controlled and maintained constant for all images generated. Fluorescence excitation was provided by an argon laser at 488 nm (FITC) and a helium-neon laser at 543 nm (Texas Red). The emission filters were a bandpass filter set at 505-550 nm and a longpass filter set at 560 nm. The detector gains were 788 (Zeiss value) and 606 (Zeiss value) for FITC and Texas Red, respectively.
Aminopyrine Uptake-Gastric acid secretion was measured according to previously described methods (7,9). Briefly, the accumulation of the weak base [ 14 C]aminopyrine (Amersham Biosciences) was used as an indicator of acid production by parietal cells. The cultured parietal cells were washed once with Earle's balanced salt solution, incubated with 0.1 Ci of [ 14 C]aminopyrine for 60 min, and then stimulated, when indicated, with histamine for the last 30 min of aminopyrine incubation. In some experiments, the cells were cultured for 16 -18 h in the presence of either EGF (10 nM) or Sonic hedgehog (0.5 g/ml) prior to the addition of histamine. The parietal cells were lysed with 500 l of 1% Triton X-100, and the radioactivity of lysate was quantified in a liquid scintillation counter as previously reported (7,9).
Gel Filtration Chromatography-Shh in the parietal cell-conditioned media was identified using previously described methods (51). Briefly, equal volumes of media derived from equal numbers of parietal cells, which were either left untreated or stimulated with 10 nM EGF for 16 h, were collected and centrifuged at 2000 rpm for 15 min. The superna-tants were concentrated using Centricon YM-10 (Millipore, Bedford, MA) and loaded on to a Superose 12 (Amersham Biosciences) gel filtration column that had been equilibrated with PBS/0.01% Nonidet P-40. The column was calibrated by denoting the elution fraction of a recombinant Shh, amino-terminal peptide standard. The fractions were collected, concentrated with Centricon YM-10, and subjected to Western blot analysis using specific rabbit anti-Shh antibodies (sc-9024).
Data Analysis-Data are expressed as means Ϯ S.E., wherein n is equal to the number of separate dog preparations from which the parietal cells were obtained. Statistical analysis was performed using Student's t test. p values Ͻ 0.05 were considered to be significant.

RESULTS
We first investigated if Shh is expressed in the gastric parietal cells. For these studies we used non-Percoll-purified gastric epithelial cells from elutriator fractions 8 and 9, which are enriched in parietal cells (50). As shown in Fig. 1 (A-C), immunofluorescence staining of the cells, with an anti-H ϩ /K ϩ -ATPase ␣-subunit primary antibody and a Cy-3-conjugated secondary antibody together with an anti-Shh primary antibody and a FITC-conjugated secondary antibody, demonstrated co-localization of Shh with the H ϩ /K ϩ -ATPase ␣-subunit in the gastric parietal cells. The specificity of the immunohistochemical staining was underscored by the observation that a few cells did not stain for both the H ϩ /K ϩ -ATPase ␣-subunit and Shh. Similar results were observed when the cells were stained with the anti-H ϩ /K ϩ -ATPase ␣-subunit antibody together with an antibody recognizing the Shh transmembrane receptor protein, ptc (Fig. 1, D-F). No staining was detected in control experiments in which the slides were incubated with either the Cy-3-or FITC-conjugated secondary antibodies without the primary antibodies (Fig. 1, G-I). Accordingly, both ptc and Shh appear to be specifically localized in the canine parietal cells.
We examined the effect of EGF on Shh expression in the gastric parietal cells using Western blots with anti-Shh antibodies. As shown in Fig. 2 (A and B), EGF (10 nM) induced a 6-fold increase in the expression of a protein of ϳ35 kDa that reacted with the anti-Shh antibody, and this effect was blocked by the addition of LY294002, a specific and well characterized inhibitor of PI3K. Both LY294002 alone and vehicle (0.1% Me 2 SO) had no independent effects on the gastric parietal cells ( Fig. 2 and data not shown). Because induction of PI3K leads to Akt activation, we examined if EGF stimulation of Shh requires the activation of Akt. As shown in Fig. 2 (C and D), EGF stimulation of Shh expression was inhibited by transduction of the parietal cells with an adenoviral vector expressing a dominant negative Akt gene. Thus, in the gastric parietal cells, EGF targets Shh through PI3K-and Akt-dependent signaling pathways. The specificity of these findings was demonstrated by the observation that EGF did not affect the expression of actin in the gastric parietal cells (Fig. 2, A and C).
We investigated if EGF stimulates Shh release from the parietal cells. For these studies the cells were either left untreated or stimulated for 16 h with 10 nM EGF. At the end of the incubation period, the media from both control and stimulated cells were collected, concentrated, and subjected to gel filtration chromatography. The column was calibrated by using recombinant Shh amino-terminal peptide standards. Fractions known to contain Shh-like immunoreactivity were collected, concentrated, and analyzed by Western blots. As shown in Fig.  3, EGF stimulated the release in the media of a protein of ϳ19 kDa that was detected by the anti-Shh antibody. Thus, EGF promotes the release of Shh from the parietal cells.
We examined if Shh regulates H ϩ /K ϩ -ATPase ␣-subunit gene expression. H ϩ /K ϩ -ATPase ␣-subunit mRNA was isolated from the parietal cells and quantitated by Northern blots using a canine H ϩ /K ϩ -ATPase ␣-subunit gene cDNA probe. As shown in Fig. 4, Shh stimulated H ϩ /K ϩ -ATPase ␣-subunit gene expression with a maximal effect observed at the dose of 0.5 g/ml.
To elucidate if Shh induces H ϩ /K ϩ -ATPase ␣-subunit gene transcription, the parietal cells were transfected with a luciferase reporter plasmid containing 619 bp of the H ϩ /K ϩ -ATPase ␣-subunit gene promoter, which was previously reported to respond to EGF in the parietal cells (7). Shh induced H ϩ /K ϩ -ATPase ␣-subunit gene transcription 2-fold (Fig. 5A). Accordingly, in the gastric parietal cells, Shh stimulates both the expression and the transcription of the H ϩ /K ϩ -ATPase ␣-subunit gene. Because, in the canine parietal cells, EGF is known to induce H ϩ /K ϩ -ATPase ␣-subunit gene transcription through a specific EGF response element or ERE, located between basis 162 and 156 of the promoter (7), we sought to investigate if Shh targets the H ϩ /K ϩ -ATPase ␣-subunit gene ERE. As shown in Fig. 5B, Shh induced the transcriptional activity of the H ϩ /K ϩ -ATPase ␣-subunit gene ERE, 2-fold, and this effect was similar to that observed in the presence of EGF. To further demonstrate that Shh regulates the activity of the ERE, we performed gel shift assays using parietal cell nuclear extracts and a 32 Plabeled ERE probe. As shown in Fig. 5C, Shh induced nuclear protein binding to the ERE, confirming the notion that Shh regulates the function of the EGF-responsive element of the H ϩ /K ϩ -ATPase ␣-subunit gene promoter.
Some of the intracellular actions of Shh are mediated by the Gli family of transcription factors, nuclear proteins known to bind to specific DNA regulatory elements present in the promoter of several genes involved in the regulation of cellular growth and differentiation (17,(25)(26)(27)(28)(29)(30). Accordingly, we investigated if Gli2, which is thought to mediate the intracellular actions of Shh, participates in the regulation of H ϩ /K ϩ -ATPase ␣-subunit gene transcription in response to EGF stimulation. First, we examined if, in the parietal cells, EGF induces signal transduction pathways that target Gliresponsive DNA regulatory elements. For these studies, the parietal cells were transfected with plasmids expressing the luciferase reporter gene under the control of eight Gli-responsive elements (8X3ЈGli-BS-Luc) (40). EGF stimulated the transcriptional activity of the 8X3Ј Gli-BS-Luc plasmid more then 2-fold 6A). In additional studies, the parietal cells were co-transfected with the luciferase reporter plasmid containing 619 bp of the H ϩ /K ϩ -ATPase ␣-subunit gene promoter, together with a vector expressing Gli2. As shown in Fig. 6B, Gli2 induced H ϩ /K ϩ -ATPase-Luciferase activity 5-fold. To define if Gli2 is involved in the stimulatory action of EGF on H ϩ /K ϩ -ATPase ␣-subunit gene transcription, we performed experiments in which the parietal cells were transfected with the H ϩ /K ϩ -ATPase-Luciferase plasmid together with one expressing a dominant negative Gli2 gene. As shown in Fig. 6C, EGF induced H ϩ /K ϩ -Luciferase activity more then 2-fold, and dominant negative Gli2 inhibited this effect. Taken together, these observations suggest that Gli2 is involved in the mediation of the stimulatory action of EGF on H ϩ /K ϩ -ATPase ␣-subunit gene transcription.
To confirm that EGF targets the H ϩ /K ϩ -ATPase ␣-subunit gene through the activation of the Shh signal transduction pathway, we examined the effect of cyclopamine, a specific inhibitor of the transmembrane receptor smoothened (52) on H ϩ /K ϩ -ATPase ␣-subunit gene transcription. As shown in Fig.  7A, EGF induced H ϩ /K ϩ -ATPase-Luciferase activity more then 2-fold and cyclopamine inhibited this effect, confirming the notion that the Shh signal transduction pathway is involved in the stimulatory action of EGF on H ϩ /K ϩ -ATPase ␣-subunit gene transcription. The specificity of this effect was confirmed by the observation that cyclopamine failed to inhibit the stimulatory action of EGF on c-fos gene transcription (Fig. 7B).
We reported that Akt regulates both H ϩ /K ϩ -ATPase ␣-subunit gene expression and Shh generation. Accordingly, we investigated the role of Akt in H ϩ /K ϩ -ATPase ␣-subunit gene transcription in response to either EGF or Shh. For these studies, the parietal cells were co-transfected with the H ϩ /K ϩ -ATPase ␣-subunit luciferase reporter plasmid together with a vector expressing a dominant negative Akt gene. As shown in Fig. 8A, dominant negative Akt inhibited EGF induction of H ϩ /K ϩ -ATPase ␣-subunit gene transcription. In contrast, dominant negative Akt failed to inhibit Shh stimulation of H ϩ /K ϩ -ATPase ␣-subunit gene transcription (Fig. 8B), suggesting that EGF but not Shh signals through Akt. These findings were further confirmed by Western blots with anti-phospho Akt antibodies, which demonstrated that EGF, but not Shh, induces the activation of Akt (Fig. 8C).
To assess the functional significance of Shh in gastric acid secretion, we tested the effect of Shh on [ 14 C]aminopyrine uptake. As depicted in Fig. 9, treatment of the parietal cells with Shh alone had no effect on [ 14 C]aminopyrine uptake. In contrast, preincubation of the parietal cells with Shh for 16 h, prior to stimulation of the cells with histamine for 30 min, led to a statistically significant enhancement of the stimulatory effect of histamine. These effects were identical to those ob- served in the presence of EGF. Thus, Shh appears to be important for the enhancement of secretagogue-stimulated gastric acid production. DISCUSSION Numerous studies have underscored the importance of the parietal cells in the biology and pathobiology of the gastric mucosa. Several reports have indicated that the parietal cells are the major site of production of growth factors and regulatory peptides in the gastric epithelium (21,31,32). In fact, in addition to TGF-␣ the parietal cells have been shown to express Shh, a peptide known to play an important role in the regulation of complex programs of cellular growth and differentiation in the stomach (20 -22).
In this study, we demonstrate that the canine parietal cells express both Shh and its transmembrane receptor protein, Ptc. In addition, we report that EGF, a growth factor that has been shown to exert important regulatory actions on the growth and differentiation of the gastric epithelium (1)(2)(3)(4), stimulates the expression and the release of Shh from the parietal cells, suggesting that this peptide might be an important mediator of the actions of EGF in the stomach through the activation of both paracrine and autocrine mechanisms.
Shh is known to undergo extensive post-translational modifications to gain biological activity (18,53). In most cells, the full-length form of Shh migrates with a mobility that corresponds to a relative mass of ϳ46 kDa (53). This protein is then converted to a species of ϳ39 kDa, which is thought to represent the signal-cleaved form of the full-length peptide. Further processing of this molecular form of Shh appears to generate smaller molecules of 19, 25, and 16 kDa, respectively. Of these, the 19-kDa amino-terminal peptide undergoes esterification at its C terminus to a cholesterol molecule prior to being secreted into the extracellular space where it mediates the physiological actions of Shh (18,53,54). In agreement with these reports, we observed that, in the parietal cells, EGF stimulates the expression of a protein of ϳ35 kDa, which is likely to represent the signal-cleaved form of full-length Shh. In addition, we noted that EGF induces the release, in the culture medium, of the 19-kDa molecular form of Shh which is thought to mediate the biological activities of Shh on target cells (18,53,54). Several mechanisms could be considered to explain these observations. One possibility would be that EGF stimulates the processing of Shh, leading to increased production of the smaller molecular forms of this peptide. Alternatively, EGF might induce either the expression of the Shh gene or the translation of the Shh protein, leading, in the end, to increased production, processing, and release of Shh. It is clear that future studies will be necessary to better define these important biochemical issues.
EGF has been shown to stimulate H ϩ ,K ϩ -ATPase ␣-subunit gene expression. Previous studies conducted in canine parietal cells in primary culture transfected with plasmids containing a series of H ϩ ,K ϩ -ATPase ␣-subunit gene promoter deletion mutants, demonstrated that the sequence comprised between bases Ϫ619 to ϩ34 of the promoter, is EGF-responsive (7). In addition, those reports identified and characterized a specific element located between bases Ϫ162 and Ϫ156 of the promoter, which appears to be sufficient to confer EGF responsiveness (7). In this report, we propose that the regulatory actions of EGF on the H ϩ ,K ϩ -ATPase ␣-subunit gene might be, at least in part, mediated by Shh. In support of this hypothesis, we demonstrate that Shh induces H ϩ ,K ϩ -ATPase ␣-subunit gene expression and that it stimulates the activation of the H ϩ ,K ϩ -ATPase ␣-subunit gene promoter EGF-responsive regulatory element. In addition, we report that both cyclopamine, a specific inhibitor of Smo activation (52), and a dominant negative Gli2 gene are able to inhibit the stimulatory actions of EGF on H ϩ ,K ϩ -ATPase ␣-subunit gene transcription, confirming the notion that the Shh signal transduction pathway is involved in the regulation of some of the actions of EGF in the parietal cells.
Gli transcription factors mediate the intracellular actions of Shh (17,(25)(26)(27)(28)(29)(30). Gli2, in particular, appears to be the principal transducer of the physiological effects of Shh in target cells (17,(25)(26)(27)(28)(29)(30). Our studies suggest that Gli2 is involved in the regulation of the stimulatory actions of EGF on the H ϩ ,K ϩ -ATPase ␣-subunit gene, because expression of Gli2 potently activates the H ϩ ,K ϩ -ATPase ␣-subunit gene promoter, whereas a dominant negative mutant of this transcription factor blocks EGF stimulation of H ϩ ,K ϩ -ATPase ␣-subunit gene transcription. These findings are further supported by the observation that, in the parietal cells, EGF induces signal transduction pathways that target a luciferase reporter plasmid containing Gli-responsive elements. Although these results strongly support the notion that Gli2 belongs to a cascade of biochemical events that link EGF to the H ϩ ,K ϩ -ATPase ␣-subunit gene promoter, they do not allow us to define if the actions of Gli2 on the H ϩ ,K ϩ -ATPase ␣-subunit gene are direct or indirect. Indeed, it would be conceivable to speculate that Gli2 might induce the sequential transcription of still undetermined, downstream nuclear proteins, which, in turn, could exert regulatory functions on the H ϩ ,K ϩ -ATPase ␣-subunit gene. Alternatively, even if the ERE is not a canonical Gli binding site, we cannot exclude that Gli transcription factors could interact directly with the H ϩ ,K ϩ -ATPase ␣-subunit ERE. Future experiments will be needed to define these intriguing possibilities.
In previous studies we observed that Akt, plays a crucial role in the process of parietal cell maturation and differentiation (9,50). Here we report that, in the parietal cells, Akt mediates the stimulatory action of EGF on the expression of Shh, underscoring the importance of this kinase in the regulation of gastric epithelial cell differentiation. Recent studies have indicated that Shh induces the activation of Akt in immortalized murine brain capillary endothelial cells (55). In contrast to these findings, our results clearly demonstrate that Shh does not induce Akt in the parietal cells. Thus, in the stomach, Akt appears to regulate the expression of Shh, an important mediator of parietal cell maturation and differentiation, in response to growth factor stimulation. These findings suggest the existence of a novel mechanism linking a growth factor, such as EGF, to Shh, through the induction of the Akt signal transduction pathway.
One important observation of our study is that incubation of the parietal cells with Shh leads to enhancement of secretagogue-stimulated gastric acid secretion, an event that appears to be identical to that observed in the presence of EGF (7,9). Thus, Shh could be one of the effectors of the stimulatory actions of EGF on H ϩ /K ϩ -ATPase ␣-subunit gene expression and gastric acid production.
In conclusion, our study demonstrates that Akt is a crucial molecular switch that regulates the expression of Shh. Moreover, the Shh signal transduction pathway appears to mediate some of the actions of EGF in the parietal cells. These findings shed new insight into the complex signal transduction pathways that mediate the actions of growth factors in the stomach, providing new clues for a better understanding of the mechanisms that regulate gastric epithelial cell growth and differentiation.