| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
HORMONES AND SIGNALING
Gastrointestinal Diseases Research Unit, Kingston General Hospital, and Department of Physiology, Queen's University, Kingston, Ontario, Canada
Submitted 19 May 2006 ; accepted in final form 26 November 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
bone morphogenetic protein 2; barrier
7 mM (33). Perfusates containing 7–10 mM Ca2+ resulted in increases of extracellular Ca2+ (Cao2+ ) from 1 to 1.9 mM in the duodenal submucosal interstitium (33). These changes are of sufficient magnitude to be detected by the G protein-coupled, extracellular calcium-sensing receptor (CaSR) (5, 6, 19). The CaSR has been shown to be expressed on the epithelia of the stomach, small intestine, and colon (8–10). We speculated that if some cells in the lamina propria expressed the CaSR, then the postprandial changes in Cao2+ occurring regularly could stimulate these cells to signal to the luminal epithelia and perhaps contribute to epithelial homeostasis (25). One such cell type that might express the CaSR is the colonic subepithelial myofibroblast, which sits under the basement membrane and is physically poised to signal to the intestinal epithelia.
The CaSR was originally cloned from the parathyroid gland but was subsequently found to be widely expressed on cell types uninvolved in Ca2+ homeostasis (6). Evidence about the CaSR's physiological role in the gut is limited to the colon, where its activation by either Cao2+ or spermine inhibits forskolin-stimulated secretion (9, 10). High Ca2+ has been shown to inhibit 
-catenin signaling on colonic cancer cell lines (7), and expression of the colonic CaSR is progressively lost during the transition from adenoma to carcinoma (7, 41). When activated in heterologous or endogenously expressing cells, the CaSR stimulates several MAPKs (MEK1, MEK2, ERK1, ERK2, p38, and JNK) using filamin A as a scaffold (21, 23, 26, 42, 43). Recently, it has been reported that CaSR activation "transactivates" the EGF receptor (EGFR) through a triple membrane-spanning signaling cascade (27).
Intestinal myofibroblasts are a resident cell type in the lamina propria (36, 37). Numerous studies have documented the constitutive secretion of a large number of growth factors from myofibroblasts with changes in growth factor secretion occuring during inflammatory bowel diseases (28, 29). The 18Co human colonic myofibroblast cell line has been extensively used to demonstrate signaling cascades from these cells in response to the proinflammatory cytokine IL-1 (31, 32) and secretes TGF- isoforms (37).
Bone morphogenetic proteins (BMPs) belong to the TGF- superfamily. BMP-4 is a canonical Wnt target gene, and BMP-2 has recently been shown to be expressed in the epithelia of surface and crypt cells in the human and mouse colon (17). Noggin, the diffusible antagonist of BMP receptor (BMPR) interactions, has also been reported to be expressed in the human and mouse colon (17). Evidence for a functional role of BMP-2 in the intestine is scant as homozygous knockout mice for BMP-2 die at birth (44). A recent report (17) has suggested that exogenous BMP-2 stimulates the upregulation of some markers of differentiation and promotes apoptosis in colonic cell lines, but in vivo data showed equal rates of proliferation and apoptosis when mice were fed chimeric Noggin.
In the present study, we show that colonic myofibroblasts (18Co cells) express the CaSR and that activation of the CaSR results in the synthesis and secretion of BMP-2. We defined the signal transduction cascade activated by the CaSR to cause BMP-2 synthesis and secretion and showed, using a scrape-migration assay of postconfluent Caco-2 cells, that this BMP-2 is biologically active in accelerating restitution. Using T-84 colonic adenocarcinoma cells, which express the CaSR (14), grown on Transwells to model the intestinal barrier, we added exogenous BMP-2 and observed increases in transepithelial resistance (TER) that occurred with cells of both low and high resistances. We further determined that CaSR activation of myofibroblasts and the intestinal epithelia substantially downregulates transcripts for BMP-4, a Wnt target, and eliminates the transcript and protein of Noggin, the BMP antagonist in both cell types. Therefore, the net effect of CaSR activation is to increase the effective concentration of BMP-2 in epithelia and subepithelia. The released BMP-2 accelerates the restitution and barrier development of the intestine. This represents a new function of BMP-2 in the adult intestine.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Cell culture. 18Co, T-84, HT-29, SW480, and Caco-2 cell lines were purchased from the American Type Culture Collection (ATCC; Rockville, MD). Stably transfected CaSR HEK cells were a gift from E. M. Brown (Harvard Medical School, Boston, MA). The 18Co line was cultured in Minimal Essential Medium Alpha supplemented with 10% FBS and 100 µU/ml penicillin and grown at 37°C in a humidified 5% CO2 atmosphere. Cells were passaged weekly with 0.25% trypsin and used for experimentation within the first six passages. HT-29 and Caco-2 cells were cultured in DMEM supplemented with 10% or 20% FBS, respectively, and 100 µU/ml penicillin. T-84 cells were seeded to reach confluency in Transwells in DMEM-F-12 supplemented with 10% FBS and 100 µU/ml penicillin. Control human colonic myofibroblasts isolated by conventional protocols (28, 29) and transformed by telomerase were a generous gift from R. Mifflin (Galveston, Texas).
Generation of short interfering RNA against the CaSR. We found 13 potential short interfering (si)RNA target sites in the extracellular domain of the CaSR. Based on a GC content of >40 but <57%, we evaluated siRNA target sites by a BLAST analysis and then selected those that gave <14 hits. The present experiments used nucleotides 371–390 (5'-AACCTTGATGAGTTCTGCAAC-3'), to which were added complementary nucleotides to a T7 promoter primer. Sense and antisense oligonucleotide templates were generated, and transcribed siRNA was produced according to the manufacturer's protocol (Silencer siRNA Construction Kit, Ambion). Nucleotides 371–390 were scrambled, and a BLAST analysis was preformed to confirm the lack of specificity of this construct. T7 sites were added to this oligonucleotide template, and it was processed as described above. 18Co cells were transfected with 25–50 nM siRNA of either the scrambled construct or siRNA against the CaSR using Superfect according to the manufacturer's instructions (Superfect, Qiagen, Valencia, CA).
Transfection of 18Co cells with dominant negative CaSR. 18Co cells were transiently transfected with dominant negative CaSR (R185Q) using Superfect according to the manufacturer's instructions (Qiagen). Cells were then incubated for 18 h in serum-free, Ca2+-free DMEM containing 4 mM L-glutamine, 0.2% BSA, and 0.5 mM CaCl2 and stimulated with high (5 mM) and low (0.5 mM) Ca2+. At 18 h of stimulation, aliquots of media were centrifuged at 12,000 rpm at 4°C for 3 min, and supernatants were used for immediate BMP-2 ELISA determination.
Western blot analysis of the CaSR in 18Co cells. Immunoblot analysis for the CaSR was performed using confluent monolayers of 18Co cells that had been cultured in 10-cm plastic plates. Monolayers were rinsed once with ice-cold PBS containing 1 mM sodium vanadate and 25 mM NaF and scraped on ice into lysis buffer containing 10 mM Tris·HCl (pH 7.4), 1 mM EGTA, 1 mM EDTA, 0.25 M sucrose, 1% Triton X-100, 1 mM dithiothreitol, and a cocktail of protease inhibitors (10 µg/ml each of aprotinin, leupeptin, calpain, pepstatin, and soybean trypsin inhibitor as well as 100 µg/ml Pefabloc). NaF, sodium vanadate, and Pefabloc were freshly prepared on the same day of the experiments. Lysates were sonicated for 5 s and centrifuged at 10,000 g at 4°C for 10 min, and the resultant total cellular lysate in the supernatant was stored at –70°C for batch analysis. Protein concentrations were measured using the Bio-Rad protein assay. Aliquots of supernatant fractions containing 80 µg protein from 18Co cells and 5 µg protein of CaSR HEK cells were mixed with an equal volume of 2x SDS-Laemmli gel loading buffer under nonreducing conditions, incubated at 65°C for 30 min, and resolved electrophoretically on 6.5% acrylamide gels. Separated proteins were then transferred to Immobilon 0.2-µm membranes and incubated with blocking solution (PBS with 0.1% Tween 20 and 5% dry milk in 20 ml) for 1 h at room temperature. Blots were subsequently incubated overnight at 4°C with PA 1-1034 affinity-purified polyclonal antiserum at a 1:100 dilution. Blots were washed five times with PBS containing 0.1% Tween 20 at room temperature for 5 min each. Blots were further incubated with a 1:2,000 dilution of horseradish peroxidase-coupled goat anti-rabbit IgG (Sigma) in blocking solution for 1 h at room temperature. Blots were then washed five times in washing solution, and bands were visualized by chemiluminescence according to the manufacturer's instructions (Supersignal, Pierce Chemical).
Phospho-Akt, total Akt, and Noggin Western blot analysis. For the determination of Akt or Noggin of 18Co cells, early passage cells were grown on six-well dishes. Cells were incubated for 18 h in serum-free, Ca2+-free DMEM (as described above). This medium was removed and substituted with the same medium supplemented with 0.5 or 5 mM Ca2+. At the end of the incubation period, cells were washed once with ice-cold PBS containing 1 mM sodium vanadate and 25 mM NaF, and 100 µl of ice-cold lysis buffer were added [20 mM Tris·HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 25 mM NaF, 1% Triton X-100, 10% glycerol, 1 mM dithiothreitol, 1 mM sodium vanadate, 50 mM glycerophosphate, and a cocktail of proteases inhibitors, as described above]. After being sonicated for 5 s, lysates were centrifuged at 10,000 g for 10 min at 4°C and processed as previously described (26, 27). Total Akt and phospho-Akt were detected by an overnight incubation with a 1:1,000 dilution of rabbit polyclonal antibodies against Akt or phospho-Akt in 1x PBS and 0.1% Triton X-100 with 5% BSA. Noggin was detected with a 1:500 dilution of primary antibody in PBS and Triton X-100 with 2.5% milk. Blots were washed for three 5-min periods at room temperature (1x PBS and 0.1% Triton X-100) and then incubated for 1 h with a secondary antibody conjugated to horseradish peroxidase (1:2,000) in blocking solution. Blots were then washed a second time (3 x 5 min). Bands were visualized by chemiluminiscence as previously described (25). Quantitation of the phosphorylation of Akt was done using an ImageQuant and a Personal Densitomenter (Molecular Dynamics).
ELISA determinations. Cells were incubated for 18 h in serum-free, Cao2+-free DMEM containing 4 mM L-glutamine, 0.2% BSA, and 0.5 mM CaCl2. For kinetic studies, this medium was removed and substituted with the same medium supplemented with CaCl2 (5 mM) for 0, 3, 6, 12, 18, and 24 h. Based on preliminary results, the well-characterized CaSR agonists GdCl3 (25 mM), neomycin sulfate (300 mM), and spermine (2 mM), inhibitors of various MAPKs, and PKC inhibitors were assessed at 18 h. At the end of the incubation period, aliquots of media were centrifuged at 12,000 rpm at 4°C for 3 min, and supernatants were used for immediate BMP-2 determinations. Similarly, HT-29, SW-480, and T-84 cells were incubated in serum-free and 0.5 mM Cao2+ DMEM over 18 h, and supernants were collected for immediate BMP-2 ELISA determination.
RT-PCR. Differential gene expression analysis was performed by semiquantitative PCR using a Mastercycler (Eppendorf, Hamburg, Germany). Total RNA of 18Co cells was isolated by TRIzol reagent by following manufacturer's instructions. The recovered RNA was quantitated by spectrophotometry, and aliquots of 1 µg total RNA from low (0.5 mM) or high (5 mM) Cao2+ were used for cDNA synthesis at 37°C for 1 h in a volume of 20 µl containing the following reagents: 5 mM dNTP mix, 10 µM oligo-dT primer (Invitrogen, Carlsbad, CA), 10 units RNase inhibitor (Roche), 4 units Omniscript reverse transcriptase (Qiagen), and 1x buffer RT (Qiagen). The PCR was prepared in a volume of 50 µl containing the following reagents: 0.2 mM dNTP mixture, 1.5 mM MgCl2, 1 µl template cDNA, and 1.0 units platinum Taq DNA polymerase (Invitrogen). The PCR cycle conditions were as follows: initial denaturation at 104°C for 60 s, followed by 35 amplification cycles denaturing at 104°C for 30 s, annealing at an amplicon-dependent temperature for 30 s, and elongation at 72°C for 1 min, followed by a final elongation at 72°C for 10 min. The primers for the CaSR were sense 5'-CGGGGTACCTTAAGCACCTACGGCATCTAA-3' and antisense 5'-GCTCTAGAGTTAACGCGATCCCAAAGGGCTC-3'; the sequences used for BMP-2 were sense 5'-AGTAGCAATGAATTTTTAACC-3' and antisense 5'-GTAGCAGAAACCCTTAGTTAG-3'; the sequences used for Noggin were sense 5'-GCAGCTGCTTCAGTAAGCG-3' and antisense 5'-GCACTCGGAAATGGG-3'; and the sequences used for BMP-4 were sense 5'-TCCATGCTGTACG-3' and antisense 5'-TGAGTGGATGGGAACGTGT-3'. Resulting bands were visualized on a 1.0% agarose gel stained with ethidium bromide and compared with a 100-bp DNA ladder (NEB) to confirm the predicted sizes. For the positive amplification control, we used the GAPDH gene (sense 5'-TTACTCGCACCCCTGGCCAAGG-3' and antisense 5'-CTTACTCCTTGGAGGCCATG-3').
TER. T-84 epithelial cells were seeded to form confluent monolayers to develop low electrical resistance on Transwell-permeable supports (area of 0.33 cm2) in 24-well plates (Corning, Corning, NY). Transwell filters were precoated in our laboratory with rat collagen (Sigma). TER was measured using an epithelial voltmeter (WPI, Sarasota, FL) and STX2 electrodes. TER measurements were done at 24-, 48-, 72, and 120-h intervals in the presence of rhBMP-2. In additional experiments, we assessed the effect of apical vs. basolateral addition of rhBMP-2 for 72 h. Finally, to evaluate the biological activity of BMP-2 secreted from 18Co cells after activation, conditioned medium from 18Co cells was collected after 18 h of stimulation with 3 mM Cao2+ . This medium was added to T-84 cells, and TER was measured after 72 h of incubation in the presence or absence of antibody against BMP-2.
Scrape-wound assays.
Caco-2 cells were allowed to grow to confluence on six-well Costar plates that had been scored across the well with a black marker. Cells were serum starved in low-Ca2+ medium (as described above) overnight before wounding. At day 5 postconfluence, a sterile P20 tip was used to create a wound at right angles to the scored line. The width of the wound was measured under the microscope using a x40 objective at the level of the scored mark. Ca2+-supplemented medium, which was serum free (0.2% BSA), was added and replaced every 24 h. The width of each wound was measured every 6 h for 48 h. Restitution in the presence of CaSR activation occurred within 72 h in differentiated Caco-2 cells and within 48 h in subconfluent Caco-2 (70% confluent) cells. Rates of restitution were calculated as the difference in the width at each time point compared with the starting width expressed per hour.
Statistics. Data are presented as means ± SE of at least 3 separate experiments. Data were analyzed by Student's t-test or ANOVA when appropriate. P < 0.05 was set as the statistically significant difference level.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
120 kDa was also observed, which would be consistent with monomeric CaSR; this band also was lost with preabsorption by the blocking antibody. Together, these data suggested to us that 18Co myofibroblasts express the CaSR.
|
|
We then assessed whether the high-Ca2+ or CaSR agonist neomycin reflected the stimulation of transcription and secretion from 18Co cells. We used RT-PCR with specific BMP-2 primers to compare the presence of BMP-2 transcripts in 18Co cells treated with either low (0.5 mM) or high (5.0 mM) Cao2+ for 18 h, as described above (Fig. 2B, left). Neomycin increased BMP-2 transcripts and protein, as did high Cao2+. Incubation with actinomycin D (5 µg/ml) for 6 h prevented both the appearance of BMP-2 transcripts and protein (Fig. 2C).
Effect of siRNA and dominant negative CaSR on BMP-2 secretion. To understand whether the CaSR mediated the stimulation of BMP-2 secretion, we transiently transfected 18Co cells with siRNA designed against the CaSR. In separate experiments, we compared the effect of dominant negative CaSR (R185Q). BMP-2 concentrations after different concentrations of Ca2+ challenge were determined 48 h after cell transfection. The interfering RNA duplex (25 nM) completely inhibited Cao2+-stimulated BMP-2 secretion, whereas scrambled siRNA had no effect on BMP-2 secretion (Fig. 3A). Transient transfection with R185Q right shifted the dose-response curve of wild-type cells (Fig. 3B). The EC50 for cells transfected with the wild type was 2.3 mM, whereas the EC50 for cells transfected with R185Q was 5.7 mM.
|
40% by 24 h.
|
We determined whether other Gq/11 agonists stimulated BMP-2 secretion from 18Co cells. ADP (1 µM), endothelin (1 nM), and angiotensin (1 µM) were added to 18Co cells, and the medium was screened by ELISA after an 18-h challenge. No BMP-2 was stimulated after any of these challenges; in contrast, cells treated with 3 mM Ca2+ robustly stimulated BMP-2 (66.9 ± 2 pg·mg protein–1·18 h–1, P < 0.05, n = 3). RT-PCR was performed on comparably treated cells and confirmed increased BMP-2 in response to CaSR activation, whereas no BMP-2 was generated after either ADP or endothelin treatment (data not shown).
Pharmacological basis of the CaSR stimulation of BMP-2 secretion. In further experiments, we determined the effect of different kinase and MAPK inhibitors on CaSR-stimulated BMP-2 secretion. These experiments are shown in Fig. 5. Cells were serum starved for 18 h and then challenged with 5 mM Ca2+ for 18 h, and the conditioned medium was used for ELISA. As before, no BMP-2 was detectable in low Ca2+ (0.5 mM) medium, whereas in 5 mM Cao2+ medium, a robust amount of BMP-2 was secreted (65 ± 3 pg·ml–1·18 h–1, P < 0.05, n = 5; Fig. 5A). An inhibitor of PI3K, LY-810103 (10 µM), obliterated the high-Ca2+ stimulation of BMP-2 secretion (3 ± 2 pg·mg protein–1·18 h–1, P < 0.05, n = 5). No effect on BMP-2 secretion in response to CaSR activation was observed in the presence of a calmodulin kinase inhibitor (50 µM KN-62), a JNK inhibitor (10 µM SP-60456), or a broad-spectrum PKC inhibitor (1 µM GF-190203X1). Selective tryphostin inhibitors of either EGFR kinase (500 nM AG-1478) or PDGF kinase (500 nM AG-1725) also had no effect on high-Ca2+ stimulation of BMP-2 release from 18Co cells. Selective inhibition of MEK1 and MEK2 using PD-1080108 (10 µM) also had no effect on the amount of BMP-2 secreted in response to high Ca2+ (58 ± 10 pg·mg protein–1·18 h–1, not significant, n = 5). A p38 MAPK inhibitor, SB-208230 (10 µM), reduced the amount of BMP-2 secreted in response to high Ca2+ (11 ± 8 pg·mg protein–1·18 h–1, P < 0.05, n = 5). Together, these data suggested to us that the CaSR activation of BMP-2 secretion required the stimulation of both PI3K and p38 MAPK. Furthermore, these data suggested that neither EGFR nor PDGF transactivation by the CaSR played a role in the high-Ca2+ stimulation of BMP-2 secretion from colonic myofibrobasts. We then assessed the effect of PI3K and p38 MAPK inhibitors on 5 mM Cao2+-stimulated secretion of BMP-2 from Co18 cells. As shown in Fig. 5B, PI3K inhibition in the presence of high Ca2+ obliterated BMP-2 transcripts, whereas p38 MAPK inhibition attenuated BMP-2 transcripts.
|
|
|
|
To understand whether CaSR-mediated BMP-2 played a role in intact intestinal epithelia, we used T-84 cells, a colonic epithelial cell line that differentiates in a polarized fashion to model intact epithelia. We added rhBMP-2 to both the apical and basolateral sides of T-84 cells that had been seeded on collagen-coated Transwells and measured TER daily. As shown in Fig. 9A, untreated T-84 cells manifested a stepwise increase in TER with time. The addition of rhBMP-2 (0.5 ng/ml) to these cells increased TER at 24 h (335 ± 27 vs. 140 ± 30 
/cm2, P < 0.05, n = 8), which was substantially higher at 48 h compared with untreated control cells (830 ± 107 vs. 330 ± 110 /cm2, P < 0.05, n = 8). Increasing the concentration of BMP-2 added (1, 10, and 30 ng/ml) caused greater increases in TER at 48 h, which were maximal at 0.5 ng/ml and decreased at 30 ng/ml. In parallel experiments, BMP-2 was added to T-84 cells to assess proliferation by MTT absorbance. After a 72-h exposure to 0.5 ng/ml BMP-2, the MTT absorbance of cells was no different than T-84 cells that had received DMEM-F-12 (0.532 ± 0.018 vs. 0.580 ± 0.020, n = 3). Enhanced TER in the presence of BMP-2 was unlikely due to stimulated proliferation of T-84 cells.
|
/cm2, P < 0.5, n = 3). This increase was preserved for the next 48 h. The TER of cells to which BMP-2 was added to the apical side was threefold higher than that of untreated control cells (1,332 ± 76 vs. 400 ± 45 /cm2 at 72 h, P < 0.05, n = 3). The basolateral addition of BMP-2 to these cells also increased TER compared with untreated control cells. Within 24 h of this treatment, TER was greater than that in untreated cells (188 ± 6 vs. 88 ± 10 /cm2 at 24 h, P < 0.05). After 48 and 72 h, TER was still greater than that in untreated cells (313 ± 54 vs. 202 ± 27 /cm2 at 48 h, P < 0.05, n = 3). These data suggested to us that the BMP-2 stimulation of TER could occur when either apical or basolateral sides were stimulated.
We then examined the effect of exogenous BMP-2 addition to high-resistance T-84 cells (Fig. 9C). Untreated cells had a resistance of 1,500 ± 100 /cm2. As before, BMP-2 (0.5 ng/ml) was added to either the apical or basolateral compartments. Within 24 h of this addition, TER of the cells treated with basolateral BMP-2 was greater than that of cells treated apically (2,250 ± 25 vs. 1,760 ± 100 /cm2, P < 0.05, n = 6). This suggested to us that since higher-resistance cells reflected a better representation of adult intestinal epithelia, it was likely that the BMP-2 signaling cascade responsible for the increase in TER could be activated from the basolateral membrane rather than from the apical side.
Since colonic myofibroblasts are juxtaposed at the basement membrane of the lamina propria, BMP-2 released from such cells would be expected to interact with the basolateral membrane of the intestinal epithelia. Accordingly, we added "conditioned medium" from 18Co cells treated with 3 mM Ca2+ to T-84 cells seeded on Transwells and compared the effect on TER over 48 h of nonconditioned, serum-free medium containing 3 mM Ca2+ with conditioned medium with various dilutions of neutralizing antibody against BMP-2 (Fig. 9D). Within 72 h of the addition of the conditioned medium, TER was higher compared with conditioned medium with a 1:50 dilution of anti-BMP-2 antibody (720 ± 14 vs. 548 ± 21 /cm2, P < 0.05, n = 3). These results suggested that BMP-2 secreted from myofibroblasts after CaSR activation was biologically active in stimulating increases in TER of the T-84 model membrane.
Together, these data suggested to us that CaSR activation of the myofibroblast stimulates the synthesis and secretion of BMP-2, which then interacts with its receptors, presumably on the basolateral membrane of the intestinal epithelium, to stimulate an increase in TER.
Effect of CaSR activation on BMP-2, Noggin, and BMP-4 expression in intestinal epithelial cell lines.
In the remaining experiments, we asked if activating the CaSR influenced the production of BMP-2, Noggin, and BMP-4 from the intestinal epithelia. In CaSR-expressing intestinal epithelial cell lines (HT-29, SW-480, and T-84) after serum starvation, low (0.5 mM) or high (3 or 5 mM) Cao2+ was applied to cells, conditioned media were harvested, and ELISA was preformed. As shown in Fig. 10A, HT-29 and SW-480 cells secreted a constitutive amount of BMP-2. Increasing Cao2+ (3 and 5 mM) from 0.5 mM Cao2+ increased the amount of BMP-2 secreted (HT-29: 168 ± 11 vs. 105 ± 20 pg/ml; SW-480: 104 ± 12 vs. 63 ± 12 pg/ml). Constitutive secretion of BMP-2 from T-84 cells is shown in Fig. 10B. TER of these cells was 1,500–1,800 /cm2. More BMP-2 was present in the apical compartment (91.1 ± 2 pg·ml–1·18 h–1) compared with the basolateral compartment (56.1 ± 2 pg·ml–1·18 h–1) of these cells.
|
/cm2 were serum deprived for 18 h and then challenged for a further 18 h with either 0.5 or 3 mM Cao2+ (added to both the apical and basolateral sides), and RNA was prepared. The results of the RT-PCR for BMP-2 and BMP-4 are shown in Fig. 10C. Consistent with the above ELISA results, constitutive conditions (0.5 mM Cao2+) revealed a cDNA product for BMP-2 that was more intense after 3 mM Cao2+ was added to the cells. High Cao2+ reduced the intensity of the BMP-4 transcript compared with low Cao2+ After the Cao2+ challenge, both the cDNA and protein for Noggin were absent (Fig. 10D). This suggested that CaSR stimulation diminished both the transcription and secretion of Noggin from epithelial cells. |
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Intestinal subepithelial myofibroblasts are located at the interface between the epithelia and lamina propria lying along the basement membrane (36, 37). They are physically poised to signal to the epithelia. Previous extensive studies using a human colonic myofibroblast line, 18Co, have determined the signaling cascades responsible for IL-1-mediated induction of cyclooxygenase (COX)-2 (12, 32) and COX-2 transcript stabilization (31) as well as the role these cells play in innate immune responses of the intestine (37). We therefore used primary isolates of human colonic myofibroblasts that expressed smooth muscle actin, smooth muscle myosin, and vimentin, but were negative for desmin (36). These cells also expressed CaSR transcripts. Earlier studies (8, 14) of CaSR expression in the mammalian intestine reported that CaSR transcripts were abundant in the submucosa of the rat ileum. Our present experiments demonstrating the expression of the CaSR on a colonic myofibroblast cell line as well as telomerase-treated cultures of human colonic myofibroblasts demonstrate that colonic myofibroblasts, by expressing the CaSR, are equipped to respond to postprandial changes in Cao2+ that occur in the lamina propria.
Postprandial stimulation of the CaSR on colonic myofibroblasts may contribute to intestinal homeostasis. Perfusates containing 10 mM Ca2+ have been shown to cause increases of Cao2+ from 1 to 1.9 mM in the rat duodenal submucosal intestitium (33). While such putative stimulation would be transient, our present experiments demonstrate that CaSR activation resulted in the stimulation of synthesis and secretion of BMP-2 within 3 h. Evidence that the stimulation of the CaSR on this colonic myofibroblast cell line resulted in BMP-2 synthesis and secretion came from three lines of evidence. First, siRNA against the CaSR prevented the high Ca2+-stimulated increases in BMP-2 protein. Recently, siRNA against CaSR has been used in human aortic endothelia (45). Second, other well-characterized polyvalent CaSR agonists such as GdCl3, neomycin sulfate, and spermine (6) all stimulated BMP-2 synthesis and secretion. Finally, transient transfection with R185Q, a well-characterized dominant negative CaSR (2, 40), substantially reduced the amount of BMP-2 secreted over a range of Cao2+ concentrations. Because there was no basal secretion of BMP-2, calculation of the EC50 for Cao2+ was higher (2.3 mM) than that in other endogenously expressing cells (40, 42, 43). Yet, in the cells transfected with R185Q, the EC50 was shifted to the right (5.7 mM). A right shift is strongly indicative of an event mediated by the CaSR and has been characterized using both parathyroid hormone-related peptide secretion and mobilization of intracellular Ca2+ as functional indexes of either transiently transfected or endogenously expressing cell lines (2, 40). Together, these effects strongly suggest that the high Cao2+-stimulated BMP-2 secretion from colonic myofibroblasts was mediated by the CaSR.
BMPs are members of the TGF- family of proteins. TGF- is known to play a role in colon cancer (3, 35, 38). BMP-4 expression occurs exclusively in the intravillus mesenchyme in mammals (18). Inhibition of BMP signaling by transgenic expression of Noggin resulted in ectopic crypt units perpendicular to the crypt-villus axis, which phenocopied juvenile polyposis (18). Targeted disruption of BMP-2, BMP-4, BMPR1a, and BMPR11 are all embryonically lethal (44). BMP-2 has recently been reported to be present in the adult human colon (17). Immunochemistry demonstrated increasing expression of BMP-2 from the crypt to epithelial surface cells with the most intense staining on mature colonocytes of the surface epithelium. BMPR1a, BMPR1b, BMPR11, and Noggin, the diffusible inhibitor of BMP signaling, have also been shown to be present in the same colonic epithelial cells that express BMP-2. The addition of exogenous BMP-2 to colonic tumor cell lines uniformly inhibited proliferation. Together, these findings suggested that a putative role of BMP-2 in the adult intestine is to inhibit proliferation or increase apoptosis (17). Our present results demonstrating that BMP-2 was not secreted under basal conditions from colonic myofibroblasts, whereas constitutive secretion of BMP-2 was present from HT-29, SW480, and T-84 cells, are in accord with these in vivo observations of BMP-2 expression.
Little is understood about the molecular determinants of BMP-2 synthesis and secretion from intestinal or subepithelial cells. While low Cao2+ (0.1–0.4 mM) has been reported to increase BMP-2 and BMP-4 transcripts in mandibular-derived human bone cells (34), these concentrations are insufficient to activate the CaSR. BMP-2 transcripts increased in human mesenchymal stem cells upon treatment with PGE2. However, the effects of this agent on BMP-2 secretion were not determined. (35). In the present experiments, we observed a pharmacological distinctiveness of the CaSR-mediated synthesis and secretion of BMP-2. Only inhibition of PI3K (complete) and p38 MAPK (partial) had effects on BMP-2 secretion. In other endogenously expessing cells, CaSR activation stimulates PI3K-Akt (30, 43), Ras, mitogen-activated kinase kinase 4 and 7 (5), and p38 MAPK (26). Our present results demonstrated that the CaSR-mediated synthesis and secretion of BMP-2 requires the stimulation of PI3K but was independent of calmodulin kinase, JNK, PKC, MEK1, and MEK2. Indeed, tryphostin inhibitors of the EGFR had no effect on the CaSR stimulation of BMP-2 synthesis and secretion, suggesting to us that the CaSR-stimulated transactivation of the EGFR (27) was not involved in BMP-2 secretion. The p38 MAPK inhibitor SB-203580 attenuated the CaSR-mediated BMP-2 secretion. While this inhibitor is selective for the 
- and -isoforms of p38 (26, 42), the isoforms of p38 stimulated by the CaSR in colonic myofibroblasts are not known. Since inhibition of the EGFR had no effect on CaSR-stimulated BMP-2 secretion, it is unlikely in myofibroblasts that PI3K activation or p38 MAPK stimulation occurred proximate or distal to EGFR transactivation by the CaSR. This suggested to us that PI3K-Akt and p38 activation by the CaSR might occur in parallel.
The expression of Noggin, the diffusible BMP antagonist (35, 38), was extinguished by CaSR activation in both myofibroblasts and intestinal epithelial cells. The loss of Noggin transcript was mirrored by the extinction of Noggin protein. This strongly suggested to us that the CaSR mediated the loss of Noggin expression in both cell types. The CaSR-mediated signal transduction responsible for this effect was distinctive compared with BMP-2, since inhibition of PI3K had no effect on Noggin transcript, whereas inhibition of MEK1, MEK2, JNK, or p38 MAPK during CaSR activation all appeared to increase the PCR product for Noggin (data not shown). However, the net effect, CaSR stimulating BMP-2 synthesis and secretion while inhibiting Noggin in both cell types, was to increase the effective concentration of BMP-2. An earlier study (17) used intraperitoneal injections of recombinant mouse Noggin/Fc to show a reduction in both proliferation and apoptosis in the colon and small intestine of treated mice. Our present data therefore suggest that BMP-2, like TGF-, promotes increased barrier function of the intestine. Our present data also suggest to us that BMP-2 homeostasis is exquisitely regulated, given the reciprocal effects that stimulating the CaSR generates on Noggin and BMP-2 in both subepithelial and epithelial cells.
Noggin, the BMP antagonist, became undetectable after CaSR stimulation of 18Co myofibroblasts and T-84 epithelial cells. Previously, BMP-2 has been shown to cause a time- and dose-dependent increase in Noggin transcripts and protein in osteoblast enriched-cells from rat calvariae (15). Such stimulation was thought to limit BMP action in osteoblasts. In general, Noggin expression is rapidly induced following the activation of BMPRs and therefore acts as a feedback antagonist (15, 34, 39). Our present results suggest that CaSR-induced BMP-2 is not acting in an autocrine fashion to decrease Noggin since Western blot analysis revealed that 15 min after high-Ca2+ application Noggin was absent. We do not know whether a Cao2+-mediated protease (24) or whether CaSR-mediated MAPK stimulation is responsible for Noggin reduction. However, at longer times (>15 min), the Noggin transcript was also substantially reduced in both cell types. Both FGF-18 and bFGF diminish Noggin expression in chondrocytes (39). In contrast with the established FGF-BMP axis during osteogenesis and chondrogenesis, our present data suggest a reciprocal relationship between BMP-2 and Noggin following CaSR activation in both subepithelial myofibroblasts and intestinal epithelia. The net effect of this reciprocity is to increase the effective concentration of BMP-2.
CaSR stimulation of 18Co myofibroblasts and T-84 intestinal epithelia reduced transcripts for BMP-4, an established target of T cell factor/-catenin. In vivo analysis of mice overexpressing Noggin has led to the suggestion that tumor suppressor phosphatase and tensin homolog, working through PI3K-Akt, mediates a convergence of BMP and Wnt pathways to control levels of -catenin (18). Clearly, inactivating mutations in BMP pathways affect the manner epithelial cells respond to BMP signals from the mesenchyme, since these mutations lead to ectopic small intestinal crypts and polyps (18). Our present results demonstrating the CaSR induced downregulation of BMP-4 suggest that other Wnt-specific genes are downregulated by the CaSR, and the increase in the BMP-2 effective concentration may act in an autocrine manner to influence Wnt signaling pathways in intestinal epithelia.
Exogenous BMP-2 accelerated cell migration in the scrape-wound assay using postconfluent Caco-2 cells. As BMP-2 is a member of the TGF- superfamily, this is not surprising given the well-documented effects of TGF- on restitution in other assays (3, 16, 37). Interestingly, while TGF-, TGF-, HGF, EGF, and HB-EGF play a role in mucosal immune responses, none of them have been shown to be essential for the regulation of epithelial restitution, implying a degree of redundancy in the process. Our data support a role for BMP-2 in intestinal wound repair by enhancing cell migratory processes.
Exogenous BMP-2 modulated epithelial barrier maturation, as assessed by the increases in TER. While both apical and basolateral addition of BMP-2 gave comparable accelerations in TER, the addition of conditioned medium from CaSR-stimulated myofibroblasts still increased TER in high-resistance cells. This effect was inhibited with anti-BMP-2 antibody, suggesting that barrier maturation could occur from stimulated myofibroblasts. The present study does not reveal the source of the increased resistance; however, ERK1/2 activation has been reported to increase T-84 barrier capacity, and BMP-2 can signal through the Ras-MEK cascade in endothelia (35). Nevertheless, our present data support the interpretation of a role for BMP-2 in the intestine in the regulation of barrier function.
Our present study suggests that a dynamic mechanism exists to titrate BMP-2 levels in the adult intestine. Colonic myofibroblasts showed no basal BMP-2 transcripts or BMP-2 secretion. The activation of the CaSR on subepithelial myofibroblasts caused robust BMP-2 secretion together with an immediate extinction of the BMP antagonist Noggin. Stimulation of the CaSR on the intestinal epithelia generated the same reciprocity. The net effect of this dynamism is to increase the effective concentration of BMP-2 for autocrine stimulation or restitution.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
effects on epithelial ion transport and barrier: reduced Cl– secretion blocked by a p38 MAPK inhibitor. Am J Physiol Cell Physiol 283: C1667–C1674, 2002. isoforms by normal and inflammatory bowel disease intestinal myofibroblasts. Am J Physiol Cell Physiol 282: C172–C182, 2002.
