AJP - GI Watch the video to see how APS reaches out to developing nations.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 292: G191-G200, 2007. First published August 31, 2006; doi:10.1152/ajpgi.00131.2006
0193-1857/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
292/1/G191    most recent
00131.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kidd, M.
Right arrow Articles by Mane, S. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kidd, M.
Right arrow Articles by Mane, S. M.

MUCOSAL BIOLOGY

Role of CCN2/CTGF in the proliferation of Mastomys enterochromaffin-like cells and gastric carcinoid development

M. Kidd,1 I. M. Modlin,1 G. N. Eick,1 R. L. Camp,2 and S. M. Mane3

1Department of Surgery, 2Department of Pathology, and 3W. M. Keck-Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, Connecticut

Submitted 22 March 2006 ; accepted in final form 28 August 2006


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Mastomys enterochromaffin-like (ECL) cell proliferation is initially gastrin driven, but once neoplasia develops, cells become gastrin autonomous. We hypothesized that CCN2 (CTGF), a mitogenic growth factor, may regulate ECL cell proliferation. A Mastomys GeneChip database was examined (dCHIP) to identify CCN2 expression levels. CCN2 in normal and tumor ECL cell preparations obtained using FACS (100 nM acridine orange) was examined by real-time PCR. CCN2 protein was identified in mucosal and ECL cell preparations by immunohistochemistry. Short-term cultured cells were stimulated with either CCN2 or CCN2 + EGF, and proliferation was measured (MTT assay). The ERK1/2 inhibitor PD-98059 (0.1–100 µM) was assessed in terms of CCN2 (1 ng/ml)-mediated proliferation and ERK1/2 phosphorylation. CCN2 transcript and protein was then examined in clinical gastric carcinoids. The ccn2 transcript was upregulated in tumor samples compared with the normal mucosa (+2.36-fold, P < 0.01). PCR demonstrated that ccn2 was not expressed in FACS-prepared (>98% pure) normal ECL cells but was elevated in tumor ECL cell fractions (41.3 ± 10.7-fold). Immunostaining of the Mastomys gastric mucosa and FACS preparations confirmed that CCN2 protein was present in ECL tumors but not in normal ECL cells. Neither CCN2 nor CCN2 + EGF stimulated normal ECL cell proliferation. CCN2 stimulated tumor proliferation (EC50 ~0.01 ng/ml); EGF significantly augmented (P < 0.01) CCN2-induced tumor cell proliferation (EC50 = 20 pg/ml). PD-98059 inhibited CCN2-induced proliferation (–12 ± 3%, P < 0.05) and ERK1/2 phosphorylation (–34 ± 5%, P < 0.05) in tumor cells. In clinical samples, both CCN2 transcript and protein were elevated in gastrin-autonomous carcinoids (P < 0.02) compared with the normal mucosa. In conclusion, CCN2 may be a proliferative regulator of Mastomys ECL neoplastic proliferation once these cells become autonomous of gastrin regulation. Identification of CCN2 in gastric carcinoid tissue may be useful both as an indicator of ECL cell transformation and may define gastrin autonomy, a criteria of gastric carcinoid malignancy.

connective tissue growth factor; epithelial growth factor

AGENTS involved in regulating cell proliferation during the development of carcinoids and neuroendocrine (NE) tumors remain an elusive issue. Little information in this area is available since few models and NE cell lines exist to evaluate the problem. The rodent Mastomys develops gastric carcinoids both spontaneously (12–18 mo) and very rapidly (8–12 wk) when acid suppression and hypergastrinemia are engendered (26, 34). In the stomach, the enterochromaffin-like (ECL) cell, which regulates acid secretion under the effect of both gastrin and PACAP (31, 40), transforms into a neoplastic phenotype when exposed to hypergastrinemia (34). Gastrin itself has no proliferative effect on ECL cells once neoplastic transformation has occurred and the ECL cells become gastrin autonomous (34). Nevertheless, proliferation continues even if the gastrin stimulus is withdrawn, indicating that an alternative proliferative agent(s) is now operative (35). The identity of these agents and the mechanism(s) by which autonomous growth occurs remain unresolved. A number of pathways have been considered in this capacity and include activation of the EGF/transforming growth factor (TGF)-{alpha} pathway and the polyamine pathway (19, 35).

Of particular interest is CCN2 (also known as CTGF) since expression of this growth factor has been identified in a variety of tumors of mesenchymal, epithelial, and lymphoid origin. Levels of CCN2 transcript and/or protein are positively correlated with bone metastasis in breast cancer (13), glioblastoma growth (29), poor prognosis in esophageal adenocarcinoma (20), aggressive behavior of pancreatic cancer cells (38), and invasive melanoma (21). This gene is also overexpressed in a mouse transgenic model of gastric NE cell carcinoma (33). No information, however, is available on the expression of CCN2 in normal or neoplastic gastric NE ECL cells or its potential role in mediating ECL cell proliferation and gastric carcinoid tumor growth. In addition, very little is known about CCN2 in the stomach apart from one study (24) demonstrating elevated CCN2 expression in the mucosa of animals with indomethacin-induced ulceration.

In the present set of studies, we reexamined datasets previously used to identify candidate genes (15) using more recently developed approaches (dChIP). One gene that was consistently overexpressed in the Mastomys model and the transgenic mouse model (33) was ccn2. We hypothesized that CCN2 may play a central role in regulating ECL cell transformation. The aims of this study were therefore to identify and confirm CCN2 overexpression during ECL cell neoplasia in the Mastomys model and then verify the proliferative effects of this growth factor under in vitro conditions. We then measured CCN2 expression in clinical samples to examine whether this defined gastrin-autonomous gastric carcinoids.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Mastomys Model

Mastomys GeneChip datasets. We obtained Mastomys GeneChip-derived datasets from the gastric mucosa obtained from animals exposed to hypergastrinemia for variable times (15). These GeneChip datasets are derived from samples hybridized to mouse U74A (11,000 potential genes) Affymetrix GeneChips (n = 6). These data have previously been examined using a GeneSpring approach to identify potential genes upregulated by hypergastrinemia (15). In the present study, these data were reanalyzed using a dChIP approach (25). Array datasets included Mastomys samples [normal fundic mucosa: n = 2; mucosa from animals with loxtidine-induced hyperplasia (8 wk treated): n = 2; ECL cell tumors from long-term loxtidine-treated animals: n = 2]. Data analysis included a sample comparison approach that identified differentially altered genes using the lower 90% confidence bound of fold change (LBFC > 1.2-fold) and the unpaired t-test (P < 0.05) (25). This was undertaken to identify novel overexpressed genes that could be compared with other models of gastric NE cell tumorigenesis that had been similarly processed. We then compared our Mastomys data with genes identified on Affymetrix mouse arrays obtained from laser capture microscopy of dissected tumors from a mouse transgenic model of metastatic gastric NE cancer (33). In this model, this tumor is initiated by the expression of the simian virus 40 large tumor antigen under control of regulatory elements from the mouse ATPase, H+/K+ exchanging, beta-polypeptide (Atp4b) gene (33). Cells from these tumors have a NE phenotype and genotype characteristics. We postulated that genes shared between these two models of NE tumors could play critical roles in the development of gastric NE tumors. This approach identified ccn2 as the single most commonly upregulated gene in both models.

Generation of ECL cell hyperplasia and neoplasia for subsequent experiments. Loxitidine, an oral irreversible histamine-2 receptor antagonist, was utilized to induce acid inhibition and hypergastrinemia (34). A total of 34 animals (3–6 mo of age) with equal sex distribution were randomly assigned to two groups: control and loxtidine (1 mg·kg–1·day–1). Animals were maintained on oral loxtidine treatment for 8–16 wk to generate hypergastrinemia, hyperplasia, and ECL tumors as described previously (15, 34, 35). Whole gastric mucosal samples were used to examine CCN2 expression by immunostaining. Cells isolated in separate experiments were used both for RNA studies (real-time PCR) and in vitro culture studies examining the effects of CCN2 on ECL cell proliferation.

Cell preparation for real-time PCR and in vitro proliferation experiments. NORMAL ECL CELLS (N = 4). ECL cells were isolated from nontreated animals (n = 6 animals/preparation; n = 4 preparations) using pronase digestion of everted gastric sacs in alternating calcium-free and respiration media followed by elutriation and Nycodenz gradient centrifugation (34). This resulted in a 65–80% ECL cell population (~1 x 106 cells, an enrichment of 32-fold over nonenriched gastric cells).


TUMOR ECL CELLS (N = 5). Preparations of tumor ECL cells were obtained from animals (n = 1 animal/preparation) treated with loxtidine (n = 5). The macroscopic gastric nodules were dissected, and the resulting cell mass was subjected to pronase (1.6 mg/ml)-collagenase (1.0 mg/ml) digestion and Nycodenz gradient centrifugation (35). This resulted in a 75–90% ECL cell population (~3 x 106 cells, an enrichment of 25-fold over a nonenriched gastric mucosal cell preparations).


FACS-SORTED NORMAL AND TUMOR ECL CELLS (N = 10). Preparations of normal ECL cells and tumor cells were stained with acridine orange (100 nM) for 30 min (room temperature) (16). The BD FACS Aria Cell-Sorting System (Yale University Department of Immunobiology and Yale Cancer Center) was used to identify and sort cells. Excitation was at 488 nm (to activate acridine orange-labeled cells); sorting was achieved by gating on side scatter (dense, small cells of an estimated ~8- to 10-µm size) and an emission of 532 ± 15 nm. Cells were collected over a 1-h time period. Cell purity was confirmed by quantitating the number of histidine decarboxylase (HDC)-positive cells in each FACS-sorted fraction using fluorescent microscopy. Cells from normal preparations and tumors were used in RNA studies and cell proliferation and signal transduction studies. FACS-sorted normal and tumor ECL cells were used in the RNA studies.

Quantitative real-time PCR studies. RNA ISOLATION. Total RNA was extracted from ECL cell preparations (n = 9) using TRIzol (Invitrogen), and the RNA quality was assessed using an Agilent Bioanalyzer (Agilent Technologies, Palo Alto, CA) to visually verify the absence of genomic DNA contamination, integrity, and the ratio of 28S and 18S bands. Only samples with a 260-to-280-nm absorbance ratio of ≥1.8 were used for PCR (15).


QUANTITATIVE RT-PCR. Total RNA (2 µg) from each preparation was reverse transcribed using the High-Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA). Quantitative RT-PCR analysis was performed (ABI 7900 Sequence Detection System) using Assays-on-Demand products for murine ccn2 according to the manufacturer's suggestions. This ccn2 primer (Ms00515790_m1) was designed to encompass exon:exon boundaries and therefore preferentially amplified cDNA, which minimized the possibility of genomic DNA contamination. Non-reverse transcriptase controls were included as negative controls for each of the samples. All reactions were performed in triplicate in 16-µl final volumes. The following standard quantitative RT-PCR conditions were used: 50°C for 2 min and then 95°C for 10 min, followed by 40 cycles at 95°C for 0.15 min and 60°C for 1 min. A standard curve was generated using cDNA obtained by pooling equal amounts from each sample. The expression level was normalized to internal murine GAPDH. Data were analyzed in Microsoft Excel (ABI, bulletin no. 2) (15).

CCN2 immunohistochemistry and quantitation. IMMUNOSTAINING. Paraffin-embedded whole tissue mucosal sections were dual stained using a modification of previously described methods (14, 15). For antigen retrieval purposes, sections were immersed in citrate buffer (10 mM sodium citrate, pH 6.0) and subjected to 1 x 10 min high-temperature, high-pressure treatment followed by a treatment with 0.3% H2O2 in methanol (30 min at 37°C) to inactivate endogenous peroxidases. Slides were incubated (24 h at 4°C) with monoclonal mouse anti-human CCN2 (1 µg/ml, R&D Systems; MAB660) and the specific ECL cell marker (rabbit anti-HDC, 1:20 dilution). Donkey anti-mouse antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) were used as a secondary reagent to identify CCN2 immunoreactivity. This signal was visualized with a fluorescent chromogen (Alexa-488/FITC). Cy5-tyramide (NEN Life Science Products)-labeled goat anti-rabbit antibodies were used for HDC. Nuclei were stained by 4',6-diamidino-2-phenylindole (DAPI; 20 mg/ml). Paraffin-embedded gastric mucosa autofluoresces in the Cy3 range (573–648 nm). Measuring this fluorescence allowed us to develop a "mask" of the whole mucosa. FACS-sorted normal and tumor ECL cell preparations (prefixed, HDC positive) were stained as described for CCN2 except that Cy5 was used to visualize CCN2 staining.


AUTOMATED QUANTITATIVE ANALYSIS. For automated analysis, monochromatic, high-resolution (1,024 x 1,024 pixels, 0.5 µm) images were obtained for each whole tissue section. A whole gastric mucosal section was created by masking the Cy3 autofluorescence signal. CCN2-immunopositive cells were distinguished by the Cy5 signal. ECL cells were identified by Alexa-488/FITC fluorescence, and DAPI was used to identify nuclei. The Cy5 signal was scored and expressed as the signal intensity divided by the Cy3 mask area. This provided a quantitation of CCN2 immunoreactivity in the total gastric mucosa. The Cy5 signal was then scored and expressed as signal intensity divided by the Alexa-488/FITC mask area. This provided a quantitation of CCN2 immunoreactivity in ECL cells within the total gastric mucosa.


WESTERN BLOT ANALYSIS. Whole tissue mucosal sections from normal (n = 3), hyperplastic (n = 3), and tumor tissue (n = 3) were prepared as previously described (35). Briefly, tissues were snap frozen in liquid nitrogen and homogenized using a mortar and pestle before the addition of RIPA buffer. Protein concentration was assayed using the Bio-Rad protein assay (Bio-Rad Laboratories). Samples were then boiled in Laemmli's reducing buffer and equal protein was resolved on 12% SDS-PAGE gels. Protein was immobilized on nitrocellulose membranes by electrotransfer. Following blockade (5% nonfat milk, 60-min RT), gels were immunoblotted with monoclonal mouse anti-human CCN2 (1 µg/ml, R&D Systems; MAB660, 4 h at room temperature) overnight at 4°C and then horseradish peroxidase-conjugated goat-anti mouse secondary antibodies (R&D Systems, 1:4,000, 30 min at room temperature). Membrane-bound antibodies were detected using a luminol-based chemiluminescence system (Roche). Blots were exposed on X-OMAT-AR film. Gels were stripped and reprobed with anti-beta-actin to confirm equivalent protein loading. Densitometry analysis of bands was performed by Scion Image (Scion, Frederick, MD).

Cell culture and proliferation studies. Freshly isolated normal or tumor ECL cells were rinsed in culture medium at 37°C (DMEM-F-12 with 12% BSA at pH 7.4), and 2 x 104 cells were added to each well (6 wells/stimulant) of a collagen type I-coated 96-well plate (Becton Dickinson) in growth medium [culture medium plus FCS (2%), ITS (0.5 mg/100 ml), hydrocortisone (10 nM), and gentamicin (0.1 mg/100 ml)] and cultured for 24 h (37°C, 5% CO2) (15).

Cultured cells were stimulated with CCN2 (Peprotech, 0.001–1 ng/ml) for an additional 12–48 h. DNA proliferation was then measured using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). Briefly, MTT was added to each well of stimulated cells for 4 h at a final concentration of 0.5 mg/ml. The converted dye (formazan) was solubilized with 0.04 M HCl in isopropanol, and the absorbance was measured at a wavelength of 595 nm with background subtraction at 650 nm on an ELISA plate reader. MTT levels (fold increases over basal) were determined for each concentration of CCN2, and a dose-response curve was generated for this agent. The EC50 and ECmax were calculated from the log of the dose-response curve. The combinatorial effect of CCN2 with a known regulator of ECL cell proliferation, EGF, was also examined. In these experiments, normal or tumor ECL cells were cultured as described above but stimulated with each agent (0–5 ng/ml) for 12–48 h or costimulated with a submaximal dose of CCN2 (0.01 ng/ml).

For signal pathway inhibitor studies, cells were preincubated with the ERK1/2 phosphorylation inhibitor PD-98059 (0.1–100 µM) or the phosphatidylinositol 3-kinase (PI3K) inhibitor wortmannin (0.01–10 µM) for 15 min prior to the addition of CCN2 (1 ng/ml), and cells were incubated overnight (24 h). Cell proliferation was measured by MTT uptake (as described above), and MTT levels (fold increase differences to CCN2 alone) were determined for each inhibitor concentration.

The effect of CCN2 on ERK1/2 phosphorylation was next measured in neoplastic cells. Studies were the same as for the inhibitor assays. Cells, preincubated with PD-98059 (1–100 µM), were stimulated with CCN2 (1 ng/ml) and cultured overnight (24 h). Thereafter, ERK1/2 phosphorylation was measured using an ELISA approach (SuperArray CASE, ERK1/2 kit) as per the manufacturer's protocol. Briefly, stimulated cells were fixed (4% formaldehyde) and stained with either phospho-ERK1/2 or ERK1/2 primary antibodies (60 min at room temperature). After a wash and secondary antibody application (60 min at room temperature), cells were incubated with color developer (10 min at room temperature), and plates were read at 450 nm. Thereafter, protein was assayed in each well (protein development: reading at 595 nm). Results were calculated as antibody (450 nm)/protein concentration (595 nm) and normalized to unstimulated cells. The phosphorylated signal was compared with the total ERK1/2 signal for CCN2 or CCN2 and each inhibitor concentration.

Effect of gastrin and TGF-beta1 on ccn2 expression in normal ECL cells. Acutely isolated normal ECL cells (1 x 106) were cultured (in triplicate) as described above. Cells were stimulated with three doses of human gastrin-17 [1, 20, and 100 pg/ml; the EC50 for proliferation = 20 pg/ml (34)] or TGF-beta1 [1, 5, and 20 ng/ml; the EC50 for ccn2 mRNA activation in fibroblasts = 5 ng/ml (9)] for 60 min. RNA was isolated (as described above), and real-time PCR was performed for the ccn2 transcript. Levels of ccn2 were normalized to GAPDH.

Human Samples

Samples for RNA studies. For RNA studies, tumor tissue was collected from eight patients (3 men and 5 women; median age: 62 yr, range: 34–79 yr) with histologically proven gastric carcinoid tumors. These patients had either undergone resection of the primary tumor between 1997 and 2003 in the Yale University Department of Surgery or the samples were obtained from the Cooperative Human Tissue Network, funded by the National Cancer Institute. Paired normal tissue samples were also obtained from the adjacent macroscopically normal, nontumor mucosa in five patients. These studies were approved by the Human Investigations Committee at Yale University School of Medicine.

Quantitative real-time PCR studies. RNA isolation and quantitative RT-PCR were performed on 13 human samples as described for Mastomys tissue except that the ccn2 primer was based on the human gene (ccn2: HS00170014_m1). To normalize the data, we used levels to the house-keeping genes asparagine-linked glycosylation 9 homolog (ALG9), transcription factor CP2 (TFCP2), and zinc finger protein 410 (ZNF410), which we determined by geNorm could provide the most reliable normalization factor (4).

CCN2 immunostaining and quantification. A gastrointestinal tumor tissue microarray (YTMA63) was used to measure CCN2 expression. This array has been previously described (18) but now includes 13 patients with histologically proven gastric carcinoid tumors [type I/II gastric carcinoid tumors (n = 7) and type III/IV (n = 6)]. Paired normal tissue samples from the adjacent, macroscopically normal gastric nontumor mucosa in eight patients were available for measurements of normal CCN2 expression. Sections from the paraffin-embedded whole tissue mucosa were dual stained using a modification of previously described methods (14, 15). Antibodies used were the same as for Mastomys tissue sections except that cytokeratin was used as a mask (14, 15). The standard automated quantitative analysis (AQUA) protocol was used to quantitate CCN2 expression in human samples (14, 15).

Statistical Evaluation

Results are expressed as means ± SE; n indicates the number of ECL cell preparations (6 normal or hyperplastic animals/ECL cell preparation or 1 tumor animal/ECL cell preparation). All statistical analyses were performed using Prism 4 (GraphPad Software, San Diego, CA). These included two-tailed Mann-Whitney and Wilcoxon rank-sum tests for nonparametric data. For AQUA analyses, two-tailed unpaired t-tests with the Welsh correction were performed. P < 0.05 was considered as significant.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Mastomys Model

Mastomys GeneChip datasets. Hyperplastic ECL cell preparations were characterized by 44 probe sets, whereas the ECL cell tumor was characterized by 22 probe sets (Table 1). Nine probe sets were common to both the gastrin-responsive and gastrin-autonomous preparations and included probes coding for lectin, myosins, matrilin, and transgelin. ccn2 was overexpressed in neoplastic samples in this model. A comparison between the neoplastic Mastomys dataset and the 191 probes sets in the laser capture microscopy-derived NEC dataset from the transgenic mouse model of NE carcinogenesis (33) revealed that ccn2 was overexpressed in both models. Based on these data as well as the known biological function of the growth factor (9, 10), we hypothesized that CCN2 would play a role in Mastomys gastric ECL cell proliferation and transformation. However, because we were concerned with the purity of the ECL cells and the possibility of contamination with other CCN2-containing cells in the mucosa, we next used a FACS approach to obtain a homogenous preparation of normal and tumor ECL cells.


View this table:
[in this window]
[in a new window]

  		Table 1. Transcripts elevated in Mastomys tumor ECL cells using dChIP analysis compared with normal mucosa

 
Purification of normal and tumor ECL cells using a FACS approach. Normal ECL cells were enriched from ~1% in total mucosal cells to 64.9–74.3% (Nycodenz gradient) to >98% using FACS (Fig. 1, A and B). HDC staining of cells confirmed that cells were 98 ± 2.3% ECL cell positive.


Figure 1
View larger version (39K):
[in this window]
[in a new window]

  		Fig. 1. FACS of viable Mastomys normal enterochromaffin-like (ECL) cells. A: green-labeled ECF cells (P5) showed 532- and 595-nm emission simultaneously, whereas red-labeled parietal cells (P4) showed mainly emission at 592 nm. y-Axis, 595-nm emission (APC-Cy7-A); x-axis, 532-nm emission (FITC). B: purity assessment of ECL cell suspensions after FACS. The dual-stained preparations [histidine decarboxylase (HDC), stained with Cy5 (red); and nuclei, stained with 4',6-diamidino-2-phenylindole (DAPI) (blue)] demonstrate that high purity (98 ± 2.3%) of cells were ECL cells.

 
CCN2 in normal ECL cells. To evaluate ccn2 expression in normal ECL cells, we used a real-time PCR approach in preparations of both standard prepared ECL cells (~65% pure) and FACS-sorted ECL cells (>98% pure). Transcript levels of ccn2 were low and absent in each of these preparations, respectively (Fig. 2A).


Figure 2
View larger version (32K):
[in this window]
[in a new window]

  		Fig. 2. CCN2 (CTGF) expression in normal ECL cells. A: ccn2 transcript was expressed at low levels in standard elutriation preparations of ECL cells (NE) or was undetectable in FACS-sorted ECL cell preparations (NFACS) compared with the normal mucosa (NMUC). Levels of ccn2 were elevated >10-fold in the normal mucosa compared with the standard elutriation protocol. *P < 0.001 vs. normal cells. Values are means ± SE; n = 3. B: in the normal Mastomys mucosa, immunostaining demonstrated that CCN2 [stained with Alexa488 (green)], while present in the normal mucosa (yellow arrows), did not colocalize in HDC-positive [stained with Cy5 (red)] cells (white arrowheads). Nuclei were stained with DAPI (blue). Magnification: x200. C: in FACS-sorted normal ECL preparations, dual staining with both anti-HDC and anti-CCN2 demonstrated that only HDC-positive [stained with FITC (green)] cells could be identified. Magnification: x600. A scale bar for the pictogram is included. Images are representative of 3 separate experiments.

 
We next examined protein expression of CCN2 in the normal gastric mucosa. Using a monoclonal mouse antibody, CCN2-positive cells were identifiable in Mastomys whole tissue sections from the normal gastric mucosa (Fig. 2B). Specific staining with the ECL cell marker HDC (rabbit polyclonal, 0.2 µg/ml, EuroDiagnostica) indicated that CCN2 immunoreactivity was not evident in normal ECL cells (Fig. 2B, inset). AQUA analysis and quantitation of CCN2 expression demonstrated that scores for this growth factor were absent (0 ± 0%) in normal ECL cells within the gastric mucosa. The absence of CCN2 was confirmed in normal ECL cells by immunostaining of FACS-sorted (>98%) preparations. No CCN2-positive cells were identified in these preparations (Fig. 2C).

CCN2 in tumor ECL cells. Using real-time PCR, ccn2 transcript was identified in preparations of both standard prepared tumor ECL cells (~65% pure) and FACS-sorted tumor ECL cells (~98% pure) (Fig. 3A). ccn2 transcript was significantly elevated >20-fold (P < 0.001) in tumor cells and 41.3 ± 10.7-fold versus the elutriated-enriched or FACS-enriched population of normal cells. These data are consistent with the dChIP analysis of the GeneChip data and demonstrate that ccn2 is overexpressed in tumor ECL cells compared with normal ECL cells.


Figure 3
View larger version (38K):
[in this window]
[in a new window]

  		Fig. 3. CCN2 expression in tumor ECL cells. A: ccn2 transcript was detectable in both standard preparations of tumor ECL cells (TE) or FACS-sorted tumor ECL cell preparations (TFACS). Expression levels of ccn2 were elevated >20-fold in elutriated tumor preparations compared with elutriated normal ECL cells. *P < 0.001 vs. normal cells. Values are means ± SE; n = 3. B–D: Mastomys tumor mucosa immunostained for nuclei [stained with DAPI (blue); B], HDC [stained with Cy5 (red); C], and CCN2 [stained with Alexa488 (green); D]. The composite is shown in E. HDC-positive, non-CCN2-staining cells (blue arrowheads), dual-stained cells (HDC positive/CCN2 positive, yellow arrowheads), and CCN2-positive non-ECL cells (white arrowheads) are all clearly indicated. HDC staining was largely cytoplasmic with few cells demonstrating nuclear staining; CCN2 staining was exclusively cytoplasmic. Magnification: x200. F: in FACS-sorted tumor ECL cell preparations, dual-stained HDC-positive [stained with FITC (green)]/CCN2-positive [stained with Cy5 (red)] cells could be indentified (orange cells, white arrowheads). Magnification: x600. A scale bar for the pictogram is included. Images are representative of 3 separate experiments.

 
An examination of CCN2 immunoreactivity demonstrated that unlike in normal ECL cells, CCN2 was present in ECL cell tumor cells (Fig. 3B). It was also identified in cells adjacent to the tumor. AQUA analysis and quantitation of CCN2 expression demonstrated that scores for this growth factor were significantly elevated in ECL cells (34 ± 17%, P < 0.02) compared with expression levels in normal ECL cells. The presence of CCN2 in tumor ECL cells was confirmed by immunostaining FACS-sorted (>98%) preparations. CCN2-positive cells (89 ± 3.8%) were identified in these preparations (Fig. 3C).

To further verify the expression of CCN2 in the Mastomys gastric mucosa, we next examined the expression in normal and hyperplastic tumor samples by Western blot analysis using the mouse monoclonal antibody. This approach confirmed that CCN2 was expressed (single band of ~38 kDa) in the mucosa and that this was overexpressed in the tumor mucosa compared with the normal mucosa (Fig. 4A). Densitometric quantitation analysis of bands (Scion Image) confirmed that CCN2 (after normalization to beta-actin) was elevated 24–67% compared with the normal mucosa but showed no major increase in expression in the hyperplastic mucosa (Fig. 4B). Having established that CCN2 was elevated during ECL cell proliferation, we next investigated whether CCN2 had a biological or regulatory function in the Mastomys neoplastic model.


Figure 4
View larger version (43K):
[in this window]
[in a new window]

  		Fig. 4. Western blot analysis of CCN2 in the normal mucosa (NM), hyperplastic mucosa (HM), and tumor mucosa (TM) samples. A: a single band of ~38 kDa was identified using an anti-human CTGF/CCN2 COOH-terminal peptide antibody (1 mg/ml). A loading control (beta-actin) was included to demonstrate equivalent protein load. B: densitometric analysis of CCN2 expression demonstrated that this was significantly increased in the tumor mucosa compared the normal mucosa and hyperplastic mucosa. *P = 0.042 vs. normal mucosa. Values are means ± SE; n = 3.

 
Cell culture and proliferation studies. We have previously noted that the EGF receptor pathway is upregulated in the transformed Mastomys ECL cell (35) and EGF itself has been shown to augment CCN2-mediated fibroblast proliferation (9). Based on these data, we evaluated the effects of either CCN2 alone or in combination with EGF on normal and tumor ECL cell proliferation. A dose-response curve using human recombinant CCN2 (Peprotech) at concentrations ranging from 0.001 to 10 ng/ml was initially generated to establish the EC50. This recombinant isoform of CTGF comprises the last 98 amino acids of CCN2, which exerts full heparin binding, cell adhesion, and mitogenic activity (Peprotech datasheet). A 24-h incubation with CCN2 resulted in stimulation of ECL cell proliferation in tumor ECL cells with an EC50 of ~0.01 ng/ml and an ECmax of 0.1 ng/ml (Fig. 5A). In contrast, normal cells failed to respond with significant proliferation to CCN2.


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 5. Effect of CCN2 and EGF on normal and tumor ECL cell proliferation. A: CCN2 significantly stimulated ECL cell tumor proliferation but not normal ECL cell proliferation. Expression levels were normalized to unstimulated (basal) proliferation. *P < 0.01 vs. normal ECL cells. B: EGF did not significantly stimulate either normal or tumor ECL cell tumor proliferation. Expression levels were normalized to unstimulated (basal) proliferation. C: CCN2 augmented EGF-stimulated tumor ECL cell proliferation but not EGF-stimulated normal cell proliferation using the MTT assay. Expression levels were normalized to basal proliferation. *P < 0.004 vs. 0 ng/ml EGF; #P < 0.0002 vs. normal ECL cells. Values are means ± SE; n = 3.

 
We then examined whether CCN2 augmented EGF-stimulated ECL cell proliferation. These experiments were based on similar studies in normal rat kidney fibroblasts (9) and use the same concentration range of EGF. An examination of EGF alone (0.1–100 ng/ml) demonstrated that this growth factor did not have a significant effect on either normal or tumor ECL cell proliferation (Fig. 5B). Preincubation of tumor cells with a submaximal concentration of CCN2 (0.01 ng/ml) prior to the addition of EGF (0–5 ng/ml) resulted in a dose-dependent increase in proliferation (Fig. 5C). This effect was significantly (P < 0.004) greater than the effect of CCN2 alone (12- vs. 20- to 22-fold), with an estimated EC50 of 20 pg/ml. No additive effect of CCN2 was noted in EGF-stimulated normal ECL cells using the same doses.

To define the mechanistic pathway by which CCN2 mediated tumor ECL cell proliferation, we next examined whether the ERK1/2 phosphorylation inhibitor PD-98059 (0.1–100 µM) or the PI3K inhibitor wortmannin (0.01–10 µM) altered CCN2 (1 ng/ml)-mediated neoplastic ECL cell proliferation. These studies demonstrated that inhibition of ERK1/2 phosphorylation (Fig. 6A), but not PI3K (Fig. 6B), resulted in a dose-dependent decrease in neoplastic ECL cell proliferation. This suggests that CCN2 mediates proliferation via activation of the MAPK pathway rather than the PI3K pathway in tumor ECL cells.


Figure 6
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 6. Effect of signal transduction inhibition on ECL cell proliferation and ERK1/2 activation. A: PD-98059 (P) dose dependently inhibited CCN2 (C)-mediated neoplastic ECL cell proliferation. *P < 0.05 vs. CCN2 (1 ng/ml) alone. B: wortmannin had no consistent effect on CCN2-mediated proliferation. C: CCN2 stimulated ERK1/2 phosphorylation in neoplastic ECL cells. This could be reversed by PD-98059. **P < 0.01 vs. unstimulated cells; *P < 0.05 vs. CCN2 alone. Con, control (unstimulated) cells. Values are means ± SE; n = 3.

 
To confirm that CCN2 activated the ERK1/2 pathway in tumor ECL cells, we next measured ERK1/2 phosphorylation in tumor ECL cells in response to CCN2 (1 ng/ml) and demonstrated that the addition of CCN2 to ECL cells resulted in activation of the ERK1/2 pathway (Fig. 6C). The addition of PD-98059 (associated with a reduction of CCN2-mediated ECL cell proliferation) significantly reduced ERK1/2 phosphorylation, demonstrating that the effects of CCN2 are mediated at least via this pathway in the neoplastic ECL cell.

Examination of the role of gastrin and TGF-beta1 in CCN2 expression in normal ECL cells. Under normal physiological conditions, the ECL cell does not produce significant ccn2 transcript or CCN2 protein (Fig. 2, A–C). Since CCN2 was increased in ECL cell tumors and these tumors arise from gastrin-driven ECL cell proliferation, we next investigated whether gastrin was responsible for the elevation of this growth factor.

ccn2 mRNA expression in the total Mastomys gastric mucosa isolated from hypergastrinemic animals (animals treated with H2 receptor blockade for 8 wk) was significantly elevated (1.77 ± 0.44 vs. 0.74 ± 0.08, P = 0.05) compared with the total mucosa from normogastrinemic animals. Hypergastrinemia is therefore associated with an elevation in gastric mucosal ccn2. To establish whether gastrin or TGF-beta1, a well-known regulator of ccn2 production (10), directly stimulated ECL cell ccn2 production, we measured ccn2 transcript in isolated, normal ECL cells stimulated with either gastrin (1–100 pg/ml) or TGF-beta1 (1–20 ng/ml) for 1 h. Quantitative RT-PCR analysis demonstrated that gastrin had no effect on ccn2 transcription (Fig. 7A), whereas TGF-beta1 stimulated ccn2 message levels with an EC50 of 10.5 ng/ml and a maximal effect of 3.2 ± 0.2-fold in the normal cell (Fig. 7B).


Figure 7
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 7. Effect of transforming growth factor (TGF)-beta1 or gastrin on ccn2 transcript in isolated cultured normal ECL cells. A: TGF-beta1 elevated ccn2 transcript with an EC50 of 10.5 ng/ml (maximal response: 3.2 ± 0.2-fold compared with unstimulated cells). *P < 0.01 vs. unstimulated cells. B: gastrin-17 had no effect on ccn2 mRNA expression after 1 h of stimulation. Values are means ± SE; n = 3.

 
Human Samples

Real-time PCR for CCN2. To assess whether CCN2 was differentially expressed between type I carcinoids and type III/IV carcinoids, we used a real-time approach and normalized ccn2 levels using three robust housekeeping genes, ALG9, TFCP2, and ZNF410, by geNorm as previously described (36). Compared with control tissue (normal gastric mucosa), CCN2 was overexpressed in type III/IV tumor carcinoids (P < 0.02; Fig. 8A). ccn2 levels were not different in the normal mucosa and type I carcinoids.


Figure 8
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 8. ccn2 transcript and CCN2 protein expression in gastric carcinoids. A: ccn2 gene expression normalized to geNormATZ demonstrates that this was overexpressed in type III/IV carcinoids (GC-III/IV) compared with type I tumors (GC-I) and the normal mucosa (GN). *P = 0.016 vs. normal mucosa; #P = 0.028 vs. type I carcinoids. B: expression levels of CCN2 determined by AQUA quantitations on YTMA63. Levels of CCN2 were identified to be significantly elevated in type III/IV gastric carcinoids and gastrointestinal stromal tumors (GISTs) compared with normal mucosa and gastric adenocarcinomas (GCA). CCN2 was also increased in malignant gastric carcinoids compared with GISTs. *P < 0.003 vs. control; **P < 0.0001 vs. normal mucosa; #P < 0.02 vs. GISTs; &P < 0.002 vs. type III/IV carcinoids. Values are means ± SE; n = 3.

 
CCN2 immunohistochemistry. CCN2 protein expression was quantified using AQUA analysis (14) on a novel gastric neoplasia array that we developed (YTMA63). Compared with control tissue (normal gastric mucosa), CCN2 was overexpressed in carcinoids (AQUA score: 23.3 ± 11.7, P < 0.004) and gastrointestinal stromal tumors (GISTs; 21.6 ± 3.9, P < 0.0001) but not in adenocarcinomas (10.7 ± 1.5; Fig. 8B). When we separated the carcinoids into "gastrin-dependent" type I and "gastrin-independent" type III/IV tumors, we found that the gastrin-autonomous tumors had significantly elevated CCN2 compared with gastrin-responsive tumors (41.8 ± 16 vs. 5.8 ± 1.8, P < 0.002).


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
We have previously examined Mastomys GeneChip data (rat and mouse chips) using GeneSpring and MAS5.0 approaches (15). This resulted in the identification of alterations in the activator protein (AP)-1 pathway as well as the H1 receptor in mediating Mastomys ECL cell proliferation and transformation (15), with the latter effect confirming our earlier in vivo studies (27). We have, however, identified that more Mastomys transcripts can be identified on mouse chips than on rat chips (22% vs. 16.5% of probe sets positive, P < 0.005). In addition, Mastomys has a closer phylogenetic affinity to the mouse than the rat (15). The degree of overlap of the Mastomys sequence on mouse chips is not known, and it is probable that some Mastomys genes would not be detected. Nevertheless, this is more robust than identifying alterations in gene expression using a rat genome approach. In addition, it has been reported that a dChIP approach may consistently perform better than MAS5.0 for the identification of differentially expressed neoplastic genes (32). We therefore reexamined the Mastomys gene chip data using this approach and compared the data with a mouse transgenic model of NE cell carcinoma (33). This approach showed that the ccn2 transcript was elevated in both the Mastomys tumor ECL cell preparations and mouse NE carcinoma cells. Based on these data, we hypothesized that CCN2 may, like gastrin, TGF-{alpha} or PACAP (23, 34, 35), play a role in modulating the proliferation of NE tumors.

We initially measured ccn2 transcript levels during tumorigenesis in the Mastomys model. Using standard elutriated ECL cells (~60% enriched), we were able to confirm the data analysis of the GeneChip studies and demonstrated that ccn2 mRNA was overexpressed in enriched ECL cell carcinoids compared with normal ECL cells. However, tumorigenesis is associated with a peritumoral response that includes inflammatory cells, myofibroblastic and desmoplastic responses, and angiogenesis that may involve cell types that express CCN2 (e.g., fibroblasts) (8). Cell contamination by any of these matrix cells could result in the measurement of elevated ccn2 transcript in tumor ECL cells compared with normal cells. This possibility led us to modify methodology initially developed by us for small intestinal EC cells (16) that would facilitate our ability to obtain ECL cell populations of substantially greater purity than the enriched ECL cell populations generated by counterflow elutriation or tumor dissection. Using ECL cell-specific immunostaining by acridine orange, we were able to enrich ECL cells >98% from both normal mucosa and gastric carcinoids. This is a similar level of purity to that recently described, which used this basic dye for FACS-sorted rat ECL cells (22). An examination of the two highly enriched preparations confirmed that ccn2 mRNA was overexpressed in tumor ECL cell preparations compared with normal ECL cells.

To examine the topographical relationship of CCN2 and ECL cells in the Mastomys mucosa and to confirm the absence of CCN2 in normal cells, we used a dual staining approach with anti-CCN2 and anti-HDC antibodies. CCN2-like immunoreactivity was consistently absent in all normal ECL cells (both in the mucosa as well as FACS-sorted preparations). In contrast, numerous HDC-positive tumor ECL cells coexpressed CCN2, indicating that the transformed ECL cell synthesizes CCN2 protein. In addition, we noted that other cell types in the normal and tumor mucosa produced CCN2 immunoreactivity. A number of possibilities exist for alternative CCN2 producing cell system(s) including stem (mesenchymal) cells, other endocrine cells, or stromal cells (fibroblasts and myofibroblasts) (6, 12). Given the known association between tumorigenesis and peritumoral responses, upregulation of CCN2 in the tumor mucosa may be related to CCN2 playing a role in NE tumor pathobiology consistent with the previously identified functions of this growth factor (2, 11, 30). Tumorigenesis and ulcerogenesis share a number of common pathways, and CCN2 expression has previously been identified in the rat gastric mucosa under ulcerogenic conditions (indomethacin-induced gastric ulcers) (24). In the latter study, the cell type(s) was not determined, but CCN2 was hypothesized to participate in regulating connective tissue formation and angiogenesis–tissue remodeling that also occurs during neoplasia.

Having demonstrated upregulation of ccn2 transcript and CCN2 protein in the transformed ECL cell, we next evaluated whether this factor in the absence or presence of EGF played any role in modulating the proliferation of either normal or tumor ECL cells. These studies demonstrated that neither CCN2 nor EGF had any significant effect on normal ECL cells. In addition, the combination of CCN2 and EGF did not significantly alter normal ECL cell growth. In contrast, in ECL tumor cells, CCN2 either alone or with EGF significantly and dose dependently elevated tumor ECL cell growth. In these studies, the effects of CCN2 and EGF were synergistic because EGF alone had no effect on tumor cells. The response (>20-fold) as well as the EC50 [>10-fold more sensitive than for fibroblasts (9)] demonstrated the sensitivity of the transformed ECL cell to this agent.

Incubating cells with PD-98059, an inhibitor of MEK1/2 (which phosphorylates ERK1/2), decreased ECL cell proliferation in response to CCN2 but preincubation with wortmannin had no effect. PD-98059 is a highly selective inhibitor of MEK1 activation (1) but may also inhibit MEK5 (37) and the kinase suppressor of Ras (3), whereas wortmannin, an irreversible inhibitor of PI3K, also inhibits other kinases including PI4K and DAG kinase (5). Studies using these inhibitors therefore have suggested that the proliferative effect of CCN2 on neoplastic ECL cells occurs via MAPK pathway activation (ERK1/2) and not via the PI3K pathway. The results in ECL cells with PD-98059 were similar to studies in hepatic stellate cells, where CCN2-stimulated DNA synthesis was associated with phosphorylation of ERK1/2 and an induction of c-fos expression (7), effects that could be blocked by inhibiting ERK1/2 with PD-98059 (7). The principal pathway that regulated CCN2-mediated proliferation in the chondrocyte (HCS-2/8) cell line was phosphorylation of ERK1/2 and could be inhibited by PD-98059 (39). We postulate that the ERK1/2 pathway may be the primary signal pathway related to CCN2-mediated ECL cell proliferation.

To investigate whether gastrin affected CCN2 expression, we evaluated the effects of this growth factor on normal ECL cell ccn2 transcript. No evidence of an elevation in ccn2 transcript was noted under in vitro conditions. We next examined whether hypergastrinemia in vivo altered CCN2. We noted that ccn2 mRNA expression in the total Mastomys gastric mucosa isolated from hypergastrinemic animals was significantly elevated compared with the mucosa from normogastrinemic animals. Protein levels of CCN2 were also increased in this mucosa, but only occasional ECL cells were also identified that expressed CCN2. These data suggested that gastrin elevated CCN2 in the gastric mucosa but did not allow us to conclude that this effect was either specific to the ECL cell or whether an alternative growth factor might be in involved in the process. TGF-beta1 seemed a likely candidate for this role since it is a well-known regulator of CCN2 production (10), and we have noted that levels of TGF-beta1 are increased (1.4-fold, P < 0.05) in the hypergastrinemic mucosa (M. Kidd and I. M. Modlin, unpublished observations). Using isolated ECL cells, we demonstrated that TGF-beta1 stimulated ECL cell ccn2 transcription in vitro. A direct effect of gastrin on CCN2 remains unclear, but the elevated ccn2 transcript and CCN2 protein levels identified in the hypergastrinemic mucosa may reflect an indirect effect of gastrin mediated by TGF-beta1. It is of note, however, that a genomic study (28) of the gastric mucosa of rats treated with omeprazole (infusion: 400 µmol·kg–1·day–1) for 10 wk failed to demonstrate elevated ccn2 expression (28). This is of interest, because over the time course of this experimental period (hypergastrinemia generated ECL cell neoplasia in the rat occurs after >2 yr of drug therapy), this model represents the effects of hypergastrinemia on the gastric mucosa of this species. Hypergastrinemia alone may not therefore be sufficient for ccn2 transcript elevation. Hypergastrinemic alterations in other growth factors, e.g., EGF/TGF-{alpha}, may activate the expression of CCN2 in the mucosa.

Since we have previously identified a substantial role for the EGF/TGF-{alpha} pathway in tumor ECL cell proliferation (35), we now are able to extend our earlier work by providing evidence that the EGF/TGF-{alpha} mechanism may, in addition, involve a CCN2 effect mediated via ERK1/2 pathway activation. Thus, a system in which CCN2 is upregulated in tumor cells can clearly be postulated to amplify the proliferative effects previously assigned to EGF/TGF-{alpha} alone (35).

Very little is known about CCN2 expression in the human gastric mucosa, and neither the expression nor role of this growth factor in the stomach is known. In the present studies, we identified that ccn2 mRNA and CCN2 protein expression specifically differentiated "gastrin-dependent" carcinoids from "gastrin-independent" carcinoids. Specifically, CCN2 was present in hypergastrinemic (type I) and normogastrinemic (type III/NE carcinomas) ECL cell carcinoids but was overexpressed in the latter compared with tumors that are responsive to gastrin. This is similar to Mastomys and suggests that CCN2 may be related to autonomous (nongastrin responsive) tumor growth. In addition, CCN2 expression also appeared to have utility in discriminating between different types of "gastrin-autonomous" neoplasia. NE and neural-derived gastric tumors (ECL carcinoids and GISTs) had very different CCN2 expression profiles compared with epithelial cell-derived adenocarcinomas. Expression of this growth factor may therefore be used to differentiate between the normal mucosa and different types of human neoplastic gastric ECL cell tumors.

In summary, in the Mastomys model of ECL cell neoplasia, ccn2 transcript and CCN2 protein are overexpressed in ECL tumor cells compared with normal ECL cells. CCN2 stimulates tumor ECL cell proliferation but not normal cell proliferation and synergizes the proliferative effects of EGF under in vitro conditions. These effects are mediated via ERK1/2 phosphorylation and can be reduced by inhibition of this pathway. These data support the identification of CCN2 in tumor ECL cells and suggest that it may play a role as a key regulator of ECL cell proliferation. We hypothesize that in humans, an analogous pathway may exist whereby ECL cell neoplasia may be mediated by the expression of CCN2 in the proliferating ECL cells and cells of the adjacent "tumor-responsive" mucosa. CCN2 either alone or in concert with additional growth factors may represent a common molecular mechanism responsible for ECL cell proliferation and gastric carcinoid neoplasia.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported in part by National Cancer Institute Grant R01-CA-097050 (to I. M. Modlin) and by the Bruggeman Medical Foundation.


   FOOTNOTES
 

Address for reprint requests and other correspondence: I. M. Modlin, Dept. of Surgery, Yale Univ. School of Medicine, TMP202, 333 Cedar St., New Haven, CT 06520-8062 (e-mail: imodlin{at}optonline.net)

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
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Alessi D, Cuenda A, Cohen P, Dudley D, Saltiel A. PD098059 is a specific inhibitor of mitogen-activated protein kinase kinase in vitro and in vivo. J Biol Chem 270: 27489–27494, 1995.[Abstract/Free Full Text]
  2. Brigstock DR. The CCN family: a new stimulus package. J Endocrinol 178: 169–175, 2003.[Abstract]
  3. Cavanaugh J, Ham J, Hetman M, Poser S, Yan C, Xia Z. Differential regulation of mitogen-activated protein kinases ERK1/2 and ERK5 by neurotrophins, neuronal activity, and cAMP in neurons. J Neurosci 21: 434–443, 2001.[Abstract/Free Full Text]
  4. Eick GN, Kidd M, Mane SM, Nadler B, Champaneria MC, Pfragner R, Modlin IM. GeneChip, geNorm and GI carcinoids: novel reference genes for real-time PCR. J Mol Diagnostics. In press.
  5. Frumman D, Meyers R, Cantley L. Phosphoinositide kinases. Annu Rev Biochem 67: 481–507, 1998.[CrossRef][ISI][Medline]
  6. Gallagher PG, Bao Y, Prorock A, Zigrino P, Nischt R, Politi V, Mauch C, Dragulev B, Fox JW. Gene expression profiling reveals cross-talk between melanoma and fibroblasts: implications for host-tumor interactions in metastasis. Cancer Res 65: 4134–4146, 2005.[Abstract/Free Full Text]
  7. Gao R, Ball DK, Perbal B, Brigstock DR. Connective tissue growth factor induces c-fos gene activation and cell proliferation through p44/42 MAP kinase in primary rat hepatic stellate cells. J Hepatol 40: 431–438, 2004.[CrossRef][ISI][Medline]
  8. Grotendorst GR. Connective tissue growth factor: a mediator of TGF-beta action on fibroblasts. Cytokine Growth Factor Rev 8: 171–179, 1997.[CrossRef][Medline]
  9. Grotendorst GR, Duncan MR. Individual domains of connective tissue growth factor regulate fibroblast proliferation and myofibroblast differentiation. FASEB J 19: 729–738, 2005.[Abstract/Free Full Text]
  10. Grotendorst GR, Rahmanie H, Duncan MR. Combinatorial signaling pathways determine fibroblast proliferation and myofibroblast differentiation. FASEB J 18: 469–479, 2004.[Abstract/Free Full Text]
  11. Hartel M, Di Mola FF, Gardini A, Zimmermann A, Di Sebastiano P, Guweidhi A, Innocenti P, Giese T, Giese N, Buchler MW, Friess H. Desmoplastic reaction influences pancreatic cancer growth behavior. World J Surg 28: 818–825, 2004.[CrossRef][ISI][Medline]
  12. Joyce JA. Therapeutic targeting of the tumor microenvironment. Cancer Cell 7: 513–520, 2005.[CrossRef][ISI][Medline]
  13. Kang Y, Siegel PM, Shu W, Drobnjak M, Kakonen SM, Cordon-Cardo C, Guise TA, Massague J. A multigenic program mediating breast cancer metastasis to bone. Cancer Cell 3: 537–549, 2003.[CrossRef][ISI][Medline]
  14. Kidd M, Eick G, Shapiro MD, Camp RL, Mane SM, Modlin IM. Microsatellite instability and gene mutations in transforming growth factor-beta type II receptor are absent in small bowel carcinoid tumors. Cancer 103: 229–236, 2005.[CrossRef][ISI][Medline]
  15. Kidd M, Hinoue T, Eick G, Lye KD, Mane SM, Wen Y, Modlin IM. Global expression analysis of ECL cells in Mastomys natalensis gastric mucosa identifies alterations in the AP-1 pathway induced by gastrin-mediated transformation. Physiol Genomics 20: 131–142, 2004.[Abstract/Free Full Text]
  16. Kidd M, Modlin IM, Eick GN, Champaneria MC. Isolation, functional characterization, and transcriptsome of Mastomys ileal enterochromaffin cells. Am J Physiol Gastrointest Liver Physiol 291: G778–G791, 2006.[Abstract/Free Full Text]
  17. Kidd M, Modlin I, Shapiro M, Camp R, Mane S, Lye K, Usinger W, Murren J. CTGF, intestinal stellate cells and carcinoid tumor fibrogenesis. World J Gastroenterol. In press.
  18. Kidd M, Modlin IM, Mane SM, Camp RL, Latic I, Zikusoka MN. The utility of molecular genetic signatures in the delineation of gastric neoplasia. Cancer 106: 1480–1488, 2006.[CrossRef][ISI][Medline]
  19. Kidd M, Tang LH, Schmid SW, Miu K, Modlin IM. A polyamine pathway-mediated mitogenic mechanism in enterochromaffin-like cells of Mastomys. Am J Physiol Gastrointest Liver Physiol 275: G370–G376, 1998.[Abstract/Free Full Text]
  20. Koliopanos A, Friess H, di Mola FF, Tang WH, Kubulus D, Brigstock D, Zimmermann A, Buchler MW. Connective tissue growth factor gene expression alters tumor progression in esophageal cancer. World J Surg 26: 420–427, 2002.[CrossRef][ISI][Medline]
  21. Kubo M, Kikuchi K, Nashiro K, Kakinuma T, Hayashi N, Nanko H, Tamaki K. Expression of fibrogenic cytokines in desmoplastic malignant melanoma. Br J Dermatol 139: 192–197, 1998.[CrossRef][ISI][Medline]
  22. Lambrecht NW, Yakubov I, Zer C, Sachs G. The transcriptomes of purified gastric ECL and parietal cells: identification of a novel pathway regulating acid secretion. Physiol Genomics 25: 153–165, 2006.[Abstract/Free Full Text]
  23. Lauffer J, Tang L, Zhang T, Hinoue T, Rahbar S, Odo M, Modlin I, Kidd M. PACAP mediates the neural proliferative pathway of Mastomys enterochromaffin-like cell transformation. Regul Pept 102: 157–164, 2001.[CrossRef][ISI][Medline]
  24. Lempinen M, Inkinen K, Wolff H, Ahonen J. Connective tissue growth factor in indomethacin-induced rat gastric ulcer. Eur Surg Res 34: 232–238, 2002.[CrossRef][ISI][Medline]
  25. Li C, Wong WH. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci USA 98: 31–36, 2001.[Abstract/Free Full Text]
  26. Modlin IM, Esterline W, Kim H, Goldenring JR. Enterochromaffin-like cells and gastric argyrophil carcinoidosis. Acta Oncol 30: 493–498, 1991.[ISI][Medline]
  27. Modlin IM, Zhu Z, Tang LH, Kidd M, Lawton GP, Miu K, Powers RE, Goldenring JR, Pasikhov D, Soroka CJ. Evidence for a regulatory role for histamine in gastric enterochromaffin-like cell proliferation induced by hypergastrinemia. Digestion 57: 310–321, 1996.[ISI][Medline]
  28. Norsett KG, Laegreid A, Langaas M, Worlund S, Fossmark R, Waldum HL, Sandvik AK. Molecular characterization of rat gastric mucosal response to potent acid inhibition. Physiol Genomics 22: 24–32, 2005.[Abstract/Free Full Text]
  29. Pan LH, Beppu T, Kurose A, Yamauchi K, Sugawara A, Suzuki M, Ogawa A, Sawai T. Neoplastic cells and proliferating endothelial cells express connective tissue growth factor (CTGF) in glioblastoma. Neurol Res 24: 677–683, 2002.[CrossRef][ISI][Medline]
  30. Perbal B. CCN proteins: multifunctional signalling regulators. Lancet 363: 62–64, 2004.[CrossRef][ISI][Medline]
  31. Sandor A, Kidd M, Lawton G, Miu K, Tang L, Modlin I. Neurohormonal regulation of rat ECL cell histamine secretion. Gastroenterology 110: 1084–1092, 1996.[CrossRef][ISI][Medline]
  32. Shedden K, Chen W, Kuick R, Ghosh D, Macdonald J, Cho KR, Giordano TJ, Gruber SB, Fearon ER, Taylor JM, Hanash S. Comparison of seven methods for producing Affymetrix expression scores based on false discovery rates in disease profiling data. BMC Bioinformatics 6: 26, 2005.[CrossRef][Medline]
  33. Syder AJ, Karam SM, Mills JC, Ippolito JE, Ansari HR, Farook V, Gordon JI. A transgenic mouse model of metastatic carcinoma involving transdifferentiation of a gastric epithelial lineage progenitor to a neuroendocrine phenotype. Proc Natl Acad Sci USA 101: 4471–4476, 2004.[Abstract/Free Full Text]
  34. Tang LH, Luque EA, Efstathiou JA, Bortecen KH, Kidd M, Tarasova NI, Modlin IM. Gastrin receptor expression and function during rapid transformation of the enterochromaffin-like cells in an African rodent. Regul Pept 72: 9–18, 1997.[CrossRef][ISI][Medline]
  35. Tang LH, Modlin IM, Lawton GP, Kidd M, Chinery R. The role of transforming growth factor alpha in the enterochromaffin-like cell tumor autonomy in an African rodent Mastomys. Gastroenterology 111: 1212–1223, 1996.[CrossRef][ISI][Medline]
  36. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3: RESEARCH0034, 2002.[Medline]
  37. Wang X, Studzinski G. Phosphorylation of raf-1 by kinase suppressor of ras is inhibited by "MEK-specific" inhibitors PD 098059 and U0126 in differentiating HL60 cells. Exp Cell Res 268: 294–300, 2001.[CrossRef][ISI][Medline]
  38. Wenger C, Ellenrieder V, Alber B, Lacher U, Menke A, Hameister H, Wilda M, Iwamura T, Beger HG, Adler G, Gress TM. Expression and differential regulation of connective tissue growth factor in pancreatic cancer cells. Oncogene 18: 1073–1080, 1999.[CrossRef][ISI][Medline]
  39. Yosimichi G, Kubota S, Nishida T, Kondo S, Yanagita T, Nakao K, Takano-Yamamoto T, Takigawa M. Roles of PKC, PI3K and JNK in multiple transduction of CCN2/CTGF signals in chondrocytes. Bone 38: 853–863, 2006.[Medline]
  40. Zeng N, Athmann C, Kang T, Lyu RM, Walsh JH, Ohning GV, Sachs G, Pisegna JR. PACAP type I receptor activation regulates ECL cells and gastric acid secretion. J Clin Invest 104: 1383–1391, 1999.[ISI][Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Gastrointest. Liver Physiol.Home page
M. Kidd, I. M. Modlin, B. I. Gustafsson, I. Drozdov, O. Hauso, and R. Pfragner
Luminal regulation of normal and neoplastic human EC cell serotonin release is mediated by bile salts, amines, tastants, and olfactants
Am J Physiol Gastrointest Liver Physiol, August 1, 2008; 295(2): G260 - G272.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
292/1/G191    most recent
00131.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow