JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M603578200 on November 7, 2006

J. Biol. Chem., Vol. 281, Issue 52, 39819-39830, December 29, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/52/39819    most recent
M603578200v1
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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kunnimalaiyaan, M.
Right arrow Articles by Chen, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kunnimalaiyaan, M.
Right arrow Articles by Chen, H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Overexpression of the NOTCH1 Intracellular Domain Inhibits Cell Proliferation and Alters the Neuroendocrine Phenotype of Medullary Thyroid Cancer Cells*

Muthusamy Kunnimalaiyaan, Abram M. Vaccaro, Mary A. Ndiaye, and Herbert Chen1

From the Endocrine Surgery Research Laboratories, Department of Surgery, and the University of Wisconsin Paul P. Carbone Comprehensive Cancer Center, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin 53792

Received for publication, April 13, 2006 , and in revised form, November 7, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The role of NOTCH1 as an oncogene or tumor suppressor appears to be cell type-specific. Medullary thyroid cancer (MTC) cells characteristically express the transcription factor ASCL1 (achaete-scute complex-like 1) as well as high levels of the neuroendocrine (NE) markers calcitonin and chromogranin A (CgA). In this study, we show that the active NOTCH1 intracellular domain is absent in human MTC tumor tissue samples and MTC-TT cells. To determine the effects of NOTCH1 expression, we created a doxycycline-inducible NOTCH1 intracellular domain in MTC cells (TT-NOTCH cells). Treatment of TT-NOTCH cells with doxycycline led to dose-dependent induction of NOTCH1 protein with corresponding decreases in ASCL1 protein and NE hormones. ASCL1 promoter-reporter assay and Northern analysis revealed that ASCL1 reduction by NOTCH1 activation is predominantly via silencing of ASCL1 gene transcription. Overexpression of ASCL1 in MTC cells indicated that CgA expression is highly dependent on the levels of ASCL1. This was further confirmed by experiments using small interfering RNA against ASCL1, in which reduction in ASCL1 led to reduction in both CgA and calcitonin. Furthermore, we demonstrate that NOTCH1 signaling activation leads to ERK1/2 phosphorylation, but that reduction in NE markers is independent of ERK1/2 activation. Activation of NOTCH1 resulted in significant MTC cell growth inhibition. Notably, reduction in MTC cell growth was dependent on the level of NOTCH1 protein present. Moreover, no increase in growth upon expression of ASCL1 in NOTCH1-activated cells was observed, indicating that the growth suppression observed upon NOTCH1 activation is independent of ASCL1 reduction. Mechanistically, we show that MTC cell growth inhibition by NOTCH1 is mediated by cell cycle arrest associated with up-regulation of p21.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Medullary thyroid cancer (MTC)2 is a neuroendocrine (NE) tumor derived from the calcitonin-producing C-cells of the thyroid gland and accounts for 3–5% of cases of thyroid cancer (1, 2). The only curative therapy for patients with MTC is surgical resection. Eighty percent of all MTCs are sporadic in nature, and the remaining 20% are familial and caused by germ line mutations in the RET proto-oncogene (2, 3). Although development of RET gene testing has allowed for early prophylactic thyroidectomy for patients with familial MTC, the majority of patients with sporadic MTC have persistent or recurrent disease after surgery because the natural history of MTC is characterized by early metastasis. Understanding the molecular pathways that control MTC and C-cell development and proliferation is essential for the development of novel therapies for patients with advanced MTC.

Like other NE tumors, MTC cells secrete various hormones and NE markers such as calcitonin and chromogranin A (CgA) (4). In addition, MTC cells express high levels of ASCL1 (achaete-scute complex-like 1, also known as human ASH1 (achaete-scute homolog-1)), an evolutionarily conserved basic helix-loop-helix transcription factor that seems to be limited to NE tumors (5–8). Several recent studies suggest that ASCL1 appears to be critical for C-cell development and may play an important role in MTC tumor growth and NE differentiation. Ash1 transgenic knock-out mice fail to develop thyroid C-cells, which suggests that mASH1 is essential for normal C-cell development (9). Furthermore, Raf-1 activation in human MTC cells also leads to significant reductions in ASCL1, CgA, calcitonin, and growth (1, 5, 10). In addition, ASCL1 plays an important role in the development of other NE cells such as adrenal chromaffin cells and pulmonary endocrine cells (11–13). These findings also indicate that pathway(s) that regulate ASCL1 may also regulate MTC cell growth and differentiation. During development, ASCL1 expression is tightly controlled by the NOTCH1 signaling pathway (for reviews, see Refs. 14 and 15).

NOTCH1 is a multifunctional transmembrane receptor that regulates cell differentiation in a variety of contexts. Recently, NOTCH1 has also been shown to play an essential role in NE differentiation of the lung and gastrointestinal tract (16–19). In human cancer cells, NOTCH1 signaling engages in a dual role as either a tumor suppressor or an oncogene. Activation of the NOTCH1 signaling pathway has been shown to inhibit growth of prostate cancer, small cell lung cancer (SCLC), and pancreatic carcinoid (8, 20–22) and to induce apoptosis of B-cells and other hematopoietic lineages in vitro (23). However, the degree of NOTCH1 activation required to achieve growth suppression is unclear because most of the studies utilized transient expression with very high levels of active NOTCH1.

In this study, we demonstrate the critical role of NOTCH1 in controlling both cell proliferation and the NE phenotype in MTC cells. We show that NOTCH1 protein is undetectable in human MTC tumor samples and cells. However, high levels of ASCL1 and CgA were found in the MTC cell line and tumor specimens. To determine the effects of NOTCH1 expression, we created a doxycycline-inducible NOTCH1 intracellular domain (NICD) construct in MTC cells (TT-NOTCH cells). Using this doxycycline-inducible in vitro model, we assessed the dose-dependent effects of NOTCH1 on MTC cell growth. Expression of NICD resulted in significant inhibition of MTC cell proliferation. Notably, the degree of tumor cell growth inhibition was directly proportional to the amount of NOTCH1 protein present. Notably, NOTCH1 activation also caused reductions in the levels of calcitonin and CgA. Furthermore, we show for the first time by transient overexpression of ASCL1 in MTC cells that CgA levels are directly correlated with ASCL1 protein levels. The fact that transfection of small interfering RNA (siRNA) against ASCL1 in MTC cells resulted in a proportional reduction in CgA further confirms that the level of CgA is tightly regulated by ASCL1. Finally, we report that activation of NOTCH1 also leads to the phosphorylation of ERK1/2 in a dose-dependent manner, but that active ERK1/2 is not required for ASCL1-mediated suppression of NE markers.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Human Tissue Samples—Human MTC tumor samples and other control tumor samples were collected, and tumor histology was verified by pathological review. Samples were snap-frozen in liquid nitrogen and kept at –80 °C for long-term storage. Tumor cell lysates were prepared from the frozen samples by grinding in liquid nitrogen. The powder was then lysed with lysis buffer as described (24). The lysates were analyzed by Western blotting for NOTCH1, ASCL1, and CgA as described below.

Cell Culture—Human MTC (TT) cells were obtained from Dr. Barry D. Nelkin (The Johns Hopkins University, Baltimore, MD) and maintained in RPMI 1640 medium (Invitrogen) supplemented with 16% fetal bovine serum (Sigma), 100 IU/ml penicillin (Invitrogen), and 100 µg/ml streptomycin (Invitrogen) in a humidified atmosphere of 5% CO2 in air at 37 °C (1, 10). Doxycycline-inducible cell lines were maintained similar to TT cells except that tetracycline-free fetal bovine serum (Clontech), 0.4 µg/ml G418 (Invitrogen), and 0.4 µg/ml hygromycin (Invitrogen) were used.

Doxycycline-inducible NOTCH1—The Tet-On expression system was obtained from Clontech. A 2.3-kb BamHI-XhoI fragment containing NICD (amino acids 1759–2556) from pTAN1-cDNA was subcloned into the pRevTRE vector at the BamHI/SalI sites. The cloned construct (pRevTRE-NOTCH1) was confirmed by DNA sequencing. To create an inducible TT-NOTCH cell line, TT cells were first transfected with plasmid pRevTet-On (Clontech) containing the Tet-responsive transcriptional activator using Lipofectamine 2000 (Invitrogen) and selected in medium containing 0.4 µg/ml G418. The resulting 12 G418-resistant, TT-Tet-on clones were screened for doxycycline-dependent inducibility of the reporter gene luciferase by transient transfection of the pRevTRE-Luc vector into TT-Tet-on clones. A clone with low background, which reproducibly induced the luciferase gene by >60-fold, was maintained and used for transfection with the response plasmid containing the NOTCH1 (NICD) gene (pRevTRE-NOTCH1) and the empty vector (pRevTRE). TT-Tet-on cells were transfected with pRevTRE-NOTCH1 and the pRevTRE empty vector using Lipofectamine 2000. Transfected cells were selected in 0.4 µg/ml hygromycin. Resistant TT-NOTCH and TT-TRE (vector alone) clones were treated with 1 µg/ml doxycycline for 48 h and screened for the presence of NOTCH1 protein by Western blot analysis.

Western Blot Analysis—Cell pellets were lysed in sample buffer as described (24). Total cellular protein concentrations were determined using a BCA assay kit (Pierce). Cell extracts (20–30 µg) were boiled with equal amounts of loading dye (24) for 10 min and separated on 10% SDS-polyacrylamide gel. Proteins were transferred onto nitrocellulose membranes (Schleicher & Schüll) by electroblotting. Membranes were blocked in milk and incubated with primary and secondary antibodies as described (24). The following primary antibodies at the indicated dilutions were used: anti-NOTCH1 (1:500); anti-HA probe (1:200; Santa Cruz Biotechnology, Inc., Santa Cruz, CA); anti-MASH1 (mammalian ASH1) to detect ASCL1 (1:1000; Pharmingen); anti-HES-1 (Hairy Enhancer of Split-1; 1:10,000; a kind gift from Tetsuo Sudo, Toray Industries, Inc., Kanagawa, Japan); anti-CgA (1:1000; Zymed Laboratories Inc.); anti-phospho-ERK1/2, anti-ERK1/2, anti-p21, anti-p27, anti-cyclin D1, anti-Cdk4, and anti-Cdc2 phospho-Tyr15 (1:1000; Cell Signaling Technology, Inc., Beverly, MA); and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 1:10,000; Trevigen, Inc., Gaithersburg, MD). Primary antibody incubations were kept overnight at 4 °C. The membranes were then washed with wash buffer (1x phosphate-buffered saline and 0.01% Tween 20) and incubated with a 1:500 dilution of goat anti-rabbit (for NOTCH1, HES-1, CgA, and GAPDH) or goat anti-mouse (for MASH1) secondary antibody (Pierce) coupled with horseradish peroxidase. The membranes were washed with wash buffer and developed with Immun-Star (Bio-Rad) for HES-1, CgA, and GAPDH or with SuperSignal West Femto chemiluminescent substrate (Pierce) for MASH1, cell cycle inhibitors, and NOTCH1 according to the manufacturers' directions.

NOTCH1 Functional Analysis—Plasmids containing four copies of either the wild-type or mutant CBF1-binding elements in pGL2pro (Promega Corp., Madison, WI) were obtained from Dr. Diane Hayward (The Johns Hopkins University). TT-NOTCH cells were maintained in RPMI 1640 medium supplemented with 16% tetracycline-free fetal bovine serum, 0.4 µg/ml hygromycin, and 0.4 µg/ml G418. Cells grown overnight were cotransfected with 1 µg of wild-type or mutant CBF1 plasmid and 0.5 µg of plasmid containing the beta-galactosidase gene driven by the cytomegalovirus (CMV) promoter (CMV-beta-galactosidase plasmid) as an internal control for transfection efficiency. The next day, cells were treated with or without doxycycline and incubated for 2 days to induce NOTCH1. The cells were then lysed, and luciferase and beta-galactosidase assays were carried out as recommended by Promega Corp. using a luminometer (Analytical Luminescence Laboratory, Ann Arbor, MI) and a plate reader, respectively. Luciferase activity was determined relative to the expression calculated by the wild type over the mutant (25). Experiments were performed in triplicate at least twice.

Cell Proliferation Assay and Calcitonin Measurement—To measure the proliferation rate, cells were plated in triplicate in 6-well plates, and the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was performed as described (20, 26). The cells were treated with or without doxycycline at various concentrations for the indicated number of days. Experiments were performed at least twice. To determine the amount of calcitonin in cell extracts, we utilized a calcitonin enzyme-linked immunosorbent assay kit (BIOSOURCE) following the manufacturer's instructions. Calcitonin values were standardized and quantified relative to control cell extracts. Samples were analyzed in triplicate.

Transfection, Reporter Assays, and siRNA Experiments—To determine the effect of ASCL1 on the level of CgA, TT-NOTCH cells were transiently transfected with plasmid pcDNA3.0 containing the CMV promoter-derived ASCL1 gene fused with an HA tag (a kind gift from Dr. Douglas W. Ball, The Johns Hopkins University) using Lipofectamine 2000 according to the manufacturer's directions. The next day, transfected cells were treated with or without doxycycline for 2 days, and cell lysates were prepared and analyzed for the levels of ASCL1 and CgA by Western analysis.

Luciferase Assay for the ASCL1 Promoter—The ASCL1 promoter DNA fragment (–7900 to +37) containing the 5'-flanking region of ASCL1 genomic DNA, including the HES-1-binding site, was cloned into the luciferase reporter gene in pGL2 (Promega Corp.) (27). Transfection of the ASCL1 promoter DNA showed a reduction in luciferase activity in the presence of HES-1 expression (27). To determine the effect of NOTCH1 activation on ASCL1 transcription, we transiently transfected the plasmid containing the luciferase gene under the control of the ASCL1 promoter (27) into TT-NOTCH cells. Plasmids pGL2-Control (positive control for luciferase) and pGL2-Basic (negative control) were also transfected into TT-NOTCH cells as controls. In addition, the CMV-beta-galactosidase plasmid was also cotransfected into TT-NOTCH cells to normalize the transfection. The next day, the medium was changed with or without doxycycline to activate NOTCH1 for 2 days. Cell lysates were then prepared and analyzed for luciferase activity as described (20).

ASCL1 siRNA—To determine the effect of ASCL1 on CgA, siRNA against ASCL1 (catalog no. sc-37692, Santa Cruz Biotechnology, Inc.) and nonspecific siRNA (catalog no. sc-37007) were transfected into TT cells using Lipofectamine 2000 as described by the manufacturer. After 2 days of incubation, cell lysates were prepared and analyzed for the levels of ASCL1 and CgA proteins by Western analysis.


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 1.
Expression patterns of NOTCH1 and NE markers in various human MTC tumor samples and cells. Shown are the results from Western blot analysis of cleaved NOTCH1 protein in human tumor samples of MTC, papillary thyroid cancer (PTC), and pancreatic adenocarcinoma (PA) and in MTC-TT cells. Note the lack of cleaved NOTCH1 protein in MTC tumors and TT cells. However, NOTCH1 is present in non-NE tissues (papillary thyroid cancer and pancreatic adenocarcinoma). Interestingly, CgA and ASCL1 are present only in NE tumors and the tumor cell line, and their levels are inversely related to NOTCH1 expression. Immunoblotting with anti-GAPDH (G3PDH) antibody confirmed the relatively equal protein loading.

 
Growth Rescue Experiment—To determine the importance of ASCL1 in NOTCH1-mediated growth suppression, we carried out the MTT assay. Briefly, TT-NOTCH cells grown overnight were transfected with plasmid pcDNA 3.0 containing the CMV promoter-derived ASCL1 gene fused with an HA tag using Lipofectamine 2000 according to the manufacturer's directions. The next day, cells were trypsinized, and equal numbers of cells were plated onto 24-well plates and incubated overnight. The following day, the cells were treated with or without doxycycline (1 µg/ml) and incubated for 4 days, and the MTT assay was then carried out as described above.

Northern Analysis—To study the effect of NOTCH1 on ASCL1 transcription, TT-NOTCH cells were treated with doxycycline (0, 0.2, 0.5, and 1 µg/ml) at indicated time points, and total RNA was isolated using an RNA isolation kit (Qiagen Inc.). Ten micrograms of this RNA was separated on denaturing 1% formaldehyde-agarose gel and transferred onto positively charged nylon membrane (Ambion, Inc., Austin, TX). DNA probes using ASCL1 and GAPDH DNAs were made using a random primer labeling kit (New England Biolabs Inc., Beverly, MA). The probes were denatured and hybridized with the membrane using a standard protocol (24).

Statistical Analysis—Analysis of variance with Bonferroni's post hoc test (SPSS Version 10.0 software, SPSS Inc., Chicago, IL) was utilized for statistical analysis. A p value of <0.05 was considered significant. Unless noted, data are represented as the means ± S.E.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Expression of NOTCH1 Components in MTC Tumors—We demonstrated previously the presence of ASCL1 mRNA in MTC, carcinoid, and pheochromocytoma cells by Northern analysis (7). However, the protein levels of ASCL1 in human tumor samples have not been determined. Fig. 1 shows significant amounts of ASCL1 protein present in six human MTC tumors (four sporadic and two multiple endocrine neoplasia type 2A patients) as well as in human MTC (TT) cells. However, ASCL1 is not present in non-NE tumor tissues such as papillary thyroid cancer samples (Fig. 1). Furthermore, as expected, expression of the NE marker CgA is restricted to NE tumors (MTC).


Figure 2
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 2.
Doxycycline-induced NOTCH1 activation in MTC cells. A, TT-NOTCH cells treated with various concentrations of doxycycline as indicated showed increased levels of NOTCH1 protein. B, TT-NOTCH cells were transiently transfected with CBF1-luciferase reporter constructs and then treated with various concentrations of doxycycline for 48 h. NOTCH1 function was measured by the degree of CBF1 binding utilizing a luciferase construct containing four CBF1-binding sites (4xCBF1-Luc) compared with background binding to a luciferase construct with mutant CBF1-binding sites (4xmtCBF1-Luc). The values represent the ratio of 4xCBF1-Luc to 4xmtCBF1-Luc expressed relative to TT-NOTCH control cells in the absence of doxycycline. Note that various concentrations of doxycycline resulted in an increase in relative luciferase activity compared with the control (no doxycycline). Statistical analysis showed significant differences in luciferase -fold induction (p < 0.001) in all groups. G3PDH, GAPDH.

 
Given the importance of NOTCH1 signaling and its role in ASCL1, we expected a lack of active NOTCH1 (NICD) protein in these NE tumors. As shown in Fig. 1, MTC tumors lack NICD protein. In contrast, NOTCH1 is expressed in papillary thyroid cancer and pancreatic adenocarcinoma tissues. Because cleaved active NOTCH1 is absent in MTC tumors and cells, we hypothesized that overexpression of active NOTCH1 (NICD) in MTC cells may down-regulate ASCL1 and possibly affect cell growth and/or differentiation. Thus, we created a doxycycline-inducible NICD (NOTCH1) system in TT cells.

Dose-dependent Functional NOTCH1 Expression in TT-NOTCH Cells—To determine the inducibility of NICD by doxycycline in TT-NOTCH cells, we carried out Western analysis after treatment with various concentrations of doxycycline. As shown in Fig. 2A, there was no detectable NOTCH1 protein in TT-NOTCH cells in the absence of doxycycline. However, treatment of TT-NOTCH cells with doxycycline led to an induction of NOTCH1 protein. Notably, the amount of NOTCH1 protein was proportional to the dose of doxycycline.

To determine the functionality of NOTCH1 protein in TT cells, we utilized a well described reporter system in which binding of NOTCH1 leads to high levels of luciferase expression (25). CBF1/RBP-J{kappa} (recombination signal-binding protein 1 for J-{kappa}) is the best characterized downstream effector of NOTCH1 in mammals (25, 28). Fig. 2B shows that NOTCH1 activation increased the relative -fold induction of luciferase activity. This increase in luciferase activity was dependent on the amount of doxycycline used. However, in the absence of doxycycline, there was a minimal amount of luciferase activity, suggesting the lack of functional NOTCH1 activity (Fig. 2B). Taken together, the results of the Western and NOTCH1 functional analyses clearly suggest that NOTCH1 protein produced after doxycycline treatment in TT-NOTCH cells is functional and that the activity is proportional to the amount of protein present.

Conservation of the NOTCH1 Signaling Pathway in MTC Cells—To determine whether the downstream targets of activated NOTCH1 are intact in MTC cells, Western blotting was performed for HES-1 protein, an ASCL1 transcriptional repressor and an immediate downstream mediator of NOTCH1 signaling. In a dose-response assay, the TT-NOTCH cells treated with various concentration of doxycycline for 2 days showed an increase in the amount of HES-1 protein (Fig. 3A). The increase in NOTCH1 expression resulted in a significant increase in the level of HES-1 protein. However, the level of HES-1 protein plateaued after treatment with 0.5 µg/ml doxycycline. This is consistent with previous findings that HES-1 protein negatively regulates its expression by binding to its own promoter (29, 30). HES-1 is also present in control cells at a minimal level, and it has been shown that, under normal conditions, HES-6, another basic helix-loop-helix protein, binds to HES-1 and blocks the binding of HES-1 to the ASCL1 promoter (31, 32). As shown in Fig. 3A, HES-6 is expressed in TT cells, and this could be the reason why we observed ASCL1 expression in the presence of HES-1 protein in untreated cells. Notably, the level of HES-6 was reduced after NOTCH1 activation. Activation of NOTCH1 has been shown to down-regulate the expression of ASCL1 and its homologs in multiple organisms. A previous study has shown that ASCL1 expression is associated with NE markers (33). Furthermore, ASCL1 transcriptional repression due to NOTCH1-mediated HES-1 expression has been well characterized (27, 34). In a dose-response assay, TT-NOTCH cells showed a progressive reduction in ASCL1 and CgA proteins with increasing NOTCH1 protein (Fig. 3A). As expected, treatment with various concentrations of doxycycline for 2 days in TT-vector alone cells did not affect the expression levels of ASCL1 and CgA (Fig. 3B).

NOTCH1 Signaling in MTC Cells Decreases Calcitonin Levels—We have shown that an increase in NOTCH1 protein leads to dose-dependent reduction in CgA (Fig. 3A). However, the most important bioactive hormone produced by MTC cells is calcitonin. Therefore, we were interested in the level of calcitonin after varying levels of NOTCH1 activation were achieved in MTC cells. As shown Fig. 3C, induction of NOTCH1 resulted in a progressive reduction in calcitonin. The relative calcitonin reductions were 5% at 0.2 µg/ml doxycycline, 30% at 0.5 µg/ml, and 47% at 1.0 µg/ml (Fig. 3C). TT-vector alone cells treated with doxycycline did not show any reduction in calcitonin levels at any time point (data not shown).


Figure 3
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 3.
Effects of NOTCH1 signaling on NE markers and its persistent requirement in MTC cells. A, TT-NOTCH cells were treated with various concentrations of doxycycline as indicated for 2 days. Total cell extracts were isolated and analyzed by Western blotting for downstream mediators of NOTCH1 signaling. Increased NOTCH1 protein production by increasing concentrations of doxycycline (Fig. 1A) led to an increase in HES-1 protein. HES-6, a basic helix-loop-helix protein that binds to HES-1 and inhibits its activity, was reduced by increasing amounts of NOTCH1 protein. HES-1 is a transcriptional suppressor of ASCL1 protein. TT-NOTCH control cells (no doxycycline) have a large amount of endogenous ASCL1. However, the ASCL1 level was reduced progressively by increasing amounts of HES-1 protein. Similarly, CgA levels were reduced by increasing concentrations of doxycycline. B, TT-vector alone cells were treated with various concentrations of doxycycline as indicated for 2 days. Cell extracts were analyzed for the presence of ASCL1 and CgA. Note that TT-vector alone cells treated with doxycycline showed no change in the levels of ASCL1 and CgA, indicating that doxycycline alone does not have any effect on MTC cells. Immunoblotting with anti-GAPDH (G3PDH) antibody confirmed equal protein loading. C, high levels of calcitonin are present in MTC cells in the absence of NOTCH1 activation (TT-NOTCH control (C)). However, activation of NOTCH1 in TT-NOTCH cells led to a progressive decrease in relative calcitonin by 5% at 0.2 µg/ml doxycycline, by 30% at 0.5 µg/ml, and by 47% at 1.0 µg/ml. ELISA, enzyme-linked immunosorbent assay. D, TT-NOTCH cells were treated with 0.5 and 1 µg/ml doxycycline (Doxy) for 2 days. The medium was removed; cells were washed with phosphate-buffered saline; and doxycycline-free medium was added up to 6 days (d). Total cell extracts were isolated and analyzed by Western blotting for the presence of downstream mediators of NOTCH1 signaling. Note that NOTCH1 (NICD) protein was present only in doxycycline-treated cells, indicating not only that constant doxycycline is required for the expression of NICD, but also the tight regulation of expression. Interestingly, ASCL1 expression was reduced when NOTCH1 was present. In contrast, CgA reduction was stable up to 2 days after doxycycline withdrawal; but at later time points, the level of CgA returned to the control cell level.

 


Figure 4
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 4.
Persistent activation of the NOTCH1 signaling pathway in TT-NOTCH cells by doxycycline treatment. TT-NOTCH and TT-vector alone cells were treated with various concentrations of doxycycline (Doxy) for 4 and 8 days, and cell extracts were prepared and analyzed for activation of the NOTCH1 signaling pathway by Western blotting. As predicted, the level of NOTCH1 protein was dependent on the amount of doxycycline at all time points. However, ASCL1 was barely detectable after 4 day in NOTCH1-activated cells. Interestingly, the NE marker CgA was reduced by day 4 and was barely detectable by day 8. Immunoblotting with anti-GAPDH (G3PDH) antibody confirmed equal protein loading.

 
Continuous Requirement of NOTCH1 Induction—To determine the duration of induction of NOTCH1 by doxycycline, TT-NOTCH cells were treated with and without doxycycline (0.5 and 1 µg/ml) for 2 days. The cells were then washed with phosphate-buffered saline and grown in complete medium without doxycycline for up to 6 days. Every 2 days, cell extracts were prepared and analyzed for NOTCH1, CgA, and ASCL1 protein levels. As shown in Fig. 3D, NOTCH1 protein was detectable only in the presence of doxycycline. Interestingly, when the medium was changed to normal medium without doxycycline, NOTCH1 protein became undetectable as early as 2 days after withdrawal of the doxycycline. This result confirms that the induction of NOTCH1 is very tightly regulated and that doxycycline is required all of the time. ASCL1 reduction also correlated with the presence of NOTCH1 protein. However, CgA reduction persisted up to 2 days after the withdrawal of doxycycline, but the level increased to normal after day 2 (Fig. 3D).

Effect of the NOTCH1 Signaling Pathway in MTC Cells—To determine the effects of long-term induction of the NOTCH1 signaling pathway, we treated TT-NOTCH cells with 0, 0.2, 0.5, and 1 µg/ml doxycycline for up to 8 days. As shown in Fig. 4, continuous doxycycline treatment persistently activated the NOTCH1 pathway in a dose-dependent manner as illustrated by the presence of NOTCH1 protein. Similarly, the level of HES-1 protein increased in NOTCH1-expressing cells compared with untreated cells in which NOTCH1 protein was absent. As mentioned above, HES-1 protein binds to its own promoter and regulates its expression. Therefore, we did not see a significant increase in HES-1 levels with increasing levels of NOTCH1 protein. Interestingly, after 8 days of treatment with doxycycline, ASCL1 protein was barely detectable in TT-NOTCH cells. As shown in Fig. 4, activation of the NOTCH1 pathway led to a significant dose-dependent reduction in CgA. The reduction in CgA was evident at day 4 and pronounced at day 8.


Figure 5
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 5.
NOTCH1 signaling inhibits tumor cell proliferation. A, shown are the results from growth assay by cell counts. TT-vector alone and TT-NOTCH cells were treated with (Doxy) or without (Control)1 µg/ml doxycycline for the indicated time periods. Viable cells were counted at each time point. TT-vector alone cells treated with or without doxycycline and TT-NOTCH cells with doxycycline treatment did not have any reduction in cell growth. Activation of NOTCH1 by doxycycline in TT-NOTCH cells reduced cell growth significantly at day 8 (58%) and day 10 (70%). B, shown are the results from dose-response growth assay. TT-NOTCH cells were treated with various concentrations of doxycycline as indicated for 12 days, and the MTT assay was carried out. At 0.1 µg/ml doxycycline, there was a slight growth reduction at day 12. Interestingly, there was a dramatic reduction in TT-NOTCH cell growth at 0.5 µg/ml and also at higher doses of doxycycline (0.8 and 1 µg/ml) compared with 0.1 µg/ml. At all time points, growth reduction was significant compared with the control (p < 0.00001). C, TT-vector only control cells treated with 0, 0.1, 0.5, and 1.0 µg/ml doxycycline did not show growth reduction at day 12. Thus, doxycycline did not have any effect on TT cell growth. At all time points, there was no difference in proliferation compared with the control.

 
NOTCH1 Activation Inhibits MTC Cell Growth—Activation of the NOTCH1 signaling pathway has been shown to inhibit growth of prostate cancer, SCLC, and carcinoid cells in vitro (8, 20–22). To determine whether NOTCH1 activation in MTC cells inhibits growth, we utilized cell counts. As shown in Fig. 5A, there were no significant differences in growth rate between TT-vector alone cells with and without doxycycline treatment and TT-NOTCH cells without doxycycline treatment. However, doxycycline-induced NOTCH1 expression led to inhibition of cell growth. This suggests that persistent activation of NOTCH1 is required to inhibit the growth of MTC cells.

NOTCH1 Dose-Response Inhibition of MTC Cell Growth—Inhibition of MTC cell growth requires high levels of NOTCH1 protein. Furthermore, dose-response experiments showed a progressive decrease in ASCL1 and CgA. However, it had not been established previously if MTC cell proliferation is directly related to levels of NOTCH1 present. To determine the dose-response effects of NOTCH1 on MTC cell growth inhibition, we treated TT-NOTCH cells with varying concentrations of doxycycline as shown in Fig. 5B and measured cell viability by the MTT assay. In the absence of doxycycline, TT-NOTCH cells had a high growth rate, whereas in the presence of 0.1 µg/ml doxycycline, there was modest growth suppression. Interestingly, increasing levels of doxycycline led to proportional increases in growth suppression. Thus, the degree of MTC tumor cell growth inhibition is directly proportional to the amount of NOTCH1 protein present. Furthermore, as a control, we also treated TT-vector alone cells with doxycycline at 0, 0.1, 0.5, and 1 µg/ml and did not see any reduction in cell growth (Fig. 5C).

NOTCH1 Inhibits Growth by Cell Cycle Arrest—It has been shown that transient activation of NOTCH1 in SCLC cells leads to cell cycle arrest (22). The decrease in MTC cell growth upon overexpression of NOTCH1 may be due to decreased cell cycle transit or increased cell death. To determine the mechanism by which growth reduction occurs, we carried out Western analysis for various cell cycle regulatory proteins. As shown in Fig. 6, the level of p21WAF1/CIP1 was significantly and dose-dependently up-regulated in TT-NOTCH cells treated with different concentrations of doxycycline. However, there was no change in the p21 level in TT-vector alone control cells treated with or without doxycycline. Interestingly, the level of p27KIP1 protein was reduced after NOTCH1 activation compared with the control. We also analyzed the expression levels of cyclin D1 and Cdk4, which are involved in cell cycle regulation. Cyclin D1 levels increased in NOTCH1-activated cells, supporting the results of the reduction in p27 protein and the increase in p21 protein. However, there was no change in the level of Cdk4 protein. These results confirm the earlier work on NOTCH1 activation-mediated cell cycle arrest in SCLC cells (22). Furthermore, we were interested in determining at which stage NOTCH1 activation induces cell cycle arrest. Phosphorylation of Cdc2 at Tyr15 is associated with cell cycle regulation especially at G1/S phase. Western analysis of NOTCH1-activated cells showed that there was an increase in the levels of Tyr15-phosphorylated Cdc2 with increasing concentrations of doxycycline (Fig. 6). Taken together, these data suggest that the growth inhibition could be due to cell cycle arrest and possibly at S phase. This is consistent with the previously published result that NOTCH1-mediated growth inhibition in SCLC cells is due to cell cycle arrest (22).


Figure 6
View larger version (60K):
[in this window]
[in a new window]

  		FIGURE 6.
NOTCH1 signaling induces cell cycle arrest. TT-vector alone and TT-NOTCH cells were treated with doxycycline (Doxy) at 0, 0.2, 0.5, and 1 µg/ml for the indicated time periods. Total cell extracts were isolated and analyzed by Western blotting using antibodies against the p21, p27, cyclin D1, Cdk4, and Cdc2 phospho-Tyr15 (p-cdc2Tyr15) proteins. The tumor suppressor protein p21 and cyclin D1 were increased by increased concentrations of doxycycline, whereas p27 was reduced. The Cdk4 protein level was not changed upon NOTCH1 expression. The level of Tyr15-phosphorylated Cdc2 was increased compared with the control. G3PDH, GAPDH.

 


Figure 7
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 7.
NOTCH1 signaling activates the ERK1/2 pathway, but phosphorylated ERK1/2 is not required for NE marker reduction. A, TT-NOTCH cells were treated with doxycycline (Doxy) at 0, 0.2, 0.5, and 1 µg/ml for the indicated time periods. Total cell extracts were isolated and analyzed by Western blotting using antibody against phospho-ERK1/2 (p-ERK1/2) protein to determine the activation of the Raf-1/MEK/ERK1/2 pathway. A dose-dependent increase in phosphorylated ERK1/2 was observed in NOTCH1-activated cells. B, TT-NOTCH cells were pretreated with 10 µM U0126 (UO) for 45 min and then with 1 µg/ml doxycycline for 4 days. During the entire treatment period, U0126 was present in the treatment groups. Total cell extracts were isolated and analyzed by Western blotting using antibodies against the phospho-ERK1/2, ERK1/2, ASCL1, and CgA proteins to determine whether Raf-1 pathway activation is required for NE marker reduction. Doxycycline-treated cells showed an increase in phosphorylated ERK1/2, whereas similar treatment in the presence of U0126 showed the absence of phosphorylated ERK1/2, indicating that the Raf-1 pathway was successfully blocked by U0126. As expected, in NOTCH1-activated cells, there was a reduction in NE markers such as ASCL1 and CgA. Interestingly and perhaps surprisingly, inhibition of the ERK1/2 pathway did not increase ASCL1 and CgA to normal levels. ERK1/2 was used as a loading control. C, NOTCH1 signaling silences ASCL1 at the transcriptional level. To determine whether NOTCH1/HES-1 suppresses ASCL1 at the transcription level, we transfected a luciferase reporter plasmid (pGL2) containing an ASCL1 promoter DNA fragment (–7900 to +37) comprising the 5'-flanking region of ASCL1 genomic DNA, including the HES-1-binding site, into TT-NOTCH cells. Luciferase and CMV-beta-galactosidase activities were measured in cell lysates from control and NOTCH1-activated cells. The relative -fold induction of luciferase activity was calculated after normalizing to beta-galactosidase activity for NOTCH1-activated cells in relation to control cells. NOTCH1 activation led to a 30% reduction in luciferase activity compared with control cells, indicating that NOTCH1 signaling silences ASCL1 transcription. Plasmids pGL2-Control and pGL2-Basic served as positive and negative controls for luciferase activity, respectively. D, total mRNA isolated from NOTCH1-activated cells was separated on denaturing 1% formaldehyde-agarose gel. Northern analysis using an ASCL1 probe showed reduction in ASCL1 mRNA after NOTCH1 activation. At day 4, the level of ASCL1 mRNA was significantly reduced. The GAPDH (G3PDH) probe showed an equal amount of mRNA present in all the lanes.

 
NOTCH1 Activates the ERK1/2 Pathway, but Is Not Essential for NE Marker Reduction—We (5) and others (10) have shown that activation of the Raf-1 pathway results in both reduction in NE markers such as ASCL1 and CgA and growth of MTC-TT cells. In addition, NOTCH1 has been shown to activate ERK1/2 in SCLC cells (22). Because we observed similar effects of NE marker reduction and growth inhibition upon NOTCH1 activation in TT cells, we hypothesized that NOTCH1-associated NE marker reduction may be mediated by ERK1/2 activation. As shown in Fig. 7A, activation of NOTCH1 led to phosphorylation of ERK1/2 in a dose-dependent manner. However, when ERK1/2 phosphorylation was blocked by U0126, a potent inhibitor, there was no up-regulation of ASCL1 and CgA in NOTCH1-activated cells (Fig. 7B). There was no change in the levels of total ERK1/2 proteins in U0126-treated cells (Fig. 7B). This suggests that NE marker reduction is independent of ERK1/2 activation in NOTCH1-activated MTC cells.

Activation of NOTCH1 Inhibits ASCL1 Expression—It has been shown previously that activated NOTCH1 degrades both endogenous and exogenous ASCL1 proteins in SCLC cells; however, overexpression of HES-1 does not result in reduction of ASCL1 in SCLC cells (30). In contrast, overexpression of HES-1 in human pulmonary carcinoid cells significantly reduces ASCL1 protein (34). In this study, we found that NOTCH1 activation reduces ASCL1 protein in MTC cells. Therefore, we were interested in the mechanism of ASCL1 reduction by NOTCH1 activation. To determine this, we transiently transfected an ASCL1 promoter-luciferase reporter plasmid as described under "Experimental Procedures." This plasmid has been shown to reduce luciferase activity upon HES-1 expression in SCLC cells (27). The results from the luciferase reporter assay showed that NOTCH1 activation resulted in a 30% decrease in luciferase activity in TT-NOTCH cells transfected with a plasmid containing the ASCL1 promoter (Fig. 7C). This suggests that NOTCH1 signaling regulates ASCL1 transcription. To further confirm this, we performed Northern analysis for the status of the ASCL1 mRNA after various levels of NOTCH1 activation as indicated (Fig. 7D). As shown in Fig. 7D, the level of ASCL1 mRNA was reduced with increasing concentrations of doxycycline. Notably, at day 4, the level of ASCL1 mRNA was significantly reduced. Together, the results from the ASCL1 promoter-luciferase reporter experiment and Northern analysis indicate that the reduction in ASCL1 protein is due predominantly to transcriptional silencing of ASCL1.


Figure 8
View larger version (27K):
[in this window]
[in a new window]

  		FIGURE 8.
Role of ASCL1 in TT cell hormone regulation and growth. A, ASCL1 siRNA was transfected into TT cells, and the cell lysates were analyzed for the level of ASCL1 and CgA proteins to further confirm that ASCL1 regulates CgA. With increasing concentrations of ASCL1 siRNA, there was a progressive reduction in ASCL1 protein. Notably, reduction in ASCL1 resulted in a decrease in the level of CgA. hASH1, human ASH1. B, to determine whether silencing of ASCL1 leads to a change in the level of calcitonin in TT cells, a calcitonin enzyme-linked immunosorbent assay (ELISA) was performed using cell lysates from the control (C), nonspecific siRNA (100 nM)-treated (NS), and ASCL1 siRNA (100 nM)-treated (ASCL1) cells. The results are shown as a relative calcitonin level compared with the control treatment. There was an almost 50% reduction in the calcitonin level in ASCL1 siRNA-treated cells, indicating that ASCL1 also regulates calcitonin levels in TT cells. C, to determine the effect of ASCL1 on CgA, the ASCL1-HA plasmid (1 and 5 µg of DNA) was transfected into TT-NOTCH cells, which were then treated with or without doxycycline (Doxy). Cell extracts were isolated and analyzed by Western blotting for the indicated proteins. As expected, NOTCH1 activation reduced ASCL1 protein levels. The presence of exogenously supplied ASCL1-HA protein was identified using anti-HA antibody. CgA levels were reduced by NOTCH1 activation. However, exogenous ASCL1 protein increased the level of CgA in a dose-dependent manner, suggesting that ASCL1 might up-regulate CgA. ImageQuant software was used to quantify the band intensity of CgA. The band intensity was normalized with GAPDH and the ratio given below the CgA band on the figure. There was an increase in the CgA level from TT-NOTCH to TT-NOTCH cells with the ASCL1 plasmid (1 > 1.2 > 1.5, first, third, and fifth lanes, respectively). D, to determine the role of ASCL1 in cell proliferation of TT cells, a growth rescue experiment by MTT assay was performed as described under "Experimental Procedures." NOTCH1-activated TT-NOTCH cells (D) and TT-NOTCH cells with vector showed similar levels of growth reduction. Notably, overexpression of ASCL1 in NOTCH1-activated TT-NOTCH cells did not reverse NOTCH1-mediated growth inhibition.

 
Effect of ASCL1 on CgA and Calcitonin—Activation of either the NOTCH1 or Raf-1 pathway results in ASCL1 reduction and concomitant reduction in CgA and calcitonin, suggesting a close correlation between ASCL1 and CgA/calcitonin levels. The finding that mAscl1 transgenic knock-out mice fail to develop thyroid C-cells suggests that ASCL1 is important for normal C-cell development (9). Similarly, blocking ASCL1 by antisense treatment in SCLC cells results in reduction of NE markers (33). However, there are no studies to date that illustrate a direct correlation between the levels of ASCL1, CgA, and other hormones. Our interest in the regulation of ASCL1 and NE hormones led us to explore the correlation between these two markers. To identify the relationship between ASCL1 and NE hormones, we transfected plasmid pcDNA3.0 containing an HA epitope tag fused to the ASCL1 gene into TT-NOTCH cells. Western analysis of cell lysates from ASCL1-overexpressing TT-NOTCH cells without doxycycline treatment showed an overall increase in the level of CgA compared with no ASCL1 expression (Fig. 8C, compare the first lane with the third and fifth lanes), suggesting that CgA expression is highly dependent upon ASCL1 levels. However, doxycycline-treated ASCL1-overexpressing TT-NOTCH cells (fourth and sixth lanes) showed reduced CgA levels compared with control cells (third and fifth lanes). Interestingly, NOTCH1-activated TT-NOTCH cell lysates containing exogenous ASCL1 protein did not show any reduction in exogenous ASCL1 as indicated by the anti-HA antibody (Fig. 8C). In contrast to this, Sriuranpong et al. (30) reported that transient NOTCH1 expression using an adenoviral vector leads to reduction in exogenous ASCL1 from the ASCL1-HA vector, suggesting that NOTCH1 degrades ASCL1. However, this confirms our above results (Fig. 7, C and D) that ASCL1 reduction by NOTCH1 activation in MTC cells is predominantly at the transcriptional level. Furthermore, siRNA against ASCL1 resulted in CgA reduction (Fig. 8A). As shown in Fig. 8A, nonspecific siRNA did not have any effect on ASCL1 or CgA protein. However, the decrease in the level of ASCL1 protein with increasing concentrations of ASCL1 siRNA resulted in a decrease in CgA in a dose-dependent manner. Furthermore, to determine the level of calcitonin after ASCL1 depletion, we carried out a calcitonin enzyme-linked immunosorbent assay using the ASCL1-depleted cell lysates from the siRNA treatment. Fig. 8B shows an ~50% reduction in the relative calcitonin level in the ASCL1 siRNA-treated group compared with both the nonspecific siRNA-treated and control groups. In conclusion, our results indicate, for the first time, that levels of NE hormones such as CgA and calcitonin are directly related to ASCL1 levels.

ASCL1 Overexpression in NOTCH1-activated TT Cells Is Unable to Rescue Growth—Activation of the NOTCH1 pathway in TT cells leads to significant growth suppression and also a marked reduction in NE marker production. We have clearly shown that ASCL1 regulates hormone production. However, it is not known whether ASCL1 is also involved in growth regulation by NOTCH1 signaling activation. Therefore, to determine whether growth suppression by NOTCH1 activation is mediated by ASCL1, we carried out the MTT cell proliferation assay in ASCL1-overexpressing TT-NOTCH cells with or without activation of NOTCH1 signaling by doxycycline treatment. Interestingly, transient expression of ASCL1 did not increase growth of NOTCH1-activated cells (Fig. 8D). Notably, NOTCH1-activated TT-NOTCH cells showed a similar reduction in growth in the presence and absence of exogenous expression of ASCL1. We have shown above that the expression of ASCL1 protein from this plasmid is functional as indicated by the increase in CgA. Therefore, the results of this experiment indicate that growth suppression by activation of NOTCH1 signaling in TT cells is independent of ASCL1 reduction or that growth is not regulated by ASCL1.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
NOTCH1 is a multifunctional transmembrane receptor that plays important roles in cell differentiation, development, proliferation, and survival (14, 15, 35). In Drosophila neural development, the most studied NOTCH1 signaling pathway, NOTCH1 maintains the neural progenitor stage and inhibits differentiation. Transgenic mice lacking the NOTCH1 ligand Dll1 (delta-like gene-1) or the intracellular mediator RBP-J{kappa} show accelerated pancreatic endocrine differentiation with a specific increase in endocrine cells. As a result of this premature differentiation, the development of the pancreas is arrested because of the reduction in precursor cells. These findings clearly demonstrate that NOTCH1 signaling is required for the normal development of the pancreas (19). Similar to the results of this study, Hes-1 knock-out mice also display severe pancreatic hypoplasia with an increase in endocrine cells (36). In a different study using a Mash1 knock-out transgenic mouse model, mice died at birth because of the lack of development of thyroid C-cells (9). Taken together, these results suggest that components of the NOTCH1 signaling pathway tightly regulate the NE phenotype in the developing lung, thyroid, and gastrointestinal tract. However, the role of NOTCH1 in cancer cells remains controversial yet interesting. Recent studies on NOTCH1 signaling in cancer biology have contributed to our understanding that NOTCH1 signaling can act either as a tumor suppressor (37, 38) or as a tumor promoter (39), suggesting that the effects of NOTCH1 signaling are cell context-specific.

Transient expression of active NOTCH1 in SCLC, pancreatic carcinoid, and prostate cancer cells inhibits cell growth in vitro. We have shown recently that stable expression of estradiol-inducible active NOTCH1 (NICD) in pancreatic carcinoid BON cells leads to growth inhibition and reduction in NE hormone production (20). However, the role of NOTCH1 signaling in MTC cells is not known. MTC derives from calcitonin-producing thyroid C-cells and produces excess amounts of calcitonin. In addition, MTC cells express high levels of ASCL1, which seems to be limited to NE tumors (5, 7, 8). Various studies using Mash1 transgenic knock-out mice revealed that ASCL1 plays an important role in the development of adrenal chromaffin cells and pulmonary endocrine cells (9, 11–13). Several reports have shown that ASCL1 levels can be reduced by NOTCH1-mediated HES-1 protein activation (8, 20, 22, 27). Despite the importance of the NOTCH1 signaling pathway in cell fate determination, the role of the NOTCH1 signaling pathway in MTC has, until now, not been described.

In this study, we have shown that active NOTCH1 (NICD) is absent in MTC tumor cells and in human MTC tumor samples. However, these MTC tumors and cell lines have high levels of ASCL1 and CgA. These findings are consistent with characteristics of other NE tumors. Active NOTCH1 dose-response induction by doxycycline led to an increase in functional NOTCH1 protein production as measured by CBF1 binding studies, resulting in activation of the NOTCH1 pathway. Furthermore, continuous NOTCH1 activation in TT-NOTCH cells inhibited tumor cell growth. Notably, this growth reduction was dependent on the levels of NOTCH1 protein present. Our findings provide the first documentation of the role of NOTCH1 signaling as a tumor suppressor in MTC cells. We have also demonstrated that activation of NOTCH1 signaling in MTC cells reduces the levels of calcitonin and CgA. For the first time, we have shown in this study that ASCL1 tightly regulates CgA and calcitonin expression and that ASCL1 is possibly not involved in growth regulation.

However, the mechanism by which NOTCH1 causes hormone reductions is not clear. raf-1 activation in MTC cells also leads to significant reductions in CgA, calcitonin, and growth (1, 5, 10, 40). NOTCH1 activates other signaling pathways such as the JAK (Janus kinase)-STAT (signal transducer and activator of transcription) pathway (41) or the Raf-1-MEK-ERK1/2 pathway (5), which regulates NE hormone production in these tumors. However, the results from this study indicate that the reduction in NE hormone production is independent of ERK1/2 activation. However, we predicted that ASCL1 might regulate the expression of CgA and other hormones in MTC cells considering that ASCL1 is a basic helix-loop-helix transcription factor and that exogenous expression of ASCL1 resulted in an increase in the level of CgA. Accordingly, loss of ASCL1 protein upon ASCL1 siRNA transfection in MTC cells resulted in a reduction in CgA as well as calcitonin. This strongly indicates that the levels of CgA and calcitonin are regulated by ASCL1 protein. Interestingly, the fact that overexpression of ASCL1 in TT-NOTCH cells led to an increase in CgA levels further confirms the involvement of ASCL1 in the regulation of CgA. However, overexpression of ASCL1 did not increase cell proliferation in NOTCH1-activated TT cells, indicating that NOTCH1-mediated growth suppression is independent of ASCL1 levels.

Although NOTCH1 signaling can act either as an oncogene or an antiproliferator, the activation of NOTCH1 signaling in NE tumor cells leads to growth inhibition. Pancreatic adenocarcinoma cells express high levels of NOTCH1, and NOTCH1 is required for tumor growth (16, 19, 42). However, MTC and other NE tumor cells lack functional NOTCH1 (NICD) protein, and activation of NOTCH1 in these cells leads to significant growth reduction. Although the growth-suppressing effects of NOTCH1 activation have been well characterized in other tumor types, to our knowledge, this is the first report of inhibition of MTC tumor cell growth by NOTCH1 signaling. Moreover, the findings of this study suggest that the strong inhibition of MTC cell growth by NOTCH1 signaling may be due to alterations in cell cycle regulators, causing G1/S arrest. p21 is a universal inhibitor of cyclin-dependent kinases, and its expression is normally regulated by the p53 tumor suppressor protein. The observations of up-regulation of p21, phospho-Cdc2, and cyclin D1 suggest that growth inhibition of MTC cells upon NOTCH1 expression could be due to cell cycle arrest.

In conclusion, this study illustrates that NOTCH1 inhibits growth and hormone production in MTC cells. Furthermore, the results further emphasize the critically important role of ASCL1-dependent hormone regulation and ASCL1-independent growth regulation by the NOTCH1 signaling pathway in MTC tumors. Further research based on our findings may lead to the development of novel therapies for patients with MTC and possibly for other NE tumors based on activation of the NOTCH1 signaling pathway.


   FOOTNOTES
 
* This work was supported in part by an American Cancer Society research scholars grant, National Institutes of Health Grants DK063015, DK064735, DK066169, and CA109053, the George H. A. Clowes, Jr., Memorial Research Career Development Award of the American College of Surgeons, and a Carcinoid Cancer Foundation research award (to H. C.) and by a University of Wisconsin Medical School grant (to M. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence should be addressed: Dept. of Surgery, University of Wisconsin, H4/750 Clinical Science Center, 600 Highland Ave., Madison, WI 53792. Tel.: 608-263-1387; Fax: 608-263-7652; E-mail: chen{at}surgery.wisc.edu.

2 The abbreviations used are: MTC, medullary thyroid cancer; NE, neuroendocrine; CgA, chromogranin A; SCLC, small cell lung cancer; NICD, NOTCH1 intracellular domain; siRNA, small interfering RNA; ERK1/2, extracellular signal-regulated kinase-1/2; HA, hemagglutinin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; CMV, cytomegalovirus; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; MEK, mitogen-activated protein kinase/extracellular signal-regulated kinase kinase. Back


   ACKNOWLEDGMENTS
 
We thank Dr. David Yu Greenblatt for helpful discussions and critical reading of the manuscript; Dr. Barry D. Nelkin for the TT cells; Dr. Douglas W. Ball for the NOTCH1 and ASCL1-HA plasmids and pTAN1; Dr. Diane Hayward for the CBF1 reporter constructs; and Dr. Tetsuo Sudo for the generous gift of anti-HES-1 antibody. We thank Amber Shada and Yi-Wei Zhang for technical assistance.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Chen, H., Carson-Walter, E. B., Baylin, S. B., Nelkin, B. D., and Ball, D. W. (1996) Surgery 120, 168–172[CrossRef][Medline] [Order article via Infotrieve]
  2. Chen, H., Roberts, J. R., Ball, D. W., Eisele, D. W., Baylin, S. B., Udelsman, R., and Bulkley, G. B. (1998) Ann. Surg. 227, 887–895[CrossRef][Medline] [Order article via Infotrieve]
  3. Moley, J. F., Lairmore, T. C., and Phay, J. E. (1999) Curr. Probl. Surg. 36, 653–762[CrossRef][Medline] [Order article via Infotrieve]
  4. Chen, H., Kunnimalaiyaan, M., and Van Gompel, J. J. (2005) Thyroid 15, 511–521[CrossRef][Medline] [Order article via Infotrieve]
  5. Sippel, R. S., Carpenter, J. E., Kunnimalaiyaan, M., and Chen, H. (2003) Surgery 134, 866–871[CrossRef][Medline] [Order article via Infotrieve]
  6. Chen, H., Biel, M. A., Borges, M. W., Thiagalingam, A., Nelkin, B. D., Baylin, S. B., and Ball, D. W. (1997) Cell Growth & Differ. 8, 677–686[Abstract]
  7. Chen, H., Udelsman, R., Zeiger, M. A., and Ball, D. A. (1997) Oncol. Rep. 4, 775–778
  8. Nakakura, E. K., Sriuranpong, V. R., Kunnimalaiyaan, M., Hsiao, E. C., Schuebel, K. E., Borges, M. W., Jin, N., Collins, B. J., Nelkin, B. D., Chen, H., and Ball, D. W. (2005) J. Clin. Endocrinol. Metab. 90, 4350–4356[Abstract/Free Full Text]
  9. Lanigan, T. M., DeRaad, S. K., and Russo, A. F. (1998) J. Neurobiol. 34, 126–134[CrossRef][Medline] [Order article via Infotrieve]
  10. Park, J. I., Strock, C. J., Ball, D. W., and Nelkin, B. D. (2003) Mol. Cell. Biol. 23, 543–554[Abstract/Free Full Text]
  11. Guillemot, F., Nagy, A., Auerbach, A., Rossant, J., and Joyner, A. L. (1994) Nature 371, 333–336[CrossRef][Medline] [Order article via Infotrieve]
  12. Lo, L., Guillemot, F., Joyner, A. L., and Anderson, D. J. (1994) Perspect. Dev. Neurobiol. 2, 191–201[Medline] [Order article via Infotrieve]
  13. Guillemot, F., and Joyner, A. L. (1993) Mech. Dev. 42, 171–185[CrossRef][Medline] [Order article via Infotrieve]
  14. Maillard, I., and Pear, W. S. (2003) Cancer Cell 3, 203–205[CrossRef][Medline] [Order article via Infotrieve]
  15. Yoon, K., and Gaiano, N. (2005) Nat. Neurosci. 8, 709–715[CrossRef][Medline] [Order article via Infotrieve]
  16. Hald, J., Hjorth, J. P., German, M. S., Madsen, O. D., Serup, P., and Jensen, J. (2003) Dev. Biol. 260, 426–437[CrossRef][Medline] [Order article via Infotrieve]
  17. Jensen, J., Heller, R. S., Funder-Nielsen, T., Pedersen, E. E., Lindsell, C., Weinmaster, G., Madsen, O. D., and Serup, P. (2000) Diabetes 49, 163–176[Abstract]
  18. Murtaugh, L. C., Stanger, B. Z., Kwan, K. M., and Melton, D. A. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 14920–14925[Abstract/Free Full Text]
  19. Apelqvist, A., Li, H., Sommer, L., Beatus, P., Anderson, D. J., Honjo, T., Hrabe, D. A., Lendahl, U., and Edlund, H. (1999) Nature 400, 877–881[CrossRef][Medline] [Order article via Infotrieve]
  20. Kunnimalaiyaan, M., Traeger, K., and Chen, H. (2005) Am. J. Physiol. 289, G636–G642
  21. Shou, J., Ross, S., Koeppen, H., de Sauvage, F. J., and Gao, W. Q. (2001) Cancer Res. 61, 7291–7297[Abstract/Free Full Text]
  22. Sriuranpong, V., Borges, M. W., Ravi, R. K., Arnold, D. R., Nelkin, B. D., Baylin, S. B., and Ball, D. W. (2001) Cancer Res. 61, 3200–3205[Abstract/Free Full Text]
  23. Morimura, T., Goitsuka, R., Zhang, Y., Saito, I., Reth, M., and Kitamura, D. (2000) J. Biol. Chem. 275, 36523–36531[Abstract/Free Full Text]
  24. Sippel, R. S., Carpenter, J. E., Kunnimalaiyaan, M., Lagerholm, S., and Chen, H. (2003) Am. J. Physiol. 285, G245–G254
  25. Hsieh, J. J., Henkel, T., Salmon, P., Robey, E., Peterson, M. G., and Hayward, S. D. (1996) Mol. Cell. Biol. 16, 952–959[Abstract]
  26. Van Gompel, J. J., Kunnimalaiyaan, M., Holen, K., and Chen, H. (2005) Mol. Cancer Ther. 4, 910–917[Abstract/Free Full Text]
  27. Chen, H., Thiagalingam, A., Chopra, H., Borges, M. W., Feder, J. N., Nelkin, B. D., Baylin, S. B., and Ball, D. W. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 5355–5360[Abstract/Free Full Text]
  28. Ronchini, C., and Capobianco, A. J. (2000) Oncogene 19, 3914–3924[CrossRef][Medline] [Order article via Infotrieve]
  29. Kuroda, K., Tani, S., Tamura, K., Minoguchi, S., Kurooka, H., and Honjo, T. (1999) J. Biol. Chem. 274, 7238–7244[Abstract/Free Full Text]
  30. Sriuranpong, V., Borges, M. W., Strock, C. L., Nakakura, E. K., Watkins, D. N., Blaumueller, C. M., Nelkin, B. D., and Ball, D. W. (2002) Mol. Cell. Biol. 22, 3129–3139[Abstract/Free Full Text]
  31. Bae, S., Bessho, Y., Hojo, M., and Kageyama, R. (2000) Development (Camb.) 127, 2933–2943[Abstract]
  32. Joseph, J., Mudduluru, G., Antony, S., Vashistha, S., Ajitkumar, P., and Somasundaram, K. (2004) Oncogene 23, 6304–6315[CrossRef][Medline] [Order article via Infotrieve]
  33. Borges, M., Linnoila, R. I., van de Velde, H. J., Chen, H., Nelkin, B. D., Mabry, M., Baylin, S. B., and Ball, D. W. (1997) Nature 386, 852–855[CrossRef][Medline] [Order article via Infotrieve]
  34. Kunnimalaiyaan, M., Yan, S., Wong, F., Zhang, Y. W., and Chen, H. (2005) Surgery 138, 1137–1142[CrossRef][Medline] [Order article via Infotrieve]
  35. Kadesch, T. (2004) Curr. Opin. Genet. Dev. 14, 506–512[CrossRef][Medline] [Order article via Infotrieve]
  36. Jensen, J., Pedersen, E. E., Galante, P., Hald, J., Heller, R. S., Ishibashi, M., Kageyama, R., Guillemot, F., Serup, P., and Madsen, O. D. (2000) Nat. Genet. 24, 36–44[CrossRef][Medline] [Order article via Infotrieve]
  37. Nicolas, M., Wolfer, A., Raj, K., Kummer, J. A., Mill, P., van Noort, M., Hui, C. C., Clevers, H., Dotto, G. P., and Radtke, F. (2003) Nat. Genet. 33, 416–421[CrossRef][Medline] [Order article via Infotrieve]
  38. Rangarajan, A., Talora, C., Okuyama, R., Nicolas, M., Mammucari, C., Oh, H., Aster, J. C., Krishna, S., Metzger, D., Chambon, P., Miele, L., Aguet, M., Radtke, F., and Dotto, G. P. (2001) EMBO J. 20, 3427–3436[CrossRef][Medline] [Order article via Infotrieve]
  39. Oishi, K., Kamakura, S., Isazawa, Y., Yoshimatsu, T., Kuida, K., Nakafuku, M., Masuyama, N., and Gotoh, Y. (2004) Dev. Biol. 276, 172–184[CrossRef][Medline] [Order article via Infotrieve]
  40. Kunnimalaiyaan, M., and Chen, H. (2006) Anti-cancer Drugs 17, 139–142[CrossRef][Medline] [Order article via Infotrieve]
  41. Kamakura, S., Oishi, K., Yoshimatsu, T., Nakafuku, M., Masuyama, N., and Gotoh, Y. (2004) Nat. Cell Biol. 6, 547–554[CrossRef][Medline] [Order article via Infotrieve]
  42. Miyamoto, Y., Maitra, A., Ghosh, B., Zechner, U., Argani, P., IacobuzioDonahue, C. A., Sriuranpong, V., Iso, T., Meszoely, I. M., Wolfe, M. S., Hruban, R. H., Ball, D. W., Schmid, R. M., and Leach, S. D. (2003) Cancer Cell 3, 565–576[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike