Neurotensin, a novel target of Wnt/β-catenin pathway, promotes growth of neuroendocrine tumor cells

Abstract

Wnt/β-catenin signaling plays a pivotal role in regulating cell growth and differentiation by activation of the β-catenin/T-cell factor (TCF) complex and subsequent regulation of a set of target genes that have one or more TCF-binding elements (TBEs). Hyperactivation of this pathway has been implicated in numerous malignancies including human neuroendocrine tumors (NETs). Neurotensin (NT), an intestinal hormone, induces proliferation of several gastrointestinal (GI) cancers including cancers of the pancreas and colon. Here, we analyzed the human NT promoter in silico and found at least four consensus TBEs within the proximal promoter region. Using a combination of ChIP and luciferase reporter assays, we identified one TBE (located approximately 900 bp proximal from the transcription start site) that was immunoprecipitated efficiently by TCF4-targeting antibody; mutation of this site attenuated the responsiveness to β-catenin. We also confirmed that the promoter activity and the mRNA and protein expression levels of NT were increased by various Wnt pathway activators and decreased by Wnt inhibitors in NET cell lines BON and QGP-1, which express and secrete NT. Similarly, the intracellular content and secretion of NT were induced by Wnt3a in these cells. Finally, inhibition of NT signaling suppressed cell proliferation and anchorage-independent growth and decreased expression levels of growth-related proteins in NET cells. Our results indicate that NT is a direct target of the Wnt/β-catenin pathway and may be a mediator for NET cell growth.

Introduction

Wnt/β-catenin signaling, an important pathway in the development and progression of numerous cancers, leads to stabilization and nuclear translocation of β-catenin, which activates T-cell factor/lymphoid enhancer-binding factor (TCF/LEF)-dependent transcription of its target genes.^{1, 2} A number of these target genes, which have one or more TCF-binding elements (TBEs) in their promoter region,^{3, 4} play crucial roles in neoplastic transformation and cancer progression.^{1, 2} In particular, aberrant activation of Wnt/β-catenin signaling by genetic mutations or epigenetic silencing of the oncogenes or tumor suppressor genes in this pathway has been closely associated with a variety of cancers including neuroendocrine tumors (NETs).^{5, 6} Several growth factors and gastrointestinal (GI) hormones (e.g., gastrin), which are downstream targets of this pathway, contribute to early tumor progression.^{1, 7, 8}

Neurotensin (NT) which is predominantly localized to endocrine cells of the small bowel,⁹ has multiple physiological functions in the GI tract, including stimulation of pancreatic and biliary secretion and growth stimulation of various normal tissues.^{9, 10} In addition to its cellular effects, NT acting through predominantly its high-affinity receptor (NTR1), induces the proliferation of several types of cancers including pancreas and colon.^{9, 11} Previously, our laboratory identified several mechanisms for control of NT gene expression (e.g., regulation of Ras and mTORC1 or DNA methylation)^12-14 and delineated intracellular mechanisms contributing to NT secretion.^{14, 15} Moreover, it was reported that NTR1 expression is regulated by Wnt/β-catenin signaling through a functional TBE and correlates with abnormal localization of β-catenin in colorectal cancers.¹⁶

In the present study, we identified a functional TBE within the human NT promoter region. We also confirmed that the expression and release of NT are directly regulated by the Wnt/β-catenin pathway in NET cells. Moreover, we showed that knockdown of NT or treatment with SR-48692, an NTR1 antagonist,¹⁷ represses NET cell proliferation, anchorage-independent growth and the expression of growth-related proteins. Together, these findings identify a novel role for the Wnt/β-catenin pathway in the regulation of NT expression and secretion.

Materials and Methods

Materials

The materials utilized in this study are described in Supplementary Materials.

Cell culture

Human NET cell lines BON and QGP-1 were maintained in DMEM and F12K in a 1:1 ratio supplemented with 5% FBS and in RPMI 1640 medium with 10% FBS, respectively. The cells were authenticated in May 2012 at Genetica DNA Laboratories (Cincinnati, OH) profiled with 17 autosomal short tandem repeat (STR) loci and the sex identity locus.

Chromatin Immunoprecipitation (ChIP) analysis

ChIP analysis was performed per the manufacturer's protocol (Millipore, Bedford, MA). Purified DNA from BON cells was amplified using the primers for potential TBEs 1-4 in the NT promoter region: TBE 1 forward (F), 5'-GAATTTCCATTAATTCTTCTC-3', and TBE 1 reverse (R), 5'-GGAAAATTATATATACTTTGC-3'; TBE 2 F, 5'-GCAATTCAAAAGCAGAGAAAAC-3', and TBE 2 R, 5'-AGCAATGGAAGCTTGAAACAC-3'; TBE 3 F, 5'-GGATTGTCTCCTTTCCAAAAG-3', and TBE 3 R, 5'-GATGACTGAACTATGTGTGCT-3'; TBE 4 F, 5'-ATGGAGGTGAAGATAGGGCAC-3', and TBE 4 R, 5'-GAGCACAGACTCCAGGAGCTG-3'. The PCR products were visualized by 2% agarose gel.

NT promoter constructs and mutagenesis

The NT promoter fragment (−2200/+100) was PCR amplified from genomic DNA isolated from BON cells using primers: NT promoter F, 5'-GCGAGCTCTAGCTTGAAGGCATTAGATTAG-3', and NT promoter R, 5'-CGCCCGGGCAGCCTTCTAACAAGCCAAGTC-3', and then cloned into the pXP1 Luciferase reporter plasmid (ATCC, Manassas, VA). Site-directed mutagenesis of TCF-binding sequences was performed by standard PCR techniques using Platinum Pfx DNA Polymerase (Invitrogen, Carlsbad, CA). All wild type and mutant promoter constructs were confirmed by sequencing.

Luciferase reporter assays

Cells were plated in 24 well plates and transiently transfected with the NT reporter or TopFlash (0.4 μg) and the Renilla reporter (0.05 μg) with or without pcDNA3.1 vectors containing Wnt/β-catenin pathway regulatory genes using Lipofectamine 2000 (Invitrogen). For Wnt3a or iCRT3 treatment, varying concentrations of the Wnt regulators were added to NET cells one day after plating. The cells were harvested and luciferase activity was measured two days after transfection.

RNA isolation, reverse transcription-polymerase chain reaction (RT-PCR) and quantitative RT-PCR (qRT-PCR) analysis

Total RNA was isolated using RNeasy kits (Qiagen, Valencia, CA) according to the manufacturer's instructions. RT-PCR analysis of NT and β-actin expression was performed using cDNA synthesized from 1 μg of total RNA, and the primers: NT F, 5'-GATGATGGCAGGAATGAAAATCCAG-3', and NT R, 5'-GTTGAAAAGCCCTGCTGTGACAGA-3'; β-actin F, 5'-TCACCAACTGGGACGACATG-3', and β-actin R, 5'-ACCGGAGTCCATCACGATG-3'. The PCR products were analyzed on a 2% agarose gel. Quantitative real-time PCR (qRT-PCR) was carried out using a TaqMan Gene Expression Master Mix (#4369016), and TaqMan probes for human NT (ID Hs00900055_m1) and human 18SrRNA (# 4333760F) according to manufacturer's protocol (Applied Biosystems, Austin, TX).

Western blot, cell proliferation and soft agar assays

Western blot, cell proliferation and soft agar assays were performed as described previously.⁶

NT enzyme immunoassay (EIA)

Cells were plated in 24 well plates at a density of 1×10⁵ cells/cm² and grown for 24 h. Cells were treated with Wnt3a in growth medium for an additional 24 h. Intracellular NT content and secreted NT peptide were measured by NT EIA kit (Phoenix Pharmaceuticals, Belmont, CA) using 50 μg of the lysates and the media as previously described.^{14, 15}

Statistical analysis

Bar and line graphs were generated to represent mean ± SD for each cell culture condition. Comparisons of quantitative measurements for luciferase activity, mRNA levels, intracellular content and secretion of NT, and number of cells and colony formation were performed using two-sample t-test or analysis of variance with contrast statements for pairwise testing or test for linear trend over time points or dose levels. p-values < 0.05 were considered statistically significant.

Results

NT is a direct target of Wnt/β-catenin signaling

Activation of Wnt target genes requires the presence of one or more TBEs in the promoter of the genes.^{3, 4} Here, we analyzed the human NT promoter in silico and found at least four consensus TBEs (5’-CTTTGNN-3’, where N represents A or T) within the promoter regions (Fig. 1A). To identify functional TBEs, we performed ChIP assays using an anti-TCF4 antibody, with IgG as a negative control. TBE 3 was immunoprecipitated efficiently, while other sites were not (Fig. 1B). To further examine whether NT is a target gene of the Wnt/β-catenin pathway, we next cloned the NT promoter (−2200/+100) into the pXP1 vector, and generated mutant constructs deleting the individual TBEs by site-directed mutagenesis (Fig. 1C). Co-expression of β-catenin-S33Y, an oncogenic stable mutant of β-catenin,¹⁸ induced activity of the wild-type NT promoter-driven luciferase reporter (Fig. 1C). Mutations of TBE 1, 2 or 4 sites did not affect the constitutively activated (CA)-β-catenin-mediated activation of the reporter gene; however, TBE 3 significantly decreased CA-β-catenin-induced NT promoter activation. As controls, cells were transfected with either the NT minimal promoter (−42/+23) or the empty pXP1 vector as described previously¹⁴; no induction of NT promoter activity was noted using either of these vectors. Taken together, these results suggest that the TBE 3 site of the NT promoter acts as a functional TBE.

Figure 1. Identification of functional TBEs in the NT promoter.

(A) Schematic representation of four potential TBEs in the NT proximal promoter. (B) ChIP analyses were performed using a TCF4-targeting antibody or IgG as a negative control. The recruitment of TCF4 to the NT promoter was quantified by gel-based PCR. PCR products were visualized by 2% agarose gel. (C) Diagram of the wild type and mutant promoter constructs, and their luciferase activities in BON cells transfected with the promoter luciferase reporter constructs together with Renilla luciferase reporter, pRL-TK, and β-catenin-S33Y expression constructs (*, P < 0.05 vs. pcDNA alone). Mutations in potential TBEs: derivatives of the wild-type NT promoter containing mutations in potential TBEs (5’-CTTTG[A/T][A/T]-3’ to 5’-CTTTGGC-3’) were constructed by standard PCR techniques. (D) NT-luciferase activity is induced by Wnt3a (left) and inhibited by iCRT3 treatment (middle) or dominant-negative (DN) mutant TCF4 cotransfection (right) in BON cells (*, P < 0.05 vs. vehicle or 0 μg DN-TCF4). (E) Cotransfection with CA-β-catenin (left) increased the activity of the NT promoter, whereas cotransfection with DN-TCF4 (right) inhibited NT induction in QGP-1 cells (*, P < 0.05 vs. 0 μg CA-β-catenin or DN-TCF4).

We further examined whether the NT promoter responds to positive or negative regulators of Wnt/β-catenin signaling. First, to identify an effective Wnt inhibitor, several compounds (e.g., XAV939, FH535, Pyrvinium tosylate, and iCRT3) were tested for their effects on TOPFlash activity and Cyclin D1 expression in NET cells (data not shown). The selective Wnt inhibitor iCRT3¹⁹ was the most potent and significantly decreased TOPFlash activity, Cyclin D1 expression and cell proliferation in a dose-dependent manner in BON and QGP-1 cells (Supplementary Fig. 1). To quantitatively measure the responsiveness of the NT promoter, the reporter plasmid was transfected into NET cells and treated with Wnt3a or iCRT3. The NT promoter was activated by Wnt3a (Fig. 1D, left) but repressed by treatment with iCRT3 (Fig. 1D, middle) or by cotransfection with dominant-negative (DN) mutant TCF4²⁰ (Fig. 1D, right) in a dose-dependent manner. We also cotransfected CA-β-catenin or DN-TCF4 in combination with the NT promoter into QGP-1 cells to confirm the NT promoter regulation and found that CA-β-catenin enhanced luciferase activity (Fig. 1E, left), while DN-TCF4 inhibited the induction of luciferase activity in these cells (Fig. 1E, right). These results suggest that the NT promoter contains a functional TBE and is regulated by Wnt/β-catenin signaling in NET cells.

NT expression is regulated by Wnt/β-catenin signaling

To examine the time course for NT induction, mRNA levels were assayed over time, following treatment of BON cells by LiCl, which activates Wnt/β-catenin signaling (Fig. 2A). Induction of NT was detectable within 2 h after LiCl treatment with maximal expression noted at 4 h after treatment (Fig. 2A). The rapid kinetics of NT induction is consistent with the notion that NT is a direct downstream target of Wnt/β-catenin signaling. We also confirmed that NT mRNA expression was increased by treatment with Wnt3a in BON (Fig. 2B, upper) and QGP-1 cells (Fig. 2B, bottom). Furthermore, expression of the NT mRNA was decreased by treatment of iCRT3 (Fig. 2B, middle).

Figure 2. Regulation of mRNA and protein expression, and intracellular content and peptide secretion of NT by Wnt/β-catenin signaling in NET cells.

(A) Time course of upregulation of NT mRNA by LiCl. Total RNA was isolated from BON cells treated with 25 mM LiCl for the indicated times and cDNA was synthesized from 1 μg of total RNA and a high capacity cDNA reverse transcription kit (Applied Biosystems). The PCR products were visualized by 2% agarose gel (upper). β-actin was used as an internal control. Quantitative RTPCR analysis also confirmed that treatment with LiCl induced NT mRNA in BON cells (bottom). The reaction was performed using a TaqMan Gene Expression Master Mix and TaqMan probes for human NT and 18S rRNA (Applied Biosystems). Expression levels were assessed by evaluating threshold cycle (Ct) values. The relative amount of mRNA expression was calculated by the comparative ΔΔCt method (*, P < 0.05 vs. 0 h). (B) The levels of NT mRNA were checked in BON cells treated with 200 nM Wnt3a for 2 h (upper) or 5 mM iCRT3 for 24 h (middle), and in QGP-1 cells treated with 200 nM Wnt3a for 2 h (bottom). β-actin was used as an internal control. (C) The protein extracts for cell lysates were analyzed with the indicated antibodies. Western blot analysis showing induction of NT and Cyclin D1 by Wnt3a treatment in BON (upper left) and QGP-1 (upper right) cells. Western blot analysis representing suppression of NT and Cyclin D1 protein expression by 25 μM and 50 μM iCRT3 treatment in BON (bottom left) and QGP-1 (bottom right) cells, respectively. β-actin was used as a loading control. (D) NET cells were treated with vehicle or Wnt3a (200 nM) for 24 h; BON (left) and QGP-1 (right) cells were lysed and protein (50 μg) was used for NT EIA (Phoenix Pharmaceuticals) to measure intracellular NT content (*, P < 0.05 vs. vehicle). (E) The parallel medium for BON (left) and QGP-1 (right) was collected and NT peptide secretion measured by NT EIA (*, P < 0.05 vs. vehicle).

We next determined whether Wnt/β-catenin signaling regulates NT protein expression by Western blot analysis. Activation of Wnt/β-catenin signaling by Wnt3a increased the expression of full length NT protein, a precursor form of NT, and Cyclin D1, the known target of the Wnt/β-catenin pathway in BON and QGP-1 cells (Fig. 2C, upper). In contrast, iCRT3 decreased the protein levels of NT and Cyclin D1 in BON and QGP-1 cells (Fig. 2C, bottom). Taken together, these results demonstrate that mRNA and protein expression of NT are regulated by Wnt/β-catenin signaling in NET cells. Additionally, we quantitated intracellular NT content and NT peptide secretion in NET cells after treatment with Wnt3a. Intracellular NT content was increased in BON and QGP-1 cells treated with Wnt3a (Fig. 2D). Similarly, Wnt3a treatment also increased NT peptide release from BON and QGP-1 cells (Fig. 2E).

NT inhibition decreases NET cell growth

Next, we used small interfering RNA (siRNA) against NT to delineate the growth effect of NT in NET cells with aberrant Wnt/β-catenin signaling. Compared with cells transfected with non-targeting control, siRNA-mediated knockdown of NT expression suppressed the proliferation of BON and QGP-1 cells (Fig. 3A). In addition, NT knockdown significantly inhibited the expression of the growth-related protein Cyclin D1, which was independent of the Wnt/β-catenin pathway (Fig. 3B). Furthermore, we performed soft agar assays to determine the effect of NT knockdown on anchorage-independent growth. The number of colonies in QGP-1 cells transfected with NT siRNA was significantly lower than those of control cells (Fig. 3C). We also delineated the effects of SR-48692, an NTR1 antagonist, on proliferation of BON cells that express NTR1 mRNA. SR-48692 inhibited the growth of BON cells in a dose-dependent manner (Fig. 3D) and decreased expression of Cyclin D1 and c-Myc (Fig. 3E). Moreover, the number of colonies in BON cells treated with SR-48692 was significantly lower than those of control cells (Fig. 3F). Similarly, the inhibition of cell growth and suppression of anchorage-independent growth were also noted in QGP-1 cells expressing NT small hairpin RNA (shRNA) compared with non-targeting control cells (Supplementary Fig. 2). Collectively, these results demonstrate that knockdown of NT or treatment with SR-48692 (in NTR1 expressing cells) suppresses cell proliferation, growth-related protein expression and anchorage-independent growth, suggesting a growth-promoting function for NT in NET cells.

Figure 3. Inhibition of NT signaling affects NET cell growth.

(A) Equal numbers of BON (left) and QGP-1 (right) cells transfected with siRNA against non-targeting control or NT were plated in 24 well plates. One day after seeding, serum free media were changed and the cell numbers were counted after an additional 48 h incubation (*, P < 0.05 vs. control siRNA). (B) Western blot analysis showing expression of Cyclin D1 and NT in NET cells transfected with control or NT siRNA; β-actin was used as a loading control. (C) The number of colonies compared with the control siRNA in soft agar assay. Colony formation of representative control or NT knockdown QGP-1 cells was assessed over a period of 4 weeks (*, P < 0.05 vs. control siRNA). The colonies were stained with crystal violet solution and quantified using AlphaEaseFC™ software (Alpha Innotech Corporation, San Leandro, CA). (D) BON cells were treated with varying concentrations of SR-48692 (0-20 μM) for 48 and 96 h and cell numbers were counted using a cell counter (*, P < 0.05 vs. vehicle for 48 h; †, P < 0.05 vs. vehicle for 96 h). (E) Western blot analysis representing expression of c-Myc and Cyclin D1 regulated by SR-48692 treatment in BON cells. β-actin was used as a loading control. (F) The number of colonies compared with vehicle in soft agar assay. Colony formation of BON cell treated with vehicle or SR-48692 was assessed over a period of 4 weeks. (*, P < 0.05 vs. vehicle)

Discussion

In this study, we assessed the regulation of NT by the Wnt/β-catenin pathway and the effects of NT inhibition on NET cell growth. First, a functional TBE was identified within the proximal promoter of the NT gene. Second, the regulators of the Wnt/β-catenin signaling pathway controlled NT expression in NET cells that express the NT gene and secrete the NT peptide. Moreover, the intracellular content and release of NT were increased by Wnt3a. Finally, knockdown of NT or treatment with SR-48692 suppressed NET cell growth, the levels of growth-related proteins and anchorage-independent growth.

Recently, we detected localization of β-catenin in the cytoplasm and/or nucleus in 25% of NET clinical samples and identified mutations of β-catenin and APC in NET cell lines and clinical tissues.⁶ We also demonstrated that negative regulators of the Wnt/β-catenin pathway were repressed by promoter methylation or histone modification leading to aberrant Wnt/β-catenin signaling in NETs.⁶ Uniquely, these NETs can produce and secrete a variety of bioactive substances, including serotonin, chromogranin A, and NT.^{21, 22} These products can elicit the carcinoid syndrome, which is characterized by flushing, diarrhea and valvular heart disease with potentially life threatening effects.^{21, 22} It has also been reported that some secretory substances (e.g., gastrin) are downstream targets of Wnt/β-catenin signaling and function as important mediators of cancer progression.^{1, 7}

Among these factors, NT acts as a primary neurotransmitter as well as a neuromodulator of other signals in the CNS and as a local hormone in the GI tract.^9-11 Its effects are mediated through three receptors: two G protein-coupled receptors, NTR1 and NTR2; and a single transmembrane receptor, NTR3.^{10, 11} In particular, the high-affinity NTR1 is found in various regions of the CNS, in the small and large intestine, and in various malignancies, and is known as a primary mediator of NT effects. Previously, it was reported that upregulation of NTR1 in colon cancer is the result of Wnt/β-catenin signaling pathway activation.¹⁶ It was shown that NTR1 expression was induced by Wnt agonists and by the association of TCF with the functional TBE identified in the NTR1 promoter. Additionally, the strong correlation noted with NTR1 expression and the β-catenin abnormal localization suggested that the activation of the Wnt/β-catenin pathway in colorectal cancers is a major cellular event for NTR1 overexpression at early stages of cell transformation.¹⁶

In addition to NTR1, we found that NT itself is a direct target for Wnt/β-catenin signaling in NET cells suggesting that the Wnt/β-catenin pathway exhibits a complex regulatory effect on NT signaling at the level of the agonist and its receptor. Many ligands and receptors have been identified as target genes of Wnt/β-catenin signaling, including EphB/ephrin-B and Wnt pathway components autoregulated by feedback.^{1, 23} Through inhibition of Wnt/β-catenin signaling, gene expression, levels for EphB2 and EphB3 receptors were decreased, but their ligand gene, ephrin-B1, was augmented.²³ However, it is unusual for NT and its receptor, NTR1, to be activated simultaneously by Wnt/β-catenin signaling implying existence of other mechanism(s) for NT signaling.

Furthermore, we evaluated the effect of NT on NET cell growth using siRNA or shRNA. Previously, it was reported that NT promoted cell growth in several types of cancers.^{9, 11} In general, the cellular effects of NT have been assessed by exogenous treatment which is similar to a paracrine effect. In contrast, we evaluated the autocrine/paracrine effect of NT by interfering with the endogenous NT in NET cells which produce and secrete NT.^{15, 21, 24} Di Florio et al.²⁵ showed that several GI hormones and neurotransmitters, including NT, have growth-inducing effects on foregut NETs and that these effects are highly dependent on transactivation of the EGF receptor by direct treatment.

In summary, we show that the mRNA and protein expression and peptide secretion of NT can be directly regulated by the Wnt/β-catenin pathway in NET cells. Furthermore, we confirmed that inhibition of NT signaling suppressed cell proliferation and anchorage-independent growth in these cells. Our results identify a novel functional link between Wnt/β-catenin and NT signaling pathways in NET cells, suggesting that components of the Wnt/β-catenin pathway may represent candidate targets for therapeutic intervention.

Novelty and impact.

We determined that the intestinal hormone neurotensin (NT), which stimulates growth of many cancers, is regulated, in part, by the Wnt/β-catenin pathway in neuroendocrine tumors (NETs). We show, for the first time, that NT mRNA, protein expression and secretion are directly regulated by the Wnt/β-catenin pathway. Furthermore, inhibition of NT signaling suppresses NET cell growth, suggesting that this novel physiological link between Wnt/β-catenin and NT signaling in NETs may represent candidate targets for therapeutic intervention.

Abbreviations

NT

neurotensin

TCF

T-cell factor

TBEs

TCF-binding elements

GI

gastrointestinal

NET

neuroendocrine tumor

NTR1

neurotensin receptor 1

siRNA

small interfering RNA

shRNA

small hairpin RNA

DN

dominant-negative

CA

constitutively activated