Inhibition of Wnt/β-Catenin Signaling in Neuroendocrine Tumors In Vitro: Antitumoral Effects

Background and aims: Inhibition of Wnt/β-catenin signaling by specific inhibitors is currently being investigated as an antitumoral strategy for various cancers. The role of Wnt/β-catenin signaling in neuroendocrine tumors still needs to be further investigated. Methods: This study investigated the antitumor activity of the porcupine (PORCN) inhibitor WNT974 and the β-catenin inhibitor PRI-724 in human neuroendocrine tumor (NET) cell lines BON1, QGP-1, and NCI-H727 in vitro. NET cells were treated with WNT974, PRI-724, or small interfering ribonucleic acids against β-catenin, and subsequent analyses included cell viability assays, flow cytometric cell cycle analysis, caspase3/7 assays and Western blot analysis. Results: Treatment of NET cells with WNT974 significantly reduced NET cell viability in a dose- and time-dependent manner by inducing NET cell cycle arrest at the G1 and G2/M phases without inducing apoptosis. WNT974 primarily blocked Wnt/β-catenin signaling by the dose- and time-dependent downregulation of low-density lipoprotein receptor-related protein 6 (LRP6) phosphorylation and non-phosphorylated β-catenin and total β-catenin, as well as the genes targeting the latter (c-Myc and cyclinD1). Furthermore, the WNT974-induced reduction of NET cell viability occurred through the inhibition of GSK-3-dependent or independent signaling (including pAKT/mTOR, pEGFR and pIGFR signaling). Similarly, treatment of NET cells with the β-catenin inhibitor PRI-724 caused significant growth inhibition, while the knockdown of β-catenin expression by siRNA reduced NET tumor cell viability of BON1 cells but not of NCI-H727 cells. Conclusions: The PORCN inhibitor WNT974 possesses antitumor properties in NET cell lines by inhibiting Wnt and related signaling. In addition, the β-catenin inhibitor PRI-724 possesses antitumor properties in NET cell lines. Future studies are needed to determine the role of Wnt/β-catenin signaling in NET as a potential therapeutic target.

In the current study, we aimed to investigate the potential role of in vitro inhibition of Wnt/β-catenin signaling as a novel strategy of molecular targeted therapy in neuroendocrine tumor cells. We investigated the in vitro effects of the PORCN inhibitor WNT974 and the β-catenin inhibitor PRI-724 in human NET cells. We provide insightful information regarding the canonical and non-canonical Wnt signaling pathway in NET cells and demonstrate inhibition of Wnt/β-catenin signaling as a potential treatment strategy in NET.
To treat NET cells with WNT974 (also named LGK974; Novartis, Basel, Switzerland) or PRI-724 (Selleckchem, Germany), the cell lines were first seeded into cell culture dishes and grown overnight prior to treatment with various doses of WNT974 (1-32 µM, dissolved in dimethyl sulfoxide (DMSO)) or PRI-724 (1-10 µM, dissolved in DMSO) for different periods of time according to the assays listed below.

Cell Viability Assay and Population Doubling Time
To determine the effect of WNT974 or PRI-724 on the regulation of NET cell viability, we performed the Cell Titer Blue®cell viability assay (Promega, Madison, WI, USA). In particular, NET cells were grown overnight and then treated with various doses of WNT974 (1-32 µM) or PRI-724 (1-10 µM) for up to 144 h. At the end of each experiment, 20 µL of Cell Titer Blue ® solution was added to the cell culture followed by further culturing for 4 h. Thereafter, the fluorescence was measured at 560/590 nm using a GLOMAX plate reader (Promega, Madison, WI, USA).
The population doubling time (PDT) in the exponential growth phase was calculated using the formula: PDT = ∆t × [ln2/(lnNt − lnN0)] [44]. All experiments, consisting of technical triplicates, were repeated at least three times. The data were expressed as a percentage of control as mean ± SD.

Flow Cytometric Cell Cycle Distribution Assay
NET cells were grown and treated with WNT974 (1-16 µM) for 72 h and then harvested through trypsinization, centrifugation and two washing steps in phosphate-buffered saline (PBS). After that, the cells were re-suspended in Nicoletti solution containing propidium iodide (Sigma-Aldrich, Taufkirchen, Germany) and then analyzed with the BD Accuri C6 flow cytometer and quantified using BD Accuri C6 Analysis software (BD Biosciences, Heidelberg, Germany) for cell cycle distribution. All experiments, consisting of technical triplicates, were repeated at least three times.

Caspase-3/7 Activity Assay
To assess changes in tumor cell apoptosis, we utilized the Apo-ONE Homogeneous Caspase-3/7 Assay (Promega, Mannheim, Germany). In brief, NET cells were seeded into 96 well plates at 10,000 cells per well and grown overnight, treated with 1 µM and 16 µM of WNT974 for 72 h, and then subjected to the Apo-ONE Homogeneous Caspase-3/7 Assay (Promega, Mannheim, Germany) according to the manufacturer's instructions. The experiments, consisting of technical duplicates, were repeated at least three times.

Wound Healing Assay
NET cells were seeded into six-well plates containing cell culture inserts (Ibidi, Munich, Germany) at a density of 120,000-140,000 cells per chamber and grown for 24 h. Afterwards the cell culture inserts were removed, fresh medium with 1% FBS was added, and the cells were treated with WNT974 for 24 h. The wound gap (created with the cell culture inserts) was observed and photographed using a Zeiss Axiovert 135 TV microscope fitted with a Zeiss AxioCam MRm camera (Zeiss, München, Germany). The NET cell migration activity was calculated by measuring the relative gap at each time point with Image-J software (NIH, Bethesda, MD, USA). The experiments, consisting of technical duplicates, were repeated at least three times.

Statistical Analysis
Our data were summarized as the mean ± SD and statistically analyzed by one-way analysis of variance (ANOVA) or two-sample t-tests using SPSS 16.0 software for Windows (SPSS Inc., Chicago, IL, USA). A p value < 0.05 indicated statistical significance.

WNT974 Reduces NET Cell Viability in a Dose-and Time-Dependent Manner
In pre-experiments, the population doubling time (PDT) was calculated as 0.895 ± 0.066 d for BON1, 1.536 ± 0.051 d for QGP-1, 1.781 ± 0.295 d for NCI-H727 cells and 15.48± 1.757 d for GOT1 cells respectively. Our results were in accordance with the short PDTs in BON1 and QGP-1 cells previously reported by Hofving et al [45], while the PDT of our GOT1 cells was even longer, with 15 days versus 5 days in the same report [45].
Based on these observations, BON1 exhibited the most pronounced response to WNT974. Due to the long PDT of GOT1 cells, and, thus, their limited availability, all further experiments were performed using only BON1, QGP-1, and NCI-H727 cells.
3.2. WNT974 induces NET Cell Cycle Arrest at the G0/G1 Phase and G2 Phase, but does not Cause Apoptosis We next used FACS and Western blot to assess the effect of WNT974 treatment on the regulation of cell cycle distribution and apoptosis in order to better understand the WNT974-induced reduction of NET cell viability (Figures 2 and 3). Treatment of NET cells with WNT974 at concentrations of 1-16 µM for 72 h resulted in the dose-dependent arrest of BON1 and NCI-H727 cells at the G1 phase of the cell cycle (Figure 2A,C). Following incubation with WNT974 (16 µM), 75.4% (vs. 61.4% of the control) and 74.0% (vs. 57.6% of the control) of the cells were observed to be in G1 phase for the BON1 and NCI-H727 cell lines, respectively. Meanwhile, the percentage of S phase cells decreased to 8.0% (vs. 14.1% of the control) and 7.4% (vs. 12.8% of the control), in BON1 and NCI-H727 cell lines, respectively. In QGP-1 cells, incubation with WNT974 (16 µM) induced the accumulation of cells in the G2 phase 25.13% (vs. 15.24% of the control cells) and subsequently decreased the percentage of cells in the G0/G1 phase to 68.07% (vs. 77.93% of the control cells) ( Figure 2B). Thus, WNT974 caused cell cycle arrest in BON1, QGP-1, and NCI-H727 cells (Figure 2A-C). Accordingly, Western blot data demonstrated that treatment with WNT974 caused dose-dependent downregulation of the expression of cyclin D1, cyclin D3, and cyclin B1, as well as cyclin-dependent kinases (CDK1, CDK4, and CDK6) and checkpoint kinase-1 (CHK1) (Figure 3). regulation of cell viability. As shown in Figure 1, in the four cell lines BON1, QGP-1, NCI-H727, and GOT1, WNT974 treatment caused a dose-and time-dependent reduction of cell viability, i.e., after 144 h incubation at a dosage of 16 µM WNT974 with values of 63.8% ± 8.5% in BON1, 74.4% ± 7.4% in QGP-1, 65.0% ± 9.4% in NCI-H727, and 69.0% ± 8.9% in GOT1 cells. The calculated IC20 (concentration of drug which causes 20% inhibition of cell viability) value was 5.4 µM for BON1, 7.3 µM for GOT1, 7.8 µM for NCI-H727, and 10.1 µM for QGP-1.  No induction of apoptosis was observed following incubation of BON1, QGP-1, and NCI-H727 cells with WNT974 at concentrations of 1-16 µM ( Figure 2). Following WNT974 treatment, the FACS analysis showed no significant increase in sub-G1 phase accumulation ( Figure 2D) and the caspase 3/7 assay showed no significant increase in caspase 3/7 activities ( Figure 2E). Surprisingly, WNT974 even suppressed caspase 3/7 activity in NCI-H727 (p < 0.001).
In summary, WNT974 induced NET cell cycle arrest at the G1/G2 phase of the cell cycle but did not induce apoptosis.

Effects of WNT974 on the Inhibition of Wnt/β-Catenin Signaling in NET Cells
Treatment of BON1 and QGP-1 cells with WNT974 at concentrations of 0.25-16 µM significantly downregulated the expression of LRP/pLRP6, DVL2, and Wnt5a/b and reduced the level of non-phosphorylated β-catenin, total β-catenin, and β-catenin phosphorylation at Ser33/Ser37/Thr41 ( Figure 4) and its downstream targeting proteins, such as c-Myc, cyclinD1, and cyclinD3 ( Figure 3). These effects of WNT974 were more prominent in BON1 and QGP-1 cells but slightly weaker effects were also observed for NCI-H727 cells. However, secreted frizzled-related protein 1 (SFRP1), an antagonist of Wnt signaling, was increased by WNT974 treatment only in NCI-H727 cells (Figure 4), suggesting cell-line-specific modulation of the Wnt signaling cascade. µM) induced the accumulation of cells in the G2 phase 25.13% (vs. 15.24% of the control cells) and subsequently decreased the percentage of cells in the G0/G1 phase to 68.07% (vs. 77.93% of the control cells) ( Figure 2B). Thus, WNT974 caused cell cycle arrest in BON1, QGP-1, and NCI-H727 cells (Figure 2A-C). Accordingly, Western blot data demonstrated that treatment with WNT974 caused dose-dependent downregulation of the expression of cyclin D1, cyclin D3, and cyclin B1, as well as cyclin-dependent kinases (CDK1, CDK4, and CDK6) and checkpoint kinase-1 (CHK1) ( Figure 3).  Effect of WNT974 on cell cycle arrest at the G1, G2/M, and sub-G1 phase. BON1, QGP-1, and NCI-H727 cells were treated with or without WNT974 for 72 h and then subjected to flow cycle metric cell cycle distribution assay. The percentage of cells at each phase of the cell cycle is shown as the mean ± SD of three independent experiments. (E). Effect of WNT974 on the regulation of caspase-3/7 activity in NET cells. BON1, QGP-1, and NCI-H727 cells were treated with or without WNT974 for 72 h and then subjected to caspase-3/7 analysis, which shows the mean percentage of caspase-3/7 activity compared to the untreated control (100%) ± SD. * p < 0.05, ** p < 0.01, and *** p < 0.001 compared with that of DMSO control.  No induction of apoptosis was observed following incubation of BON1, QGP-1, and NCI-H727 cells with WNT974 at concentrations of 1-16 µM (Figure 2). Following WNT974 treatment, the FACS analysis showed no significant increase in sub-G1 phase accumulation ( Figure 2D) and the caspase 3/7 assay showed no significant increase in caspase 3/7 activities ( Figure 2E). Surprisingly, WNT974 even suppressed caspase 3/7 activity in NCI-H727 (p < 0.001).
In summary, WNT974 induced NET cell cycle arrest at the G1/G2 phase of the cell cycle but did

Effects of WNT974 on the Inhibition of the pAKT/mTOR, MAPK/ERK, pEGFR and pIGFR Pathways in NET Cells
Because the inhibitory effects of WNT974 on canonical Wnt/β-catenin signaling in NET cells were not unique in all three NET cell lines, we further assessed whether inhibition of non-canonical Wnt/receptor tyrosine kinase signaling, i.e., PI3K/-Akt/-mTOR, could mediate the effect of WNT974. Relative expression levels (normalized to DMSO control) of treated cells were calculated in %. * p < 0.05, ** p < 0.01, and *** p < 0.001 compared with that of DMSO controls.

Effects of WNT974 on the Inhibition of the pAKT/mTOR, MAPK/ERK, pEGFR and pIGFR Pathways in NET Cells
Because the inhibitory effects of WNT974 on canonical Wnt/β-catenin signaling in NET cells were not unique in all three NET cell lines, we further assessed whether inhibition of non-canonical Wnt/receptor tyrosine kinase signaling, i.e., PI3K/-Akt/-mTOR, could mediate the effect of WNT974.
Treatment of NET cells with WNT974 at concentrations of 0.25-16 µM for 72 h caused a significant reduction in the levels of pAKT and downstream p4EBP1, and p70S6K in BON1 and QGP-1 cells, respectively ( Figure 5). Furthermore, WNT974 treatment also decreased pEGFR and pIGFR in BON1 and QGP-1 cells, and downregulated pERK in QGP-1 and NCI-H727 cells. pJNK decreased only in BON1 cells. These results further confirmed that the effects of WNT974 are cell-line-dependent ( Figure 5).
Cancers 2020, 12, 345 9 of 21 Treatment of NET cells with WNT974 at concentrations of 0.25-16 µM for 72 h caused a significant reduction in the levels of pAKT and downstream p4EBP1, and p70S6K in BON1 and QGP-1 cells, respectively ( Figure 5). Furthermore, WNT974 treatment also decreased pEGFR and pIGFR in BON1 and QGP-1 cells, and downregulated pERK in QGP-1 and NCI-H727 cells. pJNK decreased only in BON1 cells. These results further confirmed that the effects of WNT974 are cell-line-dependent ( Figure 5).

Effects of the Selective β-Catenin Inhibitor PRI-724 on NET Cell Viability and Protein Expression
As the PORCN inhibitor WNT974 demonstrated inhibition of Wnt-signaling and antitumor activity in NET cells in vitro, we explored the effects of the β-catenin inhibitor PRI-724 as well as β-catenin siRNA on NET cells.

Effects of the Selective β-Catenin Inhibitor PRI-724 on NET Cell Viability and Protein Expression
As the PORCN inhibitor WNT974 demonstrated inhibition of Wnt-signaling and antitumor activity in NET cells in vitro, we explored the effects of the β-catenin inhibitor PRI-724 as well as β-catenin siRNA on NET cells.
Cancers 2020, 12, 345 10 of 21  Western blot is shown. Equal protein loading was verified in all Western blots by normalization to the total protein staining and by the housekeeping protein β-actin. (B) Densitometric quantification of Western blot data was performed. The DMSO control was set as 1.0. Relative expression levels (normalized to DMSO control) of treated cells were calculated in %. * p < 0.05, ** p < 0.01, and *** p < 0.001 compared with that of DMSO controls. Relative expression levels (normalized to DMSO control) of treated cells were calculated in %. * p < 0.05, ** p < 0.01, and *** p < 0.001 compared with that of DMSO controls.

Effects of β-Catenin siRNA on the Regulation of NET Cell Viability and Protein Expression
We transfected β-catenin siRNA to knock down β-catenin expression. The transfection of β-catenin siRNA decreased the protein expression of β-catenin in BON1 and NCI-H727 cells to 42 ± 16% and 49 ± 8%, compared to control transfection of non-targeting control (β-actin) siRNA ( Figure  9A,B). β-Catenin siRNA significantly reduced the viability of BON1 cells ( Figure 9A) but had no effect of the viability of NCI-H727 cells ( Figure 9A). At the molecular level, the effects of the β-catenin siRNA were different from the effects of the PORCN inhibitor WNT974, e.g., regarding pLRP6, pGSK3, and cyclin D1 ( Figure 9B,C).

Effects of β-Catenin siRNA on the Regulation of NET Cell Viability and Protein Expression
We transfected β-catenin siRNA to knock down β-catenin expression. The transfection of β-catenin siRNA decreased the protein expression of β-catenin in BON1 and NCI-H727 cells to 42 ± 16% and 49 ± 8%, compared to control transfection of non-targeting control (β-actin) siRNA ( Figure 9A,B). β-Catenin siRNA significantly reduced the viability of BON1 cells ( Figure 9A) but had no effect of the viability of NCI-H727 cells ( Figure 9A). At the molecular level, the effects of the β-catenin siRNA were different from the effects of the PORCN inhibitor WNT974, e.g., regarding pLRP6, pGSK3, and cyclin D1 ( Figure 9B,C).
We studied the effects of GSK3β siRNA on the regulation of NET cell viability and protein expression. In the canonical Wnt signaling cascade, GSK3 phosphorylates β-catenin at S33/S37/T41 and subsequently causes proteasomal degradation of β-catenin. Inhibition of GSK3 activity can activate canonical Wnt/β-catenin signaling [12,46,47]. We studied the effects of GSK3β siRNA on the regulation of NET cell viability and protein expression. In the canonical Wnt signaling cascade, GSK3 phosphorylates β-catenin at S33/S37/T41 and subsequently causes proteasomal degradation of β-catenin. Inhibition of GSK3 activity can activate canonical Wnt/β-catenin signaling [12,46,47]. Cell viability assay and siRNA transfection efficacy. BON1 and NCI-H727 cells were transfected with β-catenin siRNA or non-targeting control (β-actin) siRNA for 72 h and then subjected to the cell viability assay. * p < 0.05 and ** p < 0.01 compared versus non-targeting control (β-actin) siRNA. Transfection with β-catenin siRNA caused downregulation of β-catenin protein expression, as shown by densitometric quantification of Western blot data. (B) Western blot. BON1 and NCI-H727 cells transfected with β-catenin siRNA or non-targeting control (β-actin) siRNA for 72 h and then subjected to Western blot analysis. A representative Western blot is shown. Equal protein loading was verified in all Western blots by normalization to the total protein staining and by the housekeeping protein β-actin. (C,D) Quantification of Western blot data of β-catenin siRNA with or without WNT974 for the regulation of NET cells. Densitometric quantification of Western blot data was performed. The DMSO control was set as 1.0. Relative expression levels (normalized to DMSO control) of treated cells were calculated in %. * p < 0.05 and ** p < 0.01 compared with that of non-targeting control (β-actin) siRNA control.
We transfected GSK3β siRNA to knock down GSK3β expression. The transfection of GSK3β siRNA significantly decreased the protein expression of GSK3β in BON1 and NCI-H727 cells to 10% ± 5% and 19% ± 4%, compared to control transfection of non-targeting control (β-actin) siRNA ( Figure 10A,B). As expected, GSK3β knockdown significantly reduced β-catenin phosphorylation at Ser33/37/Thr41 in NET cells ( Figure 10B) and also caused the upregulation of non-phosphorylated β-catenin S45 and of the downstream marker c-Myc ( Figure 10B). GSK3β siRNA transfection caused a significant increase in NET cell viability ( Figure 10A) and partially rescued BON1 and NCI-H727 cells from the WNT974-induced decrease in cell viability ( Figure 10A). cells from the WNT974-induced decrease in cell viability ( Figure 10A).
On the other hand, phosphorylation of GSK3 is well known for inhibiting GSK3 enzyme activity [9,[48][49][50][51]. WNT974 (Figure 4) caused strong upregulation of pGSK. The different crossways were further studied, and GSK3β siRNA with or without WNT974 downregulated pEGFR, pAKT, pERK, and pJNK, but upregulated mTOR slightly in the two cell lines, which confirmed that they are GSK3β-independent. pIGFR was also downregulated by GSK3β siRNA but was shown to be GSK3β-independent in NCI-H727 and GSK3β-dependent in BON1 ( Figure 10B). On the other hand, phosphorylation of GSK3 is well known for inhibiting GSK3 enzyme activity [9,[48][49][50][51]. WNT974 (Figure 4) caused strong upregulation of pGSK. The different crossways were further studied, and GSK3β siRNA with or without WNT974 downregulated pEGFR, pAKT, pERK, and pJNK, but upregulated mTOR slightly in the two cell lines, which confirmed that they are GSK3β-independent. pIGFR was also downregulated by GSK3β siRNA but was shown to be GSK3β-independent in NCI-H727 and GSK3β-dependent in BON1 ( Figure 10B).

WNT974 Regulation of p21 and p53 Expression
Treatment with WNT974 at concentrations of 0.25-16 µM significantly decreased protein expression of p53 and p21 in BON1 cells, and of p21 in QGP1 cells ( Figure S1).
Therefore, we investigated whether WNT974 can regulate the expression of either neurotensin (NT) or Menin. Western blot data showed expression of both NT and Menin to be modestly downregulated by WNT974 ( Figure S2).

WNT974 Suppression of NET Cell Migration and Expression of the EMT Markers
To further understand WNT974 antitumor activity, we next determined the effect of WNT974 (8 and 16 µM) on regulation of the NET cell migration capacity, and found that WNT974 only modestly reduced the migration of BON1 and QGP-1 cells and had no effect on migration in NCI-H727 cells ( Figure S3).
At the molecular level, we analyzed the expression of the mesenchymal marker vimentin and the epithelial marker E-cadherin, as well as the tight junction protein ZO-1, and found that WNT974 treatment decreased the expression level of vimentin and of ZO-1, whereas E-cadherin expression underwent no significant change in these cell lines ( Figure S4).

Discussion
The current study demonstrates that the PORCN inhibitor WNT974 reduced NET cell viability in a dose-and time-dependent manner ( Figure 1) and induced NET cell arrest at the G0/G1 and G2 phase of the cell cycle (Figure 2). At the molecular level, WNT974 treatment inhibited Wnt/β-catenin (Figures 3 and 4) signaling, but also pAKT/mTOR, pEGFR and pIGFR ( Figure 5) signaling in NET cells. In addition, treatment with the β-catenin inhibitor PRI-724 ( Figure 6) inhibited NET cell viability in vitro. Treatment with β-catenin siRNA (Figure 9) showed cell-line-specific effects. These data indicate canonical Wnt/β-catenin signaling (Figures 3 and 4) and non-canonical Wnt/receptor tyrosine kinase signaling ( Figure 5) to be involved in NET tumor cell growth regulation. These data suggest that targeting of Wnt/β-catenin signaling might be a novel molecular targeted therapeutic strategy against NETs.
The potential role of Wnt/β-catenin signaling in neuroendocrine tumors (NETs) has recently been reviewed [9,16]. Menin seems to be a negative regulator of β-catenin, and in MEN1-deficient knockout mice, β-catenin is activated in pNETs while, vice versa, conditional β-catenin knockout decreased tumorigenesis of pNETs in this in vivo model [18].
Analysis of human NET tumor samples demonstrated mutations of the MEN1 gene in up to 35% of lung carcinoids [21][22][23] and up to 37% of pNETs [6]. Mutations in other well-known negative regulators of the Wnt/β-catenin signaling cascade, such as the APC gene, were present in 6-12% of typical/atypical lung carcinoids [24] and in 8%-23.0% of SI-NETs [25,26]. A single nucleotide variation in the APC gene has been reported in QGP-1 cells but has not been found in BON1 and NCI-H727 cells [55]. Further experiments in NET cell lines with MEN1 silencing or in MEN1-deficient knockout mice [18] should be performed to evaluate the potential effect of inhibition of Wnt/β-catenin signaling in MEN1-deficient tumors due to somatic or germline MEN1 mutations. Therefore, there might be a subgroup of patients with neuroendocrine tumors who are responsive to personalized molecular targeted therapy with inhibitors of Wnt/β-catenin signaling. Further clinical studies are needed to define these subgroups for individualized precision oncology.
We revealed that the expression of pLRP6, DVL2, and Wnt5a/b was downregulated by WNT974 in NET cells (Figure 4), and correspondingly, we also found that WNT974 treatment downregulated the level of β-catenin phosphorylation at Ser33/Ser37/Thr41 (Figure 4). These findings indicate that, in NET cells, the Wnt/β-catenin canonical pathway is altered by WNT974. Axin is a concentration-limiting factor in the β-catenin degradation complex. Our current study showed that, with WNT974 treatment in NET cells, Axin1 expression initially showed a time-dependent increase, but subsequently decreased ( Figure 4).
Cell cycle progression is tightly controlled by a complex of cell cycle regulatory molecules, such as cyclin-dependent kinases (CDKs), CDK inhibitors, and cyclins. Indeed, Wnt/β-catenin signaling controls cell proliferation and migration, and an increase in β-catenin expression or activity will initiate transcriptional activation of cyclinD1/D3, CDK1/2/4, and c-Myc proteins, which control the cell cycle transition from G1 to S phase [56,57]. In the current study, we were able to show that WNT974 induced cell cycle arrest at the G1 or G2 phase of NET cells (Figure 2A-C) but did not induce NET cells to undergo apoptosis as demonstrated by sub-G1 events ( Figure 2D) and caspase 3/7 assay activity ( Figure 2E). Previous reports of the induction of cell cycle regulation and apoptosis by WNT974 have been inconsistent. While Boone JD et al. [34] reported cell cycle arrest to only to be induced in primary ovarian cancer following WNT974 treatment, a study by Tian et al. [38] reported that WNT974 enhanced apoptosis in HepG2 cells. Our results in NET cells demonstrating only cell cycle inhibition, and no apoptosis, suggest that a combination of WNT974 with other molecular inhibitors or chemotherapy might be reasonable to effectively control NETs or other cancers [37,38].
Interestingly, incubation with WNT974 caused a decrease in the expression of p53 and p21 proteins ( Figure S1). At first glance, this seems to be in contrast with WNT974 inducing cell cycle arrest in NET cells as we have previously demonstrated that upregulation of the tumor suppressor p53 and the cyclin-dependent kinase inhibitor p21 were important in NET cell cycle regulation and growth inhibition [58]. However, the interplay between p53-p21 and Wnt/β-catenin is not fully understood and might reveal differential effects on cancer cells [59,60], and p21 has also been suggested to have dual/differential effects in cancer cells, causing cell cycle arrest but also anti-apoptotic effects [61][62][63][64][65].
Inhibitory effects of WNT974 on the non-canonical Wnt/receptor tyrosine kinase signaling pathway ( Figure 5) and on GSK3 activity (Figures 4, 9B and 10B) in NETs might contribute to its antiproliferative efficacy in NET cells in vitro.
GSK3 has been implicated in the pathogenesis of various diseases, including cancer [9]. GSK3 can act paradoxically as a tumor suppressor gene in some cancer types but as an oncogene in others [9]. Phosphorylated GSK3 is the inactive form of GSK3 [9]. WNT974 (Figures 4, 9B and 10B) caused a strong upregulation of pGSK3, thus indicating inactivation of GSK3. We have previously demonstrated that inhibition of GSK3 in neuroendocrine tumor cells results in potent antiproliferative effects [48][49][50][51].
On the other hand, GSK3 is known to play a pivotal role in regulating the canonical Wnt pathway [12]. In the canonical Wnt signaling cascade, GSK3 phosphorylates β-catenin at S33/S37/T41 and subsequently causes proteasomal degradation of β-catenin. Inhibition of GSK3 activity can activate canonical Wnt/β-catenin signaling [12,46,47]. Accordingly, GSK3β siRNA significantly enhanced the viability of BON1 and NCI-H727 cells (Figure 10), while GSK3β knockdown downregulated the level of phosphorylated β-catenin, but upregulated the expression of non-phosphorylated β-catenin ( Figure 10). Our results with GSK3β siRNA might be limited as only GSK3β but not GSK3α was knocked down in our experiments, while others have demonstrated that the downregulation of GSK3α and GSK3β was necessary to obtain a functional knockdown [46]. Our findings indicate that the WNT974-mediated inhibition of NET cell viability might occur through direct inhibition of GSK3β signaling. However, as complex bidirectional regulatory mechanisms exist between GSK3 and Wnt/β-catenin signaling [66,67], a counter-regulatory mechanism of the NET cells due to WNT974-mediated inhibition of canonical Wnt/β-catenin signaling might also cause the inhibition of GSK3 activity as a rescue mechanism to reestablish β-catenin signaling.
The β-catenin inhibitor PRI-724 inhibited NET cell viability in vitro ( Figure 6) and caused downregulation of cell cycle proteins as cyclinD1, CDK1, and CHK1 ( Figure 7A,B). The PORCN inhibitor WNT974 (Figures 1-4) and further downstream β-catenin inhibitor PRI-724 (Figures 6 and 7) demonstrated similar effects on NET cell proliferation (Figure 1; Figure 6) and on pGSK3 (Figures 4  and 8) and cell cycle proteins (Figures 3 and 7). Also treatment with β-catenin siRNA (Figure 9) in BON1 cells decreased NET cell viability ( Figure 9A). However, in contrast to the significant effects on pLRP6, cyclin D1, and CDK1 expression by WNT974 (Figures 3 and 4) and by PRI-724 (Figures 7 and 8), β-catenin siRNA ( Figure 9B,C) showed no significant effects. Thus, the molecular mechanisms of action of the β-catenin inhibitor PRI-724 and β-catenin siRNA seem somehow different, however, this finding might be limited by the fact that β-catenin siRNA caused only a partial decrease of β-catenin protein expression ( Figure 9A).
Our in vitro study has several limitations. A major limitation is the limited number of human neuroendocrine cell lines investigated, and the fact that the established human neuroendocrine tumor cell lines differ from neuroendocrine tumors in vivo with respect to tumor genetics, tumor biology, proliferation rate, and population doubling time [9,45,55,68]. The WNT974 concentrations that were effective in the investigated NET cell lines were rather high compared to other more WNT974-sensitive cancer cell lines [32]. Future studies on WNT/β-catenin signaling in neuroendocrine tumors should also aim to investigate the effects in primary neuroendocrine tumor cell cultures, as has been established [45,69], but this was beyond the current scope of this work.

Conclusions
In conclusion, the PORCN inhibitor WNT974 and the β-catenin inhibitor PRI-724 exert antiproliferative activities in human NET tumor cell lines in vitro. WNT974 inhibits canonical Wnt/β-catenin signaling and exerts an inhibitory effect on the non-canonical Wnt/receptor tyrosine kinase signaling pathways, PI3K/AKT/mTOR, EGFR, and IGFR, as well as on GSK3 activity. All these mechanisms might be involved in the inhibition of NET tumor cell growth and, in future, the molecular mechanisms of WNT974 on NET tumor cells need further in-depth investigation. Targeting Wnt/β-catenin signaling by various approaches, such as through the use of PORCN inhibitors or direct β-catenin inhibitors, might be a promising molecular targeted therapeutic strategy in NETs. Potential subgroups of patients with neuroendocrine tumors who may actually be responsive to personalized molecular targeted therapy based on inhibitors of Wnt/β-catenin signaling [16] still need to be defined. Further preclinical studies and clinical trials are also needed.

Supplementary Materials:
The following are available online at http://www.mdpi.com/2072-6694/12/2/345/s1, Figure S1: Effect of WNT974 on protein expression of p53 and of p21, Figure S2: Effect of WNT974 on protein expression of neurotensin and of Menin, Figure S3: Effect of WNT974 on regulation of NET cell migration, Figure  S4: Effect of WNT974 on protein expression of epithelial to mesenchymal transition (EMT) markers. Funding: X.-F.J. has been supported by a scholarship from Chinese Scholarship Council (CSC). This study has been funded by an unrestricted research grant to C.J.A. from Novartis Pharma GmbH, Nuernberg, Germany.