The tankyrase-specific inhibitor JW74 affects cell cycle progression and induces apoptosis and differentiation in osteosarcoma cell lines

Wnt/β-catenin is a major regulator of stem cell self-renewal and differentiation and this signaling pathway is aberrantly activated in a several cancers, including osteosarcoma (OS). Attenuation of Wnt/β-catenin activity by tankyrase inhibitors is an appealing strategy in treatment of OS. The efficacy of the tankyrase inhibitor JW74 was evaluated in three OS cell lines (KPD, U2OS, and SaOS-2) both at the molecular and functional level. At the molecular level, JW74 induces stabilization of AXIN2, a key component of the β-catenin destruction complex, resulting in reduced levels of nuclear β-catenin. At the functional level, JW74 induces reduced cell growth in all three tested cell lines, in part due to a delay in cell cycle progression and in part due to an induction of caspase-3-mediated apoptosis. Furthermore, JW74 induces differentiation in U2OS cells, which under standard conditions are resistant to osteogenic differentiation. JW74 also enhances differentiation of OS cell lines, which do not harbor a differentiation block. Interestingly, microRNAs (miRNAs) of the let-7 family, which are known tumor suppressors and inducers of differentiation, are significantly upregulated following treatment with JW74. We demonstrate for the first time that tankyrase inhibition triggers reduced cell growth and differentiation of OS cells. This may in part be due to an induction of let-7 miRNA. The presented data open for novel therapeutic strategies in the treatment of malignant OS.


Introduction
Wnt/b-catenin signaling is involved in multiple biological processes, including regulation of cellular proliferation and the switch between stem cell-ness and differentiation [1][2][3][4]. Altered Wnt/b-catenin signaling has been linked to degenerative diseases, metabolic diseases, and cancer [2,[5][6][7]. The key mediator of canonical Wnt signaling, b-catenin, is found at multiple subcellular localizations, including adherence junctions where it contributes to stabilizing cell-cell contacts, and in the nucleus where b-catenin is involved in transcriptional regulation [2,4,8].
The Wnt/b-catenin signaling pathway is activated when Wnt ligand binds to Frizzled (FZD) receptors and low-density lipoprotein receptor-related proteins-5/6 (LRP5/6) coreceptors. As a result, b-catenin accumulates in the cytoplasm and subsequently translocates to the nucleus where it regulates transcription of Wnt/b-catenin target genes, in part by binding to transcription factor T-cell factor/lymphoid enhancer-binding factor (TCF/LEF) [6].
Osteosarcoma (OS) is the most common primary malignant bone cancer [24] and although the majority of patients undergo an aggressive treatment regime, often including surgery, radiotherapy, and chemotherapy, prognosis remains poor [25]. OS is characterized by the presence of abnormal osteoblasts. Thus, imbalance in the osteogenic differentiation process is central to the disease, and in agreement with this, more than 80% of OS tumors are poorly differentiated and of higher grade [26]. Wnt/b-catenin signaling is implicated in normal osteoblast differentiation and aberrant Wnt/b-catenin signaling disrupts normal bone development [6] and is frequently observed in OS [27]. Mutations in b-catenin have not been observed in OS, but instead increased b-catenin activity has been linked to increased expression of Wnt receptors or an inhibition or loss of expression of secreted inhibitors [28]. Indeed, elevated expression of the receptor LRP5 was observed in 50% of high-grade OS tumors and expression correlated with metastasis [29]. Inhibition or loss of expression of the secreted inhibitor Wnt inhibitory factor (WIF1) was observed in 76% of OS patient samples in a different study [30,31]. As elevated Wnt signaling is a common event in OS, inhibitors of Wnt/b-catenin may have therapeutic potential for OS patients [28]. In this study, we have investigated the effect of the tankyrase-specific inhibitor JW74 on OS cell lines KPD, U2OS, and SaOS-2 at the molecular and functional level.

Materials and Methods
Cell lines, culture conditions, and reagents The cell lines U2OS, SaOS-2 (both from American type culture collection [ATCC]), and KPD [32] were cultured in RPMI-1640 (Life Technologies, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) (PAA laboratories Gmbh, Pashing, Austria), glutamax, and penicillin/ streptomycin (both from Life Technologies). Short tandem repeat (STR)-DNA profiling of 15 loci and amelogenin was performed (Genetica DNA Laboratories, Cincinnati, OH) and U2OS and SaOS-2 profiles were validated by comparing to the ATCC database. The KPD STR-DNA profile was validated by matching the obtained profile with a profile from a xenograft, generated from the original patient sample. JW74 [21] was dissolved in dimethyl sulfoxide (DMSO) (10 mmol/L) and stored at 4°C for maximum 2 weeks. Dilutions in culturing medium to final concentrations of 10-0.5 lmol/L were done immediately before use.

Reporter luciferase assay
Transfection of 2000 U2OS cells plated in 96-well plates was done the following day with reporter plasmid pTA-Luc-STF and control plasmid-expressing Renilla, as in [21]. Transfected cells were incubated for 48 h in culturing media supplemented with 0.1% DMSO (control) or JW74 (1 l-10 lmol/L). Luciferase and Renilla activity were determined using Dual-Glo Luciferase Assay System (Promega, Madison, WI).

Quantitative real-time polymerase chain reaction
Isolation of total RNA and cDNA synthesis were performed using Cell-to-Ct kit for mRNA or miRNA (Ambion, Austin, TX), following the manufacturer's protocol. Quantitative real-time polymerase chain reaction (qRT-PCR) was performed with primers and master mix from Ambion, using cDNA from 100 to 500 cells/well. The detection limit was set to cycle threshold value = 36. Relative quantifications were calculated with the 2 ÀDDCt method normalizing to PGK1 or RNU44 for mRNA and miRNA analyses, respectively. PGK1 or RNU44 were used as housekeeping genes, due to their unchanged expression during treatment [33]. Data were presented relative to the DMSO-treated sample.

Cell cycle analyses
Three hundred thousand cells in T25 flasks were attached overnight and treated for 72 h with DMSO (control) or 5 lmol/L JW74. Two million treated cells were stained with 2 lg/mL Hoechst 33342 and 20 lL/test of PE-mouse anti-human Ki-67 (BD Pharmigen, San Diego, CA), as described previously [34]. Flow cytometric analyses were performed using Becton Dickinson LSRII Flow Cytometer. Minimum 100,000 cells were acquired per sample, and gating on forward scatter versus side scatter was used to exclude cell debris and doublets. Data analysis was performed using FlowJo (TreeStar, Inc., Ashland, OR).

Proliferation assay
Two to three thousand cells attached overnight in 96-well plates were treated with culturing medium containing 0.1% DMSO (control) or JW74 (10-0.1 lmol/L). Proliferation rates based on cell confluence were determined by live cell imaging (IncuCyte; Essens Bioscience, Birmingham, U.K.), as described previously [35]. Cellular viability was also determined by MTS assay (3-[4,5- , according to the manufacturer's protocol. Expression of the proliferation marker Ki-67 was performed by staining cells with PE-mouse anti-human Ki-67 (BD Pharmigen) and by analyzing the expression by flow cytometry, as described earlier.

Apoptosis assay
For Caspase-3 assay, cells were plated and treated as for the proliferation assay. In addition, Cell player reagent (5 mmol/L in DMSO) (Essens Bioscience) was included in the medium (1:1000 dilution), enabling quantitative measurement of Caspase-3 activity by fluorescence live cell imaging in the IncuCyte. Data show total number of cells with high Caspase-3 activity in each well 52 h post treatment-start. Annexin V assay was performed using the Alexa Fluor 488 annexin V and propidium iodide (PI) kit for flow cytometry (Invitrogen, Carlsbad, CA). 100,000 U2OS cells were plated in six-well plates and incubated with DMSO or 10 lmol/L JW74 for 72 h and subsequently analyzed according to the protocol provided by the manufacturer. In brief, Alexa 488labeled Annexin V binds to phosphatidyl serines exposed on the outer leaflet of the plasma membrane of apoptotic cells. PI was used to exclude necrotic cells from the assay.
Osteogenic differentiation and quantitative and qualitative assessment of the process Thirty thousand cells attached overnight in 24-well plates were incubated in culturing medium supplemented with one of four combinations: (1) 0.1% DMSO (control); (2) JW74 (10 lmol/L) only; (3) 0.1% DMSO in combination with a differentiation cocktail (10 mmol/L glycerol phosphate, 10 nmol/L dexamethasone, and 50 lg/mL ascorbic Acid), or (4) differentiation cocktail combined with JW74 (10 lmol/L). Cells were not passaged during the experiment (maximum 24 days), but medium and supplements were changed twice per week. Osteogenic differentiation was determined quantitatively, using alkaline phosphatase (ALP) activity as a marker. The ALP assay kit (Abcam) was performed as recommended by manufacturer. Data are presented relative to total protein concentration. Degree of osteogenic differentiation was also assessed by alizarin red staining (40 mmol/L alizarin red S solution for 20 min).

Statistical analyses
Statistical analyses were performed using SigmaPlot 11 (Systat Software Inc., Chicago, IL). For comparisons of two groups, normal distributions of datasets were first analyzed with the Shapiro-Wilk test. When the Shapiro-Wilk test passed (P = > 0.05), Student's t-test was performed. If the Shapiro-Wilk test failed (P < 0.05), Mann-Whitney rank sum test was applied. P < 0.05 was regarded as a statistically significant difference.

Results
The tankyrase inhibitor JW74 reduces b-catenin levels in OS cell lines We selected three OS cell lines for testing the efficacy of the tankyrase-specific inhibitor JW74. U2OS and SaOS-2 were chosen due to increased expression of LRP5 receptor and several isoforms of the FZD receptor [29], as well as reduced expression of WIF1 [30,31], resulting in aberrant activation of Wnt/b-catenin signaling. With regard to differentiation status, SaOS-2 is considered more differentiated, consistent with high-basal ALP activity [36]. On the contrary, U2OS is more undifferentiated, with resistance to undergo in vitro osteogenic differentiation, consistent with low and noninducible basal ALP levels [36,37]. Thus, the two cell lines enabled us to study the efficacy of Wnt/b-catenin inhibition in opposing differentiation contexts. From a panel of well-characterized OS cell lines [38], we also included KPD, which is a less well-studied cell line in the context of Wnt/b-catenin signaling, but like U2OS and SaOS-2, was reported to express increased AXIN2 mRNA levels [39].
Following treatment with JW74, stabilization of AXIN2 was demonstrated in all three OS cell lines by Western blotting (Fig. 1A). AXIN2 stabilization is considered a reliable marker of tankyrase inhibition in the context of the DC [16,17,40]. We also wanted to determine the TNKS1/2 protein levels in the three cell lines following JW74 treatment, as TNKS1/2 protein levels can be either stabilized or destabilized in response to tankyrase inhibition, depending on context [40]. Alterations in TNKS1/2 protein levels after JW74 treatment were varied in the OS cell lines (Fig. 1A). While KPD cells displayed a clear reduction in TNKS, TNKS levels were unaltered in U2OS cells, and in SaOS-2 cells we observed slightly increased TNKS levels (confirmed by quantification of TNKS1/2 relative to ACTIN). The drug response was sustained, as AXIN2 protein levels were strongly elevated at 24 h, and remained increased throughout 72 h incubation with 10 lmol/L JW74 (Fig. 1B). AXIN2 stabilization was dosedependent, being in U2OS cells effective across the range from 1 to 10 lmol/L JW74 (Fig. 1C, confirmed by quantification). Although AXIN2 stabilization did not alter cytoplasmic b-catenin levels in these cells as measured by Western blot, nuclear levels of total b-catenin and active b-catenin (also known as ABC) were strongly reduced in a dose-dependent manner (Fig. 2A). The reduction in nuclear b-catenin translated into reduced transcriptional activity of a TCF/LEF-based luciferase reporter (Fig. 2B). Accordingly, transcription of the b-catenin target gene AXIN2 (Fig. 2C) and C-MYC (Fig. 2D) were reduced moderately, but significantly, following 48 and 72 h incubation with JW74.

Tankyrase inhibition reduces growth, increases apoptosis, and delays cell cycle progression
Having shown that JW74 exerts molecular effects on key mediators of the canonical Wnt signaling pathway, we next wanted to evaluate the functional effects of tankyrase  inhibition. We first studied the proliferative capacity of OS cells during short-term in vitro treatment with JW74. For this purpose, we used the a live cell imaging machine (IncuCyte), which captures cellular images every second hour throughout the duration of the experiment enabling us to determine the effect of the drug on cell confluence over time. The time lapse experiment clearly showed that tankyrase inhibition had a dose-dependent growth-limiting effect on U2OS, KPD, and SaOS-2 cells (Fig. 3A). In addition to assessing proliferative capacity by live cell imaging, we tested the effect of tankyrase inhibition on cellular viability by performing an MTS assay and found that the cellular viability of U2OS cells treated for 72 h with 10 lmol/L JW74 was reduced to 80%, relative to DMSO-treated cells (data not shown). We also performed flow cytometry to determined the expression of the proliferation marker Ki-67 in U2OS following 48 h treatment with DMSO or 10 lmol/L JW74. Ki-67 expression was reduced from 97.5% in DMSO-treated cells to 86.7% in JW74-treated cells (data not shown). We next used the live cell imaging machine to perform a Caspase-3 activity assay in U2OS, SaOS-2, and KPD cells treated with the tankyrase inhibitor. Interestingly, we found that Caspase-3 activity increased in a dose-dependent manner in all three cell lines (Fig. 3B). However, as others have shown that Caspase-3 was activated in several colon cancer cell lines, without resulting in the onset of apoptosis [41], we carefully examined serial images of individual Caspase-3-positive cells (appearing as green fluorescent). We observed membrane blebbing, detachment of the cells from the surface and production of apoptotic bodies and debris, morphological changes consistent with apoptosis. To investigate the onset of apoptosis by an additional method, we performed Annexin V flow cytometric analyses of U2OS cells treated with JW74 for 72 h. Also by this method, we observed increased apoptosis following drug treatment. The percentage of apoptotic cells bound by Alexa 488-Annexin V increased from 0.8% (DMSO) to 1.6% (10 lmol/L) (Fig. 3C).
We subsequently performed flow cytometric cell cycle analyses of Hoechst-stained U2OS cells treated with 5 lmol/L JW74 for 72 h and found an increased number of cells in the G1-phase (45.5-54.8%) and a decreased number of cells in S-phase (27.4-24.0%) and G2/M (22.2-16.2%) compared to control-treated cells (Fig. 3D), indicating that a delay in G1 contributes to the reduced growth rate. We did not observe any morphological changes indicative of senescence, such as flattened cellular morphology (data not shown). In agreement with these effects on the cell cycle, we observed significantly decreased expression of CCND1 following exposure of U2OS cells to 5 lmol/L JW74 for 48 h (~twofold reduction; data not shown).

Wnt/b-catenin inhibition induces osteogenic differentiation and leads to an increase in miRNAs of the let-7 family
We subsequently went on to assess the effect of JW74 on differentiation. In agreement with previous studies, we found that U2OS cells did not spontaneously differentiate and showed only moderate signs of induced differentia-tion in the presence of osteogenic differentiation cocktail during a 24-day differentiation assay (Fig. 4A). This was determined quantitatively by measuring enzymatic ALP activity, an established osteogenic differentiation marker, and qualitatively by alizarin red staining, which marks calcium deposits generated in the mature osteoblasts on day 0, day 6, day 12, day 18, and day 24. Moderately increased ALP levels were observed in U2OS cells subjected to long-term incubation (24 days) with 10 lmol/L JW74 alone, compared to control-treated cells (DMSO) (Fig. 4A). The changes were comparable to cells treated with differentiation cocktail, neither showing signs of full differentiation. However, when JW74 was combined with the differentiation cocktail, U2OS cells showed strong and unequivocal signs of differentiation, demonstrated by significantly increased ALP activity as well as alizarin red staining (Fig. 4A). We also observed that alizarin redpositive cells had morphological characteristics consistent with osteogenic differentiation, such as the presence of a small, round-celled body and long, thin processes (data not shown). Next, we investigated whether JW74 could improve the efficiency of differentiation in SaOS-2 cells. As expected, full differentiation was observed both qualitatively and quantitatively, when SaOS-2 cells were incubated with the standard differentiation cocktail for 12 days (Fig. 4B). Intriguingly, JW74 treatment alone induced differentiation in SaOS-2 cells equally efficient as differentiation cocktail and significantly better than cells treated with DMSO only. No additive effect was seen when differentiation cocktail was combined with JW74, presumably because maximal differentiation was already achieved.
As JW74 treatment both induces osteogenic differentiation of OS cells and reduces c-MYC expression, we hypothesized that microRNA (miRNA) let-7 levels might be elevated following JW74 treatment. miRNA let-7 is a master regulator of differentiation [42], frequently reduced or lost in a range of cancers [43], and is negatively regulated by c-MYC. Indeed, we observed a solid increase in all the let-7 orthologs evaluated (Fig. 5A) following 72-h treatment of U2OS cells with 5 or 10 lmol/L JW74, as demonstrated by qRT-PCR analyses.

Discussion
In this study, we present for the first time, the impact of tankyrase inhibition on representative OS cell lines using the novel specific tankyrase inhibitor JW74. In agreement with effects observed for colon cancer [16,17,20,21,40,44], we found that the TNKS-target AXIN2 was stabilized in all three OS lines evaluated. Furthermore, this resulted in reduced levels of b-catenin in the nucleus, reduced TCF/LEF reporter activity, and decreased AXIN2 mRNA  levels as demonstrated in U2OS cells. Similar to observations in treated colon cancer cell lines [17,21,40], TCF/ LEF reporter activity was not lowered beyond 50%, indicating active feedback loops or alternative mechanisms preventing complete reduction in reporter activity. As TNKS, the primary drug target of JW74, is implicated in cellular functions beyond its role in the DC, such as telomere maintenance, glucose metabolism, and centrosome maturation [45], the observed effects may not be exclusively explained by altered b-catenin levels.
Functionally, OS cells treated with JW74 displayed reduced growth rate due to increased apoptosis and delayed cell cycle progression. This is consistent with the observed reduction in nuclear b-catenin levels and in agreement with findings in other cancer models [16,17,20,21,40,44], including synovial sarcoma [46]. In addition, we found that tankyrase inhibition strongly induced differentiation of OS cells and enabled cells with resistance to induced differentiation to overcome their differentiation block.
The majority of OS tumors are poorly differentiated and induction of differentiation may be an interesting therapeutic strategy, as cells may become more susceptible to treatment upon induced differentiation [25]. It has been suggested that OS should be considered a "differentiation disease" caused by genetic changes, which prevent full osteoblastic differentiation [47]. The therapeutic potential of OS differentiation therapy has previously been demonstrated with nuclear receptor agonists, such as peroxisome proliferator-activated receptor (PPAR)c agonists, which either on their own, or in combination with retinoids have been shown to inhibit proliferation, induce apoptosis, and most importantly, promote terminal differentiation of OS cells [48,49]. Indeed, differentiation therapy with the retinoid all-trans retinoic acid is successfully used as standard treatment of acute promyelocytic leukemia patients [50]. However, the observed differentiation induced by JW74 in this study did not correlate with an increase in PPARc mRNA levels, following 72-h incubation with JW74 (data not shown).
It has also been shown that SOX2 plays a key role in maintaining OS cells in an undifferentiated state, being essential for self-renewal and acting as an antagonist of the Wnt pathway [51]. However, JW74 treatment did not result in reduced SOX2 expression in U2OS cells. Thus, mechanisms involving SOX2 do not seem responsible for the observed differentiation in our system.
The miRNA family let-7 are tumor suppressors and key regulators of differentiation [42]. Interestingly, we observed increased expression levels of multiple let-7 orthologs following incubation with JW74. To our knowledge, neither tankyrase nor the Wnt/b-catenin signaling pathway has to date been directly linked with the let-7 systems. As we observed reduced C-MYC levels following JW74 incubation, regulation of let-7 through C-MYC is a possibility. However, further work is required to elucidate the links between tankyrase inhibition and increased let-7 levels. Interestingly, b-catenin has been described as a regulator of other miRNAs, including miR-15, miR-16, miR-375, and miR-122a [52]. However, the mechanisms through which b-catenin regulate these miRNAs are not known.
The significant upregulation of multiple let-7 orthologs in response to JW74 treatment is of particular importance in the light of therapeutic attempts to reduce the proliferative capacity and trigger differentiation of poorly differentiated cancer cells through increased let-7 levels. Let-7 replacement therapy has shown great potential as a novel cancer therapeutic in xenograft models, where the tumor regresses following introduction of let-7 [53][54][55]. Our data suggest that similar therapeutic effects may be achievable by small drug inhibitors of tankyrase, establishing tankyrase as an important druggable biotarget, regulating a molecular switch between stem cell-ness and differentiation.