Active Beta-Catenin (ABC) promotes an invasive phenotype in pediatric osteosarcoma

doi:10.21203/rs.3.rs-4979798/v1

Download PDF

Article

Active Beta-Catenin (ABC) promotes an invasive phenotype in pediatric osteosarcoma

https://doi.org/10.21203/rs.3.rs-4979798/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Osteosarcoma (OS) is an aggressive primary bone malignancy with peak incidence in children and adolescents. Despite current multimodal treatments, there has been little change in overall survival outcomes in the last two decades. The canonical Wnt/β-catenin pathway is known to be a critical pathway in OS progression. To better understand the molecular basis of OS and potentially provide target/s for new therapies or diagnostics, we investigated the relationship between β-catenin, more specifically, the transcriptionally active form of β-catenin, Activated β-Catenin (ABC), and OS progression. We previously reported an association between ABC and aggressive OS whereby, cellular/nuclear ABC levels, but not cellular/nuclear β-catenin levels, increase with the degree of aggressiveness. However, a direct role for ABC in promoting OS progression has not been shown. In order to directly determine the role/impact of ABC in OS progression, we generated a pEGFP-ABC fusion construct which simulates ABC's phosphorylation pattern. Transfection of pEGFP-ABC, pEGFP-β-catenin, or an empty vector (pEGFP) into OS cell lines showed that wnt pathway transcriptional activity in GFP-ABC-expressing cells was significantly higher than that in both GFP-β-catenin and empty-vector-transfected cells. We also show that the in vitro invasive potential of the pEGFP-ABC-transfected cells was significantly higher compared to both pEGFP-β-catenin and pEGFP-transfected cells. Immunohistochemistry of clinical pediatric OS specimens supported these findings, showing a significant correlation between high ABC levels and invasive disease. To the best of our knowledge, this is the first report that suggest that ABC drive transcriptional activity to enhance invasiveness in OS and could serve as a biomarker of aggressive or metastatic OS.

Biological sciences/Cell biology

Health sciences/Oncology

Osteosarcoma progression

Active β-Catenin (ABC)

β-Catenin

epithelial-mesenchymal transition

biomarker of osteosarcoma

Osteosarcoma (OS) is a primary bone malignancy most commonly occurring in children and adolescents (1,2). The 5-year survival rate for OS is 60% (3), which decreases to 27% when there are distant metastases (usually lungs) (3). The rate of metastasis at diagnosis is 18% (4) and an additional 28% of patients with nonmetastatic disease will develop metastases within 5 years of disease onset (5).

These relatively poor outcomes have persisted for decades (6,7), and while many potential therapeutics have been explored, there have been no significant advances in standard treatment for over 10 years (7). This may be partially attributable to the poorly understood etiology of OS. The cellular origin of OS is unclear (8) and the pathophysiology of OS metastasis is complex, likely involving the Wnt/β-catenin signaling pathway (9).

The canonical Wnt/β-catenin signaling pathway has been implicated in many cancers (10), and overexpression of β-catenin has been shown to correlate with metastasis in OS (11). In the absence of wnt, β-catenin is sequestered in the cytoplasm by a degradation complex, phosphorylated at 4 N-terminal residues: S33, S37, T41, S45 (Figure 1A) and subsequently ubiquitinated and targeted for proteasomal degradation (10). In the presence of wnt, β-catenin is dephosphorylated, localizes to the nucleus where it binds the transcription factor T-cell factor (TCF) and regulates the transcription of multiple genes. The downstream targets of the wnt pathway support tumor survival/progression/metastasis and inhibit apoptosis. These functions are regulated by the matrix metalloproteinases MMP-2 and MMP-9 (12), which have both been linked to metastatic disease in OS and poor prognosis (13,14).

Active β-catenin (ABC), a partially dephosphorylated form of β-catenin (dephosphorylated at S37 and T41) first described by Staal et al(15),is a more potent activator of TCF and therefore more transcriptionally active than β-catenin dephosphorylated at all 4 N-terminal amino acid sites [S33, S37, T41, S45]. We have previously shown that partial dephosphorylation of β-catenin in the formation of ABC is mediated by crosstalk between the phosphatidylinositol 3 kinase (PI3K) and Wnt/β-catenin signaling pathways (16), pathways which are also known to play an important role in OS metastasis (9). The different phosphorylation states of the N-terminus of β-catenin are depicted in Figure 1A. Previous studies by our group and others found that there were higher levels of ABC in vitro in metastatic OS cell lines compared to their non-metastatic parent cell lines (17,18). Fang et al have also reported that inhibition of the canonical wnt pathway inhibited OS cell proliferation, invasion, migration, and colony formation in vitro (18).

This study uses a pEGFP-ABC fusion construct to directly determine the effects of over-expressing ABC or β-catenin in OS cell lines in vitro. We demonstrate that increased expression of ABC but not β-catenin promotes increased OS cell migration, OS cell invasion, anchorage-independent growth, and transcription and activity of wnt-regulated oncogenic proteins matrix metalloproteinase MMP-2 and MMP-9. We also identified that high levels of ABC in clinical pediatric tumor specimens strongly correlated with metastasis.

2.1. Constructs and Transformation

β-Catenin gene (Fig1A) was PCR-amplified from a flag-tagged pcDNA 3.1 Zeo vector and cloned into pEGFP-C2 to create the pEGFP-β-Catenin construct (19), enabling visual assessment of subcellular localization.

The pEGFP-Active β-Catenin (ABC) construct (Fig.1A&B) replaced normally phosphorylated serine residues 33 and 45 of ABC with aspartic acid (D), mimicking constitutive phosphorylation. The normally unphosphorylated serine37 and threonine41 of ABC were replaced with alanine (A), preventing further phosphorylation. This CDNA was inserted into pEGFP-C2 (GeneArt; Invitrogen). The pEGFP-ABC construct was confirmed by DNA sequencing. Constructs used were pEGFP-C2 (Empty Vector), pEGFP-β-Catenin, and pEGFP-ABC.

Positive colonies of recombinant constructs in Max Efficiency DH5α were amplified in LB medium with kanamycin and purified using the Qiagen Plasmid Purification Kit.

2.2. Cell Culture and Transfection

SaOS2-LM7 and SaOS2 cells were kindly provided by Dr. Eugenie Kleinerman (M.D. Anderson Cancer Center). HOS, HOS-143B, and MCF7 cells were from ATCC. SaOS2 originated from an 11-year-old female's primary OS, while SaOS2-LM7, its metastatic variant, was derived through sequential injection into nude mice followed by isolation and reinjection of lung metastatic cells, repeated six times. HOS-143B was obtained by transforming HOS cells with K-Ras oncogene.

All cell lines were cultured in Minimal Essential Medium (DMEM), supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin-100μg/mL streptomycin), 1mM sodium pyruvate and 2mM L-glutamine (Gibco) at 37ºC, 5% CO2. Cells were grown to 70% confluence and transfected with one of the following plasmids: pEGFP-β-Catenin, pEGFP-ABC, pEGFP-C2. Transfections utilized 3μg plasmid DNA and Lipofectamine LTX (Invitrogen) according to the manufacturer’s protocols.

2.3. Transwell® Migration & Invasion Assay

Transwell® units (8µM) coated with Matrigel Matrix were placed in a 24-well plate. 40,000-50,000 OS cells were seeded in the upper chamber. MCF-7 cells served as the negative control, and an empty chamber acted as the blank. Cells were incubated in 0.1% FBS/DMEM at 37⁰C. 10% FBS/DMEM in the lower chamber acted as a chemoattractant. After incubation, cells that invaded into the lower chamber were fixed with 100% methanol, stained with 0.5% crystal violet, and counted using a High Content Microscope at 10X magnification using MetaExpress software.

Cell migration assay was done similarly but used uncoated Transwell® units.

2.4. Whole cell lysate

Cells at 80% confluence were detached using Trypsin-EDTA (Gibco) and lysed using lysis buffer [10mM Tris-HCl pH 7.5, 1mM EDTA, 1mM EGTA, 1% Triton X-100, 2mM PMSF, 80ng/ml aprotinin, 40ng/ml chymostatin, 40ng/ml antipain, 40ng/ml leupeptin, 40ng/ml pepstatin]. Cellular debris was removed by centrifugation.

2.5. Immunoblotting

Protein quantification was done with Bicinchoninic Acid protein assay kit (ThermoScientific). 40μg of protein were run on a 7.5% or 12% SDS-PAGE gel and transferred to a PVDF membrane (Bio-Rad) at 110V; 4ºC. Membranes were then blocked in 5% milk/TBS containing 0.1% Tween-20 then incubated with the primary antibody overnight; 4ºC.

Membranes were thereafter incubated with corresponding HRP-linked secondary antibody at room temperature (RT). Blots were visualized using SuperSignal West Femto (ThermoFisher) or Western Lighting Plus ECL (Perkin Elmer). The following antibodies were used: anti-amino-terminal-β-catenin and anti-GFP (Cell Signaling) [1:1000]; anti-ABC (Millipore) [1:500]; anti-β-actin (Santa Cruz) [1:10000].

2.6. Immunofluorescence

Cells were grown to 30-40% confluence on coverslips and fixed with 4% formaldehyde at RT. Cells were permeabilized with 100% methanol and blocked with 5% goat serum (Sigma Aldrich) in PBS-0.3% Triton X-100. Cells were then incubated with anti-β-Catenin antibody or anti-ABC [1:200] overnight; 4°C. This was followed by incubation with AlexaFluor ®555 goat anti-mouse antibody (Invitrogen) for visualization. After incubation with secondary antibody, cell nuclei were stained with 300nM 6-diamidino-2-phenylindole (DAPI) (Invitrogen). Coverslips were mounted on glass slide using Prolong antifade (Invitrogen). Imaging was done at 40X magnification (oil immersion) using Carl Zeiss Laser Scanning Microscope. Images were processed with LSM Image Browser software.

2.7. High content microscopy

This was carried out as described previously (20,21). Cells were cultured to 60% confluence and stained for immunofluorescence. Images were taken at 10X (NA 0.3) using an automated, high-content screening system [ImageXpress Micro XLS, Molecular Devices]. Analysis of fluorescence intensity was conducted using MetaXpress software. We measured AlexaFluor555 (using the Cy3 channel, 532nm) signal from the nuclear and the total cellular region for all the cells in the images where the nuclear region was defined by the DAPI signal. The average intensity of AlexaFluor555 fluorescence of each cell (nuclear and cytoplasmic) was obtained using the cellular scoring algorithm.

2.8. Quantitative Real-Time PCR

Total RNA was isolated using the RNeasy® Mini Kit (QIAGEN) as per the manufacturer’s protocol. 1µg of total RNA was used to produce cDNA via Oligo (dT) (Invitrogen) and Superscript III reverse transcription (Applied Biosystems). Real-time quantification of MMP-2 and MMP-9 was assessed using the Power SYBR Green PCR Master Mix (Applied Biosystems). GAPDH was used as the endogenous control. Samples were amplified with a precycling hold at 95˚C, 30 cycles of annealing and extension at 60˚C. The following primers were used: MMP-2: sense (5`-GGC CCT GTC ACT CCT GAG AT -3`); MMP-2: anti-sense (5`-GGC ATC CAG GTT ATG GGG GA-3`); MMP-9: sense (5’-CGA ACT TTG ACA GCG ACA AG-3’); MMP-9: anti-sense (5’-CAC TGA GGA ATG ATC TAA GCC C-3’); GAPDH: sense (5`-TCA ACG ACC ACT TTG TCA AGC TCA-3`); GAPDH: anti-sense (5`-GCT GGT GGT CCA GGG GTC TTA CT-3`). Measurement was performed with a Rotor-Gene-3000 instrument (Montreal Biotech) and analyzed using Rotor-Gene-6 Software. Gene expression was determined using the relative standard curve method normalized to GAPDH expression. Histograms are reported as fold changes compared to the control.

2.9. Transcriptional Activity

SaOS2/SaOS2-transfectants/SaOS2-LM7 & HOS/HOS-transfectants/HOS-143B cell lines were cultured to 80% confluence. TCF/LEF-induced transcriptional activity was assessed using the TCF/LEF promoter-luciferase reporter construct, pTOPFlash, as previously described (17,22), and according to the manufacturer's instructions (Promega E1500). Cells were transfected with pTOPFlash or pFOPFlash (mutated TCF-Luciferase construct served as a negative control) using Lipofectamine LTX (Invitrogen). pRL-TK Renilla luciferase was co-transfected as an internal control for transfection efficiency. Relative reporter activity was measured using a luminometer.

2.10. Immunohistochemistry (IHC)

A pilot study was carried out to assess ABC in pediatric OS samples and compare to demographics and clinical outcomes. Unstained formalin-fixed, paraffin-embedded (FFPE) biospecimen slides from 15 male and 15 female pediatric patients with or without metastatic disease were obtained; colon cancer tissue served as a positive control and healthy spleen tissue as a negative control. Tissue slides were deparaffinized and then rehydrated in xylene and graded ethanol solutions. Following antigen retrieval with sodium citrate, the slides were incubated with primary anti-ABC antibody (Millipore) overnight and then incubated with peroxide followed by secondary anti-horseradish-peroxidase (HRP) antibody (Perkin Elmer). Slides were examined and scored by a clinical pathologist who was blinded to the demographic information and clinical outcomes of the patients. ABC immunoreactivity was scored based on average nuclear staining intensity, and the intensities underwent log2 transformation to normalize error distribution. Samples were assigned as “High” (>25% nuclear positivity) or “Low” (<25% nuclear positivity) intensity groups, which was determined using the median as a cutoff.

3.1. Plasmid constructs for ABC and expression in SaOS2 and HOS cells

ABC is generated endogenously by post-translational modification of β-Catenin. We have previously shown that cellular ABC levels are mainly regulated by functional interaction/crosstalk between the PTEN/PI3K and Wnt pathways and are dependent on the phosphatase activity of the protein phosphatase PP2A (16).While we can extraneously modify cellular ABC levels by regulating the level and activity of PP2A via methods we have previously described (19), this may incur other PP2A activity-associated effects that are unrelated to ABC. To avoid this confounding issue, we designed a genetic modification of β-Catenin to simulate ABC. In our novel construct (Fig 1), the two regulatory Ser/Thr residues in the N-terminal domain that are unphosphorylated in ABC are replaced with alanine that is refractory to phosphorylation, and the two Ser residues that are phosphorylated in ABC are replaced with glutamic acid to mimic constitutively phosphorylated residues. This combination is predicted to fix ABC in the active state.

We successfully transfected plasmids pEGFP-ABC and pEGFP-β-Catenin that encode the activated and wild-type versions β-Catenin, respectively, fused to GFP, in both HOS and SaOS2 cells with a transfection efficiency of approximately 50-60%.

To confirm that the plasmids pEGFP-C2 (empty vector), pEGFP-ABC, and pEGFP--Catenin were transfected into SaOS2 and HOS cells and were expressed at sufficient efficiency, we verified the presence of the respective modified proteins (GFP, GFP-tagged -Catenin and GFP-tagged ABC), using Western blot analysis. We assessed the presence of GFP-ABC and GFP--Catenin using anti-GFP antibody in both SaOS2 (Figure 2A) and HOS cells (Figure 2C). We observed the presence of three distinct bands, one at 27kDa representing GFP and two bands at 119kDa for GFP-ABC and GFP--Catenin (Figure 2A&2C). Immunoblotting with anti-ABC antibody showed a positive band at 119kDa for the SaOS2 and HOS cells transfected with pEGFP-ABC; confirming the presence of GFP-ABC (Figure 2B & 2D). Incubation with anti--Catenin antibody showed a positive band at 119kDa for SaOS2 and HOS cells transfected with pEGFP--Catenin, confirming the presence of GFP--Catenin (Figure 2B & 2D).

3.2. GFP-ABC and GFP--Catenin colocalize with endogenous ABC and -Catenin

We have previously shown that ABC and -Catenin exhibit different intracellular localization (17). ABC localizes mainly to the nucleus and -Catenin is mostly extranuclear (plasma membrane and cytoplasm). We have also shown that there were significantly higher levels of ABC in the nuclei of the more metastatic cell lines, SaOS2-LM7 and HOS-143B when compared to SaOS2 and HOS, respectively (17). There was no significant difference in -Catenin levels and localization between the metastatic cell lines and the parental cell lines.

We validated whether the trans-proteins, GFP-ABC and GFP--Catenin, exhibited similar intra-cellular localization compared to that of the respective endogenous proteins. Our data demonstrate that when SaOS2 cells were transfected with pEGFP-ABC, both the endogenous and GFP-ABC colocalized in the nucleus (Figure 3A). When SaOS2 cells were transfected with pEGFP--Catenin, both the endogenous and GFP--Catenin colocalized throughout the cell, mostly in the extranuclear space (Figure 3A). Similar localization patterns of GFP-ABC and GFP-β-Catenin with their respective endogenous proteins were also observed in with the HOS cell line transfectant (Figure 3B).

These results confirm that the GFP-ABC and GFP--Catenin trans-proteins display similar characteristics and distribution to that of their respective endogenous proteins.

3.3. ABC increases invasiveness of OS cells

To determine whether ABC plays a role in promoting invasive phenotype in OS, we measured the in vitro invasive capacity of the SaOS2 and HOS transfectants using Boyden chamber invasion assay through Matrigel®. SaOS2-LM7 and HOS-143B cells were used as positive controls to evaluate if expression of GFP-ABC or GFP--Catenin in SaOS2 and HOS would result in a similar invasive ability as the more metastatic cell lines.

We show that overexpression of GFP-ABC significantly increased the invasive capacity SaOS2 cells compared to the GFP-SaOS2 cells (control), and was comparable to the invasive capacity of SaOS2-LM7 cells (*p<0.05) (Figure 4A). There was no significant difference in invasive potential of the GFP--Catenin-overexpressing cells compared to control (Figure 4A). There was no change in cell viability under any of the experimental conditions (Figure 4A).

Similarly, overexpression of GFP-ABC significantly increased invasion in HOS cells compared to control (*p<0.05) (Figure 4B). HOS cells overexpressing GFP--Catenin showed no change in invasive capacity compared to control (Figure 4B). There was no change in cell viability under any of the experimental conditions (Figure 4B).

We next determined the effect of ABC or -Catenin overexpression on cell migration. Figure S1A shows that overexpression of GFP-ABC increased the migratory capacity of SaOS2 cells compared to the control, but this was not comparable to SaOS2-LM7. There was no change in cell migratory potential with overexpression of GFP--Catenin. Unlike SaOS2 cells, there was no change in migratory potential of HOS cells overexpressing GFP-ABC or GFP--Catenin.

These results suggest that ABC, but not -Catenin, promotes a more invasive phenotype in OS cells in vitro (Figure 4A & 4B). However, this may be independent or only partially attributable to changes in migration capacity.

3.4. ABC promotes increase in transcriptional activity

MMP9 and MMP2 matrix metalloproteases are transcriptional targets of the canonical Wnt/-Catenin pathway. We determined whether these two target proteins played a role in our observed increase in the invasive phenotype of the OS cell lines. MMP9 and MMP2 transcriptional expression was quantified by evaluating the mRNA levels of both metalloproteases in SaOS2 and HOS cells overexpressing GFP-ABC or GFP--Catenin using RT-qPCR.

Overexpression of GFP-ABC in SaOS2 cells led to a significant increase in mRNA expression of MMP9 compared to the control (*p<0.005) (Figure 5A). These mRNA levels were similar to those seen in the SaOS2-LM7 cell line (Figure 5A). MMP9 mRNA expression was not altered in GFP--Catenin-overexpressing SaOS2 cells compared to the control (Figure 5A). Similarly, overexpression of GFP-ABC in SaOS2 cells resulted in a statistically significant increase in MMP2 mRNA expression compared to the control (**p<0.001) (Figure 5B) and this increase was similar to the mRNA levels seen in the SaOS2-LM7 cell line (Figure 5B). MMP2 mRNA expression was not altered in GFP--Catenin-overexpressing SaOS2 cells compared to control (Figure 5B).

The overexpression of GFP-ABC in HOS cells led to a significant increase in MMP9 mRNA expression compared to control (*p<0.05) (Figure 5C). However, this increase was not comparable to MMP9 mRNA expression in HOS-143B cells, which exhibited much higher MMP9 expression than the GFP-ABC overexpressing HOS cells (***p<0.001) (Figure 5C). HOS cells overexpressing GFP--Catenin did not show any alteration in MMP9 mRNA expression compared to control (Figure 5C). We also observed significant increase in MMP2 mRNA expression (*p<0.05) in HOS cells overexpressing GFP-ABC compared to control and this increase was similar to the mRNA levels seen in the HOS-143B cell line (Figure 5D). However, there was no significant difference in MMP2 mRNA expression in the HOS cells overexpressing GFP--Catenin compared to control (Figure 5D).

We next investigated alterations in wnt/-Catenin transcriptional activity of SaOS2 and HOS cells overexpressing GFP-ABC or GFP--Catenin using the TOPFlash reporter assay. We co-transfected the TOPFlash-Luciferase reporter construct with either GFP-ABC or GFP--Catenin into SaOS2 or HOS cells. The results presented in Figure 5E show that while overexpression of both GFP-ABC and GFP--Catenin resulted in an increase in TOPFlash-Luciferase activity, the increase was greater with overexpression of ABC.

3.5. ABC induces anchorage-independent growth

An important characteristic of cancer progression is the ability to grow colonies that are anchorage independent. This is due to an ability of cancerous cells to avoid undergoing anoikis (apoptosis in absence of attachment to ECM). We evaluated the effect of overexpressing GFP-ABC or GFP--Catenin on the colony forming ability of SaOS2 and HOS cells and compared them to SaOS2-LM7 and HOS-143B cells, respectively.

We show that, compared to GFP control, there was a significant increase in the number of colonies of SaOS2 cells overexpressing GFP-ABC, which was similar to the number of colonies formed by SaOS2-LM7 cells (Figure 6A). However, the SaOS2-LM7 colonies was larger in size than the SaOS2-GFP-ABC colonies. There was no significant difference in the number of colonies in the GFP--Catenin-overexpressing cells compared to the GFP control (Figure 6A). HOS cells overexpressing GFP-ABC also exhibited a significant increase in the number of colonies. However, the numbers of colonies formed by HOS-GFP-ABC cells were much lower than the number of colonies formed by HOS-143B cells. There was no significant difference in the number of colonies in the GFP--Catenin-expressing cells compared to the GFP control (Figure 6B).

These results support a role for ABC in anchorage independent growth, which is a crucial factor for cancer progression.

3.6. Pediatric OS tissue shows positive association between high levels of ABC and metastasis.

We demonstrated ABC and b-Catenin expression in pediatric OS tissues, including optimizing ABC and b-Catenin antibody reactivities in decalcified tissues (Figure 7A & 7B). We show that while ABC is localized in the nuclei of cells, b-Catenin was present mainly at the cell membranes/cytoplasm.

We carried out a pilot study using 30 archival pediatric OS specimens from BC Children’s Hospital repository, we examined ABC expression in pre-treatment biopsies comparing cases with good and poor necrosis response to neo-adjuvant chemotherapy as well as other variables (Tables 1A-C).

ABC levels were correlated to the clinicopathologic features of the tumors (grade, histologic type) and to selected clinical and demographic variables: gender (male vs female), age at diagnosis, metastases at diagnosis, metastasis at resection, and percent necrosis after neoadjuvant chemotherapy (<90% vs ³90%). The average age was 13.4 years for females and 11.6 years for males.

A total of eight cases (27%) showed high ABC nuclear staining, while 22 (73%) showed low ABC staining. There was no significant difference in staining pattern between gender (p=0.215, two tailed) or mean ages in the high vs. low ABC groups (ABC high: 13.6 years + 2.4; ABC low: 12.0 years + 2.8). Our cohort identified 15 good chemotherapy responders and 15 poor chemotherapy responders, with no significant difference in nuclear ABC levels between these groups (Table 1A). However, analysis of “aggressive” disease, determined by metastasis at diagnosis or resection, showed that a significant number (p=0.029, two tailed) of cases in the ABC high group demonstrated metastatic disease at diagnosis compared with the ABC low group (Table 1B). This significance was greater (p=0.007, two tailed) when comparing metastatic disease at the time of resection, with a stronger correlation of pre-treatment ABC high group with metastatic disease (Table 1C). These results indicate that ABC may be a biomarker of “aggressive” pediatric OS as defined by metastatic potential both at the time of diagnosis and resection.

The Wnt/β-catenin pathway plays a crucial role in skeletal development and is indispensable for osteoblast lineage determination (23). Presently known molecular events underlying the genesis and progression of OS have highlighted that Wnt signaling is a significant pathway in the pathogenesis of OS. Although previous studies have shown deregulation of Wnt signaling pathway in OS (15,24–29), the role of β-catenin itself in OS and specifically in OS progression remains unclear.

We investigated whether β-catenin, more specifically, ABC, promotes OS progression. We have previously reported that there is a strong association between nuclear-ABC levels and aggressive OS in vitro (17).

In this study, we used a GFP-ABC fusion construct, to provide direct evidence for the first time that ABC, but not -Catenin, promotes an invasive phenotype in OS. We show that overexpression of ABC increased OS cell migration, invasion, and anchorage-independent colony growth in vitro. Further, overexpression of ABC led to a significant increase in wnt pathway transcriptional activity, to a greater degree than β-Catenin. In addition, nuclear ABC levels, but not β-Catenin levels, correlated with tumor stage in pediatric OS. Cumulatively, these results suggest that ABC plays a role in OS progression and this may be distinct from the role of β-Catenin.

Most investigations do not distinguish between ABC and β-Catenin and multiple studies have reported an increase in nuclear β-Catenin in OS (17,30–35). Consistent with our results, Fang et al looked selectively at ABC and identified elevated levels of ABC in multiple OS cell lines and found that wnt inhibition led to decreased OS proliferation, migration, and invasion in two cell lines – HOS-143-B and SJSA-1 – in vitro (18). We have previously shown that nuclear ABC but not nuclear β-Catenin is associated with more aggressive OS cells (17). In contrast, Cai et al examined 52 OS biopsy specimens via IHC and found that most of the samples lacked nuclear β-Catenin (33), suggesting β-Catenin levels and therefore cell signalling patterns are not uniform across all OS tumors. However, it should be pointed out that since that the c-terminal ends of both ABC and β-Catenin are identical, and the two proteins differ only at their N-terminal, the epitope-specificity of antibody that was used to detect β-Catenin could potentially have skewed these results.

A potential mechanism by which increased ABC and the resultant increase in wnt signaling activity leads to OS progression is through upregulation of MMP-2 and MMP-9. We show that overexpression of ABC significantly increased transcription of MMP-2 and MMP-9. Zhang and Zhang found that MMP-2 levels were higher in OS compared to normal tissue and correlated with metastatic disease (13). High MMP-2 levels were also found to significantly correlate to poor overall survival in OS (23). Additionally, MMP-9 levels have been found to impact OS cell migration and invasion (24). However, Poudel et al found that MMP-9 but not MMP-2 was linked to aggressive features of OS in vitro in one cell line (25). Other groups have investigated the role of the MMP-2/9 ratio and found that higher ratios were associated with greater OS cell migration and invasion (26) as well as poor response to chemotherapy (27). Gong et al. have also recently shown that a decrease in OS aggressiveness induced by the chemotherapeutic agent Ononin (formononetin-7-O-β-D-glucoside) correlated with a reduction in both MMP-2 and MMP-9 levels (28). The mechanism by which MMP-2 and MMP-9 contribute to OS aggressiveness may be through regulation of the extra-cellular matrix (ECM). The first stage of OS metastasis involves detachment of cells from the primary tumor and their migration through the ECM through degradation of the latter. This process is largely effected by the activity of the MMP family of proteins (29).

Our results also highlight a novel role for ABC as a prognostic marker in OS in the pediatric/adolescent population in which this malignancy is most common (1,2). Up to 28% of OS patients who have localized disease at diagnosis will develop metastases within 5 years (5), which significantly reduces overall survival (3). The ability to identify individuals in this “at-risk” group could lead to changes in the approach to treatment or monitoring of these patients in order to minimize disease progression. We have previously shown nuclear localization of ABC in commercially available OS tumor tissue microarray from adult patient which showed significant correlation between positive nuclear ABC and tumor stage (17). The present study was done only on pediatric/adolescent patient tumor samples. Although the number of pediatric OS samples analysed in this study was small, we show significant correlation between nuclear ABC levels and metastatic disease, in that only patients with metastases at diagnosis (biopsy) and resection had high nuclear levels of ABC. If this finding was validated in a larger clinical cohort, a prognostic test could be developed whereby nuclear ABC levels in biopsy samples could be measured with IHC.

In addition, we propose ABC is a potential therapeutic target in OS. ABC is a form of β-Catenin that mediates signaling of the canonical wnt pathway separate from β-Catenin (15). The canonical wnt pathway is important in the development/progression of many cancers including OS and is a popular target in the development of novel therapeutics (36). Given the complex and ubiquitous role of wnt signalling throughout normal tissues, the safety and potential side effects of targeting this pathway have been called into question (37). Despite this, multiple early phase clinical trials examining wnt inhibition have been undertaken or currently underway (38–40). Given that ABC is formed through partial dephosphorylation of -Catenin that is mediated by crosstalk between the wnt and the PI3K pathways (16), targeting of this dephosphorylation process is another possible avenue for the development of specific therapeutics.

Our study is limited by the fact that our results were primarily obtained in cell culture, which does not account for other cells and extracellular components of the tumor microenvironment. The architecture of traditional cell culture is also a limitation as the cells grow primarily in two-dimensions and does not reflect the in vivo architecture. This may be especially relevant for sarcomas, which originate from mesenchymal cells that do not organize into sheets and layers as epithelial cells. There are many strengths to our study including the fact that we confirmed our findings in two different OS cell lines with known positive controls and validated the direct correlation of ABC levels with OS in pediatric tumor samples. Further investigations in three-dimensional cell culture and a murine xenograft model are underway to confirm this relationship.

This study provides compelling evidence that ABC contributes to OS progression to an invasive phenotype that support OS metastasis. In addition, ABC may be a useful prognostic biomarker for risk stratification and presents a potential therapeutic target.

Data Availability:

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Acknowledgements:

We acknowledge the technical assistance of Dr. Xuejun Sun from the Cell Imaging Facility at the Cross-Cancer Institute for the imaging experiments.

Funding:

Cancer Research Institute of Northern Alberta (CRINA)/Kids with Cancer Society (KWCS) grant to SP

Conflict of Interest:

The authors declare no potential conflicts of interest.

Cole, S., Gianferante, D. M., Zhu, B., Mirabello, L. & Osteosarcoma A Surveillance, Epidemiology, and End Results program-based analysis from 1975 to 2017. Cancer. 128 (11), 2107–2118 (2022).
Mirabello, L., Troisi, R. J. & Savage, S. A. International osteosarcoma incidence patterns in children and adolescents, middle ages and elderly persons. Int. J. Cancer. 125 (1), 229–234 (2009).
American Cancer Society. Survival Rates for Osteosarcoma. (2021).
Marko, T. A., Diessner, B. J. & Spector, L. G. Prevalence of Metastasis at Diagnosis of Osteosarcoma: An International Comparison. Pediatr. Blood Cancer. 63 (6), 1006–1011 (2016).
Aljubran, A. H., Griffin, A., Pintilie, M. & Blackstein, M. Osteosarcoma in adolescents and adults: survival analysis with and without lung metastases. Ann. Oncol. 20 (6), 1136–1141 (2009).
Allison, D. C. et al. A Meta-Analysis of Osteosarcoma Outcomes in the Modern Medical Era. Sarcoma. 2012, 1–10 (2012).
Misaghi, A., Goldin, A., Awad, M. & Kulidjian, A. A. Osteosarcoma: a comprehensive review. SICOT-J. 4, 12 (2018).
de Azevedo, J. et al. Biology and pathogenesis of human osteosarcoma (Review). Oncol. Lett. ; (2019).
Yang, C. et al. Bone Microenvironment and Osteosarcoma Metastasis. Int. J. Mol. Sci. 21 (19), 6985 (2020).
MacDonald, B. T., Tamai, K. & He, X. Wnt/β-Catenin Signaling: Components, Mechanisms, and Diseases. Dev. Cell. 17 (1), 9–26 (2009).
Xie, X. et al. Prognostic Significance of β-Catenin Expression in Osteosarcoma: A Meta-Analysis. Front. Oncol. ;10. (2020).
Wu, B., Crampton, S. P. & Hughes, C. C. W. Wnt Signaling Induces Matrix Metalloproteinase Expression and Regulates T Cell Transmigration. Immunity. 26 (2), 227–239 (2007).
Zhang, M. & Zhang, X. Association of MMP-2 expression and prognosis in osteosarcoma patients. Int. J. Clin. Exp. Pathol. 8 (11), 14965–14970 (2015).
Liu, Y. et al. Matrix metalloproteinase 9 expression and survival of patients with osteosarcoma: a meta-analysis. Eur. J. Cancer Care (Engl). 26 (1), e12364 (2017).
Staal, F. J. T., van Noort, M., Strous, G. J. & Clevers, H. C. Wnt signals are transmitted through N-terminally dephosphorylated β‐catenin. EMBO Rep. 3 (1), 63–68 (2002).
Persad, A. et al. Active β-catenin is regulated by the PTEN/PI3 kinase pathway: a role for protein phosphatase PP2A. Genes Cancer. 7 (11–12), 368–382 (2017).
Ali, N. et al. Osteosarcoma progression is associated with increased nuclear levels and transcriptional activity of activated β-Catenin. Genes Cancer. 10 (3–4), 63–79 (2019).
Fang, F. et al. Targeting the Wnt/β-catenin pathway in human osteosarcoma cells. Oncotarget. 9 (95), 36780–36792 (2018).
Ha, J. R. et al. β-Catenin is O-GlcNAc glycosylated at Serine 23: Implications for β-catenin’s subcellular localization and transactivator function. Exp. Cell. Res. 321 (2), 153–166 (2014).
Lewis, C. W. et al. Upregulation of Myt1 Promotes Acquired Resistance of Cancer Cells to Wee1 Inhibition. Cancer Res. 79 (23), 5971–5985 (2019).
Ismail, I. H. et al. The RNF138 E3 ligase displaces Ku to promote DNA end resection and regulate DNA repair pathway choice. Nat. Cell. Biol. 17 (11), 1446–1457 (2015).
Sayat, R., Leber, B., Grubac, V., Wiltshire, L. & Persad, S. O-GlcNAc-glycosylation of β-catenin regulates its nuclear localization and transcriptional activity. Exp. Cell. Res. 314 (15), 2774–2787 (2008).
Chen, S. K. & Xiong, H. S. T. Z. J. Z. X. Y. P. W. J. G. Expression of CXCR4 and MMP-2 is associated with poor prognosis in patients with osteosarcoma. Histol Histopathol from Cell Biol to Tissue Eng. ;Vol.35((2020)):863–70. (2020).
Chang, P-Y. et al. Tricetin inhibits human osteosarcoma cells metastasis by transcriptionally repressing MMP-9 via p38 and Akt pathways. Environ. Toxicol. 32 (8), 2032–2040 (2017).
POUDEL, B., KIM D-K, KI H-H & KWON Y-B, LEE Y-M KIM D-K. Downregulation of ERK signaling impairs U2OS osteosarcoma cell migration in collagen matrix by suppressing MMP9 production. Oncol. Lett. 7 (1), 215–218 (2014).
Cheng, G., Gao, F., Sun, X., Bi, H. & Zhu, Y. Paris saponin VII suppresses osteosarcoma cell migration and invasion by inhibiting MMP-2/9 production via the p38 MAPK signaling pathway. Mol. Med. Rep. 14 (4), 3199–3205 (2016).
Kunz, P. et al. Elevated ratio of MMP2/MMP9 activity is associated with poor response to chemotherapy in osteosarcoma. BMC Cancer. 16 (1), 223 (2016).
Gong, G., Ganesan, K., Xiong, Q. & Zheng, Y. Anti-Invasive and Anti-Migratory Effects of Ononin on Human Osteosarcoma Cells by Limiting the MMP2/9 and EGFR-Erk1/2 Pathway. Cancers (Basel). 15 (3), 758 (2023).
Sheng, G., Gao, Y., Yang, Y. & Wu, H. Osteosarcoma and Metastasis. Front. Oncol. ;11. (2021).
Haydon, R. C. et al. Cytoplasmic and/or nuclear accumulation of the β-catenin protein is a frequent event in human osteosarcoma. Int. J. Cancer. 102 (4), 338–342 (2002).
Kansara, M. et al. Wnt inhibitory factor 1 is epigenetically silenced in human osteosarcoma, and targeted disruption accelerates osteosarcomagenesis in mice. J. Clin. Invest. 119 (4), 837–851 (2009).
Chen, K. et al. Wnt10b induces chemotaxis of osteosarcoma and correlates with reduced survival. Pediatr. Blood Cancer. 51 (3), 349–355 (2008).
Hoang, B. H. et al. Expression of LDL receptor-related protein 5 (LRP5) as a novel marker for disease progression in high‐grade osteosarcoma. Int. J. Cancer. 109 (1), 106–111 (2004).
Iwaya, K. et al. Cytoplasmic and/or Nuclear Staining of Beta-Catenin Is Associated with Lung Metastasis. Clin. Exp. Metastasis. 20 (6), 525–529 (2003).
Cai, Y. et al. Inactive Wnt/β‐catenin pathway in conventional high‐grade osteosarcoma. J. Pathol. 220 (1), 24–33 (2010).
Zhang, Y. & Wang, X. Targeting the Wnt/β-catenin signaling pathway in cancer. J. Hematol. Oncol. 13 (1), 165 (2020).
Kahn, M. Can we safely target the WNT pathway? Nat. Rev. Drug Discov. 13 (7), 513–532 (2014).
Neiheisel, A., Kaur, M., Ma, N., Havard, P. & Shenoy, A. K. Wnt pathway modulators in cancer therapeutics: An update on completed and ongoing clinical trials. Int. J. Cancer. 150 (5), 727–740 (2022).
Whittle, S. B. Tegavivint for the Treatment of Recurrent or Refractory Solid Tumors, Including Lymphomas and Desmoid Tumors. ClinicalTrials.gov. (2024).
Li, J. & Wang, C-Y. TBL1–TBLR1 and β-catenin recruit each other to Wnt target-gene promoter for transcription activation and oncogenesis. Nat. Cell. Biol. 10 (2), 160–169 (2008).

Table 1 is available in the Supplementary Files section.

Figures 1-7 are available as a pdf file in the supplemental files section.

No competing interests reported.

Download PDF

Reviews received at journal
01 Nov, 2024
Reviewers agreed at journal
16 Oct, 2024
Reviewers invited by journal
05 Sep, 2024
Editor assigned by journal
05 Sep, 2024
Editor invited by journal
05 Sep, 2024
Submission checks completed at journal
05 Sep, 2024
First submitted to journal
26 Aug, 2024

You are reading this latest preprint version

Active Beta-Catenin (ABC) promotes an invasive phenotype in pediatric osteosarcoma

Status:

Version 1

Abstract

1. Introduction

2. Materials And Methods

3. Results

4. Discussion

Declarations

References

Tables

Figures

Additional Declarations

Supplementary Files

Status:

Version 1