Recombinant bone morphogenetic protein-2 attenuates colorectal cancer progression by orchestrating Hippo pathway activation, Yes-associated protein inhibition, and epithelial–mesenchymal transition suppression

Abstract

Bone morphogenetic protein-2 (BMP-2) regulates cell differentiation and proliferation. However, its role in colorectal cancer (CRC) remains debatable owing to potential oncogenic effects. Here, we investigated the role of BMP-2 in modulating CRC progression, particularly in regulating the Hippo signaling pathway. Exposure to recombinant human BMP-2 (rhBMP-2) resulted in a concentration-dependent decrease in CRC cell proliferation, leading to G1 cell cycle arrest. This effect was associated with the increased expression of p53, p21, and Smad4, while the levels of cyclin D1, cyclin-dependent kinase 4 (CDK4), and CDK6 decreased. Additionally, rhBMP-2 promoted apoptosis by decreasing poly (ADP-ribose) polymerase and caspase-9 expression while increasing their cleaved forms. It also activated the Hippo signaling cascade, enhancing the expression of mammalian sterile 20-like kinase 1/2, Mps One Binder 1 (MOB1), phosphorylated MOB1, and Salvador homolog, along with elevated levels of phosphorylated yes-associated protein (YAP), while concurrently suppressing total YAP expression. This resulted in cytoplasmic sequestration and subsequent degradation of YAP, thereby attenuating the transcription of YAP-responsive genes such as Connective Tissue Growth Factor. Silencing of Ras association domain family member 1 restored the rhBMP-2-induced decrease in cell viability, whereas silencing YAP further reduced the viability of CRC cell lines. Administering rhBMP-2 significantly suppressed tumor expansion in a mouse model of CRC, further supporting its potential as an antitumor agent. Collectively, these results indicate that rhBMP-2 mitigates CRC progression by activating the Hippo signaling pathway and suppressing YAP-mediated oncogenic processes, thereby highlighting its potential as a therapeutic agent that warrants further clinical evaluation.

INTRODUCTION

Colorectal cancer (CRC) is a formidable threat to global public health and is the third most commonly diagnosed malignant neoplasm worldwide and the second major contributor to cancer-related mortality [1]. The worldwide burden of CRC is anticipated to escalate, reaching an estimated 3.2 million diagnoses and 1.6 million fatalities by 2040, with the greatest burden expected in nations characterized by a high Human Development Index despite notable improvements in early detection and clinical management [2]. Its incidence is also increasing among individuals under 55 years of age at an annual rate of approximately 2% [3,4]. In parallel, over 60% of patients with CRC were diagnosed only after the disease progression to a late stage, which is frequently associated with metastatic spread and resistance to conventional therapies [5,6]. These challenges emphasize the urgent need to elucidate the molecular mechanisms driving CRC progression and explore novel targeted treatment strategies.

Among the various cellular pathways involved in CRC progression, the Hippo signaling pathway is an essential regulator of cellular proliferation, apoptosis, and tissue homeostasis [7-10]. At the core of this pathway are the transcriptional coactivators Yes-associated protein (YAP) and transcriptional coactivator with PDZ-binding motif (TAZ), whose activities are negatively regulated by phosphorylation by upstream kinases, including mammalian sterile 20-like kinase 1/2 (MST1/2) and large tumor suppressor kinase 1/2 (LATS1/2) [11-13]. When the Hippo pathway is dormant, YAP/TAZ remain unphosphorylated and migrate to the nucleus, where they enhance the expression of genes involved in epithelial-mesenchymal transition (EMT), tumor proliferation, and invasion [14-18]. Dysregulated activation of YAP is typically observed in CRC and is associated with increased malignancy and poor clinical outcomes [19]. Despite its importance in tumor biology, the upstream regulatory mechanisms controlling Hippo signaling in CRC remain unclear.

The Hippo pathway is widely acknowledged for its function in regulating cell proliferation, apoptosis, and tissue homeostasis in CRC [20,21]; nevertheless, its interaction with the bone morphogenetic protein (BMP) signaling pathway remains largely unexplored in this context. Nonetheless, emerging evidence from other tissue types and cancer models shows a potential crosstalk between these two pathways, especially through BMP-mediated modulation of YAP/TAZ activity [22,23]. BMP signaling is essential in promoting cellular differentiation and inhibiting proliferation, in part, through mechanisms that may converge on the Hippo pathway [24]. Among BMP ligands, recombinant human BMP-2 (rhBMP-2) is a clinically approved agent recognized for its potent osteoinductive activity and is widely used in orthopedic procedures [25,26]. However, its role in cancer remains debatable. Some studies have associated BMP-2 with tumor-promoting effects, particularly in osteosarcoma models [27-30]; however, large-scale clinical analyses involving rhBMP-2 have shown no clear association between its use and increased cancer risk [31]. Recent studies reported the antitumor effects of BMP-2 in certain cancers, including evidence of suppressed proliferation and metastatic potential in CRC models [32-34]. However, the underlying molecular mechanisms remain poorly understood, and currently, no studies have directly investigated the interplay between rhBMP-2 and the Hippo signaling pathway. Notably, no previous reports have examined how rhBMP-2 regulates Hippo signaling in the context of CRC. To address this gap, the aim of this study is to examine the regulatory effects of rhBMP-2 on the Hippo signaling pathway and its downstream effects on CRC cells. Specifically, we sought to determine whether rhBMP-2 modulates CRC progression by altering YAP activity via activating the Hippo pathway. Notably, rhBMP-2 has been widely used in clinical settings for bone regeneration with a well-established safety profile; nonetheless, large-scale clinical data have shown no clear association between rhBMP-2 exposure and increased cancer risk, including CRC. Thus, rhBMP-2 may be a safe candidate for modulating tumor-associated signaling pathways in CRC in addition to its potential therapeutic effects. This study revealed novel molecular interactions and provided new insights into the pathogenesis of CRC, ultimately identifying promising therapeutic targets.

METHODS

Cell lines and reagents

rhBMP-2 (DaeWoong Pharmaceutical) was used in the experiments with human (HT-29, HCT116) and murine (CT26) CRC cell lines. All cell lines were obtained from the American Type Culture Collection. HT-29 and HCT116 cell lines were chosen for their distinct genetic alterations and phenotypic behaviors, which allowed us to assess the effect of rhBMP-2 in APC-mutant- and MSI-high CRC contexts [35-37]. Cells were maintained in RPMI-1640 medium (Gibco, Thermo Fisher Scientific), supplemented with 10% fetal bovine serum (FBS) (Welgene Inc.) and 1% penicillin-streptomycin (Welgene Inc.), under standard conditions of 37°C in a humidified incubator with 5% CO₂. Various primary antibodies were used to detect protein. These included antibodies specific to Hippo pathway-related proteins, including MST1, MST2, LATS1, Salvador homolog 1 (SAV1), Mps one binder 1 (MOB1), phospho-MOB1, YAP, and phospho-YAP, all of which were provided by Cell Signaling Technology. Antibodies against poly (ADP-ribose) polymerase (PARP), caspase-9, and its cleaved form (Cell Signaling Technology) were used to analyze apoptosis-associated proteins, and anti-Ki-67 was obtained from Novus Biologicals. Key cell cycle regulators, including p21, cyclin-dependent kinase 4 (CDK4), CDK6, and cyclin D1, were detected using antibodies from the same supplier. Additional reagents included antibodies against Ras association domain family member 1 (RASSF1), connective tissue growth factor (CTGF), p53, and Smad4 (Santa Cruz Biotechnology). Antibodies targeting E-cadherin and the urokinase-type plasminogen activator (uPA) from Santa Cruz Biotechnology. The vimentin antibody was obtained from Cell Signaling Technology. Opti-MEM and Lipofectamine 3000 were obtained from Thermo Fisher Scientific. The control, RASSF1, and YAP small interfering RNAs were purchased from Santa Cruz Biotechnology.

Cell proliferation assay

1 × 10⁴ HT-29 and HCT116 cells were seeded per well in 96-well plates to prepare for the downstream experimental procedures. The existing medium was replaced with Roswell Park Memorial Institute (RPMI) supplemented with 1% FBS and increasing concentrations of rhBMP-2 (0, 0.0625, 0.125, 0.25, 0.5, and 1 μM) after incubating overnight to ensure proper cell adherence. The cells were incubated for 48 h under standard culture conditions. To assess cell viability, 10 μl of EZ-Cytox reagent (DoGen Bio) was added to each well containing 100 μl of culture medium. The plates were then incubated at 37°C for 1 h in a 5% CO₂ humidified atmosphere. The plates were gently agitated for 1 min to ensure uniform reagent distribution before measurement. The optical signal was detected at 450 nm using an Epoch reader (BioTek Instruments). Each experiment was conducted in triplicate with eight replicate wells per condition.

Anchorage-independent colony formation assay

Cells were subjected to a soft agar assay using a two-layer system to evaluate the capacity for anchorage-independent proliferation. The base layer included 2 ml of 1% agar dissolved in RPMI-1640 medium supplemented with 1% FBS. For the upper layer, 2 ml of 0.7% agarose (Sigma-Aldrich) in 1% FBS medium was mixed with 2.5 × 10⁴ CRC cells, either exposed to 1 μM rhBMP-2 or left untreated. The cultures were grown at 37°C in a CO₂-controlled humidified incubator for 4 weeks. A fresh overlay medium was added every 3 days to maintain cellular viability. After incubation, the colony formation was examined under a light microscope and photographed for documentation. Each experimental group was tested in triplicate, and experiments were conducted three times under identical conditions.

Crystal violet staining assay

HT-29 and HCT116 cells (2 × 10⁴ cells/well) were inoculated into 6-well plates and incubated overnight to promote cell attachment. Then, the cells were exposed to 1 μM rhBMP-2 for 48 h, after which the culture was continued for an additional 1–2 weeks. Cells were incubated under experimental conditions with or without rhBMP-2. After medium replacement every 3 days, the cells were gently washed with Dulbecco's phosphate buffered saline (DPBS) (Gibco, Thermo Fisher Scientific) and fixed with 100% methanol at room temperature for 20 min. They were stained using 2% crystal violet (Sigma-Aldrich) for 5 min. Excess dye was removed by rinsing the plates three times with DPBS. They were then air-dried. Experiments were independently conducted three times, and each condition was evaluated in triplicate.

Annexin V–fluorescein isothiocyanate (FITC) analysis

HT-29 and HCT116 cells (1 × 10⁶) were treated with 1 μM rhBMP-2 for 48 h. Following treatment, cells were washed twice with cold phosphate buffered saline (PBS) and gently resuspended. For staining, the cells were incubated with FITC-conjugated Membrane-Associated Protein V and propidium iodide (PI) at a 1:1 ratio in 1× binding buffer. The mixture was kept at room temperature in the dark for 15 min. After incubation, an appropriate volume of 1× binding buffer was added to dilute the samples before analysis. Apoptotic cell populations were quantified using flow cytometry (CytoFLEX; Beckman Coulter Life Sciences). The CytExpert software (Beckman Coulter Life Sciences) was used to acquire and interpret data. All experimental conditions were analyzed in triplicate, and the entire procedure was independently repeated three times to ensure reproducibility.

Wound scratch assay

Approximately 3 × 10⁵ HT-29 and HCT116 cells were distributed into 6-well plates preloaded with RPMI-1640 medium supplemented with 10% FBS. Once cells adhered and formed a confluent monolayer, proliferation was suppressed by treating with 5 μg/ml mitomycin C (Sigma-Aldrich) in 1% FBS-containing medium for 6 h to prevent overgrowth. A straight scratch was introduced using a sterile pipette tip (200 μl capacity), and cellular remnants were carefully removed by rinsing with DPBS. The culture conditions were either supplemented with or lacking 1 μM rhBMP-2 for 48 h at 37°C under standard incubation conditions after scratch formation. Wound closure was observed at 0, 24, and 48 h by using a fluorescence microscope (Olympus IX71; Olympus Corporation). Images were analyzed using the ImageJ software, and the wound width was quantified by measuring five randomly chosen areas per image. All experimental groups were evaluated in triplicate, and the entire assay was independently repeated three times.

Flow cytometric analysis

Cell cycle progression, including sub-G1 accumulation, was evaluated using flow cytometry. HT-29 and HCT116 cells (1 × 10⁶) were cultured under rhBMP-2-treated and untreated conditions for 48 h. Cells were fixed in 70% ethanol at −20 °C for 2 h, after initially washing with ice-cold DPBS. The cells were then rinsed again before incubation with RNase and 10 μg/ml PI (Sigma-Aldrich) at room temperature at 37°C in the dark for 30 min. The DNA fluorescence intensity was assessed using a CytoFLEX flow cytometer (Beckman Coulter Life Sciences). Each condition was tested in triplicate, and the entire experiment was independently replicated three times.

RNA isolation and real-time polymerase chain reaction (PCR)

Total RNA was isolated from HT29 and HCT116 cells using TRIzol reagent (Invitrogen, Thermo Fisher Scientific) and phase separated with chloroform (Millipore Sigma). The aqueous phase containing RNA was carefully transferred after centrifugation, and mixed with isopropanol (SAMCHUN) to precipitate RNA. The resulting RNA pellets were washed with 75% ethanol, air-dried, and dissolved in RNase-free water. PrimeScript RT Reagent Kit (Takara Bio Inc.) was used to synthesize complementary DNA. Quantitative real-time PCR (qRT-PCR) was conducted using an ABI Prism 7900HT system (Applied Biosystems, Thermo Fisher Scientific) with SYBR Premix Ex Taq (Takara Bio Inc.). The thermal cycling protocol is initiated at 95°C for 30 sec, followed by 40 amplification cycles with denaturation at 95°C for 15 sec and combined annealing/extension at 60°C for 1 min. Gene expression levels were normalized to those of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which was used as an endogenous control. Relative quantification was conducted using the 2–ΔΔCt method. Primer sequences used in this analysis are detailed as follows: YAP forward, 5′-GTGAGGCCACAGGAGTTAGC-3′; reverse, 5′-GGTGCCACTGTTAAGGAAAG-3′; GAPDH forward, 5′-GTCTCCTCTGACTTCAACAG-3′; reverse, 5′-ACCACCCTGTTGCTGTAGCC-3′. All reactions were conducted in triplicate with three independent biological replicates to ensure experimental reproducibility and reliability.

Western blot analysis

The cells were homogenized in Radioimmunoprecipitation Assay lysis buffer (G-Biosciences) containing protease and phosphatase inhibitors (Sigma) and incubated on ice for 30 min with gentle agitation. After centrifugation at 13,200 rpm for 30 min at 4°C, the supernatants were obtained, and protein concentrations were measured using a bovine serum albumin-based quantification kit (Thermo Fisher Scientific). Equal amounts of protein (30 μg) were separated using SDS-PAGE and transferred to polyvinylidene fluoride membranes (Amersham). The membranes were blocked with 5% nonfat milk in Tris-buffered saline with Tween 20 (TBST) for 1 h at room temperature and then incubated overnight at 4°C with primary antibodies diluted in blocking buffer. After washing with TBST, membranes were incubated for 4 h at 4°C with Horseradish peroxidase-conjugated secondary antibodies. Protein signals were detected using an enhanced chemiluminescent substrate (Millipore) and visualized using an Amersham Imager 600 system (GE Healthcare Bio-Sciences AB). Band intensities were quantified using the ImageJ software. All experiments were conducted in triplicate with three independent biological replicates to ensure reproducibility.

siRNA transduction

HT-29 and HCT116 cells were seeded into 96-well plates and incubated overnight. Transfection was conducted in the presence of 5 pmol siRNA/well in Opti-MEM (Gibco, Thermo Fisher Scientific) and Lipofectamine 3000 (Invitrogen, Thermo Fisher Scientific). Cells were incubated with the transfection mixture for 6 h, and then the medium was replaced with complete medium containing 10% FBS and 1% antibiotics for 24 h. The transfected cells were used for further experiments.

Establishment of a CT26 mouse model

The Institutional Animal Care and Use Committee of Jeonbuk National University (CBNU2020-072) approved all animal procedures and the study was conducted according to the NIH animal research guidelines. Twelve male C57BL/6 mice (6 weeks old, 20 ± 1.5 g) were randomly classified into two groups: vehicle and rhBMP-2-treated. Similarly, 12 male BALB/c mice (4 weeks old, 17 ± 1.5 g) were also randomly assigned to the same groups: vehicle and rhBMP-2. CT26 cells were pretreated with 10 μg/10 μl rhBMP-2 (equivalent to 2 mg/kg) for 1 h and then mixed with 10 μl DPBS and 20 μl Matrigel (BD Biosciences) per injection. Each mouse was subcutaneously injected into the dorsal region with 5 × 10⁶ CT26 cells with or without rhBMP-2. Intra-tumoral injections were administered on days 9 and 11, and a dose of 2 µg of rhBMP-2 in a total volume of 30 µl was administered to each tumor in BALB/c mice, while control tumors were injected with PBS. Tumor growth was monitored every 2 days. Measurements commenced 7 days post-implantation when tumor volumes approached 100 mm³. Digital calipers were used to determine the dimensions, and the volumes were calculated as (width² × length)/2. The mice were weighed every 2 days, and their health status was monitored regularly. The experiments concluded that the tumor size reached ~2,000 mm³. Euthanasia was conducted with 4% isoflurane followed by CO₂ asphyxiation.

Histopathology and immunohistochemistry staining

For histological analysis, the tumor tissue samples were fixed in 10% formaldehyde, paraffin-embedded, and sectioned at a thickness of 4 μm. Sections were deparaffinized, rehydrated, and stained with hematoxylin and eosin. After staining, the slides were dehydrated using a graded alcohol series and covered with a lipophilic stain. Then, they were allowed to dry in the air overnight. Images were captured using a light microscope.

Statistical analysis

Data analysis was conducted using GraphPad Prism 8.0 (GraphPad Software). All results are reported as the mean ± standard error of the mean (SEM). Comparisons between the two groups were conducted using an unpaired two-tailed Student’s t-test. A one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test was conducted to assess differences for analyses involving multiple groups. Statistical significance was determined using a p-value < 0.05.

RESULTS

rhBMP-2 suppresses the proliferation and clonogenic potential of CRC cells

HT-29 and HCT116 cells were treated with increasing concentrations of the rhBMP-2 (0, 0.0625, 0.125, 0.25, 0.5, and 1 μM) for 48 h to explore how rhBMP-2 influences the proliferative capacity of CRC cells. As shown in Fig. 1A, rhBMP-2 treatment resulted in a dose-dependent reduction in cell viability as measured by the EZ-Cytox assay, indicating a suppressive effect on tumor cell growth. In the crystal violet assay (Fig. 1B), cells subjected to 1 μM rhBMP-2 treatment significantly reduced in colony numbers compared to untreated controls, indicating suppression of long-term proliferative potential under standard growth conditions. Similarly, in the soft agar colony assay (Fig. 1C), rhBMP-2 treatment significantly reduced the size and number of colonies formed by HT-29 and HCT116 cells, showing an impaired clonogenic capacity in a non-adherent environment. Overall, these findings provide significant evidence that rhBMP-2 inhibits short-term proliferation and long-term clonogenic potential in CRC cells.

Fig. 1. Recombinant human bone morphogenetic protein-2 (rhBMP-2) inhibits the growth and clonogenic potential of CRC cells.

(A) HT-29 and HCT116 cells were exposed to increasing concentrations of rhBMP-2 (0, 0.0625, 0.125, 0.25, 0.5, and 1 μM) for 48 h. Cell viability was quantified via a standard proliferation assay. (B) Anchorage-dependent colony formation was assessed following a 48-h exposure to 1 μM rhBMP-2 and subsequent culture for 1–2 weeks. The colonies were subsequently stained with crystal violet and quantified. (C) Anchorage-independent growth was evaluated using a soft agar assay after treatment with 1 μM rhBMP-2 for 4 weeks. All data are presented as mean ± standard error of the mean from at least three independent experiments. Statistical significance was determined by Student’s t-test or one-way analysis of variance with Tukey’s post hoc test. **p < 0.01 vs. untreated control. CRC, colorectal cancer.

rhBMP-2 induces apoptosis in CRC cells

HT-29 and HCT116 cells were exposed to 1 μM rhBMP-2 for 48 h and then subjected to flow cytometric analysis following Annexin V-FITC/PI dual staining to assess whether rhBMP-2 triggers apoptosis in CRC cells. As shown in Fig. 2A, rhBMP-2 treatment significantly increased the proportion of early-stage apoptotic cells compared to that in untreated controls, indicating activation of apoptotic pathways. To further characterize this effect, cell cycle analysis was conducted using PI staining to quantify the sub-G1 population, which is a marker of DNA fragmentation associated with apoptosis. There was a significant increase in the sub-G1 fraction observed in both cell lines following rhBMP-2 treatment (Fig. 2B), corroborating the pro-apoptotic activity. Western blotting was conducted to assess the expression of key apoptotic proteins to evaluate the molecular mechanisms underlying the rhBMP-2-induced apoptosis. As shown in Fig. 2C, rhBMP-2 treatment reduced the levels of full-length PARP and caspase-9 while increasing their cleaved forms, indicating caspase activation and the progression of programmed cell death. These data collectively show that rhBMP-2 induces apoptosis in CRC cells through phenotypic and molecular mechanisms.

Fig. 2. rhBMP-2 induces apoptosis in CRC cells.

(A) Apoptosis was evaluated by Annexin V-fluorescein isothiocyanate/propidium iodide (PI) staining followed by flow cytometric analysis in HT-29 and HCT116 cells treated with or without 1 μM rhBMP-2 for 48 h. (B) Cell cycle analysis was conducted using propidium iodide staining to assess the sub-G1 population indicative of apoptotic DNA fragmentation. (C) Western blotting was performed to detect poly (ADP-ribose) polymerase (PARP), cleaved PARP, caspase-9, and cleaved caspase-9, with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) serving as a loading control. Densitometric quantification was performed using ImageJ. GAPDH served as the loading control. Data are expressed as mean ± standard error of the mean from three independent experiments. Statistical differences were calculated using Student’s t-test. *p < 0.05, **p < 0.01 compared to control. rhBMP-2, recombinant human bone morphogenetic protein-2; CRC, colorectal cancer.

rhBMP-2 attenuates migration and modulates the expression of EMT markers in CRC cells

A wound healing assay was conducted following treatment with 1 μM rhBMP-2 for 48 h to examine the impact of recombinant rhBMP-2 on the migratory capacity of CRC cells. As shown in Fig. 3A, rhBMP-2 treatment significantly reduced the wound closure rate in HT-29 and HCT116 cells compared to those in untreated controls, indicating an inhibitory effect on cell migration. To elucidate the molecular mechanisms underlying this reduced motility, Western blot analysis was conducted to evaluate the expression of key proteins associated with EMT. Both cell lines showed significant upregulation of the epithelial marker E-cadherin after rhBMP-2 treatment, coupled with concurrent downregulation of the mesenchymal markers uPA and vimentin (Fig. 3B). These alterations in protein expression show that rhBMP-2 promotes a shift toward an epithelial phenotype, consistent with EMT suppression. Collectively, these findings show that rhBMP-2 inhibits CRC cell migration, at least in part, by modulating EMT-associated molecular pathways.

Fig. 3. rhBMP-2 attenuates migration and modulates the expression of epithelial-mesenchymal transition (EMT) markers in CRC cells.

(A) Wound healing assays were conducted on HT-29 and HCT116 cells with or without 1 μM rhBMP-2 treatment. Wound closure was monitored at 0, 24, and 48 h, and migration rates were quantified. (B) EMT marker expression, including E-cadherin, vimentin, and uPA, was examined by Western blot after 48-h treatment. Densitometric quantification was performed using ImageJ. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. Data are shown as mean ± standard error of the mean (n = 3). Statistical significance was evaluated using a two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001 vs. untreated cells. rhBMP-2, recombinant human bone morphogenetic protein-2; CRC, colorectal cancer; uPA, urokinase-type plasminogen activator.

rhBMP-2 induces G1 phase cell cycle arrest in CRC cells

Given rhBMP-2 observed cytotoxicity in CRC cells, its effect on cell cycle progression was further explored using flow cytometry. HT-29 and HCT116 cells were exposed to 1 μM rhBMP-2 for 48 h, followed by cell cycle phase distribution analysis. As shown in Fig. 4A, rhBMP-2-treated cells showed a pronounced increase in the percentage of cells in the G1 phase compared to untreated controls, indicating G1 phase arrest. To investigate the regulatory mechanisms driving this cell cycle alteration, Western blotting was conducted to evaluate the expression of key proteins involved in the G1 phase. As shown in Fig. 4B, exposure to rhBMP-2 resulted in substantial downregulation of CDK4, CDK6, and cyclin D1, all of which contribute to regulating the G1-S phase transition. Concurrently, the protein levels of Smad4, p53, and the CDK inhibitor p21 were notably elevated in response to rhBMP-2 treatment, indicating the activation of growth-inhibitory signaling pathways. These results imply that rhBMP-2 induces G1 phase arrest in CRC cells by suppressing cyclin D1–CDK4/6 complexes and upregulating the Smad4/p53/p21 axis.

Fig. 4. Recombinant human bone morphogenetic protein-2 (rhBMP-2) induces G1 phase cell cycle arrest in CRC cells.

(A) Flow cytometry analysis of cell cycle distribution was conducted after treating HT-29 and HCT116 cells with 1 μM rhBMP-2 for 48 h. Representative histograms and quantitative data of cell cycle phases are presented. (B) Western blotting was used to determine the expression of key G1 phase–associated regulators. Densitometric quantification was conducted using ImageJ. GAPDH was included as a loading control. Results are presented as mean ± standard error of the mean from three independent experiments. Statistical analysis was conducted using Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001 vs. control. CRC, colorectal cancer; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

rhBMP-2 suppresses YAP activity and promotes activation of Hippo signaling pathway in CRC cells

The expression of YAP, a primary mediator of this pathway, was first assessed to further investigate whether rhBMP-2 exerts its anti-proliferative effects on CRC cells by modulating the Hippo signaling cascade. qRT-PCR analysis showed a significant reduction in YAP messenger RNA (mRNA) levels in HT-29 and HCT116 cells following treatment with 1 μM rhBMP-2 for 48 h (Fig. 5A). To further elucidate the involvement of the Hippo pathway, protein levels were measured to assess the expression of upstream regulatory components. Treatment with rhBMP-2 notably upregulated the protein expression of core Hippo pathway mediators, including RASSF1, MST1, MST2, SAV1, LATS1, MOB1, and its phosphorylated form p-MOB1, in both cell lines (Fig. 5B). Consistent with the qRT-PCR findings, total YAP protein levels and its downstream transcriptional target, CTGF, were significantly reduced, whereas phosphorylated YAP levels were significantly increased, indicating YAP inactivation and impaired nuclear translocation (Fig. 5C). These findings showed that rhBMP-2 activates the Hippo pathway, resulting in YAP phosphorylation and inhibition, which aids in suppressing CRC cell proliferation.

Fig. 5. Activation of the Hippo signaling cascade by recombinant human bone morphogenetic protein-2 (rhBMP-2) in CRC cells.

(A) YAP mRNA levels were measured by quantitative PCR following treatment with 1 μM rhBMP-2 in HT-29 and HCT116 cells. (B) Expression of Hippo pathway–related proteins (Ras association domain family member 1 [RASSF1], mammalian sterile 20-like kinase 1/2 [MST1/2], large tumor suppressor kinase 1/2 [LATS1/2], Mps one binder 1 [MOB1], salvador homolog 1 [SAV1], and phosphorylated-MOB1) was assessed by Western blot. (C) Western blot analysis was conducted to assess pathway activity, examining total YAP, phosphorylated YAP, and the downstream effector connective tissue growth factor (CTGF). Densitometric quantification was performed using ImageJ. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the loading control. Data represent mean ± standard error of the mean from three independent experiments. Statistical comparisons were made using Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001 vs. untreated control. CRC, colorectal cancer; YAP, yes-associated protein.

Regulation of cell proliferation by rhBMP-2 through RASSF1 and YAP in the Hippo pathway in CRC cells

Cell viability was measured using transfection with siRNA against YAP and siRASSF1 to further test whether the function of BMP-2 is mediated through the YAP and Hippo pathways. As shown in Fig. 6, RASSF1 knockdown significantly increased the proliferation of HT-29 and HCT116 cells. However, when RASSF1 knockdown was combined with rhBMP-2 treatment, the increase in cell proliferation induced by RASSF1 knockdown was significantly reduced by rhBMP-2 treatment. In the presence of rhBMP-2, the siRASSF1-induced increase in cell viability was abrogated, supporting the hypothesis that RASSF1 mediates the upstream activation of Hippo signaling by rhBMP-2. Furthermore, YAP knockdown significantly reduced cell viability, and cotreatment with rhBMP-2 further decreased cell viability. The effect reached statistical significance in HCT116 but not in HT-29 cells, despite HT-29 displaying the same directional tendency. The additive effects observed following the combined rhBMP-2 treatment and YAP depletion aligned with rhBMP-2 and YAP contributing in parallel to the regulation of cell viability, instead of rhBMP-2 acting solely upstream of YAP via a linear pathway. Under rhBMP-2 + siYAP conditions, additional knockdown of RASSF1 induced a slight rescue effect and restored cell viability to levels similar to those observed with rhBMP-2 treatment alone; nevertheless, this difference was not statistically significant. Collectively, these results indicate that rhBMP-2 suppresses cell growth through the Hippo pathway, with RASSF1 and YAP acting as key mediators. However, the observed responses may show parallel or compensatory mechanisms instead of a direct upstream–downstream relationship.

Fig. 6. Recombinant human bone morphogenetic protein-2 (rhBMP-2) negatively regulates YAP activity and triggers activation of the Hippo signaling cascade in CRC cells.

HT-29 and HCT116 cells were transfected with control small interfering RNA (siRNA) (siCtrl), small interfering Ras association domain family member 1 (siRASSF1), or small interfering yes-associated protein (siYAP), with or without rhBMP-2 treatment. Cell viability was assessed using a cell proliferation assay 48 h post-transfection. Treatment with rhBMP-2 significantly reduced cell viability compared to the control. Transfection with siRASSF1 alone significantly increased cell viability relative to the control. In the presence of rhBMP-2, siRASSF1 restored cell viability, suggesting that RASSF1 mediates the upstream activation of Hippo signaling by rhBMP-2. Transfection with siYAP alone markedly decreased cell viability compared to the control. Co-treatment with rhBMP-2 and siYAP further decreased cell viability in colorectal cancer cells, consistent with the role of YAP as a downstream effector of the Hippo pathway. Furthermore, siRASSF1 treatment was found to rescue the reduced cell viability. Data are presented as mean ± standard error of the mean. Statistical significance was determined using two-way analysis of variance followed by post-hoc tests. *p < 0.05, **p < 0.01; ns, not significant. The p-value represents the comparison between the control and the experimental group, or between the experimental groups, as indicated. CRC, colorectal cancer.

rhBMP-2 inhibits tumor growth in mouse animal model

The antitumor efficacy of rhBMP-2 was evaluated using an in vivo murine allograft model (Fig. 7A) to validate these in vitro observations. To determine whether treatment with rhBMP-2 (equivalent to 2 mg/kg) for 1 h affects initial cell viability due to potential toxicity, we assessed cell viability and apoptosis. There were no significant differences, indicating that rhBMP-2 pre-treatment did not affect cell viability (Supplementary Fig. 1). As shown in Fig. 7B, tumors derived from rhBMP-2-treated cells showed substantially slower growth than those derived from the control group. Quantitative assessment confirmed a substantial reduction in tumor volume and terminal tumor mass in the rhBMP-2-treated group (Fig. 7C, D), corroborating the inhibitory effect of rhBMP-2 on tumor progression in vivo. No significant differences were detected in body weight between the two groups throughout the experimental period (Fig. 7E), indicating the absence of overt systemic toxicity associated with rhBMP-2 administration. Western blotting further showed that rhBMP-2 upregulated MST1/2, SAV1, p-YAP, and MOB1 and downregulated YAP expression, similar to the in vitro findings (Fig. 7F). Serum biochemical markers of hepatic and renal function were analyzed to further assess the potential adverse effects. The levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), and creatinine in the rhBMP-2-treated group remained within normal physiological ranges and were similar to those in the control group (Fig. 7G–J). In addition to the allograft model, we further validated the antitumor activity of rhBMP-2 in a syngeneic CRC model, employing both pretreatment and intra-tumoral administration, with 2 μg of rhBMP-2 injected directly into each tumor on days 9 and 11 (Fig. 8A). The inclusion of intra-tumoral injection, alongside pretreatment, enabled a more precise assessment of the local pharmacodynamic effects of rhBMP-2 within the tumor microenvironment and strengthened the rigor of our in vivo evaluation across both the allograft model and an immunocompetent syngeneic model. Treatment with rhBMP-2 significantly reduced tumor weight and volume in this model (Fig. 8B–D). Histological examination of the tumor revealed a higher rate of cell death—including both apoptosis and necrosis—in the rhBMP-2–treated group compared to the control group (Fig. 8F), highlighting rhBMP-2’s effectiveness in suppressing tumor cell proliferation in vivo. In tumor tissues, rhBMP-2 treatment led to increased expression of MST1/2, SAV1, p-YAP, and MOB1 and downregulation of YAP (Fig. 8G), consistent with observations from the allograft model. Serum levels of ALT, AST, BUN, and creatinine in the rhBMP-2-treated group were not significantly different from those in the control group (Fig. 8H–K), indicating that rhBMP-2 did not induce detectable hepatotoxicity or nephrotoxicity under the experimental conditions. Collectively, these findings show that rhBMP-2 exerts potent antitumor activity in a mouse CRC model, attributable to Hippo pathway activation and YAP inhibition, with no evidence of systemic adverse effects.

Fig. 7. Recombinant human bone morphogenetic protein-2 (rhBMP-2) suppresses tumor growth in a murine allograft model.

(A) CT26 cells pretreated with rhBMP-2 (10 μg/10 μl) or phosphate-buffered saline were subcutaneously injected into c57BL/6 mice on day 0. Mice were sacrificed on day 40, and tumors and serum samples were collected. (B) Representative tumor images were captured at the endpoint. (C) Tumor weights were significantly reduced in mice treated with rhBMP-2. (D, E) Tumor volume and body weight were monitored throughout the experiment, with no significant differences in body weight between groups. (F) Protein levels of mammalian sterile 20-like kinase 1/2 (MST1/2), Salvador homolog 1 (SAV1), Mps one binder 1 (MOB1), yes-associated protein (YAP), and phosphorylated-YAP were measured in control and rhBMP-2-treated tumor samples by Western blotting. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control, and band intensities were quantified in association with the control. (G–J) Serum biochemical markers (alanine aminotransferase [ALT], aspartate aminotransferase [AST], blood urea nitrogen [BUN], and creatinine) were measured to assess hepatic and renal function, with all values remaining within physiological ranges. Data are shown as mean ± standard error of the mean. *p < 0.05, **p < 0.01 vs. control; ns, not significant.

Fig. 8. Recombinant human bone morphogenetic protein-2 (rhBMP-2) suppresses tumor growth in a murine syngeneic model.

(A) rhBMP-2 (10 μg/10 μl) or PBS-pretreated CT26 cells were subcutaneously injected into BALB/c mice on day 0. Mice were euthanized on day 13 for the collection of samples; (B) representative tumor images were captured at the endpoint. (C) rhBMP-2 treatment significantly reduced tumor weight. (D, E) Tumor volume and body weight were monitored, with no significant differences in body weight between groups. (F) Representative images of tumors from the rhBMP-2– and vehicle-treated groups. Apoptotic (arrows) and necrotic areas were observed in tumors from the rhBMP-2–treated group. Sections were stained with hematoxylin and eosin (bar = 100 μm). (G) Protein levels of mammalian sterile 20-like kinase 1/2 (MST1/2), Salvador homolog 1 (SAV1), Mps one binder 1 (MOB1), yes-associated protein (YAP), and phosphorylated-YAP were measured by Western blot in control and rhBMP-2-treated tumor samples. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control, and band intensities were quantified in association with the control. (H–K) Serum biochemical markers (alanine aminotransferase [ALT], aspartate aminotransferase [AST], blood urea nitrogen [BUN], and creatinine) indicated normal hepatic and renal function. Data are shown as mean ± standard error of the mean. *p < 0.05, **p < 0.01 vs. control; ns, not significant. PBS, phosphate buffered saline; i.t., intratumoral injection.

DISCUSSION

CRC remains a formidable challenge in oncology, characterized by its substantial heterogeneity, high tendency to metastasize, and often acquired resistance to conventional therapeutic regimens, collectively contributing to its status as a leading cause of cancer-related mortality worldwide [38-40]. The complex biological profile of rhBMP-2 is a growth factor widely used in bone surgery owing to its osteoinductive properties [41,42]. However, its osteoinductive properties necessitate rigorous evaluation of its safety and efficacy in oncological contexts [29,43]. Our comprehensive study unequivocally shows that rhBMP-2 exerts robust tumor-suppressive effects in CRC. Specifically, it inhibits cell proliferation, migration, colony formation, and tumorigenicity in vitro and in vivo. These effects are mediated through the coordinated activation of the Smad and Hippo signaling pathways and are accompanied by suppressing YAP activity. Importantly, our findings address concerns regarding the potential oncogenic risks associated with rhBMP-2 in CRC. No evidence of increased risk of CRC initiation or promotion of metastasis was observed, consistent with tumor suppression in a murine allograft and syngeneic model. Furthermore, the maintenance of normal serum biochemical markers (ALT, AST, BUN, and creatinine) throughout the study indicates an absence of systemic toxicity. Unlike some reports suggesting the oncogenic potential of BMP-2 in other biological contexts, the consistent tumor-suppressive profile of rhBMP-2 in CRC, coupled with its well-established safety record in approved bone-regeneration applications, strongly supports its low risk of inducing or promoting cancer. These results show that rhBMP-2 is a promising and safe therapeutic candidate for advanced CRC, addressing the urgent clinical need for novel treatments capable of overcoming metastasis and therapeutic resistance while concurrently reinforcing its safety for existing clinical applications, including bone surgery.

BMP signaling is well established for its indispensable roles in embryogenesis and maintaining tissue homeostasis [44,45]. However, its involvement in tumor biology is complex and context-dependent. Earlier investigations showed a pro-tumorigenic role for BMP-2 in certain malignancies, including lung and prostate cancers [28,46]; nevertheless, a growing body of evidence indicates a significant tumor-suppressive function in various other cancers, including esophageal [47], colorectal [48], and gastrointestinal carcinomas [32,49,50]. Consistent with these reports, our findings showed that rhBMP-2 treatment significantly suppressed the proliferation and clonogenic potential of HT29 and HCT116 cells, which represent distinct molecular subtypes of CRC. Mechanistically, rhBMP-2 inhibits cell proliferation by inducing G1 phase arrest. This arrest was associated with a notable reduction in the expression of cyclin D1, CDK4, and CDK6, along with increased levels of Smad4, p53, and p21, indicating that rhBMP-2 may induce G1 arrest via the Smad4/p53/p21-dependent pathway. Previous research supports this, showing that Smad4 mediates the cytostatic effects of TGF-β family ligands in CRC cells [51,52]. In addition to its effects on cell proliferation, our data unequivocally showed that rhBMP-2 significantly impaired the migratory capacity of CRC cells. This is mechanistically associated with suppressing EMT, an essential event in the metastatic cascade. Western blotting confirmed this by revealing the upregulation of the epithelial marker E-cadherin and the concomitant downregulation of the mesenchymal markers vimentin and uPA. These findings robustly align with prior research implicating Smad4 in inhibiting EMT through the transcriptional repression of mesenchymal genes [52-55]. Furthermore, our investigation extends beyond Smad-dependent mechanisms, revealing that rhBMP-2's anti-CRC effects are intricately linked to activation of the Hippo signaling cascade. The Hippo pathway is a crucial evolutionarily conserved signaling network that functions primarily as a tumor suppressor by meticulously governing cell proliferation, programmed cell death, and organ development [56,57]. Its regulatory core, comprising MST1/2 and LATS1/2 kinases, ultimately dictates the activity of the oncogenic transcriptional co-activators YAP and TAZ [58,59]. Given the established role of the Hippo pathway as a master regulator of cell growth and survival, its potential involvement provides a compelling logical extension to understand rhBMP-2's observed anti-proliferative and anti-metastatic effects. Therefore, our study delineates a novel and crucial functional connection between rhBMP-2 and the Hippo signaling cascade in CRC. We observed that rhBMP-2 stimulation resulted in the coordinated upregulation of key Hippo pathway components, including MST1/2, SAV1, LATS1, and MOB1, with the concomitant phosphorylation of MOB1, indicating robust pathway activation. Moreover, enhanced expression of RASSF1, a known scaffold that stabilizes MST1/2 and anchors Hippo activation [60-62], supports the hypothesis that rhBMP-2 may signal upstream through the RASSF1/MST1 axis to initiate Hippo signaling. Concomitantly, YAP, the central transcriptional effector of the Hippo pathway, was significantly suppressed at the mRNA and protein levels, reinforcing the significant role of Hippo pathway activation in rhBMP-2's antitumor effects. Specifically, increased p-YAP leads to its inactivation and cytoplasmic sequestration, which, in turn, results in the downregulation of CTGF, a well-established YAP transcriptional target associated with cell proliferation and survival [63,64]. These comprehensive findings strongly indicate that rhBMP-2 possesses potent tumor-inhibitory properties by actively restoring Hippo signaling activity. The in vivo antitumor effectiveness of rhBMP-2 was further validated using a murine CT26 allograft and syngeneic model, which consistently showed significant reductions in tumor size, volume, and weight. Importantly, our in vivo studies showed no significant changes in animal body weights or serum biochemical markers (AST, ALT, BUN, and creatinine), indicating that rhBMP-2 administration did not lead to systemic toxicity at the tested concentrations or delivery methods. This finding is particularly noteworthy because previous concerns regarding the potential oncogenic activity of BMP-2 in certain orthopedic applications have limited its broader therapeutic exploration. Our results robustly showed that rhBMP-2 may provide substantial therapeutic benefits in CRC without overt toxicity when used at a defined concentration through an appropriate delivery method. Despite these encouraging findings, some limitations must be considered. We observed robust activation of the Smad and Hippo pathways; however, the precise mechanistic regulatory hierarchy and potential interplay between these two crucial signaling cascades in response to rhBMP-2 remain unclear. Overall, our results highlight the potential of rhBMP-2 as a therapeutic agent in CRC. It functions by simultaneously stimulating the tumor-suppressive Smad and Hippo pathways to inhibit tumor cell growth, motility, and viability. These findings underscore the therapeutic potential of targeting dual tumor-suppressive pathways using recombinant BMP-2 and offer a robust foundation for future translational studies evaluating rhBMP-2's clinical utility in CRC treatment.

SUPPLEMENTARY MATERIALS

Supplementary data including one figure can be found with this article online at https://doi.org/10.4196/kjpp.25.369