Targeting cell-matrix induced chemoresistance with regorafenib in a 3D model of osteosarcoma
Data files
Nov 10, 2025 version files 4.07 GB
-
R_Rao_et_al_JBMR-A-25-0496.zip
4.07 GB
-
README.md
5.80 KB
Abstract
Over the past four decades, there has been little advancement in treatment strategies for osteosarcoma (OS), the predominant primary bone tumor in the pediatric patient population. Current therapy involves multiple rounds of chemotherapy and surgical resection, which are associated with significant morbidity and suboptimal survival rates. A key challenge in developing new treatments is the difficulty in replicating the OS tumor microenvironment, particularly cell interactions with the extracellular matrix (ECM). This study uses an in vitro model of OS to investigate the cell response to collagen type I, the primary component of the osteosarcoma ECM. After seven days of culture within three-dimensional collagen hydrogels, OS cells displayed a more elongated cellular morphology and reduced sensitivity to the standard chemotherapy used for OS treatment compared to cells grown on two-dimensional substrates. To test whether this model could be used to study treatment strategies used for high-risk OS patients, we applied a metronomic regimen combining regorafenib, a multi-tyrosine kinase inhibitor, with front-line chemotherapy to overcome cell-matrix-induced chemoresistance. We identified overexpression of the ATP-binding cassette transporter ABCG2, a drug efflux pump, as a potential mechanism of resistance in 3D culture. Regorafenib’s inhibitory effect on ABCG2 suggests a mechanistic basis for its ability to restore chemosensitivity in 3D culture. Altogether, these findings highlight the importance of cell-matrix interactions in in vitro OS models, provide valuable insights into a matrix-induced mechanism of OS chemoresistance, and suggest an approach to its treatment.
Dataset DOI: 10.5061/dryad.w3r228128
Description of the data and file structure
All raw data files are compressed into the .zip file - R_Rao_et_al_JBMR-A-25-0496.zip
Final Figures and Supplemental Data, and Tables are provided in the Zenodo Supplemental Information.
Figure 1 - Osteosarcoma cell behavior across 2D and 3D culture. (b) These data depict the microarchitecture of collagen (COL) hydrogels in 2D and 3D configurations obtained through second harmonic generation imaging. (c) Quantification of mesh size from second harmonic generation images. (d) Cell viability images (green = live, red = dead) in the 2D tissue culture plastic (TCP), 2D COL, and 3D COL conditions. (e) Quantification of cell count through seven days of culture. (f) Cell morphology by F-actin staining (green) and nuclear (blue) visualization. (g) Quantification of cell density from the cell morphology images.
Figure 2 - Increased expression of ALP and activation of the PI3K/AKT pathway in osteosarcoma cultures within 3D COL matrices. (a) Alkaline phosphatase (ALP) activity of cells cultured in 2D TCP, 2D COL, and 3D COL over a 7-day culture period. Data is normalized to DNA content. (b) Gene expression was analyzed through RT-PCR of AKT/Protein Kinase B expression. (c) Protein expression through Western blot of phosphorylated-AKT (active) and total AKT. (d) Quantification of Western blot.
Figure 3 - Decreased chemosensitivity in OS cultures within 3D COL matrices. (a-c) Dose response data after 24 hours of treatment with methotrexate, doxorubicin, and cisplatin in the three culture conditions. (d-f) Extrapolation of half-maximal inhibitor concentrations (IC50) to each drug.
Figure 4 - Analysis of diffusivity, mesh size at day 7 of culture, and drug penetration in COL matrices. (a) Fluorescence recovery after photobleaching images. (b) Quantification of the diffusion coefficient from each fluorescently labeled particle within 3D COL hydrogels. The associated MATLAB script accompanies the data. (c) Second harmonic generation images of osteosarcoma cells cultured on 2D COL and 3D COL after 7 days of culture, with quantification of mesh size. (d) Qualitative assessment of drug penetration of doxorubicin through 2D TCP, 2D COL, and 3D COL culture conditions.
Figure 5 - Pre-treatment with regorafenib (REG), a tyrosine kinase inhibitor used in metastatic and relapsed osteosarcoma patients. (a) Live/Dead images and (b) quantification of cell viability. (c) Second harmonic generation images of cell-seeded 3D COL hydrogels after 7 days of culture with and without REG. (d) Quantification of the mesh size of the hydrogel. (e) Dose response data with and without REG pre-treatment after 24 hours of treatment with doxorubicin. (f) Dose response data in 2D TCP, 2D COL, and 3D COL with REG pre-treatment. (g) Extrapolation of IC50 values with and without REG pre-treatment.
Figure 6 - Elevated ABCG2, a drug efflux pump, expression in osteosarcoma cells cultured in 3D COL matrices. (a) Gene expression data of ABCG2 in the three culture conditions. (b) Protein expression of ABCG2 through Western blot. (c) Quantification of Western blot data.
Figure S1 - Characterization of COL matrix properties. (a) Rheology data of a temperature ramp over 3 minutes. (b) Frequency sweep of a fully formed COL hydrogel.
Figure S2 - Second harmonic generation imaging of acellular and cell-laden 2D COL and 3D COL hydrogels. Images of acellular cells (blue = nuclei, green = F-actin) in 2D and 3D COL culture. Quantification of mesh size.
Figure S3 - Quantification of cell viability on day 7 of culture in 2D TCP, 2D COL, and 3D COL.
Figure S4 - Images of cell viability of osteosarcoma cells on days 1, 3, and 7 in 3D COL. Green = live, red = dead.
Figure S5 - Montage of z-stack of cells within 3D COL. Green = live, red = dead.
Figure S6 - Quantification of COL mesh size from second harmonic generation images over a 7-day culture period.
Figure S7 - Quantification of metabolic activity on day 7 culture after REG pre-treatment (0 μM, 1 μM, and 5 μM).
Figure S8 - Impact of REG pre-treatment on osteosarcoma cells cultured on 2D TCP and 2D COL. (a) Live/Dead images (green = live, red = dead) with and without REG pre-treatment. (b) Cell viability quantification with and without REG pre-treatment.
Figure S9 - ABC transporter gene expression. Quantification of gene expression of (a) ABCC1 and (b) ABCB1.
Files and variables
All data is organized in the "Raw Data" folder and separated into folders corresponding to the Figures of the manuscript. Files are organized as raw Excel, raw images, and analyzed images.
Excel files are tabulated as "Raw" and "Analyzed" sheets to allow for reproducibility in data analysis. The "Raw" sheets are obtained from the source of data curation (spectrophotometer or imaging apparatus). "Analyzed" data are labeled with conditions and relevant formulas to allow for reanalysis.
Images are labeled in folders labeled as "Final" - indicating these are the images used for the publication - and "Raw" - indicating these images were used for data analysis.
File types
- Images as .tif, .tiff, and .PNG files
- Data sheets as .xlsx
Abbreviations
- 2D = two-dimensional
- 3D = three-dimensional
- COL = collagen
- SHG = second harmonic generation
- ALP = alkaline phosphatase
- AKT = protein kinase B
- pAKT = phosphorylated AKT (active)
- FRAP = fluorescence after photobleaching
- REG = regorafenib
- ABC = ATP-Binding Cassette
Code/software
Open software via MATLAB script used for Figure 4 FRAP data, referenced in the Methods section
Access information
NA