(C-110) Increased Collagen Fiber Density Associated with Metastatic Ovarian Cancer Promotes Tumor Cell Adhesion

Introduction:: High grade serous ovarian cancer (HGSOC) is the most common and deadly gynecologic cancer with a 5-year survival rate of < 50%. In HGSOC, cells metastasize as single-cells and spheroids in a transcoelomic fashion from the primary tumor to distal metastatic sites within the peritoneal cavity¹. These distal sites undergo alterations with disease progression; specifically, fibrillar collagen, fibronectin (FN), and stiffness are all increased in metastatic tissue². The impact of these alterations to the distal site ECM on incoming HGSOC single-cell and spheroid behaviors has yet to be investigated. Therefore, we will interrogate how Collagen I (Coll) impacts single-cell adhesion and test whether the biochemical or biophysical aspects unique to Coll can increase adhesion to accelerate metastasis. To this end, we aim to model the increase in fibrillar collagen using representative in-vitro models to investigate how Coll impacts HGSOC adhesion during transcoelomic metastasis. We modeled the presence of fibrillar collagen by using a gelatin methacrylate (GelMa) and Collagen I (GelMa/Coll) interpenetrating network (IPN) model³. Using the GelMa backbone, we can regulate stiffness while adding equal concentrations of fibrillar Coll vs denatured Coll (gelatin, which has the same sequence but is non-fibrillar). To delineate the biochemical and biophysical aspects of collagen we developed synthetic polyethylene glycol (PEG) constructs functionalized with adhesive ligands representative of adhesive sites unique to fibrillar collagen (GFOGER). Using these representative constructs, we will be able to interrogate the effect of fibrillar collagen presentation on incoming HGSOC cell adhesion and understand what aspects of fibrillar collagen drive adhesion.


Materials and Methods:: The GelMa/Coll and GelMa/Gel IPN constructs were developed to be at equal stiffness (~12 kPa) with 2.5 mg/ml of Coll I or gelatin, respectively, added to a 5% w/v GelMa background. To make synthetic PEG constructs we utilized a 4-arm PEG maleimide that was functionalized with cysteine-terminated adhesive peptides and subsequently crosslinked with a PEG-dithiol crosslinker. Using this formulation, we developed “mimetic” PEG gels by modeling the concentration of GFOGER in Coll I (37 µM), or RGD in gelatin (2 mM), or a blend of the two peptides that is representative of the GelMa/Coll IPN. To test the impact of increasing concentrations of adhesion ligands, we functionalized PEG gels with 100 µM or 200 µM of either RGD or GFOGER as well as a 1:1 blend at 200 µM total. These PEG gels and GelMa gels were then seeded with three representative HGSOC lines (OVCAR3, OVCAR8 and OV90) at 100,00 cells/cm²for three hours. Adhesion was evaluated by washing twice with PBS, followed by staining for the nuclei with Hoechst and counting the number of cells adhered per 10x frame.


Results, Conclusions, and Discussions:: Single-cell HGSOC adhesion was increased on GelMa/Coll substrates compared to GelMa/Gel across the three representative HGSOC lines (Fig. 1 A). HGSOC single-cell morphology at 24 hours also displayed significantly more spread morphology on GelMa/Coll as compared to GelMa/Gel. Single-cell adhesion on PEG constructs displayed no significant differences in response to changes in ligand identity (RGD or GFOGER), however, with increases in ligand concentration there were increases in adhesion in some lines (Fig. 1B). These data suggest that Coll fiber presentation helps to promote metastatic progression by supporting single-cell adhesion. Furthermore, from our adhesion experiments on PEG gels devoid of fibers we found that Coll-specific ligand (GFOGER) presentation does not confer increased HGSOC adhesion as compared to adhesion on PEG gels functionalized with equal concentrations of RGD. Therefore, we can conclude that Coll fiber presentation may promote HGSOC adhesion through a mechanism driven by the biophysical factors unique to Coll fibers. In future studies we will develop a model for the Coll biophysical fiber structure that incorporates the gelatin adhesion sites. Specifically, a gelatin/PCL nanofiber construct will be developed and subsequently embedded within a GelMa background to replicate our GelMa/Coll fiber network without the Coll-specific adhesive sequences. We would also like to examine other ECM proteins that are altered due to disease progression, such as FN, along with inclusion of other key cell types typically implicated in HGSOC progression.


Acknowledgements (Optional): : Funding for this study was provided by NIH 5R01CA232517. UW Soft Materials Characterization Lab and UW Optical Imaging Core were used to help analyze samples.

References (Optional): : (1) Tan, D. S., (2006). The Lancet. Oncology, 7(11), 925–934 (2) Pearce, O. M. T., (2018). Cancer discovery, 8(3) (3) Berger, A. J., (2017). Biomaterials, 141, 125–135.