Investigating the Role of Collagen-Binding Integrins in 3D Breast Cancer Spheroid Migration Models

Introduction:: The growth and metastatic spread of malignant tumors is controlled by the stromal microenvironment and the extracellular matrix (ECM). During tumor growth, matrix metalloproteinases (MMPs) secreted by various cell types remodel collagen in the ECM leading to the exposure of cryptic integrin binding sites. These sites play a role in tumor cell adhesion, migration, and proliferation through their interaction with cell surface receptors such as integrins. Recently, one such binding site termed HU177 was discovered in collagen types I-V and is recognized by integrin α10β1. The HU177 murine monoclonal antibody that can target the HU177 site is unable to bind to normal intact collagen, ensuring great binding specificity to its target. Preliminary phase-1 clinical trials demonstrated the possibility of anti-tumor activity while targeting this site. Therefore, a 3D spheroid model is currently being developed using MDA-MB-231 triple-negative breast cancer cells to explore the HU177 site in a system that closely mimics the human body through its cell-cell and cell-ECM contacts. The goal of this project is to determine how D93, the humanized version of HU177, impacts migration of cells in this 3D model.


Materials and Methods:: Spheroids were generated at an initial seeding density of 4,000 MDA-MB-231 cells per well with 2.5% Matrigel using the centrifugation method. They were embedded 72 hours after generation in 150 µL of 2 mg/mL collagen type I hydrogel in a clear, flat bottom 96 well plate (1 spheroid/well) using the sandwich method. The hydrogel contained an equal volume of acid-solubilized rat tail collagen type I and neutralizing solution (100 mM HEPES, pH 7.3 in 2X PBS) with the correct collagen concentration achieved by adding 1X PBS. The gel was allowed to fully polymerize for 90 minutes in an incubator (37°C, 5% CO2) before the addition of 150 µL of DMEM/F12 cell media and either D93 for the D93 experimental group (20 µg/mL), human IgG1 isotype control antibody for the antibody control group (20 µg/mL), or PBS for the non-antibody control group (n=5 for each group). Using a Cytation 5 Multimode Reader with a carbon dioxide regulator, a 3D stack of widefield fluorescent images was captured every 12 hours for a total duration of 72 hours. A single image projection was produced using the focus stacking method for each time point. A custom ImageJ macro was created to quantify the migration distance and area. Samples were fixed at the end of 72 hours with 4% paraformaldehyde and fluorescently labeled with DAPI and for D93 with Alexa Fluor 488 before imaging with the Cytation 5 using the DAPI and GFP filter cubes, respectively.


Results, Conclusions, and Discussions:: To ensure cellular migration was influenced only by the given treatment, it was vital to ensure spheroids are a consistent size. Using collagenase type 1 to break down the Matrigel binding the cells together, it was determined that spheroid composition was uniform at the time of embedment into the collagen type 1 hydrogels (Figure 1). During experimentation, the sandwich method of embedding the spheroids has proven effective with a 95.6% success rate of proper embedment (n=45) and cell migration observed at all angles from the spheroid. In our trials, there has been significant reduction of cellular migration of the D93 group compared to the IgG antibody control group at 24 hours while the non-antibody control group showed significant reduction in cellular migration compared to both the D93 and IgG antibody control groups at all time points (Figure 2). Furthermore, D93 labeling was achieved with complex patterns observed (Figure 3). In conclusion, a 3D spheroid model of human breast cancer has been developed and validated for use in exploring the effects of D93 on cellular migration. The significant early time reduction of cellular migration in the D93 group compared to the IgG antibody control group demonstrates high feasibility of D93 as a breast cancer therapeutic and ongoing studies are underway for hypothesis testing. We plan to expand our imaging acquisition to include labeling the cryptic site with a fluorophore and imaging via Second Harmonic Generation (SHG) microscopy, a label-free, collagen specific imaging modality. This would allow the individual sites to be examined in more detail using a co-localization image analysis approach. Lastly, with the recent addition of the carbon dioxide regulator for the Cytation 5, the plate can remain in the instrument for the total duration of the experiment (temperature control is already present). Thus, we plan on increasing our sampling speed to gain the potential of single cell tracking in the model.


Acknowledgements (Optional): : The authors would like to thank Drs. Jennifer Caron, Xianghua Han, and Allison M. Cross for their support. This work was funded by the Graduate School of Biomedical Science and Engineering and Agilent Research Gift #4624.


References (Optional): :

1. Gehler, S., Crit Rev Eukaryot Gene Expr. 2013;23(2):139-157.


2. Boraschi-Diaz, I., Front Phys. 2017;5.


3. Freimark, B., Mol. Immunol. 2007;44(15):3741-3750.


4. Caron, J. M., J. Pathol. 2016;186(6):1649-1661.


5. Charoen, K., Biomaterials. 2014;35(7):2264-2271.