Back to 2016 Annual Meeting

Background: A crucial step in the progression of cancer involves the transendothelial migration of tumor cells into vessels, a process that is facilitated by the interactions of malignant cells with their environment. Surrounding ECM in vivo has a profound effect on malignant invasion of cancer cells. Present studies suggest extensive ECM remodeling by malignant cells promote metastatic invasion. Traditional two-dimensional (2D) cell culture systems employed to study metastatic behavior have been limited by non-anatomic arrangement of cells. These in vitro models fail to account for the presence and interaction of vascular cells in proper anatomic configuration in their surrounding extracellular matrix (ECM). Thus, we aim to use our 3D platform to recapitulate proper in vivo vascular physiology and alter collagen bulk densities to study factors that influence tumor progression, vascular remodeling and the effect on ECM mechanics. By increasing collagen protein concentration to alter ECM stiffness, we can study how malignant tumor cells will behave in different microenvironments due to their surrounding ECM.
Methods: “U-shaped” Pluronic F127 fibers, 1.5 mm in diameter, were sacrificed in type I collagen, creating a central looped microchannel with 1x10⁶ cells/mL encapsulated MDA-MB-231 cells in the collagen bulk. Constructs were fabricated with various densities. Twenty-four hours following fiber sacrifice, a 100μL suspension of 5 x 10⁶ cells/mL of normal human dermal fibroblasts (NHDF) and 5 x 10⁶ cells/mL of human aortic smooth muscle cells (HASMC) was seeded into the microchannel. The following day, an additional 100μL cell suspension of 5 x 10⁶ cells/mL of human umbilical vein endothelial cells (HUVEC) and human placental pericytes (HPLP) was seeded into the microchannel. After 3, 7, and 14 days of culture, scaffolds were fixed and processed for histology and multiphoton microscopy (MPM). Mechanical testing was completed by ElectroForce-3200 Series III.
Results: After all time points, control constructs (MDAMBneg) exhibited microchannels consisting of anatomically correct robust vascular channel lining with increasing proliferation. However, in cancer constructs (MDAMBe) after 7 days, degradation of the vascular lining and aberrantly organized HUVEC and HASMC were present. IHC and MPM imaging revealed the presence of MDA-MB-231 cells invading the endoluminal lining. In constructs fabricated with a lower density collagen bulk, cancer cells migrated further from the collagen bulk into the endoluminal lining and had greater matrix disruption. Constructs with cancer cells exhibited a higher elastic modulus (greater stiffness) than constructs lacking cancer cells and cancer cells significantly increased the elastic modulus in lower collagen bulk density constructs.
Conclusion: Using our novel platform, we have demonstrated the interactions between tumor cells and vascular cells, as well as the mechanical changes tumor cells induce in the extracellular matrix. This model overcomes the limitations of previous 2D and 3D culture models and may be used to investigate any type of tumor cell and can lead to the further study of breast cancer signaling pathways, as well as provide an effective patient-specific platform for the study of metastatic invasion.