Poster Q4 - Chemical or Mechanical: How Cancer-Associated Fibroblasts Influence Angiogenesis in the Tumor Microenvironment

Introduction: The growth of blood vessels off pre-existing vasculature, also know as angiogenesis, is an essential function involved in wound-healing and development; however, angiogenesis can also promote tumor progression. Several cancer therapies target this process to stop disease spread, but many of these treatments fail to significantly increase overall survival. Our project focuses on how mechanical activity present in the tumor microenvironment (TME) may promote vessel growth, bypassing anti-angiogenic drugs used to treat cancer. These therapies target either vascular endothelial growth factor (VEGF) or VEGF receptor 2 (VEGFR-2), which are major promoters of angiogenesis along with vasculogenesis, or the growth of blood vessels without pre-existing vasculature. Highly contractile cancer-associated fibroblasts (CAFs) present within the TME cause strain on surrounding endothelial cells (ECs), which may mechanically activate VEGFR-2 present on the ECs. This strain causes phosphorylation of VEGFR-2 tyrosine residues including Y1054, Y1059, and Y1214, thus regulating activity of the receptor. To further understand how strain regulates VEGFR-2 phosphorylation and angiogenesis, we examined secretion of VEGF in 3D microtissue models of the TME used for angiogenesis studies and measured how long-term (>24hr) strain altered phosphorylation of VEGFR-2 and blood vessel growth. The objective of this study was to determine how VEGF from CAF conditioned media in combination with strain stimulation altered VEGFR-2 regulation and vasculogenesis.

Materials and

Methods: A VEGF ELISA was performed using standard protocols with media samples from human breast cancer derived CAFs and human umbilical vein endothelial cells (HUVECs) grown on tissue culture plastic. Furthermore, conditioned media was also harvested 3D fibrin gels seeded with HUVECs at either 100k or 200k cells per sample. All samples were initially fed with EGM-2 (Lonza, CC-3162) without any exogenous VEGF. Media samples were collected after 48hr of culture for VEGF analysis.

A Flexcell system was used to apply uniaxial strain at 9% elongation and 0.3Hz to human microvascular endothelial cells (HMECs) for 0 (No Strain Controls), 24, or 72hr before being seeded in 3D fibrin gels. To generate CAF conditioned media, CAFs were strained for either 0 (No Strain Controls) or 72hr before seeding at 50k cells per sample in 3D fibrin gels. Media from these samples was harvested as conditioned media to feed to HMECs as post-strain treatments. HMEC samples were grown for 7 days before being fixed and stained for CD31, VEGFR-2 pY1054/Y1059, and DAPI. FIJI was used to determine the positive areas of pY1054/Y1059 normalized to DAPI. AngioTool was used to quantify vessel growth marked by CD31.

Results, Conclusions, and Discussions: In a VEGF ELISA, CAFs showed a significantly higher secretion of VEGF compared to 2D HUVEC or 3D-100k HUVEC samples (p < 0.05) (Figure 1). Additionally, there was not a significant difference in the 100k and 200k HUVEC samples for total VEGF secretion. The result of HUVECs in 2D and 3D secreting less than 10% the amount of VEGF compared to CAFs suggests that HUVECs co-cultured with CAFs will react to mainly chemical stimuli from CAFs as opposed to other endothelial cell secretions.

CAFs were strained for 0 or 72hr then seeded into fibrin gels to create either No Strain Control (0ε) or Strain (ε) CAF conditioned media. HMECs were also strained for 0, 24, or 72hr before being seeded into fibrin gels. HMEC samples were fed with the CAF conditioned media or EGM-2 (unconditioned media). Our results show that phosphorylation of VEGFR-2 at Y1054/Y1059 does not appear to be significantly affected by CAF conditioned media, particularly in groups of HMECs that were not strained (0hr) or only strained for 24hr (Figure 2A,B). Furthermore, vessel growth was not significantly increased in HMECs treated with either CAF conditioned media compared to unconditioned media (Figure 2C). However, when HMECs were strained for 72hr, vessel growth and pY1054/Y1059 trended downwards in groups treated with unconditioned media compared to samples treated with both 0ε and ε CAF conditioned media. Overall, this data highlights that CAF secretions by themselves, whether from unstrained or strained CAFs, are insufficient to increase HMEC VEGFR-2 pY1054/Y1059 or vessel growth. This leads to the conclusion that other stimuli, such as mechanical forces produced by CAFs, are required for full VEGFR-2 activity and vasculogenesis. To further study this, we have started initial studies utilizing the contractility inhibitor verteporfin to decrease CAF contractility and identify the effects that strain by CAFs have on VEGFR-2 phosphorylation and vessel growth.

Acknowledgements (Optional):