Poster S6 - A 3D Biofabricated Vascular Model for Elucidating Endothelial Cell Roles in Breast Cancer Progression Following Cardiovascular Disease

Introduction: Cancer and cardiovascular disease (CVD) are the leading causes of death in the U.S., affecting a combined 1.3 million people in 2021 (1). Recently, it has been shown in retrospective clinical trials that the two diseases are intertwined and new interest has arisen in understanding how a cardiovascular event increases risk of cancer development and recurrence (2,3) A central hypothesis has emerged that circulating factors released due to a cardiovascular event enter the bloodstream and create a pro-tumorigenic environment at the distal cancer site (4–6). The role that cells in the tumor microenvironment, such as endothelial cells (ECs), play in creating this pro-tumorigenic environment has not been well studied. ECs line the surface of blood vessels and are one of the first cell types the CVD circulating factor encounters at the tumor site. The inflammatory state of ECs changes in response to fluid flow and substrate stiffness (7,8). 3D biofabricated microfluidic vessels can be utilized to recreate the in vivo flow and stiffness conditions. Placing endothelialized vessels alongside cancer spheroids in a 3D matrix allows for the study of how ECs modulate cancer phenotypes after exposure to CVD circulating factors. The objective of this study is to design an in vitro model incorporating a 3D biofabricated vessel alongside breast epithelial spheroids to understand how CVD circulating factors impact the pro-inflammatory state of ECs and how these ECs modulate cancer progression. This model system will advance our understanding of the role of ECs in the link between CVD and cancerous states.

Materials and

Methods: To determine the optimal hydrogel matrix for EC growth, hydrogels consisting of 4% RGD alginate, 5% gelatin methacrylate (GelMA), or 10% GelMA were fabricated. After crosslinking, human umbilical vein endothelial cells (HUVECs) at 1x10^5 cells/mL were seeded on top of hydrogels. Cell viability of HUVECs was assessed via calcein-AM, ethidium homodimer-1, and Hoechst staining on the hydrogels. To fabricate the hollow vessel, a rectangular PDMS compartment was bonded to a glass slide. GelMA hydrogel precursor solution was placed in the compartment and a needle was inserted into the center. Following crosslinking, the needle was removed to create a hollow channel (needle-template method). Channel formation was confirmed by perfusion of a dye in the channel. For vessel endothelialization, HUVECs were added to both sides of the hollow channel and vessels were rotated three times in the incubator for 15-30 minutes and placed on a rocker after an hour. Cell coverage was assessed 48hrs post-seeding with calcein-AM and Hoechst staining. HUVEC phenotype in the channels was visualized via fluorescence microscopy with staining for actin (phalloidin), VE-cadherin, and Hoechst. Imaging of stained cells was performed via a widefield microscope for the hydrogel disks and a confocal laser scanning microscope for the endothelialized channels. To precisely place spheroids near the endothelialized vessel, the Sartorius CellCelector Flex robotic cell placement system was utilized to place 200μm diameter MCF10A breast epithelial spheroids in 5% uncrosslinked GelMA. After spheroid placement, GelMA was crosslinked and stained with Hoechst to visualize and quantify spheroid location.

Results, Conclusions, and Discussions: HUVEC cell viability analyses on the surface of hydrogel disks showed high cell viability (>95%) on the 5% and 10% GelMA hydrogels. Hydrogels consisting of 4% RGD alginate showed significantly lower viability (Figure 1A-B). A hollow perfusable channel was successfully fabricated using the needle-template method with 10% GelMA (Figure 1C). Perfusion of the channel and complete lumen formation was validated using a purple dye (Figure 1D). Three cell seeding concentrations in a 30G channel were assessed for endothelial cell coverage and it was observed that the higher cell seeding concentration resulted in increased cell coverage and endothelialization of the channel (Figure 1E). Furthermore, fluorescence staining of endothelialized channels showed actin cytoskeleton filaments and expression of endothelial marker VE-cadherin. Spheroid placement using the CellCelector Flex was found to be accurate at distances of 500μm. These results show that a GelMA hydrogel vessel can be fabricated and maintain high HUVEC viability. Furthermore, breast epithelial spheroids can be accurately placed within 500μm away from the channel wall.

A 3D biofabricated vessel was successfully fabricated using a GelMA hydrogel that supports high cell viability, complete lumen formation, vessel endothelialization, and normal endothelial cell phenotype. This endothelialized vessel can be utilized to understand the response of ECs to CVD vascular insults which could include pro-angiogenic effects, enhanced permeability, and secretion of inflammatory signals. The CellCelector Flex was used to precisely place breast epithelial spheroids at distances of 500μm or less which will enable us to interrogate the effect of distance in the crosstalk between spheroid and vessel. Development of this model will allow us to study the pro-tumorigenic mechanisms of endothelial cells after a vascular insult and the downstream crosstalk with breast cancer spheroids. This will enable a broader understanding of the role of endothelial cells and CVD circulating factors in the link between CVD and cancer progression that can be used to improve screening standards for patients with a CVD diagnosis. Ongoing and future work will include optimization of channel seeding and spheroid placement techniques enabling us to study endothelial-cancer crosstalk in a 3D model.

Acknowledgements (Optional): 1. Xu et al. NCBI Data Brief. 2022;(456).
2. Aboumsallem et al. AHA. 2020;9(2):e013754.
3. Koelwyn et al. JMCC. 2022;163:1-8.
4. Meijers et al. Circ. 2018;138(7):678-691.
5. Avraham et al. Circ. 2020;142:670–683
6. Koelwyn et al. Nat Med. 2020;26(9):1452-1458.
7. Chiu et al. Physiol Rev. 2011;91(1):10.1152/physrev.00047.2009.
8. Karki et al. Pulm Circ. 2018;8(2):2045894018773044