Associate Professor University of Maryland, Maryland, United States
Introduction: Tumor vascularization is an essential facet of cancer development and provides nutrients to cancer cells, while acting as a highway for cancer dissemination. Traditionally, this relationship between cancer cells and endothelial cells (ECs) in the metastatic initiation is investigated in vivo. However, these models can lack relevance to human systems, prevent real-time observation of mechanisms at play, and produce confounding results due to model heterogeneity. As a result, there is a lack of foundational knowledge regarding the mechanistic interaction underlying this phenomena that results in a 10% success rate when translating pre-clinical in vivo trials to patients; thus, more relevant, and viable, models for studying metastatic initiation must be created to overcome these limitations. Numerous in vitro models have been created to study the functional interplay between metastasis and tumor-associated angiogenesis in a controlled, sustainable, and relevant environment. However, these in vitro models can often oversimplify the relationship between tumors and their recruited ECs. Though these models allow for efficient testing, their oversimplification of the TME hinders mechanistic translatability to patients. Thus, a more complex in vitro model should be developed to replicate the cues associated with tumor vascularization and intravasation. We have created a spheroid-associated angiogenic vascularization system (SAVS) that incorporates these relevant spatial, structural, and mechanical cues in a more complex, controllable, and easily observable in vitro environment to elucidate this understudied facet of cancer progression. After prototyping, optimization, and validation, the SAVS was employed to delineate mechanistic phenomena of cancer intravasation into endothelial vessels.
Materials and
Methods: Cell culture and treatments: GFP-tagged MDA-MB-231 cells were cultured in DMEM supplemented with fetal bovine serum and pen/strep. RFP-tagged human umbilical vein endothelial cells (HUVECs) were cultured in PromoCell EGM-2, supplemented with 1% pen/strep. Spheroid preparation: Spheroids were prepared by adding 4000 cells into a round bottom 96-well plate. These cells were cultured in DMEM supplemented with methylcellulose to increase viscosity, and thus compaction of the spheroids. Compaction was additionally induced by centrifuging the plate. After five days of culture, these cells were used for invasion experiments. SAVS preparation: Positive molds for the SAVS were 3D printed and adhered to a silicone wafer. The SAVS were cast in PDMS from the positive mold and irreversibly adhered to a glass coverslip via conventional microfluidic device preparation. These devices were prepared for addition of cells by polymerizing collagen around two needles inserted into the SAVS. HUVECs were then added to a cell inlet to create an endothelial vessel surrounded by collagen. After allowing cells to adhere to the vessel wall, the device was submerged in media and placed on a rocker in an incubator to replicate flow. Two days later, spheroids were resuspended in a collagen solution, pushed through the vessel on the opposite inlet of the endothelial vessel, and polymerized. Imaging: The SAVS were imaged in a FV3000 confocal microscope equipped with live cell imaging capabilities. Following the initiation of cell invasion, timelapse images were acquired every 30 minutes for 24 hours.
Results, Conclusions, and Discussions: Numerous prototyping iterations of the SAVS were developed to fit the desired engineering requirements. Alterations were made to the size, shape, fabrication, cell seeding, and care of cells within the SAVS to allow for the observation of cellular behaviors associated with metastatic initiation. The most recent SAVS prototype is compatible with a twelve well plate and the cellular components can be viably cultured for at least three weeks, with vessel maturation, and visible flow (Figure 1A-C). When spheroids and ECs were cultured independently of the other within the SAVS, they exhibited significant increases in unilateral invasion in response to chemotactic gradients (Figure 2). MDA-MB-231 spheroids and HUVEC vessels were successfully co-cultured within the SAVS. For functional validation, cellular phenomena within the SAVS were compared to predefined in vivo cellular behaviors: 1) cellular invasion decreases when the collagen is polymerized at a higher density, 2) invasive capacity of spheroids and the endothelial cells are positively correlated to the metastatic potential of the cells comprising the spheroids, and 3) invasion of spheroids and endothelial cells given the distance between the vessel and spheroids (Figure 3). These findings validate the relevance of the SAVS as an in vitro model capable of recapitulating numerous cellular processes associated with the initiation of metastasis. Finally, the process of cancer cell intravasation in the endothelial vessel was observed within the SAVS, which will be further studied to provide insights into the understudied facet of metastasis. MDA-MB-231 co-cultured with HUVECs displayed the formation of clear and defined invasive protrusions polarized towards the endothelial vessel. From these protrusions, GFP-tagged MDA-MB-231s could be found within the endothelial vessel, demonstrating successful intravasation (Figure 4). These GFP expressing cancer cells were found inside the endothelial vessel in abundance, yet they do not readily extravasate. In short, the SAVS recapitulates numerous cellular phenomena associated with the initiation of metastasis which can be pharmacologically inhibited, while providing insight into the progression of underlying mechanisms of metastatic initiation.
Acknowledgements (Optional): NIH National Institute of General Medical Sciences - Maximizing Investigators' Research Award, The International Foundation for Ethical Research Graduate Fellowship
