(C-109) Novel 3D Ex-Vivo Model for Delivering Radial Compressive Stress

Breast cancer is a leading cause of cancer-related deaths worldwide. In 2020, it surpassed lung cancer as the most diagnosed cancer globally, with 2.3 million cases. The 5-year survival rate for women diagnosed with late-stage, distantly metastasized breast cancer is only 30%. Research into understanding why cancerous breast epithelial tissue becomes aggressive and invasive is critical to improving patient outcomes.

Abnormal cell migration can contribute to the spread of cancer cells and the development of metastases. There is growing evidence that physical cues play a significant role in tumor progression and metastasis. As a tumor progresses, two types of stresses develop in the tumor microenvironment (TMI): solid stresses from either abnormal growth or external stress, and stress from interstitial pressure. These stresses can cause blood and lymphatic vessels at the center of the tumor to collapse, creating a hypoxic and acidic TMI that can lead to more invasive phenotypes. However, the effect of these stresses on tumor phenotype is not well understood.

We are developing a novel ex-vivo method for testing compressive radial stresses on a three-dimensional cell suspension in a collagen matrix, which expands upon current two-dimensional models. By studying how normal and breast cancer cells respond to these stresses, we hope to gain insights into how physical cues contribute to tumor progression and metastasis, potentially leading to new treatments that improve patient outcomes.

Prototyping was achieved using SolidWorks® to create a negative mold ( Fig. 1) made of TPU-95A filament. The mold was printed on a LulzBot Taz Workhorse Edition Printer with a 0.2 mm layer height and 20% infill density. We sprayed MG Chemicals® Silicone Modified Conformal Coating Spray to ensure the PDMS was fully cured. SYLGARD™ 184 Silicone Elastomer (Dow Corning®) was mixed at a 10:1 wt. % ratio and placed in the desiccator for an hour. MCF10A breast epithelial cells with nuclear GFP spheroids were seeded at 1.5k cells with 20 μl of volume in an ultra-low adhesion (ULA) 96-well plate (Eppendorf®) and allowed to form overnight. The PDMS was cured at 65℃ for two and half hours and then bonded to the surface of a 12-well plate lid using an oxygen plasma treatment for five minutes. PDMS was cleaned using 1X PBS and 10 minutes of UV crosslinking, plasma cleaning for 5 minutes. 50 μl of 2 mg/ml collagen was deposited and allowed to polymerize for 30 minutes at 37°C. Then, two spheroids were seeded into the gel with the remaining 50 μl of collagen, allowing 45 minutes for polymerization (Fig. 2). 500 μl of cell culture media was then added into the wells and 4-inch zip ties were secured around the wells and pulled until taut. Cells were imaged using a Zeiss AxioObserver microscope (Carl Zeiss Microscopy, Germany). 

3D printing has revolutionized prototyping, allowing for rapid and cost-effective methods for creating models from CAD files. Fused deposition modeling enabled testing of multiple models before refining the mold. Experiments were conducted to determine the best method for generating the cell culture well plates. We tested PETG, PLA, and TPU-95A 3D-printing filaments with three treatment groups: no treatment, oxygen plasma treatment, and silanization. A flexible TPU-95A filament with silanization worked best to release the PDMS well plate (Table 1).

We aimed to create spheroid clusters of MCF10A cells by testing different cell groups and volumes to determine optimal conditions for spheroid formation. We first tested 10K cells at 35 μL, but they were fragmented. Thus, we decided to lower the amount of cells used in the plate, testing 1.5k and 1k cells per spheroid in 20, 25, and 35 microliters of media each (Fig. 3A). A Countess 3 Automated Cell Counter (Thermo Scientific®) was used to determine the live cell count and proper dilutions. After forming the spheroids, we carefully transferred them using a cut 100 μl pipette tip (Fig. 3B). We found that seeding 1.5K cells at 20 μl was optimal for forming spheroids using ULA plates and MCF10As.

The eventual purpose of our model is to apply specific amounts of compression to normal and cancerous cell spheroids within a 3D collagen matrix, observing their behavior using time-lapse microscopy allowing us to study the effects of radial compression on tumor progression and metastasis in a three-dimensional environment. We are currently collecting data of compressed/uncompressed systems using zip ties to identify any morphological and/or migratory changes in the cells. Future iterations of the device will incorporate tunable radial compression through compressed air, along with improving the imageability of the PDMS wells.

With these experiments, we plan to examine ECM deposition; cross-linking; fiber alignment, length, and size; cellular changes to actin, myosin, and Rho-ROCK signaling; and traction forces in response to varying compression with fluorescent beads and PIV. These studies will provide valuable insights into explaining the mechanisms by which radial compression contributes to cell behavior.

The Center for Engineering Mechanobiology (CEMB), the National Science Foundation (NSF), Ye Lim Lee¹, Akash Shaji², Miranda Gamino-Ornelas¹, and Amit Pathak1,2

 1. Washington University, Dept. of Biomedical Engineering, St. Louis, MO; 2. Washington University, Dept. of Mechanical Engineering & Materials Science, St. Louis, MO