Cell Subtype-Dependent Generation of Breast Tumor Spheroids within 3D Mechanically-Tunable Microgels and Their Variable Responses to Chemotherapeutics

Abstract

Introduction Tumor spheroids have been considered valuable miniaturized three dimensional (3D) tissue models for fundamental biological investigation as well as drug screening applications. Most tumor spheroids are generated utilizing the inherent aggregate behavior of tumor cells, and the effect of microenvironmental factors such as extracellular matrix (ECM) on tumor spheroid formation has not been extensively elucidated to date. Herein, uniform-sized spherical microgels encapsulated with different subtypes of breast tumor cells, based on tumor aggressiveness, are developed by flow-focusing microfluidics technology. Mechanical properties of microgel are controlled in a wide range via polymer concentration, and their influence on tumor physiology and spheroid formation are shown to be highly dependent on cell subtype. Specifically, the formation of polyploid giant cancer cells is a key early step in determining initial proliferation and eventual tumor spheroid generation within microgels with varying mechanics. In addition, chemotherapeutic screening performed on these tumor spheroids in microgels also display significantly variable cytotoxic effects based on microgel mechanics for each cell subtype, further highlighting the importance of microenvironmental factors on tumor spheroid physiology.

Method Microfluidic fabrication of cell-laden microgels The flow-focusing microfluidic device was used to generate cell-laden droplets, which were photocrosslinked in situ to develop cell-laden microgels. The detailed device fabrication procedure is provided in detail elsewhere. Aqueous solution phases 1 and 2 (Aq1 and Aq2) both consisted of MGel and 0.2 % (w/v) Irgacure 2959® in phosphate buffered saline (PBS, pH 7.4). The channel geometry allowed Aq1 to enter Aq2 prior to droplet generation, with Aq1 becoming the core of a droplet (Fig. 1a). Oil phase (O) consisted of 20 % Span®80 (Sigma Aldrich) as a surfactant in mineral oil (Sigma Aldrich). In Aq1, breast tumor cells, MDA-MB-231, MCF-7, or SK-BR-3 (Korean Cell Line Bank,http://cellbank.snu.ac.kr), at 1 × 107 cells mL-1 were dispersed. The fluids were injected into the microfluidic device using electronic pumps (Legato®100, KD Scientific). Varying the ratio of flow rates of Aq and O resulted in the change in size of droplets (Aq1 and Aq2 were kept at the same flow rate). Here, the flow rates of Aq and O were 100 and 500 μL hr-1 respectively, resulting in 100 μm average diameter. The droplets generated from the microfluidic device was immediately irradiated with UV for 2 minutes (intensity: 200 mW, distance: 5 cm, emission filter: 250-450 nm, Model S1500, Omnicure®) to photo-crosslink the droplets to develop cell-laden microgels. The microgels were washed extensively with PBS to remove residual oil, and incubated in the cell culture medium at 37 °C under 5 % atmospheric CO2. The cell behavior and subsequent spheroid formation within microgels was monitored visually with inverted optical microscope at various times(XDS-3FL, Optika). The cell culture medium was RPMI1640 supplemented with 10 % heat inactivated fetal bovine serum (FBS), 2 mM L-glutamine, 25 mM HEPES and 25 mM NaHCO3 (all purchased from Thermo Fisher) for all cell types.

Result

The cytotoxicity of paclitaxel or cisplatin against MDA-MB-231 cells within microgels with varying rigidity (C2-C5) was assessed 1 day or 3 days after treatment by measuring their cell viability. For paclitaxel, the decrease in viability became noticeable above 50 nM at day 1 for all microgels except C2. At 100 nM, there was greater cytotoxicity at all microgels, as expected. Interestingly, there was a biphasic trend in cell viability, in which the viabilities significantly decreased for C3 and C5, 42 % and 34 %, respectively, compared to that of 2D monolayer culture, whereas the viability at C4 did not decrease as much, similar to that of 2D monolayer cu