Optimization of Factor Combinations for Stem Cell Differentiations on a Design-of-Experiment Microfluidic Chip

Directed differentiation of stem cells plays a vital role in cell replacement therapy. Many activators and inhibitors targeting different signaling pathways have been identified to contribute to each step of differentiation. Most studies relied on empirically optimizing the combinations of the aforementioned factors for each step to optimize the efficiency of differentiation, which are time-consuming and nonsystematic. Design-of-experiment (DOE) is a powerful strategy to identify the critical combinations from multiple factors systematically. However, it is prohibitively complicated for typical laboratories, given a large number of potential combinations. Here, we develop a multilayer polymethyl methacrylate-based, reusable microfluidic chip to directly facilitate the DOE in the differentiation of stem cells. The chip consists of an inlet layer and multiple disperse layers. Different solutions are injected simultaneously to the chip through the inlet layer. Subsequently, the channels in the disperse layers split and recombine the flow streams to generate solution combinations based on hard-wired DOE designs. We demonstrated that it is in quantitative agreement with the designs using fluorescent dyes. Moreover, we constructed a human-induced pluripotent stem reporter cell line to improve the consistency of the cellular state measurements and use the chip to identify critical factors for cell differentiation to definitive endoderm (DE). We found that the differentiation efficiencies under various factor combinations are significantly different, and CHIR99201 and GDF8 are the most critical factors for differentiation to DE. Our method is potentially applicable to the optimization of factor combinations for multi-step stem cell differentiation and combinatorial drug screening.