A high throughput system reveals distinct segmentation clock phase responses in hiPSC-derived organoids

<jats:p>During somitogenesis, the vertebrate body axis segments into transient periodic structures known as somites. Somite formation is regulated by a multicellular molecular oscillator known as the segmentation clock. Recent advances in human induced pluripotent stem cell (hiPSC) culture have shown that hiPSC-derived somitogenesis organoids (somitoids) exhibit segmentation clock oscillations and can be produced at scale, making them an excellent model system for high throughput investigation of the mechanisms underpinning the segmentation clock. However, somitoids interact both biochemically and mechanically, and accurate high-throughput sampling of segmentation clock phases is required to exploit this system effectively. Here we address these challenges using an image-based, high-content screening workflow. Individual hiPSC-derived somitoids carrying a segmentation clock reporter are cultured in 384-well plates, and a programmable feeding schedule is used to initiate oscillations that are monitored using fluorescence microscopy. We develop an automated pipeline for image segmentation and data analysis, represent oscillations using a compact set of parameters, and construct predictive mathematical models to interpret data. We find that: (i) a staggered feeding schedule that sequentially initiates oscillations yields large numbers of somitoids at defined stages of the segmentation clock cycle; (ii) media exchange in established oscillations induces a Type 0-like phase response, resetting the segmentation clock to a state characterised by low Notch pathway transcription; and (iii) control wells in media exchange experiments exhibit a Type 1 phase response in which the segmentation clock is delayed non-uniformly across the cycle. Using a mathematical model of segmentation clock dynamics along the anterior-posterior axis, we show that periodic activation of a Type 1 phase response could segment a continuous phase gradient - an insight with potential implications for the determination of somite boundaries in vivo</jats:p>