Label-free imaging of 3D pluripotent stem cell differentiation dynamics on chip

• Published in: Cell reports methods Vol. 3; no. 7; p. 100523
• Main Authors: Atwell, Scott, Waibel, Dominik Jens Elias, Boushehri, Sayedali Shetab, Wiedenmann, Sandra, Marr, Carsten, Meier, Matthias
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 24-07-2023; Elsevier
• Subjects: 3D cell culture technology; AI imaging; Cell Culture Techniques - methods; Cell Differentiation; cell state prediction; Humans; Induced Pluripotent Stem Cells; microfluidics; Microfluidics - methods; Pluripotent Stem Cells; stem cells-on-chip; tracking single cells; 3D cell culture technology; AI imaging; tracking single cells; CP: Systems biology; CP: Imaging; stem cells-on-chip; cell state prediction; microfluidics
• ISSN: 2667-2375; 2667-2375
• DOI: 10.1016/j.crmeth.2023.100523

Massive, parallelized 3D stem cell cultures for engineering in vitro human cell types require imaging methods with high time and spatial resolution to fully exploit technological advances in cell culture technologies. Here, we introduce a large-scale integrated microfluidic chip platform for automated 3D stem cell differentiation. To fully enable dynamic high-content imaging on the chip platform, we developed a label-free deep learning method called Bright2Nuc to predict in silico nuclear staining in 3D from confocal microscopy bright-field images. Bright2Nuc was trained and applied to hundreds of 3D human induced pluripotent stem cell cultures differentiating toward definitive endoderm on a microfluidic platform. Combined with existing image analysis tools, Bright2Nuc segmented individual nuclei from bright-field images, quantified their morphological properties, predicted stem cell differentiation state, and tracked the cells over time. Our methods are available in an open-source pipeline, enabling researchers to upscale image acquisition and phenotyping of 3D cell culture.
[Display omitted]
•Large-scale integration of 3D stem cell cultures on a microfluidic chip•An AI imaging tool is developed to predict nuclei from bright-field image stacks•Prediction of stem cell states from bright-field images along a time trajectory•Tracking of single nuclei in 3D during stem cell differentiation
Our work aims to meet the increasing demand for high-resolution single-cell phenotyping in 3D stem cell cultures. During the parallelization of 3D cell cultures, we faced the problem of acquiring high-resolution or high-time-resolved image data for phenotyping. However, single-cell resolution is required to characterize the heterogeneous cell populations as, for example, encountered within stem cell differentiation experiments. To address this challenge, we employed label-free imaging technology to stain nuclei in situ from 3D bright-field image sets, which reduces image acquisition time and enables the inference of cell types and velocity. The presented approach thus enables dynamic large-scale screening of 3D stem cell cultures based on bright-field imaging.
Atwell et al. develop an AI imaging tool to predict nuclei within 3D stem cell cultures from bright-field images. Combining the AI imaging strategy with a microfluidic large-scale integration cell culture chip platform, we are able to predict cell states, track single nuclei, or decrease image acquisition time for cell screening.