Abstract
High-throughput dynamic imaging of cells and organelles is essential for understanding complex cellular responses. We report Mantis, a high-throughput 4D microscope that integrates two complementary, gentle, live-cell imaging technologies: remote-refocus label-free microscopy and oblique light-sheet fluorescence microscopy. Additionally, we report shrimPy, an open-source software for high-throughput imaging, deconvolution, and single-cell phenotyping of 4D data. Using Mantis and shrimPy, we achieved high-content correlative imaging of molecular dynamics and the physical architecture of 20 cell lines every 15 minutes over 7.5 hours. This platform also facilitated detailed measurements of the impacts of viral infection on the architecture of host cells and host proteins. The Mantis platform can enable high-throughput profiling of intracellular dynamics, long-term imaging and analysis of cellular responses to perturbations, and live-cell optical screens to dissect gene regulatory networks.
Significance Statement Understanding the dynamics and interactions of cellular components is crucial for biological research and drug discovery. Current dynamic fluorescence microscopy methods can only image a few fluorescent labels, providing a limited view of these complex processes. We developed Mantis, a high-throughput 3D microscope that maps interactions among components of dynamic cell systems. Mantis combines light-sheet fluorescence imaging of multiple fluorophores with quantitative label-free microscopy and is complemented by shrimPy, our open-source software for high-throughput data acquisition and high-performance analysis. Mantis enabled simultaneous 3D time-lapse imaging of 20 cell lines and quantitative analysis of responses to perturbations like viral infection at single-cell resolution. This approach can accelerate the analysis of cellular dynamics and image-based drug discovery.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
This is an author's version of a revised manuscript and has been improved based on feedback from anonymous reviewers. Key changes relative to v1 of preprint: 1. We edited the manuscript to clarify that this work reports the Mantis microscope and shrimPy smart microscopy engine. We report the training and performance of robust virtual staining models in a separate preprint (https://www.biorxiv.org/content/10.1101/2024.05.31.596901). We cite the robust virtual staining preprint here. 2. We report an online PSF calibration method for aligning the light-sheet arm and a deconvolution algorithm to correct residual aberrations. These improvements led to a 60% increase in the spatial resolution (Figure 2). 3. We illustrate higher spatial resolution with a new 6-channel movie of dividing A549 cells. The 6 channels are phase, orientation, mitochondrial protein labeled with fluorescent protein, lysosomes labeled with vital dyes, virtually stained nuclei, and virtually stained membrane (Figure 1). 4. We extend the analysis of the infected cells and demonstrate the clustering of the infected and uninfected cells with image-based phenotypes (Figure 5). 5. All figures and movies are refined based on the feedback of the reviewers.
