Computable Early C. elegans Embryo with a Data-driven Phase Field Model

Xiangyu Kuang, Guoye Guan, Ming‐Kin Wong, Lu-yan Chan, Zhongying Zhao, Chao Tang, Lei ZhangbioRxiv 2020 · DOI

TL;DR

This paper presents a data-driven phase-field model that accurately simulates the early development of C. elegans embryos. By integrating experimental 3D imaging data, the model reproduces cell behavior and reveals the critical roles of cell division timing, orientation, and cell-cell attraction in robust embryonic morphogenesis.

What Problem Does This Paper Address?

Understanding how the complex, dynamic process of embryonic morphogenesis arises from the coordinated actions of individual cells remains a fundamental challenge in developmental biology. While molecular mechanisms are increasingly understood, how cell proliferation, migration, and mechanical interactions collectively drive the formation of precise three-dimensional structures is less clear. The C. elegans embryo, with its invariant cell lineage and reproducible development, serves as an excellent model system to investigate these principles. Previous computational models have often relied on numerous parameters difficult to measure or have oversimplified cellular behavior. This work aims to bridge this gap by developing a more comprehensive, data-driven model that can accurately predict embryonic development and uncover the underlying physical factors.

Methods

The authors employ a data-driven phase-field model to simulate the early development of C. elegans embryos. This approach uses phase fields to represent the boundaries and states of individual cells within a computational domain. The model incorporates several physical forces:

  • Cell surface tension: Drives cells towards a spherical shape to minimize surface energy.
  • Cell-eggshell and cell-cell repulsion: Prevents cells from overlapping.
  • Cell-cell attraction: Models adhesive forces between cells, crucial for maintaining contact and tissue integrity.
  • Cell volume constriction: Ensures cells maintain their prescribed volume.

Crucially, the model is data-driven, meaning key parameters like cell division timing, orientation, and volume segregation ratios are derived from experimental 3D time-lapse imaging of C. elegans embryos. Cell-cell attraction is initially set to zero and then progressively introduced and refined to match experimental observations. Cell division is modeled as an instantaneous bisection of the phase field, with the orientation and ratio determined by experimental data. The model is validated by comparing simulated cell locations, shapes, and contact relationships against experimental measurements.

Datasets

The study utilizes 3D time-lapse confocal microscopy imaging data from previously collected datasets of wild-type C. elegans embryos. Specifically, data from 18 embryo samples were used, including:

  • Embryos with GFP-labeled nuclei for cell tracing and lineage.
  • Embryos with both mCherry-labeled membranes and GFP-labeled nuclei for detailed 3D cell morphology and position.

This data was used to quantify cell-resolved developmental properties such as cell location, shape, volume, surface area, and cell-cell contact relationships.

Key Insights and Ideas

  • The "computable embryo" is within reach: By integrating experimental data into a physics-based model, the authors demonstrate that early C. elegans embryogenesis can be accurately simulated, suggesting a path towards truly computable developmental systems.
  • Cell division timing is critical for structural integrity: Postponing cell division can lead to a "planarization" defect, where the 3D structure collapses into a 2D state, highlighting the importance of timely cell division for maintaining robustness.
  • Asymmetric cell-cell attraction is key: The model reveals that not all cell-cell attractions are equal. Specific asymmetric attractions, particularly between the ABpl and EMS/P2 cells, are essential for correctly forming the 4-cell stage structure and influencing later developmental paths.
  • A "mechanoconnectome" governs development: The authors propose this concept to describe the conserved contact map among specific cells, which evolves and drives morphogenesis through cell-cell mechanical interactions, guided by programmed cell division timing, orientation, and attraction.
  • Cell division orientation is a powerful regulator: The direction and ratio of cell volume segregation during cytokinesis significantly impact the developmental path, demonstrating its role in establishing spatial patterns.

Key Results

  • The phase-field model successfully reproduced the morphological evolution and cell-cell contact relationships of C. elegans embryos from the 1-cell to the 8-cell stage.
  • The model predicted and experimental validation confirmed that asymmetric cell-cell attraction is crucial for accurate cell-cell contact area formation, particularly between EMS and P2 cells.
  • Simulations showed that timely cell division, triggered by reaching a quasi-steady state, is essential for maintaining the 3D structure and preventing planarization defects.
  • The study identified cell division timing, cell division orientation, and cell-cell attraction matrix as three key physical factors that regulate the developmental path.
  • The model accurately captured the time scale of embryonic development, with a high goodness of fit (R² = 0.9993) between simulated and experimental durations.

Limitations and Future Work

  • Computational cost: As the number of cells increases in later developmental stages, enumerating all possible developmental paths becomes computationally prohibitive. The authors suggest that future work could explore computational approaches for constructing solution landscapes to identify all possibilities.
  • Model simplification: The current model simplifies cell division into an instantaneous event, ignoring the elongation and cytokinesis process. For later stages with more cells, more complex biochemical and biophysical mechanisms might need to be incorporated.
  • Experimental validation of specific mechanisms: While the model provides strong predictions, further in vivo experiments are needed to fully elucidate the detailed functions of genes like pad-1 and to precisely quantify adhesion asymmetries.
  • Generalizability: The authors propose that the phase-field framework could be extended to simulate other multicellular systems, including synthetic biology applications, and to investigate more complex phenomena like contraction and polarization in larger embryos.

Summary updated

Abstract

SUMMARY Morphogenesis is a precise and robust dynamic process during metazoan embryogenesis, consisting of both cell proliferation and cell migration. Despite the fact that much is known about specific regulations at the molecular level, how cell proliferation and migration together drive the morphogenesis at the cellular and organismic levels is not well understood. Here, using Caenorhabditis elegans as the model animal, we present a data-driven phase field model to compute the early embryonic morphogenesis within a confined eggshell. By using three-dimensional time-lapse cellular morphological information generated by imaging experiments to set the model parameters, we can not only reproduce the precise evolution of cell location, cell shape and cell-cell contact relationship in vivo , but also reveal the critical roles of cell division and cellcell attraction in governing the early development of C. elegans embryo. In brief, we provide a generic approach to compute the embryonic morphogenesis and decipher the underlying mechanisms.
BibTeX
@misc{kuang2020computable,
  author = {Xiangyu Kuang and Guoye Guan and Ming‐Kin Wong and Lu-yan Chan and Zhongying Zhao and Chao Tang and Lei Zhang},
  title = {{Computable Early <i>C. elegans</i> Embryo with a Data-driven Phase Field Model}},
  year = {2020},
  doi = {10.1101/2020.12.13.422560},
  howpublished = {bioRxiv},
}

Loading related content...

Figure