Physics of the Early Embryonic Cell Divisions: Feedbacks, Flows and Information

By Claudio Hernández-López and Aditya Singh Rajput

After a taxi ride from Heathrow to Buxted Park, the first day of the conference started on a high note as everyone was treated to a hearty lunch with a view of the sprawling gardens. Later, having found our way through the many corridors of the hotel, we had a few words from the organizers of the conference, Lendert Gelens and Julia Kamenz, and also from Laura Hankins on behalf of The Company of Biologists. All the speakers introduced themselves and pitched their research interests and out-of-the-lab hobbies. And so, the stage was set for the first session of the conference: actin and its biochemical regulation

Actin I: Swirling biochemical waves in natural and artificial cortices

Andrew Goryachev, professor at the University of Edinburgh, introduced everyone to Rho-Actin (RhoA) waves in the cortex of Starfish oocytes. A critical part of the cell cycle is cytokinesis, i.e. the ability of the cell to constrict its membrane and form two separate cells. These waves are a result of cortical excitability before this process. Andrew showed us their mesmerizing spatio-temporal evolution in fluorescence microscopy movies, displaying correlated activity in regions with sizes of about 10 microns [1]. Biochemical reaction-diffusion systems are able to produce spatio-temporal patterns and spread waves. However, depending on the molecular interactions between the chemical species, the properties of these waves can change. Two competing models have been proposed in the literature to explain these waves, differing in the presence or absence of an explicit chemical species that inhibits RhoA. Current experimental data measures either the total RhoA concentration or active RhoA concentration in space and time, and these two models cannot be distinguished from the observed dynamics of one concentration alone. Andrew presented new experimental data from simultaneous space-time recordings of total and active RhoA. These measurements indicate that changes in active RhoA precede changes in total RhoA, in agreement with a model including an explicit RhoA inhibitor. 

A major push in the field is also to study and recapitulate the intrinsic propensity of the actomyosin cortex to form patterns. In this direction, the lab of Jennifer Landino, an assistant professor at Dartmouth College, has been working on developing reconstituted cortices to study the emergence of spatio-temporal patterns by an interplay of F-actin and Rho. These Supported Lipid Bilayers (SLBs) [2] display waves and localized oscillations that have previously been observed in frog and starfish oocytes. This provides strong evidence of the self-organized nature of these patterns, with the mutual feedback between F-actin and Rho rendering the cortex excitable. The observed dynamics in SLBs present crucial differences compared to living cells, in particular, lacking periodic traveling waves. Changes in material properties after Xenopus extract addition allowed more than a single wavefront to propagate, thus bringing attention to the importance of the composition and design of these reconstituted systems.

Actin II: Geometry, flows, and deformations in early development

The second part of the session on actin started with Claudio Hernández-López, who recently finished his PhD at ENS Paris and is currently a postdoc in AMOLF Amsterdam, discussing the early development of Drosophila Melanogaster. The pre-gastrulation embryo, i.e. from cell cycles 1-14, is a syncytium: all the nuclei share a common cytoplasm. Before gastrulation, the nuclei migrate towards the cortex, forming their cellular membranes during cell cycle 14. The synchrony of the cell cycle along the 500-micron embryo depends on the nuclear density in space, hence, it is important that the nuclei achieve a uniform distribution in the embryo before cellularization. Previous experiments have shown that the expansion of the nuclear cloud along the anterior-posterior axis is mediated by cytoplasmic flows driven by cortical actomyosin contractions [3]. Claudio presented a new modeling framework comprising two fluids: an active gel (actomyosin), and a passive cytosol [4]. A mechanochemical coupling between the position of the nuclei and localized activation of myosin-II at the cortex reproduced previous experimental measurements on the flows and nuclear positioning. Remarkably, this self-organized positioning is robust, meaning that no matter the location of the initial nucleus after fertilization, the final nuclear distribution will remain uniform.

Aditya Singh Rajput, a PhD student at ICTS-TIFR, Bengaluru, continued the session by discussing asymmetric ingression in embryos. This phenomenon seems to be highly conserved amongst multiple phyla, with comb jellies being one of the major examples. The discussion began by looking at ingression in the nematode C. elegans and highlighting the emergence of myosin inhomogeneities that lead to this asymmetry. He then discussed his work on understanding this from the lens of active matter physics and treating the actomyosin cortex as an active fluid surface. Due to an emergence of differing timescales of ingression and cortical flow, the cytokinesis can be symmetric or asymmetric depending on the cortical contractility. The predictions from this theory also seem to hold true when compared to the dynamics of the first cleavage in the C. elegans embryo. 

Hervé Turlier, a research group leader at the Collège de France, discussed his group’s work to build physical and computational models for the early development of multicellular organisms. The group has been developing general computational tools to simulate [5], as well as reconstruct [6] cell and tissue surfaces, which could help understand, for example, cleavage patterns in early embryos. One of the foremost open problems in the field of developmental mechanobiology is the experimental measurement of forces in tissues as they undergo growth and morphogenesis. To this end, Dr. Turlier’s group has been working on analysing fully three-dimensional networks of cellular contacts to infer cellular stresses. This method, called foambryo, relies on mesh reconstruction to accurately capture the cellular geometry and then infer the relative tensions from the junctional lengths. These physical and computational tools not only provide a unique window into the mechanical behavior of early embryos but are also demonstrated to be generalizable and scalable to different species and various stages of development.

Information processing across scales

Switching gears, the session on information was kicked off by Rob Phillips with a discussion on language, words, and the information therein. By comparing the information content in examples from daily life- works of literature, encyclopedias, and entire libraries – with the information content of genomes, Rob brought to light the vast unknown regions of genome sequences that remain understudied till now. With his lab, he has been developing novel tools [7] to understand the transcriptional crosstalk between genes in a high-throughput manner to develop a genomic Rosetta stone that helps to bridge our understanding of the genotype-phenotype map. He talked about how this work attempts to infer the underlying gene regulatory networks by studying thermodynamic interactions and binding affinities and combining this with tools from information theory. In addition to the problem of deciphering the genotype-phenotype map itself, Rob highlighted how the many-to-one nature of this map comes with added complexities of different genes having varying degrees of impact on their associated phenotypes, which is also another aspect this toolbox can help bring to light in a more quantitative manner. 

Sophie de Buyl, professor at the Vrije Universiteit Brussel, closed this session by talking about the development of the Ascidian embryo. These filter feeders present two characteristics that make them particularly attractive for experimental studies in developmental biology. First, their cleavage pattern is invariant. Second, there is no cell migration, death, or embryo growth during their development. Previous theoretical work performed in her lab focused on the differentiation of neural tissue in the very early stages of development of this species. This process depends on the activation of ERK, which is regulated by the concentration of external signaling cues. In particular, the ERK activator is localized at the basal side of the outermost cells, and the ERK repressor is localized at the cell-cell junctions. Hence, the geometry of the cells is a relevant variable to study in relation to cell differentiation [8]. One overarching question in developmental biology is how living systems integrate signals from their environment to decide cell fate robustly, and the tools of information theory allow us to cast this problem in a tractable mathematical form. As a key result of this new study, Sophie showed that maximizing the information transmission from the external input to cell expression yields a predicted cell geometry that closely matches experimentally measured values. Furthermore, this maximal information transmission supports reliable differentiation between four different possible cell fates [9].

Finishing thoughts

The early embryonic cell divisions arise from an interplay between biochemical and mechanical cues, spanning multiple temporal and spatial scales. Given that regulatory pathways are generally less active at that stage, we feel that a comprehensive dialogue between theory and experiments on early development can be fertile ground for a broader understanding of not only cell division but also the emergence of complex behaviors in metazoan cells. Going beyond, perhaps a more biophysical understanding of the embryonic divisions can tell us something about the evolution of the diverse regulatory networks that we observe in extant organisms, potentially learning more about cell division in the first multicellular organisms.

References

1. Chomchai, Dominic A., Marcin Leda, Adriana E. Golding, George von Dassow, William M. Bement, and Andrew B. Goryachev. “Testing models of cell cortex wave generation by Rho GTPases.” bioRxiv
   (2024): 2024-04.

2. Landino, Jennifer, Marcin Leda, Ani Michaud, Zachary T. Swider, Mariah Prom, Christine M. Field, William M. Bement, Anthony G. Vecchiarelli, Andrew B. Goryachev, and Ann L. Miller. “Rho and F-actin self-organize within an artificial cell cortex.” Current Biology 31, no. 24 (2021): 5613-5621.

3. Deneke, Victoria, Alberto Puliafito, Daniel Krueger, Avaneesh V. Narla, Alessandro De Simone, Luca Primo, Massimo Vergassola, Stefano de Renzis, and Stefano di Talia. Self-organized nuclear positioning synchronizes the cell cycle in Drosophila embryos. Cell, 177(4) (2019), 925-941.

4. Hernández-López, Claudio, Alberto Puliafito, Yitong Xu, Ziqi Lu, Stefano Di Talia, and Massimo Vergassola. “Two-fluid dynamics and micron-thin boundary layers shape cytoplasmic flows in early Drosophila embryos.” Proceedings of the National Academy of Sciences 120, no. 44 (2023): e2302879120.

5. da Rocha, Hudson Borja, Jeremy Bleyer, and Hervé Turlier. “A viscous active shell theory of the cell cortex.” Journal of the Mechanics and Physics of Solids 164 (2022): 104876.

6. Ichbiah, Sacha, Fabrice Delbary, Alex McDougall, Rémi Dumollard, and Hervé Turlier. “Embryo mechanics cartography: inference of 3D force atlases from fluorescence microscopy.” Nature Methods 20, no. 12 (2023): 1989-1999.

7. Pan, Rosalind Wenshan, Tom Röschinger, Kian Faizi, Hernan G. Garcia, and Rob Phillips. “Deciphering regulatory architectures of bacterial promoters from synthetic expression patterns.” PLOS Computational Biology 20, no. 12 (2024): e1012697.

8. Bettoni, Rossana, Clare Hudson, Géraldine Williaume, Cathy Sirour, Hitoyoshi Yasuo, Sophie De Buyl, and Geneviève Dupont. “Model of neural induction in the ascidian embryo.” PLOS computational biology 19, no. 2 (2023): e1010335.

9. Bettoni, Rossana, Geneviève Dupont., Aleksandra Walczak, and Sophie de Buyl. “Optimizing information transmission in neural induction constrains cell surface contacts of ascidian embryos.” arXiv preprint (2024) arXiv:2410.18143.

(No Ratings Yet)
Loading…