Spatial Self-Organization Resolves Conflicts Between Individuality and Collective Migration

Xiongfei Fu, Setsu Kato, Junjiajia Long, Henry H. Mattingly, Caiyun He, Dervis Can Vural, Steven W. Zucker, Thierry Emonet

Preprint posted on April 05, 2018

Keeping the runners and tumblers together: Spontaneous spatial organization allows for the collective migration of phenotypically diverse groups

Selected by Amanda Haage

Categories: cell biology, microbiology

Why This Is Cool – Populations of E. coli cells can migrate collectively up chemoattractant gradients and they do so with a biased random walk produced by straight “runs” and random “tumbles.” It has also been previously established that E. coli that are genetically identical present with variable swimming phenotypes i.e. some cells run more and some cells tumble more than others. What Fu et al. set out to answer was how do these groups of cells travel together despite their individual differences in chemotactic ability? To answer this, they had to observe both collective and individual behaviors, so they used the smartly designed microfluidic device pictured below. The narrow channel allowed E. coli cells to migrate in bands up a gradient, towards an open chamber where cells could disperse and individual behavior could be tracked. Using this set-up, they were able to establish that although there is some selection for less tumbling in collectively migrating bands, phenotypic diversity is definitely still present. They then expanded a classic mathematical model1 for collective chemotaxis to account for individual diversities. This expanded model allowed them to predict how different swimming phenotypes could still travel together. Spatial sorting of cells, so that high tumbling coincides with a steeper gradient, allows the group to compensate for diversity. So the less efficient, more tumbley cells get sorted to the back of the group where the gradient is steeper, because the front runner cells, which perform better with small changes in gradient, metabolize a certain amount of the attractant, leaving a steeper gradient behind them. The steeper gradient keeps the tumblers more focused and they are able to keep up to a certain cut-off point. Fu et al. were then able to confirm these predictions experimentally with mixed levels of expression of the chemotaxis regulating phosphatase CheZ. In addition, they identified a possible role for oxygen availability in controlling this spatial sorting. E. coli need oxygen to metabolize these chemoattractants, but oxygen becomes less available with more cells, such as the conditions of the middle of the band (Figure 1). More oxygen is available at the back due to lower cell density, which increases metabolism, thus increasing the local gradient steepness even more and possibly allowing even the worst tumblers to keep up.

Figure 1. Collective migration of a phenotypically diverse clonal population. A) When concentrated at the bottom of a nutrient channel, motile E. coli cells emerge from the high cell density region and travel in bands along the channel by following gradients produced by their own attractant consumption. B) Microfluidic device used to quantify the band migration. Control gates along the channel (black vertical lines) are initially open (top), and later closed to capture different bands of cells in the observation chamber (bottom), where single cells are tracked to quantify the distribution of phenotypes within the band.


Why I Selected It – I was looking for preprints on the collective migration of eukaryotic cells when I “tumbled” across this work from Fu et al., but I am so glad I stuck with it. I didn’t know bacteria even underwent collective migrations. This work uses an excellent and high-throughput model system for studying phenotypic diversity and I think their results are widely applicable, as they state – “For example in migrating neural crest cells and in fish shoals, many organisms may follow a few more informed individuals2.” Reading this preprint made me question the phenotypic diversity I see in my own research, a phenomenon not often addressed in eukaryotic cell biology.

Open Questions –

  • Would you expect the spatial sorting of cells to be any different based on the type of chemotactic cue? Are there any non-consumable cues that could be tested, presumably to see a disruption in this process?
  • Do you think this spatial sorting is an active process? If you reversed the direction of the gradient mid-migration, would the organization of the band change?
  • If you continually removed a certain phenotype, say removed all the cells that fell off the band over multiple cycles of migration, would you eventually get a homogenous band with no spatial sorting? Would this impede or improve collective migration?

Related References –

  1. Classic model of collective bacteria cell chemotaxis
    1. Keller EF, Segel LA. Traveling bands of chemotactic bacteria: a theoretical analysis. J Theor Biol 30, 235-248 (1971).
  2. Collective cell migration review
    1. Mayor R, Etienne-Manneville S. The front and rear of collective cell migration. Nat Rev Mol Cell Biol 17, 97-109 (2016).
  3. Similar mechanisms operating in eukaryotic collective cell migration
    1. Tweedy L, Knecht DA, Mackay GM, Insall RH. Self-Generated Chemoattractant Gradients: Attractant Depletion Extends the Range and Robustness of Chemotaxis. PLoS Biol 14, e1002404 (2016).

Tags: mathematical model, migration

Posted on: 24th April 2018

Read preprint (No Ratings Yet)

  • Author's response

    Thierry Emonet shared

    1. We expect spatial sorting to be a generic outcome when cells of different gradient-climbing abilities climb the same gradient, whether it be a gradient of consumable chemoattractant, non-consumable chemoattractant, oxygen, temperature, etc. However, spatial sorting in an attractant gradient created by consumption is special in that the fastest cells can’t run far ahead of the slower ones because they need those cells to help create the gradient; without them, there is no gradient to climb.
    2. We think this spatial sorting is active in the sense that if the gradient reversed direction (not too fast), then the direction of sorting would reverse, as well. Any fast gradient climbers located at the bottom of the new gradient would quickly catch up with and pass the slow gradient climbers ahead of them, leading to reorientation of the sorting. However, although this is an active process, it is not hard-coded in the cells; spatial sorting emerges from the differences in individual gradient-climbing capabilities.
    3. Eliminating phenotypic diversity would be very difficult because every time a cell divides it will give rise to daughter cells that are not entirely identical. If under special circumstances we managed to form a band composed of a narrow distribution of phenotypes that is nearly homogeneous, we would expect cell diffusion to smear out spatial sorting, making the band effectively well-mixed. To what extend this band might perform better than a more diverse one is unclear and  a question we are actively investigating.


    Have your say

    Your email address will not be published. Required fields are marked *

    Sign up to customise the site to your preferences and to receive alerts

    Register here

    Also in the cell biology category:

    The cytoskeleton as a smart composite material: A unified pathway linking microtubules, myosin-II filaments and integrin adhesions

    Nisha Mohd Rafiq, Yukako Nishimura, Sergey V. Plotnikov, et al.

    Selected by Coert Margadant

    Quantitative, real-time, single cell analysis in tissue reveals expression dynamics of neurogenesis

    Cerys S Manning, Veronica Biga, James Boyd, et al.

    Selected by Teresa Rayon

    Profiling the surface proteome identifies actionable biology for TSC1 mutant cells beyond mTORC1 signaling

    Junnian Wei, Kevin K. Leung, Charles Truillet, et al.

    Selected by Rob Hynds


    Optogenetic dissection of mitotic spindle positioning in vivo

    Lars-Eric Fielmich, Ruben Schmidt, Daniel J Dickinson, et al.

    Selected by Angika Basant


    Excitable RhoA dynamics drive pulsed contractions in the early C. elegans embryo.

    Jonathan B Michaux, Francois B Robin, William M McFadden, et al.

    Selected by Sundar Naganathan

    Moving beyond P values: Everyday data analysis with estimation plots

    Joses Ho, Tayfun Tumkaya, Sameer Aryal, et al.

    Selected by Gautam Dey


    A limited number of double-strand DNA breaks are sufficient to delay cell cycle progression.

    Jeroen van den Berg, Anna G. Manjon, Karoline Kielbassa, et al.

    Selected by Leighton Daigh

    Optogenetic manipulation of medullary neurons in the locust optic lobe

    Hongxia Wang, Richard B. Dewell, Markus U. Ehrengruber, et al.

    Selected by Ana Patricia Ramos

    JNK-mediated spindle reorientation in stem cells promotes dysplasia in the aging intestine

    Daniel Hu, Heinrich Jasper

    Selected by Maiko Kitaoka

    ER-to-Golgi trafficking of procollagen in the absence of large carriers.

    Janine McCaughey, Nicola Stevenson, Stephen Cross, et al.

    Selected by Gautam Dey


    Template switching causes artificial junction formation and false identification of circular RNAs

    Chong Tang, Tian Yu, Yeming Xie, et al.

    Selected by Fabio Liberante

    Mechanosensitive binding of p120-Catenin at cell junctions regulates E-Cadherin turnover and epithelial viscoelasticity

    K. Venkatesan Iyer, Romina Piscitello-Gómez, Frank Jülicher, et al.

    Selected by Ivana Viktorinová

    Targeting light-gated chloride channels to neuronal somatodendritic domain reduces their excitatory effect in the axon

    Jessica Messier, Hongmei Chen, Zhao-Lin Cai, et al.


    High-efficiency optogenetic silencing with soma-targeted anion-conducting channelrhodopsins

    Mathias Mahn, Lihi Gibor, Katayun Cohen-Kashi Malina, et al.

    Selected by Mahesh Karnani



    Lydia R Heasley, Jennifer G DeLuca, Steven M Markus

    Selected by Hiral Shah

    An atlas of the aging lung mapped by single cell transcriptomics and deep tissue proteomics

    Ilias Angelidis, Lukas M Simon, Isis E Fernandez, et al.

    Selected by Rob Hynds


    Peculiar features of the plastids of the colourless alga Euglena longa and photosynthetic euglenophytes unveiled by transcriptome analyses

    Kristina Zahonova, Zoltan Fussy, Erik Bircak, et al.

    Selected by Ellis O'Neill