News & Views | Published:

Systems biology

How bacteria choose a lifestyle

Nature volume 503, pages 476477 (28 November 2013) | Download Citation

In a bacterial population, some cells stay single and motile, whereas others settle down and form chains. A study now investigates the mechanisms that determine these outcomes. See Article p.481

Cells can switch identity several times during development. How do they decide to switch? There is much debate about the extent to which identity switching is a cell-autonomous decision as opposed to being driven by environmental signals. In this issue, Norman et al.1 (page 481) take an unusual approach to address this issue. They watch the soil bacterium Bacillus subtilis growing in an unchanging environment in which switching cannot be driven by extracellular signals. They focus on a simple switch — the transition from a single-cell swimming (motile) state to a sessile state that allows the bacteria to form a chain. Their findings provide invaluable insight into how an individual cell makes up its mind.

The authors grew B. subtilis in a microfluidic device consisting of several channels, each designed to support bacterial growth for days in a constantly replenishing medium that washes away any extracellular signals2. The bacterial strains studied express fluorescent 'reporter' proteins for both motile and sessile states, enabling the researchers to quantify the frequency and duration of cell-fate switching events under constant environmental conditions.

Norman and colleagues' detailed and precise characterization of hundreds of switching events reveals a critical difference in the transition from the motile to the sessile state and the switch in the other direction. The shift from the motile to the sessile state seems to be completely random and independent of how long the bacterium has been in the motile state. This motile state, therefore, is 'memoryless'.

The switch from the sessile to the motile state, however, is not random and is tightly timed: cells remain in the sessile state for roughly eight generations. The authors suggest that this memory serves a cellular function, ensuring that switching to the motile state, which breaks the chain almost immediately, does not occur too soon or with too much delay, which could result in some chains overflowing with millions of cells. The transition to the sessile state probably represents a trial period of multicellularity, which could be reinforced by environmental signals to commit the cells to forming a biofilm.

Norman and co-workers also explore the molecular mechanism that controls the cell-fate switch. It seems to involve a simple circuit consisting of only three proteins3 (Fig. 1). Specifically, the protein SinR represses the gene encoding another protein, SlrR; in turn, SlrR binds to and titrates SinR. Thus, these two proteins form a double-negative feedback switch. When SinR wins, the cell enters a motile state; when SinR loses, the cell becomes sessile. The third protein, SinI, affects which outcome wins by binding to, and inactivating, SinR.

Figure 1: To be sessile or to swim.
Figure 1

Norman et al.1 examine how single cells of the bacterium Bacillus subtilis choose between a motile (swimming) state and a sessile state. This choice is controlled by a simple double-negative feedback circuit involving three proteins: SinR represses SlrR, which represses SinR in turn. When SinR dominates, transition to the motile state occurs, whereas when SlrR dominates, the cells become sessile and form chains. A third protein, SinI, can initiate the switch to the sessile state by binding to and inactivating SinR.

The circuit seems to be modular, as the authors find that SinI is responsible for the memoryless entry into the sessile state. Once the bacteria are in the sessile state, however, SinI is no longer relevant, and the memory is set by the SlrR–SinR feedback loop. Such modularity has also been observed in another B. subtilis circuit that controls a developmental switch. Under stress conditions, B. subtilis can transiently enter a competent state, allowing it to take up external DNA4. The core circuitry that controls entry into the competent state has only a few components, similar to the SlrR–SinR–SinI network. The competence circuit is modular because one component regulates the frequency of transitions into the competent state, whereas another component determines how long a cell remains in this state4.

It is unclear what advantage, if any, such modularity has for the cell. Can having independent control of the initiation and duration of differentiation events enable the cell to adapt to independent selective pressures during evolution? And it remains to be seen whether such modularity is a general feature of circuits that control cell-identity switching.

The authors also raise questions about how the SlrR–SinR–SinI circuit controls cell-fate switching in B. subtilis. How noise, or variability, in one of the circuit components drives initiation into a sessile state remains unclear. Although initiation requires SinI, it is not known which circuit component exhibits random fluctuations to drive the random switch into the sessile state, or how these fluctuations are generated. It would be interesting to test the hypothesis that memory in state switching allows a trial window of multicellularity that is reinforced by environmental signals. One approach could be to examine what effect extending or reducing the memory of the sessile state has on biofilm formation.

Norman and colleagues' SlrR–SinR–SinI circuit joins a growing list of bacterial simple genetic circuits that have been shown to control surprisingly complex cellular dynamics. Such circuits often consist of only three or four proteins but can generate pulses5, excitable dynamics4 and robust oscillations6. Might simple genetic circuits generate a similar wealth of regulatory dynamics in plants and animals? Results in cows suggest7 that the concept of memory in state switching could be quite general. Research honoured with the 2013 Ig Nobel prize in probability showed that, in cows, the standing (motile) state is memoryless, whereas the lying down (sessile) state is timed, just as in B. subtilis.


  1. 1.

    , , & Nature 503, 481–486 (2013).

  2. 2.

    et al. Curr. Biol. 20, 1099–1103 (2010).

  3. 3.

    , , & Genes Dev. 24, 754–765 (2010).

  4. 4.

    , , , & Science 315, 1716–1719 (2007).

  5. 5.

    , , , & Science 334, 366–369 (2011).

  6. 6.

    , & Adv. Genet. 74, 13–53 (2011).

  7. 7.

    , , , & Appl. Anim. Behav. Sci. 124, 1–10 (2010).

Download references

Author information


  1. James C. W. Locke is in the Sainsbury Laboratory, University of Cambridge, Cambridge CB2 1LR, UK.

    • James C. W. Locke


  1. Search for James C. W. Locke in:

Corresponding author

Correspondence to James C. W. Locke.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing