Reverse and flick: Hybrid locomotion in bacteria.

M any bacteria are motile. They use one or more helical flagella as propellers, rotating them like the corkscrew on a wine bottle opener. Despite the limited morphological repertoire of the propulsive system, radically different movement strategies have evolved, likely reflecting the diversity of physicochemical conditions among bacterial habitats. In PNAS, Xie et al. (1) report on a newly discovered mechanism for turning used by Vibrio alginolyticus, an inhabitant of the coastal ocean: These monotrichous (“singlehaired”) bacteria change direction with a “flick” of their flagellum. Intriguingly, Xie et al. (1) show that less can be more when it comes to bacterial flagella: With its single flagellum, V. alginolyticus outperforms the multiflagellated Escherichia coli in climbing nutrient gradients (“chemotaxis”), suggesting that the flick is part of an advanced chemotaxis system. Our understanding of bacterial locomotion has long been driven and biased by the wealth of knowledge on E. coli, commonly found in animal intestines. E. coli is peritrichous, having four to eight flagella emerging from random points on its 2 × 1 μm hotdog-shaped body (2). Each flagellum is powered by a reversible rotary motor. When all motors spin counterclockwise (as seen from behind), hydrodynamic interactions cause the flagella to form a bundle that propels E. coli forward in a nearly straight “run” at ∼30 μm/s. When one or more motors switch direction, the bundle comes apart, causing a change in direction (“tumble”) before a new run begins. The angle of reorientation during a tumble is nearly random, with the new run only slightly biased in the direction of the old one. This “run-and-tumble” movement pattern is common among peritrichous bacteria, including the pathogen Salmonella typhimurium and the soil-dwelling Bacillus subtilis. Other bacteria, like V. alginolyticus, have a single flagellum and thus lack E. coli’s tumbling mechanism. When this flagellum rotates counterclockwise, it pushes the cell forward; clockwise, it pulls it backward. This “run-and-reverse” swimming is prevalent in the ocean: Shewanella putrefaciens, Pseudoalteromonas haloplanktis, and Deleya marina—all monotrichous—exhibit 180° reversals (3), as do ∼70% of marine isolates (4). At least one of these run-and-reversers, P. haloplanktis, outperforms E. coli in responding to nutrient patches, climbing gradients more rapidly and thus enhancing its exposure tohighnutrient concentrations (5). How can a simple back-and-forth movement result in high-performance chemotaxis, rather than causing the bacterium to endlessly retrace its steps? The answer might lie in a previously undetected component of the run-andreverse motion: the flick. Xie et al. (1) show that V. alginolyticus executes a cyclic, three-step motion: forward, reverse, and flick (Fig. 1). As expected, motor reversal after the forward segment induces an ∼180° reorientation. However, motor reversal after the backward segment is followed by a flick of the flagellum. Fluorescent labeling revealed that the flexible base of the flagellum develops a small kink whose angle is rapidly amplified by flagellar rotation. Although not a steering mechanism (which would imply active control over the change in direction), the flick induces a fast (<0.1 s) reorientation and the flagellum thus acts as a rudder, not just as a propeller. The change in cell orientation induced by a flick has a broad Gaussian distribution centered at 90°, and hence is very effective at randomizing swimming direction. This, then, is a hybrid swimming mode: partly run-and-reverse, partly random tumble. The resulting movement pattern is a random walk (red in Fig. 1) with dead-end forays, as can be seen by considering each step in the walk as a forward-backwardflick sequence (Fig. 1). Retracing its steps in the dead ends, after reversing and before flicking, reduces V. alginolyticus’s effectiveness in exploring its environment. Its diffusivity [210 μm/s, considering its 45 μm/s swimming speed and 0.31 s net run length (1)] is less than half that of E. coli [450 μm/s, assuming a 30 μm/s speed and 1 s run length (2)]. This comes at over twice the energetic cost for V. alginolyticus, because energy scales quadratically with swimming speed. If exploration were the main goal of the hybrid strategy, V. alginolyticus would be wasting its time and energy in the dead ends. It is possible that these drawbacks are outweighed by the savings of having just one flagellum, particularly in resource-poor environments such as the ocean. However, the fact that V. alginolyticus outperforms E. coli in terms of chemotaxis (1) implies otherwise. A similar observation for P. haloplanktis (5), along with evidence that this species also uses flicks (1), suggests that hybrid movement kinematics are more favorable for chemotaxis than run-and-tumble kinematics. Alternatively, these monotrichous marine bacteria possess considerably faster chemosensory responses than E. coli. Here I briefly explore these two hypotheses. Mathematical modeling has shown that the distribution of reorientation angles can markedly affect the quest for nutrients, as measured by the chemotactic velocity— the speed at which bacteria climb nutrient gradients (6). Different movement patterns can be modeled in terms of the persistence parameter, α, the mean of the cosine of the reorientation angle between runs. For positive persistence, the new direction is more likely to lie in the forward hemisphere: For example, E. coli’s tumbles are biased forward, with a mean angle of 68° and α = 0.33 (2). Entirely random tumbling results in α = 0 (no persistence), whereas a perfect reverser has α = cos(180°) = −1 (maximum negative persistence). In a linear nutrient gradient, the chemotactic velocity is predicted to increase with α (except near α ∼ 1) (6): Reversers should perform worse than E. coli. V. alginolyticus’s hybrid strategy increases persistence, because α = 0 for flicks [mean reorientation ∼90° (1)]. The alternation of reversals and flicks then suggests that α ∼ −0.5 for V. alginolyticus Fig. 1. Schematic of the hybrid movement pattern of Vibrio alginolyticus, which alternates reversals and flicks. Reversals are 180° reorientations, whereas flicks result in a broad distribution of reorientation angles with a mean of ∼90°. The effective trajectory is a random walk (red path) with some dead-end forays (after reversals and before flicks).