Summary Neutrophils are major inflammatory cells that rapidly infiltrate wounds to provide antimicrobial functions. Within the damaged tissue, neutrophil migration behavior often switches from exploratory patrolling to coordinated swarming, giving rise to dense clusters that further disrupt tissue architecture. This aggregation response is self-organized by neutrophil paracrine chemoattractant signaling (most notably of the inflammatory mediator leukotriene B4 [LTB4]). The coordination mechanism and possible evolutionary benefits of neutrophil swarms are elusive. Here, we show that neutrophil swarms require mutual reinforcement of damage signaling at the wound core. New biosensors and live imaging in zebrafish revealed that neutrophil chemoattractant synthesis is triggered by a sustained calcium flux upon contact with necrotic tissue that requires sensing of the damage signal ATP. This “calcium alarm” signal rapidly propagates in the nascent neutrophil cluster in a contact-dependent manner via connexin-43 (Cx43) hemichannels, which are mediators of active ATP release. This enhances chemoattractant biosynthesis in the growing cluster, which is instrumental for coordinated motion and swarming. Inhibition of neutrophil Cx43 compromises clearance of wound-colonizing P. aeruginosa bacteria and exacerbates infection-induced morbidity. Thus, cooperative production of alarm signals among pioneer clustering neutrophils fuels the growth of dense antimicrobial cell masses that effectively seal off breached tissue barriers from opportunistic pathogens.

Immune cells congregate at specific loci to fight infections during inflammatory responses, a process that must be transient and self-resolving. Cell dispersal promotes resolution, but it remains unclear how transition from clustering to dispersal is regulated. Here we show, using quantitative live imaging in zebrafish, that differential ligand-induced trafficking of chemokine receptors such as Cxcr1 and Cxcr2 orchestrates the state of neutrophil congregation at sites of tissue damage. Through receptor mutagenesis and biosensors, we show that Cxcr1 promotes clustering at wound sites, but is promptly desensitized and internalized, which prevents excess congregation. By contrast, Cxcr2 promotes bidirectional motility and is sustained at the plasma membrane. Persistent plasma membrane residence of Cxcr2 prolongs downstream signaling and is required for sustained exploratory motion conducive to dispersal. Thus, differential trafficking of two chemokine receptors allows coordination of antagonistic cell behaviors, promoting a self-resolving migratory response.

Migrating cells must interpret chemical gradients to guide themselves within tissues. A long-held principle is that gradients guide cells via reorientation of leading-edge protrusions. However, recent evidence indicates that protrusions can be dispensable for locomotion in some contexts, raising questions about how cells interpret endogenous gradients in vivo and whether other mechanisms are involved. Using laser wound assays in zebrafish to elicit acute endogenous gradients and quantitative analyses, we demonstrate a two-stage process for leukocyte chemotaxis in vivo: first a “search” phase, with stimulation of actin networks at the leading edge, cell deceleration, and turning. This is followed by a “run” phase, with fast actin flows, cell acceleration, and persistence. When actin dynamics are perturbed, cells fail to resolve the gradient, suggesting that pure spatial sensing of the gradient is insufficient for navigation. Our data suggest that cell contractility and actin flows provide memory for temporal sensing, while expansion of the leading edge serves to enhance gradient sampling.

No abstract

Migrating cells must interpret chemical gradients to guide themselves within tissues. A general premise has been that gradients influence the direction of leading-edge protrusions. However, recent evidence indicates that actin flows correlate better with directed motion than protrusions. A unified chemotaxis model that integrates the role of protrusions and actin flows and that accounts for in vivo cell motion patterns is lacking. Here, we provide direct experimental evidence of how neutrophils sense gradients in real-time in vivo and demonstrate a two-stage process: first a "search" phase, that requires actin network expansion by Arp2/3, whereby cells slow down, explore the environment and execute path corrections. This is followed by a "run" phase, that requires Myosin-II-driven contractility and fast actin flows, whereby cells accelerate and persist in the source direction. Thus, protrusive forces support chemotaxis by enhancing signal detection, while actin flows act as the ultimate output of gradient processing.