Spatial control of actin polymerization during neutrophil chemotaxis

Abstract

Neutrophils respond to chemotactic stimuli by increasing the nucleation and polymerization of actin filaments, but the location and regulation of these processes are not well understood. Here, using a permeabilized-cell assay, we show that chemotactic stimuli cause neutrophils to organize many discrete sites of actin polymerization, the distribution of which is biased by external chemotactic gradients. Furthermore, the Arp2/3 complex, which can nucleate actin polymerization, dynamically redistributes to the region of living neutrophils that receives maximal chemotactic stimulation, and the least-extractable pool of the Arp2/3 complex co-localizes with sites of actin polymerization. Our observations indicate that chemoattractant-stimulated neutrophils may establish discrete foci of actin polymerization that are similar to those generated at the posterior surface of the intracellular bacterium Listeria monocytogenes. We propose that asymmetrical establishment and/or maintenance of sites of actin polymerization produces directional migration of neutrophils in response to chemotactic gradients.

Figure 1. Polarization of a neutrophil in response to a gradient of chemoattractant

a–d, Nomarski images of an unpolarized neutrophil responding to a micropipette containing 10 μM FMLP (white circle) at a, 5 s, b, 30 s, c, 81 s, and d, 129 s.

Figure 2. Spatial distribution of incorporation of TMR–actin in a chemoattractant-stimulated permeabilized neutrophil

a–c, Neutrophil exposed to a uniform concentration (20 nM) of FMLP for 60 s. Scale bar represents 5 μm. a, Phalloidin stain, representing pre-existing filaments and those that incorporated actin during the assay. Note that because phalloidin is excluded by some actin-binding proteins, such as cofilin, phalloidin staining does not necessarily represent all actin filaments. b, TMR–actin stain, representing newly incorporated TMR–actin only. c, Colour overlay, showing phalloidin stain in red and TMR–actin stain in green. Arrows in c indicate sites of new actin incorporation at the tips of finger-like actin bundles. d, A neutrophil stimulated and permeabilized as in a–c but in the presence of 0.2 μm cytochalasin D. Colour scheme is as in c. Arrowheads indicate the absence of TMR–actin incorporation at the tips of finger-like actin bundles. Perinuclear actin incorporation parallels the subcellular distribution of granules (data not shown). Because bright perinuclear but not pseudopodial TMR–actin incorporation is observed when cytochalasin D is present during the permeabilization reaction, we conclude that incorporation at the pseudopodial surface represents new actin polymerization and that perinuclear incorporation results from G-actin-binding proteins or structures, as has been reported for permeabilized fibroblasts. e, f, Three-dimensional reconstruction of the boxed region of the pseudopodium shown in c. The bottom two panels indicate the relative orientation of the region of the pseudopodium from c as it is rotated along its x-axis. The scale bar in e represents 2 μm. Scale bar in d represents 5 μm.

Figure 3. Spatial distribution of TMR–actin incorporation in a neutrophil with two pseudopodia

Images represent maximum-intensity projections of three-dimensional immunofluorescence data. a, Phalloidin staining, representing total actin. b, TMR–actin staining, representing newly incorporated actin. c, Colour overlay, with phalloidin staining in red and newly incorporated actin in green. Arrow indicates a pseudopodium with minimal new actin incorporation. Arrowhead indicates a pseudopodium that is predominant in terms of new actin incorporation. The intense perinuclear TMR–actin stain does not represent new actin polymerization (Fig. 2). Scale bar represents 5 μm.

Figure 4. Response of neutrophils expressing Arp3–GFP to a stationary or moving chemotactic micropipette

a–e, Response to a stationary micropipette. f–j, Response to a motile micropipette. Images are single optical sections from near the bottom of a cell. a, Image taken immediately after exposing neutrophils to a chemotactic micropipette (white circle). b–e, Same group of neutrophils at b, 72 s, c, 166 s, d, 196 s, and e, 240 s of exposure to the chemotactic micropipette. f, A polarized neutrophil responding to a moving chemotactic micropipette (white circle). The white asterisk represents a fixed reference point. g–h, Same cell as that shown in f at g, 78 s, h, 109 s, l, 193 s, and j, 305 s of exposure to the micropipette. Scale bar represents 5 μm.

Figure 5. Immunofluorescence localization of endogenous Arp2/3 complex and actin in human neutrophils and the relationship of Arp2/3 localization to sites of actin polymerization

Images are single optical sections from near the midsection of cell. a, Actin immunostaining. Scale bar represents 5 μm. b, p21–Arc immunostaining. c, Colour overlay, showing actin in red and p21–Arc in green. Arrowheads indicate the localization of p21–Arc at the tips of actin fingers. d–g, Detail of an actin finger, from a cell permeabilized in the presence of TMR–actin to detect sites of actin polymerization and then processed for p21–Arc and actin immunostaining. d, Anti-actin staining, showing pre-existing and newly incorporated actin filaments. Scale bar represents 0.5 μm. e, TMR–actin staining, showing newly incorporated actin. f, p21–Arc staining. g, Triple overlay of d–f, showing total actin shown in light blue, newly incorporated actin in red, and p21–Arc in green. The green staining co-localizes with red and appears yellow.

Figure 6. Model of actin polymerization in response to a chemotactic signal

Top, Nomarski images of an unpolarized neutrophil exposed to a chemotactic micropipette (just to left of field) for a, 5 s, b, 30 s, c, 81 s, and d, 129 s. Bottom, the model. a, A neutrophil exposed to a gradient of chemoattractant (purple concentric circles) generates an asymmetric distribution of polymerization foci. b–d, The force generated by polymerization of actin (red lines) propels the polymerization focus forward and pushes the membrane outwards. The preferential activation of polymerization foci nearest to the chemoattractant could result in directional migration of neutrophils in response to chemotactic gradients and would be consistent with the behaviour of the pseudopod in response to a changing direction of chemoattractant. Note that this figure represents a single optical section of a neutrophil responding to a chemotactic gradient; the three-dimensional organization of the sites of actin polymerization and actin projections is shown in Fig. 2e, f.