Pivotal role of VASP in Arp2/3 complex-mediated actin nucleation, actin branch-formation, and Listeria monocytogenes motility

Abstract

The Listeria monocytogenes ActA protein mediates actin-based motility by recruiting and stimulating the Arp2/3 complex. In vitro, the actin monomer-binding region of ActA is critical for stimulating Arp2/3-dependent actin nucleation; however, this region is dispensable for actin-based motility in cells. Here, we provide genetic and biochemical evidence that vasodilator-stimulated phosphoprotein (VASP) recruitment by ActA can bypass defects in actin monomer-binding. Furthermore, purified VASP enhances the actin-nucleating activity of wild-type ActA and the Arp2/3 complex while also reducing the frequency of actin branch formation. These data suggest that ActA stimulates the Arp2/3 complex by both VASP-dependent and -independent mechanisms that generate distinct populations of actin filaments in the comet tails of L. monocytogenes. The ability of VASP to contribute to actin filament nucleation and to regulate actin filament architecture highlights the central role of VASP in actin-based motility.

Figure 2.

Mutations in the actin monomer-binding region and proline-rich repeats do not alter levels of ActA. (A) Schematic diagram of ActA derivatives. Functional domains of ActA are labeled: (SS) signal sequence, (A) acidic stretch, (AB) actin monomer-binding region, (C) cofilin homology sequence, (LR) long repeats, (TM) transmembrane domain, black boxes represent proline-rich repeats. (B) Surface-associated ActA from actA mutants visualized by Western blotting. Surface proteins were extracted from an equivalent number of late log phase bacteria grown in Luria-Bertani broth. Proteins were separated on a 7.5% polyacrylamide gel. ActA was detected by Western blotting using polyclonal anti-ActA antibodies made to the first 18 residues of the mature protein (Mourrain et al., 1997). All lanes shown are from a single exposure of one blot. (C) Metabolic labeling of ActA from mutant bacteria grown within host cells. Strains expressing derivatives of ActA (containing an equal number of methionine residues) were grown in J774 macrophages. Host protein synthesis was inhibited with cyclohexamide and anisomycin and bacterial proteins were labeled with [³⁵S] methionine. SDS-extracted proteins were resolved on a 7.5% gel and visualized by autoradiography. ActA and its phosphorylated forms are labeled with arrows.

Figure 1.

Effects of mutations in the actin monomer-binding region of ActA. (A) Point mutations in the actin monomer-binding region prevent sequestration of actin monomer. The effect of the indicated concentrations of ActA and ActA derivatives on the polymerization kinetics of 2 μM actin was monitored using pyrene-actin polymerization assays. The fold inhibition was calculated by dividing the maximal rate of actin polymerization by the maximal rate of actin polymerization in the presence of ActA and ActA derivatives. (B) Mutations in the actin monomer-binding region of ActA decrease the efficiency of Arp2/3 complex– mediated actin nucleation. Graphs of fluorescence intensity (measured in arbitrary units) versus time in pyrene-actin polymerization assays. Fluorescence of pyrene-labeled actin increases when it is incorporated into actin polymer. Thus, an increase in fluorescence represents an increase in actin polymer mass. Initiation of polymerization of 2 μM actin was induced at time 0 in the presence or absence of 20 nM Arp2/3 complex and 20 nM ActA derivatives.

Figure 3.

In the absence of actin monomer-binding activity, intact proline-rich repeats of ActA are required for intercellular spread of L. monocytogenes. (A) Plaque assays. Images of plaques formed in L2 mouse fibroblast monolayers after 4 d of infection with the indicated L. monocytogenes strains. Media contains 0.7% agarose and 10 μg/ml gentamicin. Live cells are stained with neutral red. Bar, 10 mm. (B) Quantification of intercellular spread. Plaque diameters were measured for each strain and are expressed as percent of wild type. Error bars represent the SD between five and twelve independent assays.

Figure 4.

Ability of ActA to induce actin polymerization in vivo requires actin monomer-binding activity or intact proline-rich repeats. PtK2 cells infected for 3.5 h with wild-type L. monocytogenes or mutants expressing the indicated derivatives of ActA. F-actin was visualized by staining with rhodamine-conjugated phalloidin and bacteria were detected by indirect immunofluorescence using polyclonal anti–L. monocytogenes primary antibodies followed by FITC-conjugated secondary antibodies. Bar, 10 μm.

Figure 5.

VASP recruitment is impaired in mutants expressing ActA with mutated proline-rich repeats. HeLa Cells were infected for 3.5 h with wild-type or indicated mutant strains. After fixation, VASP was detected with affinity-purified polyclonal anti–human VASP antibodies, followed by rhodamine-conjugated secondary antibodies. L. monocytogenes were detected by staining with FITC-conjugated polyclonal anti–L. monocytogenes antibodies. Bar, 10 μm.

Figure 6.

Purified VASP rescues the actin nucleation defect of an ActA derivative deficient for actin monomer binding. (A) Pyrene-actin polymerization assays carried out in the presence of VASP and profilin. Kinetics of 2 μM actin polymerization was monitored in the presence or absence of 100 nM ActA, ΔAB derivative, VASP, or profilin and in the presence or absence of 20 nM Arp2/3 complex as indicated. The fold increase in polymerization rate was determined by calculating the maximal rate of polymerization and dividing that value by the rate of spontaneous actin polymerization and is presented in parentheses. (B) VASP rescue of actin nucleation defect saturates at a ratio of 2:1 with ΔAB. Pyrene-actin polymerization assays were carried out with 2 μM actin and 20 nM Arp2/3 complex, with either 20 nM ActA or ΔAB in the presence of the indicated concentration of VASP. The fold increase in polymerization rate is presented in parentheses. (C) Actin filaments do not rescue the actin nucleation defect of the ΔAB derivative. Pyrene-actin polymerization assays were carried out with 2 μM actin in the presence or absence of 20 nM Arp2/3 complex, ActA or ΔAB, 40 nM VASP, or 200 nM phalloidin-stabilized F-actin seeds, as indicated. The fold increase in polymerization rate is presented in parentheses. (D) VASP stimulates actin nucleation by ActA and the Arp2/3 complex Pyrene-actin polymerization assays were carried out with 2 μM actin in the presence or absence of 20 nM Arp2/3 complex, ActA or ΔAB, 40 nM VASP, or 200 nM phalloidin- stabilized F-actin seeds, as indicated. The fold increase in polymerization rate is presented in parentheses.

Figure 7.

VASP recruitment decreases actin branch formation by ActA and the Arp2/3 complex. (A) Images of actin filament structures nucleated in the presence or absence of VASP. 4 μM actin was polymerized in the presence of 20 nM Arp2/3 complex (top) or 20 nM Arp2/3 complex and 200 nM VASP (bottom) with no ActA (a and e), 200 nM ActA (b and f) , 200 nM ΔAB (c and g), and 200 nM A263 (d and h). Actin structures were stabilized and labeled with addition of 4 μM rhodamine- phalloidin and visualized by direct microscopic observation. (B) Quantification of actin branches formed in presence of VASP. The values are presented as a percent of the total population of actin filaments that appear in branched structures. All of the filaments in a given field were counted; when two or more actin filaments were found to intersect, each was scored as branched. The data represent a minimum of 800 filaments for each condition from two to five independent experiments.