Molecular mechanism of Arp2/3 complex activation by nucleation-promoting factors and an actin monomer

Figures

• 

  Fig. 1.. Overview of structures of Arp2/3 complex. (A and B) Ribbon diagrams. (A) Inactive conformation with the Arp2 and Arp3 subunits held apart in a “splayed” conformation (pdb id: 6UHC). (B) Active state from the branch junction. Arp2 and Arp3 are half overlapped like two subunits along the short-pitch helix of an actin filament (pdb id: 7TPT). K1-2 is the equilibrium constant for the transition between these two states, while k+1 and k−1 are the rate constants for the two reactions. (C) Space-filling model of the branch junction with Arp2/3 complex in the short-pitch conformation on the side of the “mother” filament (pdb id: 7TPT).

• 

  Fig. 2.. Arp2/3 complex crosses an energy barrier of 20 ± 6 kcal/mol between the inactive state and the active state in the branch junction. (A) A linear combination of four distances d1 to d4 was used to define the path for pathCV-based free energy sampling (Eqs. 1 and 2). The distances are d1 between centers of mass of SD3 and SD4 of Arp2 and Arp3, d2 between centers of mass of Arp2 and ARPC3, d3 between centers of mass of Arp3 and ARPC4 and d4 between ARPC1 and ARPC3. (B) Free energy profile of Arp2/3 complex along the pathway to branch formation obtained by averaging PMF from metadynamics, umbrella and equilibrium MD simulations along the reweighted pathCV from the active to the inactive states. On the Bottom x-axis, a CV value of 1 corresponds to the active, short-pitch state and 12 corresponds to inactive, splayed state, while 2 to 11 are intermediate states. The numbers on the Top x-axis give the distributions of distances d1 of the active (41.0 to 42.0 Å), transition (50.7 to 52.2 Å), partially active (51.6 to 53.8 Å), and inactive (53.5 to 54.5 Å) states. The inactive and active states are separated by (i) a major barrier of 20 ± 6 kcal/mol (mean ± SD) and (ii) a minor free…

• 

  Fig. 3.. MD simulations over 3.4 μs of two copies of human N-WASP CA bound to inactive human Arp2/3 complex starting with the cryo-EM structure pdb 6UHC (24). (A and B) Space-filling models of subunits of Arp2/3 complex with ribbon diagrams of the CA motifs (C in cyan and A in magenta) rendered every 560 ns. (A) Arp2 (red) and ARPC1 (green). (B) Arp3 (orange). Contacts of C helix (cyan) with Arp2 and Arp3 and the EWE motif with ARPC1 and Arp3 are relatively stable, but the D/E loops of the A motifs (magenta) adopt multiple conformations. (C) Numbers of salt bridges between aspartate and glutamate residues in the A motif and arginine and lysine residues in Arp2/ARPC1 and Arp3 subunits during 1 μs of MD simulation. Lighter traces are averages across three independent simulations and solid lines are their running averages with a window of 50 ns. (D) Lennard-Jones and Coulombic interaction energies between CA and Arp2/3 complex subunits at short range confirm that electrostatic interactions are more favorable with Arp2/ARPC1 than Arp3.

• 

  Fig. 4.. Interactions of N terminus of the C helix of CA with C-terminal tail of Arp3. (A) Time courses of the number of steric contacts calculated to be within 4 Å between N terminus of the C-helix and C terminal tails of the Arp2 (red) and Arp3 (orange). Lighter traces are average across three independent simulations and solid lines are their running averages with a window of 50 ns. (B) Probability distributions of distances (d5) between the centers of mass of the W-loop (E182-Y184) (purple spheres) and the C-terminal tail of Arp3 (P412-V416) (green); Insets (i) and (ii) are examples of two conformations with distances marked on the graph. (C) Ribbon diagrams show the C-helix (cyan) displacing the C-terminal tail of Arp3 (orange) from its starting position (red) every 560 ns during the simulation (green). (D) Ribbon diagrams comparing the conformations of Arp3 in the cryo-EM structure 6UHC (gray) and the end of the 3.4 μs simulation (orange). The N terminus of the C-helix (cyan) displaces the C-terminal tail of Arp3 from the barbed end groove creating space for the D-loop of an actin monomer (blue) to bind.

• 

  Fig. 5.. The conformation of inactive, CA-bound human Arp2/3 complex (pdb id: 6UHC) shifts toward the short-pitch state during 3.4 μs of MD simulation. (A) During the simulation, the distance between centers of mass of subdomains 3 and 4 of Arp2 and Arp3 (blue) change from 54.7 Å to a plateau at 52.7 Å and the clamp dihedral angle (red) declines from −45.2° to −33.1°. Lighter traces are values during the simulations and solid lines are their running averages with a window of 50 ns. (B) Ribbon diagram with the definition of the distance between the centers of mass of subdomains 3 and 4 of Arp2 (red) and Arp3 (orange). (C) Ribbon diagram with the clamp dihedral angle defined by four points (centers of mass of residues K18 and I244 of ARPC2 (cyan space filling model) and R32 and S147 of ARPC4 (blue space filling model) that measures the rotation of Arp2 block relative to Arp3 block. (D) Conformational change in the complex during the 3.4 μs MD simulation. Ribbon diagrams of Arp2/3 complex: the outline is the starting structure; and the colored ribbon diagram is the conformation at the end of 3.4 μs. The arrow indicates the 12.1° rotation of the Arp2 block. The transparent space filling model…

• 

  Fig. 6.. MD simulations based on the transition state structure [pdb id: 7T5Q (35)] without CapZ and the two CA motifs. (A) (Top) Time courses of the distance between the centers of mass of subdomains 3 and 4 of Arp2 and Arp3 in Arp2/3 complex with (blue) and without (red) bound actin. In two replica runs without the actin monomer, Arp2/3 complex transitioned rapidly toward the distance in the splayed, inactive state (black dashed line). Space filling diagrams show the two starting structures. (Bottom) Time course of a transient increase in the dihedral angle of Arp3 after removing the actin monomer (red). Dihedral angle is constant when the actin monomer is bound to Arp3 (blue). (B) Orientations of the C-terminal helices of ARPC2 and ARPC4 changed relative to the ARPC2 globular domain before and after removing actin. Rotation of the helices is calculated by superposition of the ARPC2 globular domain (gray space filling model). The conformation of the transition state with actin is shown with lighter shades and the relaxed structure without actin at the end of the simulation is shown with darker shades. Movement of the ARPC2 helix was measured between vectors connecting the centers of…

• 

  Fig. 7.. Pathways from free, inactive Arp2/3 complex (1, Upper Left) to an actin filament branch (7, Upper Right). Each reactant is numbered. The dashed boxes identify reactions studied in this paper. The sizes of the arrows are proportional to measured or estimated rate or equilibrium constants for each reaction. The ratios of rate constants (k) are the equilibrium constants (K). Reactants 1 and 3 have splayed conformations. Reactants 2, 4, 6, and 7 have short-pitch conformations. Reactant 5 has an intermediate conformation. The text explains how the rate and equilibrium constants for each reaction were measured or estimated.