Fig 2: Arp3 isoforms confer different cellular properties to Arp2/3. (A) Immunofluorescence images of vaccinia-induced actin tails labeled with Alexa Fluor 568 phalloidin (magenta) together with quantification of their length and number at 8 h after infection in HeLa cells stably expressing RNAi-resistant GFP-tagged Arp3 (blue) or Arp3B (red) and treated with Arp3 siRNA. (B) Immunoblot analysis of GFP-Trap pulldowns from HeLa cells lysates stably expressing GFP, GFP-Arp3, or GFP-Arp3B demonstrates that Arp3 and Arp3B associate with all Arp2/3 complex subunits. The Arp3 antibody does not detect Arp3B. (C) Immunoblot analysis of Arp2/3 complex subunits in HeLa cells treated with Arp3 and Arp3B siRNAs (Fig. S1 A shows RT-PCR analysis of the level of Arp3 and Arp3B mRNA relative to control in HeLa cells after knockdown with the individual siRNA from the siGenome pools against Arp3 and Arp3B). (D) Immunofluorescence images of virus (magenta) induced actin tails labeled with Alexa Fluor 488 phalloidin (green) together with quantification of their length in cells treated with the indicated siRNAs. All error bars represent SEM from n = 3 independent experiments in which the length of 100 tails or their number in six cells was analyzed per condition. Tukey’s multiple comparisons test was used to determine statistical significance; ****, P < 0.0001; ***, P < 0.001; *, P < 0.05. Scale bars = 5 µm.

Fig 3: Arp3- and Arp3B-containing complexes have the same actin nucleating activity. (A) In vitro pyrene actin assays using 12.5 nM of the indicated recombinant Arp2/3 complexes reveal Arp3 and Arp3B containing complexes nucleate actin equally efficient, regardless of the ARPC1A/ARPC5 and ARPC1B/ARPC5L background. The error bars represent SEM from n = 2 independent experiments. (B) The top row shows representative in vitro TIRF microscope images of branched actin formation using 2.5-nM Arp2/3 complexes containing Arp3B, ARPC1A, and ARPC5 at the indicated times (see Video 1). n = 2 independent experiments. Scale bar = 15 µm. The bottom row shows the same panels after automatic detection of filament branches (yellow nodes) and ends (red nodes) with AnaMorf ImageJ plugin. (C) Quantification of mean branch number and filament length for the indicated Arp2/3 complexes from one of the two independent TIRF experiments. Each data point represents the average of 14 and 15 actin “structures” for Arp3B-C1B-C5L and Arp3-C1B-C5L, respectively, although there are fewer structures to analyze at the beginning of the experiment than at the end.

Fig 4: Arp3 and Arp3B knockdown and rescue. (A) RT-PCR analysis showing the level of Arp3 (blue) and Arp3B (red) mRNA relative to control in HeLa cells after knockdown with the individual siRNA from the siGenome pools against Arp3 and Arp3B. Error bars represent SEM from two independent experiments. (B) Quantification of actin tail lengths and immunoblot analysis of the level of Arp3 in HeLa cells treated with the indicated siRNA. The error bars represent SEM from n = 2 independent experiments in which the length of 100 tails was analyzed per condition. Tukey’s multiple comparisons test was used to determine statistical significance; ****, P < 0.0001; **, P < 0.01. (C) Immunoblot analysis showing that stable expression of RNAi-resistant GFP-tagged Arp3 (blue) and Arp3B (red) restore control levels of all Arp2/3 subunits in HeLa cells treated with Arp3 siRNA. (D) Coomassie-stained gel of recombinant isoform-specific Arp2/3 complexes and corresponding immunoblot. (E) RT-PCR analysis of the level of Arp3B mRNA in HeLa cells stably expressing Cherry-GFPPA-ß-actin and treated with Arp3B siRNA. Error bars represent SEM from three independent experiments. (F) Immunoblot analysis of lysates from HeLa cells stably expressing GFPPA-tagged Arp3 (blue) and Arp3B (red).

Fig 5: Residues 293/295 are surface exposed in activated Arp2/3. (A) Top: 3.9-Å cryo-EM structure of S. pombe Arp2/3 in its short-pitch conformation after activation by Dip1 (Shaaban et al., 2020; Protein Data Bank accession no. 6W17). The structure is visualized as molecular surface, and the Arp2/3 subunits, Dip1, and the daughter actin filament subunits (indicated as n, n+1…etc) are color coded as in Protein Data Bank accession no. 6W17. Middle: The structure is rotated ~180° in the direction indicated by the black curved arrow to allow visualization of residues of Leu304 and Pro306 of Schizosaccharomyces pombe Arp3 (highlighted in dark blue) that correspond to Met293 and Ser295 in human Arp3B. The dashed rectangle is enlarged at the bottom of the figure. (B) Top: 9.0-Å cryo-EM structure of human Arp2/3 at the branch point between a mother and daughter filament (Fäßler et al., 2020; Protein Data Bank accession no. 7AQK). The structure is visualized as molecular surface, and the Arp2/3 subunits, mother filament subunits (indicated as M1–M8), and daughter filament subunits (indicated as n, n + 1…etc.) are color coded as in Protein Data Bank accession no. 7AQK. The red arrow below the branch point indicates the point of view represented in the middle panel after a ~90° rotation in the direction shown by the black curved arrow. This structure contains Arp3; hence, the position of Thr293 and Pro295 is highlighted in dark blue and indicated by a dashed black line. The dashed rectangle is enlarged at the bottom of the figure.