Figure S4. IRSp53 binds to actin via... | Rockefeller University Press

$IRSp53 binds to actin via charged interactions, and its bundling activity requires hydrophobic interactions. (A and B) PSD complexes bundle actin filaments via IRSp53. Representative fluorescence images showing that 4× PSD complex with WT IRSp53 significantly promotes actin bundling (A), whereas the mixture without IRSp53 could not (Bi). The mixture of 4× PSD with IRSp53 will not undergo LLPS in the absence of G-actin (Bii). Each PSD component was at 0.5 μM, and actin was at 1 μM. (C) Effect of increasing protein concentration on the actin bundling activity of IRSp53. 1 μM G-actin was incubated with IRSp53 at indicated concentrations for 1 h at room temperature before centrifugation at 10,000 g. Amount of actin in the pellet fraction is quantified from three independent batches of experiments. Error bar indicates ± SD. (D) Effect of increasing KCl concentration on the actin bundling (i) and binding (ii) activities of IRSp53. 1 μM actin was incubated with 4 μM of IRSp53, in the presence of indicated amount of salt in the assay buffer, for 1 h at room temperature before centrifugation at 10,000 g (i) or 100,000 g (ii). Amount of actin (i) or actin-bound protein (ii) in the pellet fraction is quantified from three independent batches of experiments. Error bar indicates ± SD. (E) SDS-PAGE showing some IRSp53 protein present in the pellet fraction during the process of ultracentrifugation, in the absence of G-actin. This portion of non–actin-induced sedimentation was subtracted for correction. Note that the pellet, obtained after ultracentrifugation, was resuspended in one quarter of the initial mixture volume for concentration. (F) Multiple sequence alignment of residues 470–500 of IRSp53 from different mammalian species showing the conservation of several aromatic residues (Phe and Tyr highlighted by red arrows) throughout evolution. These residues were subjected to mutation to Ala to eliminate potential hydrophobic interactions, and the mutant was referred to as IRSp53_FYA. (G) SDS-PAGE showing that the FYA mutation in IRSp53 significantly diminished actin bundle formation (right panels) but barely interfered with its actin-binding ability (left panels). 1 μM actin was mixed with 4 μM of IRSp53 constructs. Note that the pellet, obtained after ultracentrifugation (at 100,000 g), was resuspended in one quarter of the initial mixture volume for concentration. The amount of protein sedimented independent of actin filaments (last four rows on SDS-PAGE on LHS) was subtracted for correction. The corrected pellet ratio of protein was further divided by the pellet ratio of F-actin for normalization. The amount of actin in the pellet fraction obtained after low-speed centrifugation (at 10,000 g), which represents the bundled actin filaments, was quantified. All data are obtained from three independent batches of experiments. Quantification is presented as mean ± SD. ****, P < 0.0001 using unpaired t test (protein fraction bound to actin) or one-way ANOVA with Dunnett’s multiple comparison test (actin in bundle fraction). (H) miniDawn analysis showing that the FYA mutation does not affect IRSp53 dimerization in solution. Superose 12 10/300 column was used for size-exclusion chromatography. Calculated MW of IRSp53_FYA is 57.0 kD. Analysis was performed at 100 μM protein concentration. Source data are available for this figure: SourceData FS4.$

Figure S4.

IRSp53 binds to actin via charged interactions, and its bundling activity requires hydrophobic interactions. (A and B) PSD complexes bundle actin filaments via IRSp53. Representative fluorescence images showing that 4× PSD complex with WT IRSp53 significantly promotes actin bundling (A), whereas the mixture without IRSp53 could not (Bi). The mixture of 4× PSD with IRSp53 will not undergo LLPS in the absence of G-actin (Bii). Each PSD component was at 0.5 μM, and actin was at 1 μM. (C) Effect of increasing protein concentration on the actin bundling activity of IRSp53. 1 μM G-actin was incubated with IRSp53 at indicated concentrations for 1 h at room temperature before centrifugation at 10,000 g. Amount of actin in the pellet fraction is quantified from three independent batches of experiments. Error bar indicates ± SD. (D) Effect of increasing KCl concentration on the actin bundling (i) and binding (ii) activities of IRSp53. 1 μM actin was incubated with 4 μM of IRSp53, in the presence of indicated amount of salt in the assay buffer, for 1 h at room temperature before centrifugation at 10,000 g (i) or 100,000 g (ii). Amount of actin (i) or actin-bound protein (ii) in the pellet fraction is quantified from three independent batches of experiments. Error bar indicates ± SD. (E) SDS-PAGE showing some IRSp53 protein present in the pellet fraction during the process of ultracentrifugation, in the absence of G-actin. This portion of non–actin-induced sedimentation was subtracted for correction. Note that the pellet, obtained after ultracentrifugation, was resuspended in one quarter of the initial mixture volume for concentration. (F) Multiple sequence alignment of residues 470–500 of IRSp53 from different mammalian species showing the conservation of several aromatic residues (Phe and Tyr highlighted by red arrows) throughout evolution. These residues were subjected to mutation to Ala to eliminate potential hydrophobic interactions, and the mutant was referred to as IRSp53_FYA. (G) SDS-PAGE showing that the FYA mutation in IRSp53 significantly diminished actin bundle formation (right panels) but barely interfered with its actin-binding ability (left panels). 1 μM actin was mixed with 4 μM of IRSp53 constructs. Note that the pellet, obtained after ultracentrifugation (at 100,000 g), was resuspended in one quarter of the initial mixture volume for concentration. The amount of protein sedimented independent of actin filaments (last four rows on SDS-PAGE on LHS) was subtracted for correction. The corrected pellet ratio of protein was further divided by the pellet ratio of F-actin for normalization. The amount of actin in the pellet fraction obtained after low-speed centrifugation (at 10,000 g), which represents the bundled actin filaments, was quantified. All data are obtained from three independent batches of experiments. Quantification is presented as mean ± SD. ****, P < 0.0001 using unpaired t test (protein fraction bound to actin) or one-way ANOVA with Dunnett’s multiple comparison test (actin in bundle fraction). (H) miniDawn analysis showing that the FYA mutation does not affect IRSp53 dimerization in solution. Superose 12 10/300 column was used for size-exclusion chromatography. Calculated MW of IRSp53_FYA is 57.0 kD. Analysis was performed at 100 μM protein concentration. Source data are available for this figure: SourceData FS4.

This Feature Is Available To Subscribers Only