CK-666 and CK-869 differentially inhibit Arp2/3 iso-complexes

Abstract

The inhibitors, CK-666 and CK-869, are widely used to probe the function of Arp2/3 complex mediated actin nucleation in vitro and in cells. However, in mammals, the Arp2/3 complex consists of 8 iso-complexes, as three of its subunits (Arp3, ArpC1, ArpC5) are encoded by two different genes. Here, we used recombinant Arp2/3 with defined composition to assess the activity of CK-666 and CK-869 against iso-complexes. We demonstrate that both inhibitors prevent linear actin filament formation when ArpC1A- or ArpC1B-containing complexes are activated by SPIN90. In contrast, inhibition of actin branching depends on iso-complex composition. Both drugs prevent actin branch formation by complexes containing ArpC1A, but only CK-869 can inhibit ArpC1B-containing complexes. Consistent with this, in bone marrow-derived macrophages which express low levels of ArpC1A, CK-869 but not CK-666, impacted phagocytosis and cell migration. CK-869 also only inhibits Arp3- but not Arp3B-containing iso-complexes. Our findings have important implications for the interpretation of results using CK-666 and CK-869, given that the relative expression levels of ArpC1 and Arp3 isoforms in cells and tissues remains largely unknown.

Keywords: Arp2/3 Iso-Complexes; CK-666; CK-869; Inhibition; SPIN90.

Figure 1. CK-869, but not CK-666 fully inhibits vaccinia-induced actin polymerisation.

(A) Representative immunofluorescence image of a HeLa cell infected with vaccinia virus at 9 h post-infection and labelled with an antibody against cortactin and stained with phalloidin to visualise F-actin. Extracellular viruses attached to the plasma membrane are labelled with an antibody against the viral protein B5. Scale bar = 10 μm. (B) Representative images showing localisation of cortactin in HeLa cells infected with vaccinia at 9 h post-infection after 1 h treatment with CK-666 or CK-869 at the specified concentrations. Scale bar = 10 μm. The inset colour images show 2x magnifications of the yellow boxed regions with cortactin (green) and extracellular virus (magenta). (C) The graph shows quantification of the mean number of extracellular viruses co-localising with cortactin from three independent repeats with the error bar representing standard deviation. 10 cells and 150 virus particles were analysed in each experiment. A two-tailed unpaired t-test was used to analyse the statistical significance between different drugs at the same concentration as well as between the drug and DMSO control. ns not significant. *p value <0.05. **p value <0.01. ****p value <0.0001. Source data are available online for this figure.

Figure 2. CK-666 does not inhibit ArpC1B-containing Arp2/3 complexes.

(A) Representative plots showing the polymerisation of pyrene-labelled actin (Fluorescence Intensity) when Arp2/3 iso-complexes containing ArpC1A/C5L (left) or ArpC1B/C5L (right) are activated by GST-N-WASP-VCA in the absence (blue) or presence of 100 µM CK-666 (yellow) or CK-869 (green). (B) Representative TIRF images showing mother filaments (magenta) and new daughter branches (cyan) 2 min after Arp2/3 iso-complexes containing ArpC1A/C5, ArpC1A/C5L, ArpC1B/C5, or ArpC1B/C5L were activated by GST-N-WASP-VCA in the absence (DMSO) or presence of 100 µM CK-666 or CK-869. Scale bar = 5 μm. The graph shows the quantification of the mean branching rate for each condition from three independent experiments with the error bar representing the standard deviation. Two-tailed paired t-test has been applied to analyse the statistical significance. *p value <0.05. **p <0.01. ***p value <0.001. ns not significant. (C) The branching rate of the Arp2/3 complex containing ArpC1A/C5L and ArpC1B/C5L were measured at the indicated concentrations of CK-666 (top) and CK-869 (bottom). The data was fitted with equation Y = Bottom + (Top-Bottom)/(1 + (X/IC₅₀)) to calculate the half-maximal inhibitory concentration of the drugs (IC₅₀), which is indicated (ND not determined). The points and error bars represent the mean and the standard deviation from at least two independent measurements. Source data are available online for this figure.

Figure 3. CK-666 has no adverse impact on murine bone marrow-derived macrophages.

(A) Treatment of murine bone marrow-derived macrophages (BMDM) with 100 µM CK-869 but not DMSO or CK-666 induces cell rounding. The graphs show the mean ratio of length and width as well as cell thickness from five independent experiments with error bars representing the standard deviation. In each experimental condition, at least 30 cells were randomly chosen and analysed. A two-tailed unpaired t-test was used to analyse the statistical significance. (B) Representative phase images of BMDM treated with DMSO, CK-666, or CK-869 over 20 min. Scale bar = 20 µm. (C) Quantification of the mean migration speed of BMDM treated with DMSO, CK-666, or CK-869 from five independent experiments with error bars representing the standard deviation. In each experimental condition, at least 30 cells were randomly chosen and analysed. A two-tailed, unpaired student t-test was used to analyse the statistical significance. (D) The graph shows the quantification of the phagocytosis efficiency of zymosan by murine bone marrow-derived macrophages treated with DMSO, CK-666 or CK-869 over 120 min. In each experimental condition, 10⁴ cells were analysed using flow cytometry. Each point represents the mean of three independent technical repeats and error bars represent the standard deviation. A two-way ANOVA test has been applied to analyse the statistical significance. *p value <0.05. **p value <0.01. ***p value <0.001. **** means p value <0.0001. ns not significant. Source data are available online for this figure.

Figure 4. The impact of CK-666 and CK-869 on SPIN90 activation of Arp2/3 complexes.

(A) The maximum rate of pyrene actin polymerisation stimulated by SPIN90-Arp2/3 iso-complexes containing ArpC1A/C5L and ArpC1B/C5L in the presence of the indicated concentrations of CK-666 (top) and CK-869 (bottom). The data was fitted with equation Y = Bottom + (Top-Bottom)/(1 + (X/IC₅₀)) to calculate the half-maximal inhibitory concentration of the drugs (IC₅₀), which is indicated. The points and error bars represent the mean and the standard deviation of at least two independent measurements. (B) Representative TIRF images of actin filament assembly induced by Arp2/3 complexes containing ArpC1A/C5L or ArpC1B/C5L after activation by SPIN90-Cter in the absence (DMSO) or presence of 100 µM CK-666 or CK-869 at the indicated times. Scale bar = 10 μm. The graph shows the quantification of the mean filament density at 240 s from three independent experiments, with the error bar representing the standard deviation. Two-tailed unpaired t-test was used to analyse the statistical significance and ***p value <0.001. (C) The left panels show immunoblots of GST-SPIN90-Cter (Ponceau - red) pulldown assays of the indicated Arp2/3 iso-complexes (detected by ArpC2 antibody) in the absence (DMSO) or presence of 100 µM CK-666 or CK-869. The right panel shows the quantification of the mean bound Arp2/3 complex normalised to the DMSO control for three independent pulldown assays with the error bar representing standard deviation. Two-tailed paired t-test was used to analyse the statistical significance. ns not significant. *p value <0.5. Source data are available online for this figure.

Figure 5. Impacts of drugs on Arp3B-containing complexes.

(A) Representative TIRF images showing mother filaments (magenta) and newly polymerised daughter actin branches (cyan) 2 min after Arp2/3 complex containing Arp3B/ArpC1B/C5L was activated by GST-N-WASP-VCA in the absence (DMSO) or presence of 100 µM CK-666 or CK-869. Scale bar = 5 μm. The graph shows the quantification of the mean branching rate for each condition from three independent experiments with the error bar representing the standard deviation. A two-tailed paired t-test was used to analyse the statistical significance and ns not significant. (B) Representative TIRF images of actin filament assembly induced by Arp2/3 complex containing Arp3B/ArpC1B/C5L after activation by SPIN90-Cter in the absence (DMSO) or presence of 100 µM CK-666 or CK-869 at the indicated times. Scale bar = 10 μm. The graph shows the quantification of the mean filament density at 240 s from three independent experiments with the error bar representing the standard deviation. A two-tailed unpaired t-test was used to analyse the statistical significance. ns not significant. *** presents p value <0.001. (C) The maximum rate of pyrene actin polymerisation stimulated by Arp2/3 complex containing Arp3B/ArpC1B/C5L activated by either GST-N-WASP-VCA or SPIN90-Cter at the indicated CK-869 concentration. The points and error bars represent the mean and the standard deviation of at least two independent measurements. (D) The top image shows that in the Arp2/3 complex (PDB: 3ULE), CK-869 has interaction with Leu 112 and Thr 126 of Arp3. The bottom image shows that in Arp3B, Leu 112, and Thr 126 are replaced by Met 112 and Leu 126, the latter of which is predicted to have clashes with CK-869. Source data are available online for this figure.

Figure EV1. CK-869, but not CK-666, fully inhibits Vaccinia-induced actin polymerisation.

Immunofluorescence images of the actin cytoskeleton (visualised with phalloidin) of HeLa cells infected with Vaccinia virus at 9 h post-infection after 1-hour incubation with indicated concentrations of CK-666 and CK-869. The images correspond to the actin cytoskeleton of the cortactin images in Fig. 1B. Scale bar = 10 μm.

Figure EV2. Coomassie-stained protein gels of the recombinant proteins and Arp2/3 iso-complexes.

Molecular weight markers and the protein names are indicated.

Figure EV3. Inhibition of actin branches by CK-666 and CK-869.

(A) Representative plots showing the polymerisation of pyrene actin (Fluorescence Intensity) when Arp2/3 complexes containing ArpC1A/C5 (left) or ArpC1B/C5 (right) are activated by GST-N-WASP in the absence (blue) or presence of 100 µM CK-666 (yellow) or CK-869 (green). (B) The branching rate of the Arp2/3 complex containing ArpC1B/C5 and ArpC1B/C5L (same as shown in Fig. 2C) were measured at the indicated CK-869 concentration. The data were fitted with equation Y = Bottom + (Top-Bottom)/(1 + (X/IC₅₀)) to calculate the half-maximal inhibitory concentration (IC₅₀) of the CK-869. The points and error bars represent the mean and the standard deviation of at least two independent measurements.

Figure EV4

Representative FACS plots of murine bone marrow-derived macrophage phagocytosis after treatment with DMSO (control) or 100 µM CK-666 or CK-869.

Figure EV5. Analysis of Arp2/3 complex binding to actin filaments or VCA.

(A) The left panel shows representative Coomassie-stained protein gels of the co-sedimentation of Arp2/3 complexes containing ArpC5L together with ArpC1A (top) or ArpC1B (bottom) with 15 µM actin (P pellet) in the absence (DMSO) or presence of 100 µM CK-666 or CK-869. The supernatant (S) contains unbound Arp2/3 complexes and non-pelleted G-actin. MK represents the molecular weight markers. The right graph shows the quantification of the mean co-sedimentation fraction of the indicated Arp2/3 complexes in the absence (DMSO) or presence of CK-666 or CK-869 from three independent experiments at 3 and 15 µM actin, with the error bars indicating the standard deviation. Two-tailed paired t-test was used to analyse the statistical significance. *p value < 0.05. ns not significant. (B) Immunoblots (left panels) using an ArpC2 antibody demonstrate that 100 µM CK-666 or CK-869 does not impact the interaction of Arp2/3 complexes ArpC5L together with ArpC1A (top) or ArpC1B (bottom) with GST-N-WASP-VCA (Ponceau - red). The right graph shows the quantification of the mean pull-down fraction of the indicated Arp2/3 complexes in the absence (DMSO) or presence of CK-666 or CK-869 from three independent experiments, with the error bars indicating the standard deviation. Two-tailed paired t-test was used to analyse the statistical significance. ns not significant.

Figure EV6. Inhibition of SPIN90-Arp2/3 nucleated linear actin filaments.

Representative plots showing the polymerisation of pyrene actin (Fluorescence Intensity) when Arp2/3 complexes containing ArpC1A/C5L (left) or ArpC1B/C5L (right) are activated by SPIN90-Cter in the absence (blue) or presence of 100 µM CK-666 (yellow) or CK-869 (green).