QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 24 April 2007
doi: 10.1242/jcs.007641

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in JCS Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cai, L. Articles by Bear, J. E. Search for Related Content PubMed PubMed Citation Articles by Cai, L. Articles by Bear, J. E.

F-actin binding is essential for coronin 1B function in vivo

¹ Lineberger Comprehensive Cancer Center and Department of Cell & Developmental Biology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, 27599-7295, USA
² Department of Microbiology & Immunology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, 27599-7295, USA

View larger version (52K): [in this window] [in a new window]   Fig. 1. Coronin 1B induces F-actin cables in Drosophila S2 cells independently of the Arp2/3 complex. (A) S2 cells were transfected with GFP-tagged coronin 1B or coronin 2A expression constructs, plated on ConA-treated coverslips, and stained for F-actin using phalloidin. (B) S2 cells transfected with GFP-tagged coronin 1B expression constructs were immunostained for Arp2, a subunit of the Arp2/3 complex. (C) Endogenous Drosophila p20Arc was depleted by siRNA for 7 days. S2 cells were transfected with coronin 1B expression constructs and stained with phalloidin. Arrowheads show a colocalization of F-actin and coronin 1B in p20Arc depleted S2 cells. (D) Diagram of coronin-1B–coronin-2A chimeras showing cable-formation ability of each construct. Fragments from coronin 1B are in black, and fragments from coronin 2A are in white. Red numbers indicate the start position of each region according to amino acid sequence of coronin 1B. Corresponding coronin 2A regions were chosen by sequence alignment using ClustalX 1.83.

View larger version (64K): [in this window] [in a new window]   Fig. 2. A conserved surface-exposed arginine residue, R30, in coronin 1B is responsible for actin-cable formation. (A) Charged, surface-exposed residues within the first and fourth blades of the coronin 1B -propeller were mutated to alanine or an amino acid with an opposite charge (i.e. RD). Cable-formation capability was scored as follows: +++, >99% of transfected cells contain GFP-positive actin cables; +/–, <50% of transfected cells contain actin cables, the rest have phalloidin-positive, small GFP aggregates; –, <1% of transfected cells contain phalloidin-positive GFP aggregates, the majority show high levels of cytoplasmic GFP. (B) S2 cells were transfected with the indicated GFP-tagged coronin 1B mutants and stained with phalloidin. (C) Multiple sequence alignment shows that R30 is a highly conserved charged residue in coronins, and locates between the first and second -sheet in the propeller structure. M-1B, Mus musculus coronin 1B (gi:12229769); R-1B, Rattus norvegicus coronin 1B (gi:12229732); H-1B, Homo sapiens coronin 1B (gi:21263481); H-1C, Homo sapiens coronin 1C (gi:7656991); Celegans, Caenorhabditis elegans coronin (gi:3121874); Spombe, Schizosaccharomyces pombe coronin (gi:3121869); Scerev, Saccharomyces cerevisiae coronin (gi:3121873); DICDI, Dictyostelium discoideum coronin (gi:116950). The multiple alignment was generated using ClustalX 1.83 and illustrated using ESPRIPT. (D) Top view of coronin 1B homology model. N-terminus is blue, C terminus is red. The side chain of R30 is presented in stick form. The homology model was generated in InsightII-2005 using HOMOLOGY model, and evaluated by Profile_3D function (see supplementary material Fig. S1). Illustration was generated using Pymol (http://www.pymol.org). (E) Side view of coronin 1B structural model showing the postions of R30 and K73.

View larger version (52K): [in this window] [in a new window]   Fig. 3. Coronin 1B preferentially binds to ATP/ADP-Pi–F-actin and mutation of R30 abolishes binding. (A) Representative Coomassie-Blue-stained gel showing the near-saturation binding of coronin 1B to ATP/ADP-Pi–F-actin (1 µM total actin). Upper band, coronin 1B; lower band, actin. Numbers below bands indicate protein concentrations in µM as determined by densitometry. (B) Representative Coomassie-Blue-stained gel showing F-actin binding of coronin 1B at indicated KCl concentrations (1.1 µM total actin, 1.3 µM coronin 1B). (C,D) Representative immunoblot showing F-actin bound coronin 1B in the pellets of high-speed co-sedimentation experiments (top panel); Coomassie-Blue-stained gel show the corresponding actin pellets (bottom panel). Identical amounts of coronin 1B (0.1 µM) were used and concentrations of total ADP- and ATP-G-actin used in each lane are indicated above the blots. In this experiment, the ATP G-actin used was recharged from ADP–G-actin as described previously (Pollard, 1986). (E) Equilibrium binding of coronin 1B to ADP or ATP/ADP-Pi–F-actin filaments. Bound coronin 1B was calculated from the depletion of coronin 1B from the supernatant fraction of experiments similar to those shown in C and D, and quantified by densitometry. Three different protein preparations were used to generate the data points, which are presented as means ± standard errors of the mean (±s.e.m.). (F) Representative Coomassie-Blue-stained gels comparing F-actin-binding capability of wild-type coronin 1B and R30D mutant. (G) Equilibrium binding of wild-type coronin 1B or R30D mutant to ATP/ADP-Pi–F-actin (1.5 µM F-actin in pellets).

View larger version (34K): [in this window] [in a new window]   Fig. 4. The R30D mutation does not perturb other known coronin 1B molecular interactions. (A) HEK293 cells were co-transfected with Myc-tagged coronin 1B and various GFP-tagged coronin 1B mutants: wild type (WT), R30D mutant, coil-coiled domain (CC, 451-489) alone or a deletion mutant lacking the coil-coiled domain (CC, 1-450). We immunoprecipitated Myc-tagged coronin 1B with Myc antibodies and blotted with either anti-GFP or Myc antibodies. The results demonstrate that the coiled-coil domain of coronin 1B is both necessary and sufficient for oligomerization and that the R30D mutation does not affect oligomerization. (B) Rat2 fibroblasts were infected with the knockdown/rescue lentivirus expressing the coronin 1B shRNA and either wild-type coronin 1B or the R30D mutant. Cells were subjected to the in vivo phosphorylation and dephosphorylation assay described in Materials and Methods. (C) In vitro direct-binding assay using purified coronin 1B immobilized on Ni-NTA beads and purified bovine Arp2/3 complex in solution (5 nM). (D) Bound Arp2/3 complex on beads from experiments described in C was quantified by densitometry and normalized to the amount of coronin 1B. Results from three independent experiments are presented as the mean ± s.d. (E) Cells as described in B were lysed and GFP-tagged coronin 1B was immunoprecipitated using anti-GFP antibodies. To visualize the co-immunoprecipitated Arp2/3 complex, the R30D lane was intentionally overloaded. Arrowhead, GFP-tagged coronin 1B; arrow, endogenous coronin 1B.

View larger version (38K): [in this window] [in a new window]   Fig. 5. Coronin 1B protects filaments from cofilin-induced depolymerization and binds antagonistically with cofilin to F-actin. (A) Time course of actin polymer concentration assembled from rabbit smooth muscle ATP G-actin, 30% pyrene labelled, after dilution from 5 µM to 0.1 µM, either alone, with phalloidin or in the presence of coronin 1B and cofilin. (B) Same experimental conditions as in A with phalloidin, coronin 1B and cofilin, or in the presence cofilin alone. (C) Representative Coomassie-Blue-stained gels showing the F-actin binding ability of coronin 1B in the presence of 2 µM cofilin, with or without pre-incubating actin filaments with 5 µM phalloidin. (D) Equilibrium binding of cofilin to actin filaments (assembled from 2 µM ATP G-actin and stored at 4°C overnight) in the presence of 1 µM coronin 1B. (E) Equilibrium binding of coronin 1B to 2 µM actin filaments in the presence of 1 µM cofilin. (F) Representative Coomassie-Blue-stained gel showing the high-speed actin co-sedimentation assay using increasing concentrations of the coronin 1B CC mutant with 2 µM actin and 1 µM cofilin.

View larger version (147K): [in this window] [in a new window]   Fig. 6. Coronin 1B can bundle actin filaments in vitro. (A) Electron micrograph of actin filaments, inset showing 4x magnification. Actin filaments (1 µM) were negatively stained and examined by transmission electron microscopy. Arrowheads indicate areas where filaments lie with close apposition; bar, 100 nm. (B) Electron micrograph of actin filaments (1 µM) with coronin 1B (10 nM); bar, 100 nm. (C,D) Electron micrograph of actin filaments (1 µM) with either wild-type coronin 1B or the R30D mutant (1 µM); bar, 1 µm. (E) Representative Coomassie-Blue-stained gel showing low-speed co-sedimentation of coronin 1B with F-actin. Samples were prepared as described in C and D, then subjected to centrifugation for 5 minutes at 13,000 g.

View larger version (61K): [in this window] [in a new window]   Fig. 7. F-actin binding is required for efficient and stable leading edge localization of coronin 1B. (A) Rat2 fibroblasts expressing GFP-tagged R30D mutant and lacking endogenous coronin 1B were EGF-stimulated (2 minutes) and stained for the Arp2/3 complex (p34Arc subunit), cortactin, F-actin (phalloidin) and VASP. (B) Rat2 fibroblasts lacking endogenous coronin 1B and expressing GFP-tagged wild-type protein (WT) or R30D mutant protein were stained for the Arp2/3 complex (p34Arc subunit) and cortactin following recovery from ATP depletion. (C) The lamellipodial colocalization analysis was performed on the protruding areas of the Rat2 fibroblasts prepared as in B. Fluorescence intensity values were normalized, combined from multiple cells. N, number of cells analyzed per treatment; data are given as the mean ± s.e.m. (Top) Endogenous coronin 1B, soluble GFP and Cortactin. (Middle and bottom) GFP-tagged coronin 1B (WT or R30D), p34Arc (Arp2/3) and cortactin. The indexes of relative leading edge enrichment of GFP signals from various cells were calculated and presented as the percentile mean ± s.d. in the center of the graphs. (D) Representative Rat2 fibroblast, whose endogenous coronin 1B was replaced with GFP-tagged wild-type protein, was subjected to fluorescence recovery after photobleaching (FRAP) analysis. A differential interference contrast (DIC) image is presented on the left and individual frames from time-lapse imaging of GFP are presented on the right. Bleached regions are circled; 1, lamellipodia; 2, cytoplasm). (E,F) Representative fluorescent intensity profiles from the two-spot FRAP analysis of Rat2 fibroblasts whose endogenous coronin 1B was replaced with GFP-tagged wild-type protein or GFP-tagged R30D mutant. (G) Statistical analysis of the FRAP immobile fraction across multiple cells. Data are presented as the mean, error bars indicate 95% confidence intervals. Dunnett multiple comparison test was performed as the post-test for one-way ANOVA using the first column as control. N, number of cells analyzed per treatment; **P<0.01, ***P<0.001.

View larger version (39K): [in this window] [in a new window]   Fig. 8. The R30D mutant cannot rescue depletion phenotypes of coronin 1B. (A) Rat2 fibroblasts were infected with lentivirus expressing coronin 1B shRNA (KD-1B) without or with GFP-tagged wild-type or GFP-tagged R30D coronin 1B or with control shRNA (NS). Cells were subjected to single-cell tracking. Data are presented as mean cell speed, error bars indicate 95% confidence intervals. (N, number of cells tracked per treatment; **P<0.01). (B-D) Cells described in A were subjected to kymography analysis. Protrusion parameters were determined from at least six randomly selected cells; n, number of protrusion events analyzed per treatment. Data are presented as the mean, error bars indicate 95% confidence intervals. Dunnett multiple comparison test was performed as a post-test for one-way ANOVA using the first column as control; *P<0.05, **P<0.01.