Role of the Adapter Protein Abi1 in Actin-associated Signaling and Smooth Muscle Contraction*

Background: Neuronal Wiskott-Aldrich syndrome protein (N-WASP) regulates smooth muscle contraction by affecting actin polymerization. Results: Abi1 activated by c-Abl regulates N-WASP and smooth muscle contraction, and reciprocally controls c-Abl activation. Conclusion: The adapter protein Abi1 is essential for the regulation of actin cytoskeleton signaling and force development. Significance: The Abi1-mediated actin cytoskeleton process is a novel mechanism for the regulation of smooth muscle contraction. Actin filament polymerization plays a critical role in the regulation of smooth muscle contraction. However, our knowledge regarding modulation of the actin cytoskeleton in smooth muscle just begins to accumulate. In this study, stimulation with acetylcholine (ACh) induced an increase in the association of the adapter protein c-Abl interactor 1 (Abi1) with neuronal Wiskott-Aldrich syndrome protein (N-WASP) (an actin-regulatory protein) in smooth muscle cells/tissues. Furthermore, contractile stimulation activated N-WASP in live smooth muscle cells as evidenced by changes in fluorescence resonance energy transfer efficiency of an N-WASP sensor. Abi1 knockdown by lentivirus-mediated RNAi inhibited N-WASP activation, actin polymerization, and contraction in smooth muscle. However, Abi1 silencing did not affect myosin regulatory light chain phosphorylation at Ser-19 in smooth muscle. In addition, c-Abl tyrosine kinase and Crk-associated substrate (CAS) have been shown to regulate smooth muscle contraction. The interaction of Abi1 with c-Abl and CAS has not been investigated. Here, contractile activation induced formation of a multiprotein complex including c-Abl, CAS, and Abi1. Knockdown of c-Abl and CAS attenuated the activation of Abi1 during contractile activation. More importantly, Abi1 knockdown inhibited c-Abl phosphorylation at Tyr-412 and the interaction of c-Abl with CAS. These results suggest that Abi1 is an important component of the cellular process that regulates N-WASP activation, actin dynamics, and contraction in smooth muscle. Abi1 is activated by the c-Abl-CAS pathway, and Abi1 reciprocally controls the activation of its upstream regulator c-Abl.

Despite its important role, the mechanisms that regulate smooth muscle contraction are not completely understood.
Myosin activation by phosphorylation is a key cellular process that regulates force development in smooth muscle. Myosin regulatory light chain phosphorylation at Ser-19 by myosin light chain kinase activates myosin ATPase, and induces sliding of contractile filaments and smooth muscle contraction (1,2). In addition, recent studies have shown that a pool of actin monomers is assembled onto actin filaments in smooth muscle in response to contractile stimulation. Inhibition of actin polymerization by pharmacological tools and molecular approach attenuates smooth muscle contraction without affecting myosin phosphorylation (3)(4)(5)(6). Dynamic changes in the actin cytoskeleton may allow smooth muscle cells to adjust their contractile status upon changes in external environments (3). These studies suggest that both myosin activation and reorganization of the actin cytoskeleton are necessary for force development in smooth muscle. Myosin may serve as an "engine" for smooth muscle contraction whereas the actin cytoskeleton may function as a "transmission system" in smooth muscle. There is considerable information about the regulation of myosin activation (1,2,7). In contrast, our knowledge regarding modulation of the actin cytoskeleton in smooth muscle is limited.
Actin polymerization in smooth muscle may be regulated by neuronal Wiskott-Aldrich syndrome protein (N-WASP), 2 a member of the WASP/WAVE protein family (8,9). In the unstimulated state, N-WASP exists in an autoinhibited conformation, wherein the verprolin-cofilin-acidic domain in the C terminus of N-WASP is masked by an intramolecular interaction with the N-terminal GTPase-binding domain (9). Contractile stimulation induces a conformational change, resulting in the release of the C-terminal part of N-WASP, which activates actin polymerization and branching mediated by the Arp2/3 complex (6, 8 -10). c-Abl interactor 1 (Abi1) is an adapter protein that has been implicated in the regulation of actin dynamics. In vitro biochemical studies suggest that Abi1 directly binds to N-WASP, which activates the N-WASP and Arp2/3-dependent actin polymerization (11). Moreover, Abi1 has been shown to modulate cell adhesion and migration, which are associated with dynamic changes in the actin cytoskeleton (12,13). Additionally, c-Abl tyrosine kinase regulates smooth muscle force development by controlling actin dynamics (8,14,15). Furthermore, CAS (Crk-associated substrate) has been shown to participate in the regulation of smooth muscle contraction and signaling (8, 16 -19). However, the interaction of Abi1 with c-Abl and CAS has not been investigated.
The objective of this study was to evaluate the role of Abi1 in N-WASP activation, actin polymerization, and contraction in smooth muscle. Furthermore, we also assessed whether c-Abl and CAS regulate the activation of Abi1, or vice versa in smooth muscle in response to contractile activation.

EXPERIMENTAL PROCEDURES
Cell Culture-Human airway smooth muscle (HASM) cells were prepared from human airway smooth muscle tissues that were obtained from the International Institute for Advanced Medicine. Human tissues were non-transplantable and consented for research. This study was approved by the Albany Medical College Committee on Research Involving Human Subjects. Briefly, muscle tissues were incubated for 20 min with dissociation solution (130 mM NaCl, 5 mM KCl, 1.0 mM CaCl 2 , 1.0 mM MgCl 2 , 10 mM Hepes, 0.25 mM EDTA, 10 mM D-glucose, 10 mM taurine, pH 7, 4.5 mg of collagenase (type I), 10 mg of papain (type IV), 1 mg/ml of BSA, and 1 mM dithiothreitol). All enzymes were obtained from Sigma. The tissues were then washed with Hepes-buffered saline solution (composition in mM: 10 Hepes, 130 NaCl, 5 KCl, 10 glucose, 1 CaCl 2 , 1 MgCl 2 , 0.25 EDTA, 10 taurine, pH 7). The cell suspension was mixed with Ham's F-12 medium supplemented with 10% (v/v) fetal bovine serum (FBS) and antibiotics (100 units/ml of penicillin, 100 g/ml of streptomycin). Cells were cultured at 37°C in the presence of 5% CO 2 in the same medium. The medium was changed every 3-4 days until the cells reached confluence, and confluent cells were passaged with trypsin/EDTA solution (20 -23). Smooth muscle cells within passage 5 were used for the studies.
Immunoblot Analysis-Cells were lysed in SDS sample buffer composed of 1.5% dithiothreitol, 2% SDS, 80 mM Tris-HCl, pH 6.8, 10% glycerol, and 0.01% bromphenol blue. The lysates were boiled in the buffer for 5 min and separated by SDS-PAGE. Proteins were transferred to a nitrocellulose membrane. The membrane was blocked with bovine serum albumin or milk for 1 h and probed with the use of primary antibody followed by horseradish peroxidase-conjugated secondary antibody (Fisher Scientific). Proteins were visualized by enhanced chemiluminescence (Fisher Scientific) using the LAS-4000 Fuji Image System. Abi1 antibody was purchased from Sigma. Antibodies against N-WASP, phosphomyosin light chain (Ser-19), myosin light chain, c-Abl, and phospho-Abl (Tyr-412) were purchased from Santa Cruz Biotechnology. CAS antibody was purchased from BD Biosciences and glyceraldehyde-3-phosphate dehy-drogenase (GAPDH) antibody was purchased from Fitzgerald (Acton, MA). The levels of total protein or phosphoprotein were quantified by scanning densitometry of immunoblots (Fuji Multigauge Software). The luminescent signals from all immunoblots were within the linear range.
Co-immunoprecipitation Analysis-Protein-protein interactions and protein complex formation were evaluated by coimmunoprecipitation analysis as previously described (8,21,24) with minor modifications. Briefly, cell extracts were incubated overnight with corresponding antibodies and then incubated for 2-3 h with 125 l of a 10% suspension of protein A-Sepharose beads. Immunocomplexes were washed four times in buffer containing 50 mM Tris-HCl, pH 7.6, 150 mM NaCl, and 0.1% Triton X-100. The immunoprecipitates were separated by SDS-PAGE followed by transfer to nitrocellulose membranes. The membranes of immunoprecipitates were probed with the use of corresponding antibodies.
Lentivirus-mediated RNAi in Cells-For Abi1 knockdown (KD), lentiviruses encoding Abi1 shRNA or control shRNA were purchased from Santa Cruz Biotechnology. HASM cells were infected with control shRNA lentivirus or Abi1 shRNA lentivirus for 12 h. They were then cultured for 3-4 days. Positive clones expressing shRNAs were selected by puromycin. Immunoblot analysis was used to determine the expression levels of Abi1 in these cells. Abi1 KD cells and cells expressing control shRNA were stable at least five passages after the initial infection. The experimental procedures for generating c-Abl KD cells were previously described (22,23).
FRET Analysis-HASM cells cultured in glass-bottom dishes were transfected with plasmids encoding the N-WASP sensor using the FuGENE HD transfection reagent kit (Promega), and analyzed 2-3 days after transfection by using a laser scanning confocal microscope (Zeiss 510 Meta). Briefly, cells were excited at wavelength of 458 nm, and emission of CFP and YFP was simultaneously collected every 50 s. Exposure time was between 1 and 3 s. When exposed to 458-nm wavelength, fluorescent intensity of CFP is 12-fold higher than YFP (absorbance) based on spectrum analysis (Molecular Probes, Fluorescence SpectraViewer). Appropriate microscope setting (laser power level, detector gain and amplifier gain) was used to minimize the potential bleed-through. The same microscope setting was used for the experiments. Thus, if any YFP emission occurs, the majority of the fluorescence signals should stem from FRET (25,26). For quantification of FRET efficiency, the region of interest of cells were positioned and the fluorescent intensity of each channel was measured by Zeiss Analysis software. CFP/YFP fluorescent ratios were used to assess the FRET efficiency. For in vitro analysis, extracts of HASM cells transfected with plasmids for the N-WASP sensor were collected 48 h after transfection. Cdc42 and Abi1 proteins were produced as previously described (20,27). Cell extracts were treated with 10 g/ml of Abi1 or Cdc42 plus GTP (100 M), or were left untreated. Cell extracts were placed in a PTI fluorospectrometer for measurement of the emissions of CFP and YFP.
Analysis of F-actin/G-actin Ratios-The content of F-actin and G-actin in smooth muscle was measured using a method as previously described (8,10,28). Briefly, smooth muscle cells were treated with F-actin stabilization buffer (50 mM PIPES, pH 6.9, 50 mM NaCl, 5 mM MgCl 2 , 5 mM EGTA, 5% glycerol, 0.1% Triton X-100, 0.1% Nonidet P-40, 0.1% Tween 20, 0.1% ␤-mercaptoethanol, 1 mM ATP, 1 g/ml of pepstatin, 1 g/ml of leupeptin, 10 g/ml of benzamidine). The supernatants of the protein extracts were collected after centrifugation at 151,000 ϫ g for 60 min at 37°C. The pellets were resuspended in ice-cold H 2 O plus 1 M cytochalasin D and then incubated on ice for 1 h to dissociate F-actin. The resuspended pellets were gently mixed every 15 min. The supernatant of the resuspended pellets was collected after centrifugation at 16,100 ϫ g for 2 min at 4°C. Equal volumes of the first (G-actin) or second (F-actin) supernatant were subjected to immunoblot analysis using ␣-actin antibody. The amount of F-actin and G-actin was determined by scanning densitometry.

Measurement of Smooth Muscle
Contraction-Human bronchial rings (diameter, 5 mm) were placed in physiological saline solution at 37°C in a 25-ml organ bath and attached to a Grass force transducer connected to a computer with A/D converter (Grass). For lentivirus-mediated RNAi in tissues, the epithelium layer of human bronchial rings was removed by using forceps. They were then transduced with lentivirus encoding Abi1 shRNA or control shRNA for 3-4 days. Force development in response to contractile activation was compared before and after lentivirus transduction. For biochemical analysis, human tissues were frozen using liquid nitrogen and pulverized as previously described (16,29).
Statistical Analysis-All statistical analysis was performed using Prism 4 software (GraphPad Software, San Diego, CA). Comparison among multiple groups was performed by oneway analysis of variance followed by Tukey's multiple comparison test. Differences between pairs of groups were analyzed by Student-Newman-Keuls test or Dunn's method. Values of n refer to the number of experiments used to obtain each value. p Ͻ 0.05 was considered to be significant.

RESULTS
Contractile Activation Increases the Association of Abi1 with N-WASP in HASM Cells/Tissues-N-WASP is known to regulate force development in smooth muscle in response to con-tractile activation (8,9,28). The role of Abi1 in smooth muscle has not been investigated. We hypothesized that contractile stimulation may enhance the interaction of Abi1 with N-WASP, which may facilitate the activation of N-WASP in smooth muscle. To test this, HASM cells were treated with acetylcholine (ACh) or left unstimulated. Extracts of smooth muscle cells were immunoprecipitated with the use of Abi1 antibody, and blots of the precipitates were probed using antibodies against N-WASP and Abi1. In unstimulated cells, the amount of N-WASP in Abi1 immunoprecipitates was relatively low. In contrast, the amount of N-WASP in the precipitates was increased in response to activation with ACh. The ratios of N-WASP/Abi1 were higher in ACh-stimulated cells than in control cells (Fig. 1A). In addition, we found that N-WASP was not detected in Abi1 precipitates from Abi1 KD cells (See below), suggesting the specificity of the assay (Fig. 1B).
To verify this, we used reverse co-immunoprecipitation analysis. Cell extracts were immunoprecipitated using N-WASP antibody, and blots of the immunoprecipitates were probed using antibodies against Abi1 and N-WASP. The level of Abi1 in the precipitates from stimulated cells was higher as compared with unstimulated cells (Fig. 1C).
Furthermore, we evaluated the effects of contractile activation on the interaction of Abi1 with N-WASP at the tissue level. Human bronchial rings were treated with ACh, and Abi1 immunoprecipitates of tissue extracts were subjected to immunoblot analysis. The level of N-WASP in Abi1 precipitates was higher in stimulated tissues than in unstimulated tissues (Fig. 1D).

Effects of ACh on N-WASP Activation in Live Smooth
Muscle Cells-To monitor N-WASP activation in live cells, we constructed a fluorescence resonance energy transfer (FRET)based N-WASP sensor. We fused CFP to the N terminus lacking the first 30 amino acids, and YFP to the C terminus of N-WASP ( Fig. 2A) (30). Because Cdc42 is known to activate N-WASP in in vitro studies and in smooth muscle (9, 10), we evaluated the basal and Cdc42-treated FRET signal of the sensor. Extracts of unstimulated HASM cells expressing the N-WASP sensor were treated with Cdc42 plus GTP, or left untreated. The emission ratios of YFP/CFP were measured using a spectrofluorometer. The basal FRET signal of the sensor was higher, whereas the Cdc42-mediated FRET efficiency was lower (Fig. 2B). The results demonstrate that changes in the FRET signal represent the activation state of N-WASP.
To assess whether Abi1 directly regulates N-WASP activation, extracts of unstimulated cells expressing the N-WASP probe were treated with Abi1. The emission ratios of YFP/CFP were then determined. Treatment with Abi1 reduced FRET efficiency of the biosensor, suggesting an important role of Abi1 in activating N-WASP in vitro (Fig. 2B).
We then evaluated the effects of contractile stimulation on N-WASP activation in live cells. HASM cells expressing the N-WASP sensor were treated with ACh, and emissions of CFP and YFP were monitored live by laser-scanning confocal microscopy. Before stimulation, the emission ratios of CFP/ YFP were relatively lower. In response to ACh stimulation, the CFP signal was increased, whereas the YFP signal was decreased (Fig. 2C). The CFP/YFP ratios gradually increased during the course of contractile activation (Fig. 2D). These results suggest that contractile stimulation activates N-WASP in live smooth muscle cells. In addition, the time course of N-WASP activation is consistent with that of F/G-actin ratios and contraction (supplemental Fig. S1).
To further characterize the biosensor, cells expressing single-probe constructs (CFP-tagged construct or YFP-tagged construct) were treated with ACh and emission of CFP and YFP were evaluated using the confocal microscope under the same experimental condition. For cells expressing the CFP construct, CFP emission was high, whereas YFP emission was undetected. In cells expressing the YFP construct, neither CFP nor YFP emission was detected. In addition, ACh stimulation did not alter CFP/YFP ratios of V215G N-WASP mutant (to disrupt Cdc42 binding) or P320L/P330L N-WASP mutant (to impair Abi1 binding) (supplemental Figs. S2 and S3). The results validate the selectivity and sensitivity of the biosensor.

KD of Abi1 Attenuates Activation of N-WASP in Smooth
Muscle Cells/Tissues-To reveal the functional role of Abi1 in N-WASP activation, we evaluated the effects of Abi1 KD on the FRET signal in smooth muscle cells. Stable Abi1 KD cells were generated by using lentivirus-mediated RNAi. Immunoblot analysis showed that the protein level of Abi1 in cells infected with virus for Abi1 shRNA was lower compared with control cells. However, the expression level of GAPDH was similar in these cells. Ratios of Abi1/GAPDH were lower in Abi1 KD cells than in control cells (Fig. 3A).
Stable Abi1 KD cells and cells expressing control shRNA were transfected with plasmids encoding the N-WASP sensor. The effects of ACh stimulation on N-WASP activity in these

Abi1 Regulates Actin Dynamics and Smooth Muscle Contraction
cells were determined 2-3 days after transfection. In cells expressing control shRNA, the ratios of CFP/YFP were increased in response to stimulation with ACh. However, the ACh-induced ratios of CFP/YFP were reduced in Abi1 KD cells (Fig. 3B).
We also evaluated the effects of Abi1 KD on Arp2/N-WASP ratios (another indication of N-WASP activation) (8,9) in tissues. We developed a lentivirus-mediated RNAi to inhibit the expression of Abi1 in smooth muscle tissues. Briefly, human bronchial rings were transduced with lentivirus encoding control shRNA or Abi1 shRNA for 3-4 days. Immunoblot analysis confirmed lower Abi1 expression in these tissues (Fig. 4A). More importantly, the levels of Arp2 in N-WASP precipitates in response to ACh stimulation were lower in tissues infected with virus encoding Abi1 shRNA than in uninfected tissues and tissues transduced with virus encoding control shRNA (Fig. 4B).

Abi1 KD Inhibits Increases in F/G-actin Ratios, but Not Myosin Light Chain Phosphorylation at Ser-19 elicited by
ACh-Because both actin polymerization and myosin phosphorylation are critical cellular mechanisms that regulate smooth muscle contraction (1-3, 6, 28), we evaluated the effects of Abi1 knockdown on actin dynamics and myosin activation by using the fractionation assay and immunoblot analysis, respectively. The increase in F/G-actin ratios in response to ACh stimulation was reduced in Abi1 KD cells compared with uninfected cells and cells expressing control shRNA (Fig. 5A). However, myosin light chain phosphorylation at Ser-19 was not different among uninfected cells, cells infected with virus for control shRNA, and Abi1 KD cells (Fig. 5B).
Force Development in Response to ACh Stimulation Is Reduced in Abi1-deficient Bronchial Rings-We assessed the role of Abi1 in smooth muscle contraction. Briefly, the contractile responses of human bronchial rings were determined, after which lentivirus encoding control shRNA or Abi1 shRNA was transduced into the tissues as described above. Force development in bronchial rings was then determined. In uninfected tissues and tissues treated with virus encoding control shRNA, the contractile response was 75% of preincubation force. However, contractile force in tissues transduced with virus for Abi1 shRNA was reduced to 35% of the preincubation level (Fig. 5C).
Activation with ACh Induces Formation of the Multiprotein Complex Including c-Abl, CAS, and Abi1-The regulation of Abi1 in smooth muscle during contractile activation has not been investigated. Because c-Abl and CAS have been shown to regulate smooth muscle functions (8, 14, 17-20, 31), we hypothesized that contractile stimulation may trigger forma-tion of the multiprotein complex containing c-Abl, CAS, and Abi1, which may promote the activation of Abi1. To test this, blots of CAS immunoprecipitates from unstimulated and AChstimulated cells were probed with the use of antibodies against c-Abl, CAS, and Abi1. The levels of c-Abl and Abi1 in CAS precipitates were increased in stimulated cells as compared with unstimulated cells (Fig. 6).
KD of c-Abl Attenuates the Interaction of CAS with Abi1 and the Activation of Abi1-Thus far, we have discovered that ACh stimulation promotes the assembly of the multiprotein complex containing c-Abl, CAS, and Abi1. Because c-Abl is a known upstream regulator of CAS (8), this raises the possibility that c-Abl may regulate the association of CAS with Abi1 and the activity of Abi1. To test this, we determined the effects of c-Abl KD on association of CAS with Abi1 and Abi1/N-WASP coupling (indication of Abi1 activation) in smooth muscle cells by co-immunoprecipitation analysis. c-Abl KD cells and cells expressing scramble shRNA were generated by using lentivirus-mediated RNAi as previously described (22,23). The interaction of CAS with Abi1, and Abi1/N-WASP coupling during ACh stimulation were reduced in c-Abl KD cells as compared with uninfected cells and cells expressing scramble shRNA (Fig. 7).

Down-regulation of CAS Inhibits the Coupling of Abi1 with N-WASP upon ACh Stimulation-Our previous studies have
shown that CAS is required for smooth muscle contraction (17,18). To determine whether CAS controls the activation of Abi1, we assessed the effects of CAS antisense (17,18) on the activation of Abi1. The association of Abi1 with N-WASP in response to ACh stimulation was lower in CAS KD cells than in untreated cells and cells treated with CAS sense (Fig. 8).
Silencing of Abi1 Attenuates c-Abl Phosphorylation and the Association of c-Abl with CAS-It has been suggested that Abi1 is a target of c-Abl in in vitro studies and NIH 3T3 cells (32,33). To assess whether Abi1 conversely affects c-Abl activation, we assessed the effects of Abi1 KD on c-Abl phosphorylation at Tyr-412 (an indication of c-Abl activation) (8, 34) by immunoblot analysis. c-Abl phosphorylation during contractile activation was lower in Abi1 KD cells than in uninfected cells and cells expressing control shRNA (Fig. 9A). Immunoblot analysis showed that the phospho-Abl band was barely detected in Abl KD cells, suggesting the specificity of the phospho-Abl antibody under the experimental condition (Fig. 9B). To further investigate this, the effects of Abi1 KD on the coupling of c-Abl with CAS (another index of c-Abl activation) were determined by co-immunoprecipitation analysis. The interaction of c-Abl with CAS in response to contractile activation was also reduced by Abi1 KD (Fig. 9C).

DISCUSSION
Actin polymerization has recently emerged as a critical cellular process that regulates smooth muscle contraction. Our knowledge regarding how actin dynamics is regulated in smooth muscle just begins to accumulate. Our present studies suggest that the adapter protein Abi1 is an essential component of the cellular process that regulates dynamics of the actin cytoskeleton in smooth muscle during contractile activation. Furthermore, we have identified a novel activation loop, in which  c-Abl tyrosine kinase and CAS modulate the activation of Abi1, and Abi1 conversely affects the activation of c-Abl.
We have previously shown that contractile stimulation of smooth muscle activates N-WASP as evidenced by co-immunoprecipitation (8,28). To monitor N-WASP activation in live cells, we constructed a FRET-based N-WASP sensor. To characterize the biosensor, we used extracts from unstimulated cells expressing the N-WASP sensor because these cells have a lower activation level of N-WASP (see Fig. 2, C and D). The addition of Cdc42 plus GTP, which is equal to active Cdc42, decreases the YFP/CFP ratios. The results suggest that changes in the FRET signal represent the activation state of N-WASP. In addition, our results suggest that Abi1 activates N-WASP in vitro (Fig. 2B). Although endogenous Abi1 (Ͻ1 g/ml) exists in cell extracts, the addition of a high concentration of Abi1 (10 g/ml) may easily have access to the biosensor, thus activating N-WASP. Furthermore, our results suggest that Abi1 or Cdc42 itself is suffi-cient to activate N-WASP in vitro, which is consistent with previous biochemical results published by others (11).
Properties of our N-WASP biosensor are similar to the N-WASP biosensors constructed by others despite minor structural differences (30,35). We conjugated CFP to the N terminus lacking the first 30 amino acids, and YFP to the C terminus of N-WASP, whereas Ward et al. (30) fused YFP to the N terminus without the first 30 residues, and CFP to the C terminus. In addition, Lorenz et al. (35) used full-length N-WASP for the construction. These independent studies demonstrate that FRET analysis is a novel technology to study N-WASP activation.
In the present study, we provide the first evidence that contractile stimulation activates N-WASP in live smooth muscle cells by using the N-WASP biosensor. The results also provide direct evidence that contractile stimulation is able to induce conformational changes of N-WASP from a "closed" structure to an "open" conformation. When in the closed state, the verprolincofilin-acidic domain at the C terminus of N-WASP binds to its GTP-binding domain, masking its binding motif for the Arp2/3 complex and inhibiting the activity of N-WASP. When in the open state, N-WASP exposes the binding motif for the Arp2/3 complex and initiating actin polymerization and branching mediated by the Arp2/3 complex (8,9,28).
Abi1 is an adapter protein that has a role in the regulation of cell adhesion, migration, and endocytosis (11,12). In vitro biochemical studies suggest that the SH3 domain of Abi1 directly binds to the proline-rich domain of N-WASP, activating the N-WASP and Arp2/3-dependent actin polymerization (11). However, the role of Abi1 in smooth muscle contraction has not been investigated. In this report, we provide several lines of evidence to suggest that Abi1 may activate N-WASP in smooth muscle cells/tissues upon contractile activation. First, contractile activation induced an increase in the interaction of Abi1 with N-WASP. Second, Abi1 was able to directly activate N-WASP in vitro. Third, Abi1 knockdown by RNAi attenuated the activation of N-WASP in live smooth muscle cells as evidenced by FRET analysis of the N-WASP biosensor. Fourth, Abi1 knockdown also inhibited the interaction of N-WASP  with Arp2 (another indication of N-WASP activation) in human bronchial tissues in response to contractile activation.
Contractile activation of smooth muscle induces actin filament polymerization, and inhibition of actin polymerization attenuates smooth muscle force development (3,4). In this report, we found that silencing of Abi1 diminished actin polymerization and contractile force in smooth muscle, suggesting an important role of Abi1 in controlling actin dynamics and contraction in smooth muscle. Because ACh stimulation increases the association of Abi1 with N-WASP, and Abi1 silencing inhibits both N-WASP activation and actin dynamics, we propose that agonist stimulation may activate Abi1, which subsequently activates N-WASP and induces actin polymerization mediated by the Arp2/3 complex and contraction in smooth muscle. Furthermore, Abi1 KD did not affect increases in myosin light chain phosphorylation upon contractile stimulation, indicating that Abi1 is not involved in the regulation of myosin activation.
Actin polymerization may facilitate force development by several mechanisms. Actin polymerization may enhance the linkage of actin filaments to integrins strengthening the transduction of mechanical force between contractile units and extracellular matrix (3,6,10,28,36,37). In addition, actin polymerization may promote the "latch" formation of contractile elements, supporting force maintenance under the condition of lower cross-bridge phosphorylation (5,38). Finally, actin filament assembly may also increase the numbers of contractile units and the length of actin filaments, providing more and efficient contractile elements for force development (39).
CAS is a protein that participates in the regulation of smooth muscle contraction. Knockdown of CAS in smooth muscle inhibits actin polymerization and force development during contractile activation (8, 16 -19). Furthermore, c-Abl tyrosine kinase is required for smooth muscle contraction and actin dynamics. c-Abl regulates smooth muscle force development in part by controlling CAS (8,14,15). c-Abl undergoes phosphorylation at Tyr-412 in smooth muscle in response to contractile activation. Tyr-412 is located at the activation motif of the c-Abl kinase domain. When unstimulated, the activation motif of the kinase domain folds into the active site, thereby prevent-  ing binding of both the substrate and ATP. Phosphorylation at Tyr-412 induces conformation changes; the activation motif no longer blocks the active site, which leads the increase in kinase activity (8,34). Activated c-Abl may interact with CAS and activate it during contractile stimulation (8,15,21).
How c-Abl and CAS regulate smooth muscle contraction is incompletely understood. Here, we found that contractile activation induced formation of the protein complex containing c-Abl, CAS, and Abi1 in smooth muscle. Moreover, knockdown of c-Abl or CAS diminished the activation of Abi1 elicited by ACh. These findings strongly suggest that Abi1 is activated by the c-Abl-CAS pathway during contractile activation of smooth muscle.
Changes in the physiological state of a substrate may conversely affect the activation of its upstream regulator. In this report, silencing of Abi1 inhibited phosphorylation at Tyr-412 of c-Abl (an index of c-Abl activation), and the association of c-Abl with CAS (another indication of c-Abl activation). Because Abi1 is activated by the c-Abl-CAS cascade and Abi1 reciprocally affects the activation of c-Abl, we propose that c-Abl, CAS, and Abi1 form a unique activation loop in smooth muscle in response to agonist activation. The presence of Abi1 in the multiprotein complex may stabilize the conformation of c-Abl, rendering c-Abl in active status during agonist stimulation. This novel activation loop may assist smooth muscle cells to efficiently utilize their energy upon external activation.
Thus, we propose that in addition to myosin activation, agonist stimulation induces formation of the multiprotien complex including c-Abl, CAS, and Abi1, which subsequently activates N-WASP, actin polymerization, and smooth muscle contraction. Furthermore, Abi1 may stabilize the conformation of c-Abl, rendering c-Abl active (Fig. 10).