Am J Physiol Heart Circ Physiol 305: H1614 –H1623, 2013. First published September 16, 2013; doi:10.1152/ajpheart.00477.2013.
Phosphatidylinositol 4,5-bisphosphate regulates CapZ1 and actin dynamics in response to mechanical strain Jieli Li and Brenda Russell Department of Physiology and Biophysics, University of Illinois at Chicago, Chicago, Illinois Submitted 12 June 2013; accepted in final form 10 September 2013
Li J, Russell B. Phosphatidylinositol 4,5-bisphosphate regulates CapZ1 and actin dynamics in response to mechanical strain. Am J Physiol Heart Circ Physiol 305: H1614 –H1623, 2013. First published September 16, 2013; doi:10.1152/ajpheart.00477.2013.—Mechanical stress causes filament remodeling leading to myocyte hypertrophy and heart failure. The actin capping protein Z (CapZ) tightly binds to the barbed end of actin filaments, thus regulating actin assembly. The hypothesis is that the binding between CapZ and the actin filament is modulated through phosphatidylinositol 4,5-bisphosphate (PIP2) and how the COOH-terminus of CapZ1 regulates this binding. Primary neonatal rat ventricular myocytes (NRVMs) were strained at 10% amplitude and 1-Hz frequency. Dot blotting measured the PIP2 amount, and affinity precipitation assay assessed the direct interaction between PIP2 and CapZ1. Fluorescence recovery after photobleaching of green fluorescent protein-CapZ1 and actin-green fluorescent protein after 1 h of strain shows the dynamics significantly increased above the unstrained group. The increases in CapZ and actin dynamics were blunted by neomycin, suggesting PIP2 signaling is involved. The amount of PIP2 dramatically increased in NRVMs strained for 1 h. With a ROCK or RhoA inhibitor, changes were markedly reduced. Subcellular fractionation and antibody localization showed PIP2 distributed to the sarcomeres. More PIP2-bound CapZ1 was found in strained NRVMs. Less PIP2 bound to the CapZ1 with its COOHterminus intact than in the COOH-terminal mutant of CapZ1, suggesting some inhibitory role for the COOH-terminus. Myocyte hypertrophy normally induced by 48 h of cyclic strain was blunted by dominant negative RhoA or neomycin. This suggests that after many hours of cyclic strain, a possible mechanism for cell hypertrophy is the accumulation of thin filament assembly triggered partially by the increased PIP2 level and its binding to CapZ. cardiac hypertrophy; actin assembly; phosphoinositide; mechanotransduction; CapZ tentacle
an adaptation in response to pressure or volume overload. However, sustained biomechanical stress may drive changes in myocyte hypertrophy resulting in heart failure (8, 19). Therefore, the mechanistic responses to mechanical loading need to be better understood. Key sites for mechanotransduction are at the costamere and Z disc (30). Critical to the response is the dynamic nature of the cytoskeletal networks and their ability to undergo assembly and disassembly. In vertebrate-striated muscle, the barbed ends of the actin thin filaments are capped at the Z disc by a protein named capping protein Z (CapZ), which regulates turnover of actin subunits, thereby stabilizing the cytoskeleton (27, 33). CapZ is a dimer of ␣- and -subunits with a known crystal structure (42, 43), and truncation of the terminal tentacle on the CapZ dramatically decreases the actin binding affinity (45). The 1-isoform is located at the Z disc, but 2 is thought to
CARDIAC HYPERTROPHY IS
Address for reprint requests and other correspondence: B. Russell, Dept. of Physiology and Biophysics, Univ. of Illinois at Chicago, MC 901, 835 South Wolcott Ave., Chicago, IL 60612-7342 (e-mail:
[email protected]). H1614
localize at the intercalated disc and the myocyte periphery (14). The CapZ1 content is reduced in transgenic mice overexpressing CapZ2, leading to cardiac hypertrophy and death weeks after birth (14). In cultured myotubes, injection of a CapZ antibody inhibits the actin binding of CapZ1, as does expression of deletion of COOH-terminus of the CapZ1 subunit that caps actin poorly (17, 35). Thus the COOHterminus is a critical part of the actin-binding site for the CapZ1 subunit. Taken together, the evidence suggests that CapZ1 is critical for maintaining normal thin filament assembly, myocyte size, and cardiac performance. Many actin-binding proteins are regulated by specific interactions with phosphoinositides modulating activity and localization (23, 46). In vitro experiments show that phosphatidylinositol 4,5-bisphosphate (PIP2) regulates CapZ (16). Addition of PIP2 to actin filaments in a polymerization reaction leads to an increased assembly consistent with complete and rapid uncapping (20, 41). Myofilament tension generation is depressed by extraction of CapZ from cardiac myofilaments with PIP2, and tension can be restored to control level by addition of CapZ (28). Hypertrophic remodeling to neurohumoral stimulation involves PIP2 signaling and CapZ1 (15). Furthermore, according to the computational analysis, the CapZ-subunit has hydrophobic domains that might permit lipid binding to alter its interaction with actin (37). However, after mechanical stimulation, it is still not known whether PIP2 is directly involved in regulating CapZ, nor whether there is any subsequent control of actin assembly. In striated muscle, the COOH-terminal extended ends of the ␣- and -subunits of CapZ are the major actin-binding sites and act like two “tentacles” that are critical for actin capping (43). In biochemical studies, CapZ containing either COOHterminal tentacle alone is able to cap (9). The deletion of the COOH-terminal of -tentacle significantly weakens the actin capping and increases CapZ dynamics, thus elevating the off rate of actin (21). Therefore, to test the hypothesis that the -tentacle and PIP2 have a role in detachment, or “uncapping,” neonatal rat ventricular myocytes (NRVMs) were transfected with the green fluorescent protein-COOH-terminal deletion of CapZ1 (GFP-CapZ1-⌬C). The affinity precipitation by PIP2conjugated beads investigated the CapZ1-⌬C binding to PIP2. The present study examined the role of PIP2 in neonatal myocytes after mechanical strain. Results indicated that mechanical signaling involved an alteration in dynamics of both CapZ1 and actin through PIP2, which increases the turnover of actin filament. The COOH-terminal of CapZ1 was not itself the binding site of PIP2, but its removal appeared to release an inhibition enabling the remainder of the CapZ molecule to bind more PIP2. Furthermore, the production of PIP2 was regulated by the RhoA/ROCK pathway in response to mechanical stimulation. The hypertrophy seen after a 48-h
0363-6135/13 Copyright © 2013 the American Physiological Society
http://www.ajpheart.org
H1615
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
continuous strain was blunted by intervention with PIP2 and RhoA, suggesting the importance of these molecules in regulation of physiological responses to strain. MATERIALS AND METHODS
Ethical approval. All experiments were conducted following National Institutes of Health guidelines. All experiments were approved by the University Committee on the Use and Care of Animals at the University of Illinois at Chicago and the Biologic Resources Laboratory. All euthanasia was performed following the recommendations of the University Committee on the Use and Care of Animals and the Biologic Resources Laboratory and the Committee for the Humane Use of Animals. Cell culture. Hearts were removed and cells isolated from 1- to 2-day-old Sprague-Dawley rats with collagenase type II (Worthington, Lakewood, NJ) as previously described (6). NRVMs were resuspended, filtered through a metal sieve to remove large material, and plated at high density (1,000 cells/mm2) in PC-1 medium (Biowhittaker/Cambrex) on fibronectin coated (25 g/ml) FlexCell silicone membranes (200,000 cells/cm2). Cells were left undisturbed for 24 h in a 5% CO2 incubator. Unattached cells were removed by aspiration, and PC-1 media was replenished. Myocytes were incubated for another 24 h before the experiment. Mechanical strain. Cyclic mechanical strain was generated with a Flexcell Strain Unit (Model FX-4000, Flexcell International, Hillsborough, NC). NRVMs in PC-1 medium were strained at 10% biaxially at 1 Hz for 1 (for signaling studies) or 48 h (for myocyte size) (36). Adenovirus. Recombinant adenoviruses used were GFP, GFPCapZ1, GFP-CapZ1-⌬C, and hemagglutinin-tagged dominant-negative RhoA (Asn19) (HA-RhoA-DN), as described in the original articles (4, 11, 17, 21). NRVMs were infected with CapZ1, GFPCapZ1-⌬C (multiplicity of infection, 20), or HA-RhoA-DN (multiplicity of infection, 20) for 60 min at 37°C diluted in PC-1 medium. Viral medium was then replaced with virus-free PC-1 medium for 24 h as previously described (15, 21). Actin-GFP expression. Actin-GFP expression was induced by CellLight Reagents BacMam 2.0 Actin-GFP (Invitrogen). As previously described (21), NRVMs were transfected according to the manufacturer’s instructions. Infected NRVMs were returned to the culture incubator for at least 24 h. Transfected NRVMs exhibited an actin organization with a striated structure in myocytes. Furthermore, colocalization of actin-GFP and phalloidin indicated that the actin-GFP was successfully incorporated into all the F-actin structures within NRVMs (21). Thus actin-GFP can be considered to reflect the structure and behavior of F-actin accurately in transfected myocytes. Inhibitors. Neomycin has a high affinity for PIP2 and readily sequesters any available PIP2 in the cellular milieu, rendering it unavailable for its normal physiological roles (2). C3 transferase
Table 1. Recovery kinetics (Kfrap) for CapZ1 and actin under experimental conditions
Fig. 1. Increase in dynamics of capping protein Z 1 (CapZ1) after 1 h of cyclic strain of neonatal rat ventricular myocytes (NRVMs) depends on the phosphatidylinositol 4,5-bisphosphate (PIP2) pathway. A: microscopic images of living NRVMs infected with adenovirus-green fluorescent protein (GFP)-CapZ1 in no-strain, strained, neomycin no-strain, and neomycin-strained groups. Fluorescence recovery after photobleaching (FRAP) of the region of interest (ROI) for GFP-CapZ1 before, immediately after, and 30 min after photobleaching. B: average kinetic constant (Kfrap) of GFP-CapZ1 with all groups normalized to no-strain group showed GFP-CapZ1 in the strained group had increased kinetic rates. With the treatment of neomycin, the increased kinetic rate of GFP-CapZ1 in strained myocytes was significantly reduced. Means ⫾ SE; n ⬎ 3. *P ⬍ 0.05. Bar ⫽ 10 m.
Experimental Conditions
CapZ Kfrap (10⫺3·s⫺1)
Actin Kfrap (10⫺4·s⫺1)
Unstrained Strained Neomycin Neomycin ⫹ strain Y27632 Y27632 ⫹strain C3 transferase C3 transferase ⫹ strain
1.97 ⫾ 0.17 3.96 ⫾ 0.52* 1.90 ⫾ 0.68 1.73 ⫾ 0.60# 1.42 ⫾ 0.17 1.10 ⫾ 0.45# 1.70 ⫾ 0.65 1.24 ⫾ 0.40#
5.70 ⫾ 0.71 9.66 ⫾ 0.58* 5.68 ⫾ 1.86 5.20 ⫾ 0.49# 5.72 ⫾ 0.94 3.76 ⫾ 0.98# 6.08 ⫾ 1.69 6.93 ⫾ 0.84#
Values are means ⫾ SE. CapZ, capping protein Z. *P ⬍ 0.05 vs. unstrained neonatal rat ventricular myocytes; #P ⬍ 0.05 vs. strained neonatal rat ventricular myocytes.
AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
H1616
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
inhibits Rho proteins by ADP-ribosylation on asparagine 41 in the effector binding domain of the GTPase, thereby blocking their action. NRVMs were pretreated with C3 transferase (0.25 g/ml) or ROCK inhibitor Y27632 (10 mol/l) for 1 h, or neomycin (500 M) for 4 h, and then strained for 1 h. Dot blots. Whole cell lysates were extracted from NRVMs in each experimental condition, spotted onto nitrocellulose membrane, probed with PIP2 antibody (1:500, mouse IgG, Abcam, Cambridge, MA), GAPDH (1:1,000, Sigma), and detected using a horseradish peroxidase-conjugated secondary antibody (anti-mouse, horseradish peroxidase, Cell Signaling Technology, Boston, MA) and enhanced chemiluminescence. Subcellular fractionation and localization of PIP2. For subcellular fractionation of myocytes, the Calbiochem ProteoExtract Subcellular Proteome Extraction Kit was used (EMD Millipore, Billerica, MA). This method uses a previously described detergent-based protocol (5). Cellular proteins were sequentially extracted into four compartments: cytosolic, membrane/organelles, nuclei, and cytoskeleton. Briefly, NRVMs (⬃1 ⫻ 106/well) were seeded in Flexcell plates. After 1 h cyclic strain, NRVMs were permeabilized with Digitonin-EDTA in buffer I and kept at 4°C for 10 min (releasing cytosolic proteins without solubilizing membrane proteins). The supernatant was collected as the cytosolic fraction. Subsequently, the NRVMs were extracted with Triton-EDTA in buffer II and kept at 4°C for 30 min.
The supernatant contained the membrane fraction. The NRVMs were next recovered with Tween 20-deoxycholate-benzonase in buffer III and kept at 4°C for 10 min. The supernatant contained the nuclear fraction. Finally, the cytoskeleton remaining was immunostained with a PIP2 antibody (1:200, mouse IgG, Abcam) and ␣-actinin antibody (1: 200, Cell Signaling Technology, Danvers, MA). Cells were briefly washed three times in PBS between each extraction fraction to prevent cross contamination. After each fraction, cells were observed by microscopy to ensure that they were still attached to the dish. Cell integrity is maintained throughout the fractionation process. The accuracy of the fractionation method was verified with antibodies to well-documented subcellular distribution markers (heat shock protein 70 for cytostol, 1:200; -integrin for membrane, 1:1,000; and Histone H2B for nuclei, 1:500) (anti-heat shock protein 70 from Santa Cruz Biotechnology, Dallas, TX; anti--integrin from EMD Millipore; and anti-Histone H2B from Imgenex, San Diego, CA). PIP2 bead affinity precipitation. NRVMs were transfected with adeno-GFP, adeno-GFP-CapZ1, or GFP-CapZ1-⌬C. The myocytes were washed in PBS and lysed and sonicated in 500 l of wash/bind buffer containing 10 mM HEPES (pH 7.4), 0.25% Nonidet P-40, and 150 mM NaCl. After sonication and centrifugation at 10,000 g for 10 min in 4°C, 50 l of 50% PIP2-conjugated agarose beads (Echelon Bioscience, Salt Lake City, UT) in slurry were added to the supernatant. After overnight incubation at 4°C, beads were
Fig. 2. Increased dynamics of CapZ1 in NRVMs after 1 h of cyclic strain depends on the RhoA/ROCK pathway. A: microscopic images of living NRVMs infected with GFP-CapZ1, RhoA-inhibited (C3 transferase), ROCK inhibited (Y27632) in no-strain or strained groups. FRAP of ROI for GFP-CapZ1 before, immediately after, and 30 min after photobleaching showed that both C3 transferase and Y27632 attenuated responses in the strained groups. B: Kfrap with all groups normalized to no-strain group showed GFP-CapZ1 in the strained group had increased kinetic rates. With the treatment of C3 transferase or Y27632, the increased kinetic rate of GFP-CapZ1 in strained myocytes was significantly reduced. Means ⫾ SE; n ⬎ 3. *P ⬍ 0.05. Bar ⫽ 10 m. AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
washed three times in wash/bind buffer. The proteins were eluted from the PIP2 beads by heating at 50°C for 10 min in 2⫻ SDS-PAGE buffer. CapZ1 or CapZ1-⌬C was detected by anti-GFP (mouse, 1:1,000, Enzo Life Sciences). The bands of Western blot analysis are detected with an imager (Bio-Rad, Hercules, CA). Analysis of fluorescence recovery after photobleaching. Recently, several microscopic techniques, such as analysis of fluorescence recovery after photobleaching (FRAP) (29, 15), have begun to yield important qualitative and quantitative information on the processes that promote and regulate actin assembly in living myocytes. For FRAP of GFP-CapZ1, at least five beating and well-striated myocytes (as evidenced by GFP-CapZ tag) were randomly selected for each experiment. The GFP fusion protein was irreversibly bleached by laser excitation (488 m) at full power in a uniform square region of interest (ROI) lying midway between the myocyte nucleus and periphery. The intensity of the ROI was observed both before (t0) and immediately after (t1) bleaching and intermittently every 2 min until the end of the 30 min. Images were analyzed using Zeiss Imaging Browser. Plotted intensity values are given as a percentage of the difference between ROI (t0) and ROI (t1). Curves were fitted using nonlinear regression for two binding states in OriginPro (OriginLab, Northampton, MA): ROI共t兲 ⫽ 1 ⫺ C1e⫺Koff1t ⫺ C2e⫺Koff2t
H1617
CapZ1 dynamics induced by mechanical strain in NRVMs was examined using FRAP of CapZ1 transfected myocytes (Fig. 2A). C3 transferase markedly inhibited the strain-induced increases of CapZ1 dynamics (1.24 ⫾ 0.40 vs. 3.96 ⫾ 0.52, ⫻10⫺3·s⫺1, P ⬍ 0.05) (Fig. 2B). Signaling by ROCK was assessed. A RhoA downstream serine kinase involved in the actin filament organization, mediated strain-induced increased CapZ1 dynamics in NRVMs. The increased CapZ1 dynamics induced by strain were markedly reduced after Y27632 treatment (1.10 ⫾ 0.45 vs. 3.96 ⫾ 0.52, ⫻10⫺3·s⫺1, P ⬍ 0.05), but no change was observed in Y27632 treated myocytes that were not strained (1.10 ⫾ 0.45 vs. 1.42 ⫾ 0.17, ⫻10⫺3·s⫺1). Increased actin dynamics induced by mechanical strain depend on both PIP2 and RhoA/ROCK pathways. The actin-GFP had strong striations in NRVMs, and signals were detected in ROI (Fig. 3, A and B). After 1 h of 10% 1-Hz cyclic strain, the actin-GFP had a faster dynamic protein exchange in strained compared with the unstrained myocytes (9.66 ⫾ 0.58 vs. 5.70 ⫾ 0.71, ⫻10⫺4·s⫺1, P ⬍ 0.05) (Fig. 3C). The increased dynamics of
(1)
Average kinetic constant (Kfrap) for dynamics was calculated using the following formula: Kfrap ⫽ C1Koff1 ⫹ C2Koff2
(2)
For FRAP of actin-GFP, the ROI was observed both before (t0) and immediately after (t1) bleaching, and intermittently every 20 s until the end of the 8-min study, referred to the recovery time of actin-GFP in previous study (3). Also, actin has one binding state. Thus the equation for curve fitting using nonlinear regression in OriginPro was ROI共t兲 ⫽ 1 ⫺ C1e⫺Koff1t
(3)
And Kfrap was calculated using the following formula: Kfrap ⫽ C1Koff1
(4)
Statistics. Data are presented as means ⫾ SE. Statistical significance was determined by one-way ANOVA. Significance was taken at P ⬍ 0.05. RESULTS
Increased CapZ1 dynamics induced by mechanical strain depend on the PIP2 pathway. The GFP-CapZ1 had strong striations in NRVMs (Fig. 1A), and signals were detected in ROI (Fig. 1B). The FRAP experiments revealed differences after 1 h of 10% 1-Hz cyclic strain (Fig. 1B, and Table 1). The GFP-CapZ1 had faster dynamics in strained myocytes compared with the unstrained group (3.96 ⫾ 0.52 vs. 1.97 ⫾ 0.17, ⫻10 ⫺3·s⫺1, P ⬍ 0.05), meaning that a faster protein exchange was occurring in strained myocytes. Notably, strained myocytes treated with neomycin (a known PIP2 scavenger), had dynamics of GFP-CapZ1 that were significantly slower than strained myocytes (1.73 ⫾ 0.60 vs. 3.96 ⫾ 0.52, ⫻10 ⫺3·s⫺1, P ⬍ 0.05), but no significance was found in unstrained myocytes treated with neomycin alone (1.73 ⫾ 0.60 vs. 1.90 ⫾ 0.68, ⫻10⫺3·s⫺1) (Fig. 1B). Thus PIP2 inhibition abrogated strain-induced alterations in CapZ1 kinetics, suggesting the rate of CapZ1 exchange after mechanical stimulation was dependent on the PIP2 level and signaling. Increased CapZ1 dynamics induced by mechanical strain depend on the RhoA/ROCK pathway. The effect of C3 transferase (RhoA inhibitor) or Y27632 (ROCK inhibitor) on
Fig. 3. Increased dynamics of actin-GFP in NRVMs after 1 h of cyclic strain depends on both PIP2 and RhoA/ROCK pathway. A: microscopic images of a living NRVMs infected with actin-GFP. B: FRAP of the ROI for actin-GFP. C: Kfrap with all groups normalized to no-strain group showed actin-GFP in the strained group had increased kinetic rates. With the treatment of neomycin, C3 transferase or Y27632, the increased kinetic rate of actin-GFP in strained myocytes was significantly reduced. Means ⫾ SE; n ⬎ 3. *P ⬍ 0.05 vs. no-strain control group; #P ⬍ 0.05 vs. strained group. Bar ⫽ 10 m.
AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
H1618
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
actin-GFP seen with neomycin treatment of strained myocytes was markedly reduced compared with untreated, strained myocytes (5.20 ⫾ 0.49 vs. 9.66 ⫾ 0.58, ⫻10 ⫺4·s⫺1, P ⬍ 0.05). With the inhibition of RhoA or ROCK pathway (treated with C3 transferase or Y27632), the dynamics of actin-GFP were significantly slower than strained myocytes (6.93 ⫾ 0.84 or 3.76 ⫾ 0.98 vs. 9.66 ⫾ 0.58, ⫻10 ⫺4·s⫺1, P ⬍ 0.05). The FRAP experiments demonstrated that the dynamic exchange of actin-GFP depended on PIP2 and the RhoA/ROCK pathways after cyclic strain. Increased PIP2 production with mechanical strain depend on the RhoA/ROCK pathway. The PIP2 production was significantly increased in NRVMs after a 1-h period of cyclic strain (2.22 ⫾ 0.26 normalized to control of untreated, unstrained cells) (Fig. 4, A and B). The strain-induced increase in PIP2 (2.22 ⫾ 0.26) was markedly reduced by preventing RhoA/ ROCK signaling through pharmacological inhibition with C3 exoenzyme (1.33 ⫾ 0.03) or the ROCK-specific inhibitor Y-27632 (0.98 ⫾ 0.01). However, no change was observed in basal PIP2 expression in neomycin-treated, unstrained myocytes (Fig. 4, A and B). The RhoA inhibitor C3 exoenzyme inhibits not only RhoA but also RhoB and RhoC. To further evaluate the involvement of RhoA in strain-induced PIP2 production, NRVMs were tran-
siently transfected with adenovirus of HA-RhoA-DN. Myocytes expressing RhoA were identified by positive anti-HA-tag staining (data not shown). The control group of NRVM was transiently transfected with the GFP adenovirus. Most transfected myocytes show clear striations when stained for actin, and only a few nonstriated fibroblasts were transfected (data not shown). Thus the majority of transfected cells were myocytes in both unstrained and experimental groups. Expression of RhoA-DN significantly attenuated (1.09 ⫾ 0.18) the amount of PIP2 induced by mechanical strain (2.85 ⫾ 0.32, P ⬍ 0.05) (Fig. 4, C and D). Thus these data strongly suggested a role for RhoA in regulating strain-induced PIP2 production in cardiac myocytes. Sarcomeric distribution of PIP2 in cultured neonatal myocytes. The fractions were collected in the order of cytosolic, membranes/ organelles, and nuclei. For a biochemical analysis, subcellular fractions of cardiac myocytes were analyzed by Western blot analysis using proteins with known subcellular localization for verification (Fig. 5A). Detergent-based subcellular fractionation could be successfully used to isolate the main compartments of cultured myocytes while maintaining cell attachment and integrity. Redistribution of PIP2 was determined under the experimental conditions of cyclic strain. The cytoskeleton remained after
Fig. 4. Increased PIP2 production in NRVMs after 1 h of cyclic strain depends on the RhoA/ROCK pathway. Whole cell lysates were extracted from NRVMs in each experimental condition, spotted onto nitrocellulose membrane, probed with PIP2 antibody (1:500), GAPDH (1:1,000), and detected using a horseradish peroxidase-conjugated secondary antibody and enhanced chemiluminescence. A and B: after 1 h cyclic strain, PIP2 production dramatically increased in the strained compared with the no-strain group, whereas treatment with C3 transferase or Y27632 attenuated the PIP2 to the control level. GAPDH was used to normalize the dot blot density. Means ⫾ SE; n ⫽ 3. *P ⬍ 0.05 vs. no-strain control group; #P ⬍ 0.05 vs. strained group. C and D: NRVMs were transfected with adenovirus of RhoA-dominant negative (DN) or GFP as control. After 1 h cyclic strain, the increased PIP2 production in the strained group was significantly reduced by RhoA-DN. Means ⫾ SE; n ⫽ 3. *P ⬍ 0.05 vs. GFP-control; #P ⬍ 0.05 vs. GFP-strain. AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
H1619
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
Fig. 5. Subcellular distribution of PIP2 in NRVMs. Briefly, NRVMs (⬃1 ⫻ 106/well) were seeded in Flexcell plates. After 1 h cyclic strain, NRVMs were permeabilized with digitonin-EDTA to remove the cytosol. TritonEDTA was then used to remove the membrane-organelle fraction. Finally, Tween 20deoxycholate-benzonase removed the nuclei. A: Western blot analysis of controls for subcellular fractionation: cytostol fraction (a), membrane/organelle fraction (b), and nuclear fraction (c). Heat shock protein 70 (Hsp70) was only found in the cytosolic fractions, 1integrin was only found in the membrane fraction, and histone 2B was only found in the nuclear fraction. B: in both unstrained and strained groups, the remaining cytoskeletal fraction was stained for PIP2 (green) and ␣-actinin (red). After 1 h of cyclic strain, more PIP2 colocalized with ␣-actinin in the Z disk. Bar ⫽ 10 m.
removal of membranes, cytosol, and nuclei and was observed by microscopy to determine the subcellular location of PIP2. Indeed, PIP2 was distributed in a striated pattern colocalized with ␣-actinin at the Z disc (Fig. 5B). CapZ1 binding to PIP2. Strong interaction between PIP2 and CapZ1 was found using PIP2 beads, (Fig. 6A), whereas GFP alone did not interact with PIP2 beads. Significantly more GFP-CapZ1 bound to PIP2 in strained (1.47 ⫾ 0.17) than in unstrained cardiac myocytes (P ⬍ 0.05) (Fig. 6B). Thus CapZ1 was a PIP2-interacting protein, which depended on mechanical strain. To test the effect of the -tentacle on PIP2 binding, GFPCapZ1-⌬C was transfected in NRVMs and proteins were affinity precipitated with PIP2-conjugated beads. Over threefold more GFP-CapZ1-⌬C bound to PIP2, compared with GFP-CapZ1 (Fig. 7, A and B). Attenuation of myocyte hypertrophy induced by mechanical strain with inhibition of RhoA or PIP2. To determine if the effects of PIP2 on cardiac hypertrophy could be attenuated, NRVMs were pretreated with HA-RhoA-DN, GFP, or neomycin and subjected to cyclic strain for an additional 48 h. HA-RhoA-DN or neomycin alone had no effects on myocardial size or phenotype (data not shown), indicating that the effect of HA-RhoA-DN or neomycin was not secondary to a toxic cellular effect. Cyclic strain (48 h) induced approximately a 38% increase in myocyte size, which was inhibited by
neomycin pretreatment (Fig. 8). Similarly, the marked increase in myocyte size induced by cyclic strain was also significantly inhibited by HA-RhoA-DN (Fig. 8). These results suggested that HA-RhoA-DN or neomycin attenuated the cardiac myocyte hypertrophy induced by cyclic strain. DISCUSSION
The present study demonstrates that 1 h of cyclic strain increases the dynamics of CapZ1 and actin in NRVMs and that these changes depend on the PIP2 pathway. Furthermore, PIP2 production increases after 1 h of cyclic strain and is dependent on the RhoA/ROCK pathway. PIP2 redistributes to the sarcomeric cytoskeleton, and more PIP2 is bound to CapZ1 after cyclic strain. With deletion of COOH-terminal of CapZ1, more CapZ1-⌬C binds to PIP2, which shows that the tentacle is not the major binding site of PIP2. However, its removal appears to release an inhibition enabling higher affinity PIP2 binding to the remainder of the CapZ molecule, suggesting the interaction of PIP2 and CapZ1 affects the COOH-terminal binding affinity to CapZ, thereby affecting actin assembly. Cardiac hypertrophy caused by 48 h cyclic strain is blunted by inhibition of PIP2 pathways. A potential scheme for how PIP2 regulates actin assembly under mechanical strain is shown diagrammatically in Fig. 9. After a 1-h period of cyclic strain, focal adhesion kinase (FAK) activates
AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
H1620
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
PIP2 is a phospholipid component mainly located in the surface or internal membrane bilayers. However, PIP2 is also bound to hydrophobic protein pockets, such as CapZ. Both crystallography and total internal reflection fluorescence microscopy predicted four lipid binding domains, two in CapZ␣ and two in CapZ (20, 37, 42). A cluster of conserved basic residues on the surface of CapZ (215–232 region) includes PIP2 binding at Arg-225, which could bind to hydrophobic protein pockets (20). Moreover, another Z-disc protein with hydrophobic regions able to bind PIP2 is ␣-actinin, since mutations in its PIP2 binding regions caused abnormal actin filament formation throughout the cell (12). This study demonstrates that more PIP2 is in the Z disc after cyclic strain (Fig. 5), which further supports that PIP2 distributes not only to membranes but also to the sarcomeric cytoskeleton. These studies were done to gain insight into how PIP2 binds CapZ1. The computational analysis indicates that the COOHterminal half of CapZ -subunit could contribute to lipid interaction/insertion (37). However, deletion of the -tentacle (CapZ1-⌬C) resulted in more not less PIP2 binding (Fig. 7), which suggests the COOH-terminal of CapZ1 is not the major regulatory binding site of PIP2, and, therefore not the region by
Fig. 6. PIP2 binding to CapZ1 increases after 1 h of cyclic strain. A: PIP2conjugated beads were used to affinity precipitate GFP-CapZ1. The bound CapZ1 was detected with anti-GFP antibody by Western blot analysis. Total CapZ1, as well as GFP, in whole lysate were detected with anti-GFP antibody in no-strain and strained NRVMs. After 1 h cyclic strain (10%, 1 Hz), more CapZ1 bound to PIP2. IP, immunoprecipitation. B: Western blot analysis normalized to the no-strain NRVMs. Means ⫾ SE; n ⫽ 3. *P ⬍ 0.05.
the RhoA/ROCK pathway, which activates phosphatidylinositol 4-phosphate 5-kinase (PIP5K) to produce more PIP2. Some of the extra PIP2 binds to CapZ increasing the CapZ dynamics, and resulting in a more rapid actin exchange. This suggests that after many hours of cyclic strain, a possible mechanism for cell hypertrophy is the accumulation of thin filament assembly triggered partially by the increased PIP2 level and its binding to CapZ. The mechanical signals are transmitted from the extracellular matrix via integrins to the cytoskeleton. PIP2 has been implicated in the regulation of focal adhesions during changes in cell attachment to the extracellular matrix. PIP2 activates several key focal adhesion components including vinculin and talin (7, 13, 32). These signaling pathways are activated by forces delivered from the outside of the cell or to those generated internally with both directions leading to actin filament assembly at the focal adhesions (18). PIP2 is correctly assumed to be mainly located in the membrane fraction of cells, since its structure is a phosphinositol head and two fatty acid chains making it highly hydrophobic and lipophilic. Regulation of actin filament assembly elsewhere in the cell may also involve PIP2 signaling, even though
Fig. 7. PIP2 binding increases with the CapZ terminal deletion (CapZ1-⌬C). A: PIP2-conjugated beads were used to affinity precipitate GFP-CapZ1 or GFP-CapZ1-⌬C. The bound CapZ1 or CapZ1-⌬C was detected with anti-GFP antibody by Western blot analysis. Total CapZ1, CapZ1-⌬C, or GFP in whole lysate were detected with anti-GFP antibody in NRVMs. More CapZ1-⌬C bound to PIP2, compared with CapZ1. B: Western blot analysis normalized to the CapZ1 group. Means ⫾ SE; n ⫽ 3. *P ⬍ 0.05.
AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
Fig. 8. Cardiac myocyte area decreases following inhibition of RhoA or PIP2 pathways. Myocyte area normalized to no-strain group (n ⬎ 20 myocytes each). NRVM strained for 48 h at 10% and 1 Hz had increased area, which was attenuated by RhoA-DN or neomycin treatment. HA, hemagglutinin. Means ⫾ SE. *P ⬍ 0.05 vs. GFP-control; #P ⬍ 0.05 vs. GFP-strain.
which PIP2 regulates CapZ and actin kinetics. However, the removal of the tentacle enhances PIP2 binding to the capping portion of CapZ1. The COOH-terminus inhibits the interaction of PIP2 and CapZ1. Thus the interaction of PIP2 and
H1621
CapZ1 affects the actin binding affinity, which may be partially responsible for changes in the dynamics of CapZ and actin measured by FRAP. The mechanism of structure of CapZ1 binding with PIP2 needs to be further studied. PIP2 has a direct role in regulating actin assembly by interaction with partnering proteins in many cells. In platelets, the half-life for a capped filament is 28 min, whereas the half-life to remain uncapped is only 0.2 s. Interestingly, the addition of PIP2 in these platelets reduces the half-life of the capped filament to 46 s, suggesting PIP2 regulation of actin filament capping dynamics (34). Gelsolins (calcium-dependent actin-binding proteins) exposed to high lipid concentrations in vitro can bind PIP2 molecules (40). Also in vitro, PIP2 affects actin dynamics by direct association with the actin-binding protein villin, resulting in unstable F-actin (38). PIP2 also binds to ␣-actinin and inhibits ␣-actinin bundling activity (10, 12). More PIP2 colocalizes with ␣-actinin at Z disc after mechanical strain (Fig. 5). Thus PIP2 is critical to actin filament regulation by interacting with actin binding proteins in many situations. Cyclic strain of cardiomyocytes evokes many other intracellular signals that work in parallel to PIP signaling that lead to cardiomyocyte hypertrophy (1). One key component of the integrin complex is FAK, which is also part of the mechanotransduction signaling pathway (30). In mechanically strained cardiac myocytes, FAK phosphorylation at Y397 rapidly activates RhoA (39). PIP2 is generated by PIP5K which phosphorylates phosphatidylinositol 4-phosphate at the D-5 position of
Fig. 9. Diagram of proposed model for the mechanism of PIP2 signaling in cyclic mechanical strain that leads to cardiac myocyte hypertrophy. Left: in unstrained spontaneously beating NRVMs, the production of PIP2 is at a basal level and the RhoA/ROCK pathway is minimally activated. Right: cyclic mechanical strain of NRVM activates the RhoA/ROCK pathway via focal adhesion kinase (FAK). RhoA/ROCK activates phosphatidylinositol 4-phosphate 5-kinase (PIP5K), which phosphorylates phosphatidylinositol 4-phosphate [PI(4)P] to produce PIP2 and increases the PIP2 level. Some of the increased PIP2 colocalizes with ␣-actinin in the Z disc. More PIP2 binds to CapZ, resulting in an increase in the kinetics of CapZ and also increase in the kinetics of actin. This suggests that after many hours of cyclic strain a possible mechanism for cell hypertrophy is the accumulation of thin filament assembly triggered partially by the increased PIP2 level and its binding to CapZ. AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
H1622
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
the inositol ring. PIP5K is also directly or indirectly activated by RhoA and its effector ROCK (26, 44). Cellular and molecular biology studies have indicated a pivotal role of RhoA/ ROCK signaling in cardiac hypertrophy. RhoA and ROCK are rapidly activated by mechanical strain in cardiac myocytes (39). The RhoA/ROCK pathway is important for setting the PIP2 level (Fig. 4). However, the direct mechanism of RhoA on regulating PIP2 may be very complex and need to be further studied. To date, at least 15 RhoA effector proteins have been identified. Some of these are putative protein kinases and lipid kinases, such as 1,2-diacylglycerol kinase, phosphatidylinositol 3-kinase, and PIP5Ks (22). Nonetheless, after mechanical strain, activated RhoA/ROCK produces PIP2, which interacts with CapZ and subsequently alters cardiomyocyte cell size by mechanisms that are blocked by Rho/ROCK inhibition. A complication in understanding the response to strain is that it is known to increase the intracellular Ca2⫹ concentration through strain-activated membrane channel opening (31). Ca2⫹ is an essential cofactor for both MAPK and protein kinase C signaling pathways that induce cardiac hypertrophy (24). However, CapZ is insensitive to Ca2⫹ (25); therefore, changes in dynamics cannot be directly due to altered calcium. It is clear that PIP2 plays a major role in hypertrophic responses, since in our present study, removal of PIP2 blocked the increased in cell size normally found by two days of cyclic strain. A brief, 1-h period of cyclic strain is sufficient to initiate PIP2 binding to CapZ, modulate actin dynamics, and thereby regulate actin cytoskeleton assembly. While the Z disc is not a lipid membrane, it does have the capacity to bind PIP2 since there is evidence in the present study that PIP2 locates there. The presence of its extensive transverse architecture could permit remodeling of the thin filaments throughout the entire cross section of the muscle cell and not only at the peripherally located-myocyte membranes at the costameric focal adhesions or the intercalated discs. ACKNOWLEDGMENTS We thank Dr. Allen M. Samarel (Loyola University Chicago Stritch School of Medicine, Maywood, IL) for the gift of the CapZ DNA constructs and Dr. Viswanathan Natarajan (Department of Pharmacology College of Medicine, University of Illinois at Chicago) for the gift of Rho A-DN constructs. We thank Erik Swanson and Dr. Ke Ma for expert assistance with the FRAP experiments. GRANTS This work was supported by National Heart, Lung, and Blood Institute Grant HL-62426. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS J.L. conception and design of research; J.L. performed experiments; J.L. analyzed data; J.L. and B.R. interpreted results of experiments; J.L. prepared figures; J.L. drafted manuscript; B.R. edited and revised manuscript; B.R. approved final version of manuscript. REFERENCES 1. Aikawa R, Nagai T, Kudoh S, Zou Y, Tanaka M, Tamura M, Akazawa H, Takano H, Nagai R, Komuro I. Integrins play a critical role in mechanical stress-induced p38 MAPK activation. Hypertension 39: 233–238, 2002.
2. Anderson KE, Dart AM, Woodcock EA. Inositol phosphate release and metabolism during myocardial ischemia and reperfusion in rat heart. Circ Res 76: 261–268, 1995. 3. Balasubramanian S, Mani SK, Kasiganesan H, Baicu CC, Kuppuswamy D. Hypertrophic stimulation increases beta-actin dynamics in adult feline cardiomyocytes. PLoS One 5: e11470, 2010. 4. Birukova AA, Birukov KG, Adyshev D, Usatyuk P, Natarajan V, Garcia JG, Verin AD. Involvement of microtubules and Rho pathway in TGF-beta1-induced lung vascular barrier dysfunction. J Cell Physiol 204: 934 –947, 2005. 5. Boateng SY, Belin RJ, Geenen DL, Margulies KB, Martin JL, Hoshijima M, de Tombe PP, Russell B. Cardiac dysfunction and heart failure are associated with abnormalities in the subcellular distribution and amounts of oligomeric muscle LIM protein. Am J Physiol Heart Circ Physiol 292: H259 –H269, 2007. 6. Boateng SY, Hartman TJ, Ahluwalia N, Vidula H, Desai TA, Russell B. Inhibition of fibroblast proliferation in cardiac myocyte cultures by surface microtopography. Am J Physiol Cell Physiol 285: C171–C182, 2003. 7. Chandrasekar I, Stradal TE, Holt MR, Entschladen F, Jockusch BM, Ziegler WH. Vinculin acts as a sensor in lipid regulation of adhesion-site turnover. J Cell Sci 118: 1461–1472, 2005. 8. Chien KR. Stress pathways and heart failure. Cell 98: 555–558, 1999. 9. Cooper JA, Sept D. New insights into mechanism and regulation of actin capping protein. Int Rev Cell Mol Biol 267: 183–206, 2008. 10. Corgan AM, Singleton C, Santoso CB, Greenwood JA. Phosphoinositides differentially regulate alpha-actinin flexibility and function. Biochem J 378: 1067–1072, 2004. 11. Eble DM, Strait JB, Govindarajan G, Lou J, Byron KL, Samarel AM. Endothelin-induced cardiac myocyte hypertrophy: role for focal adhesion kinase. Am J Physiol Heart Circ Physiol 278: H1695–H1707, 2000. 12. Fraley TS, Pereira CB, Tran TC, Singleton C, Greenwood JA. Phosphoinositide binding regulates alpha-actinin dynamics: mechanism for modulating cytoskeletal remodeling. J Biol Chem 280: 15479 –15482, 2005. 13. Gilmore AP, Burridge K. Regulation of vinculin binding to talin and actin by phosphatidyl-inositol-4 –5-bisphosphate. Nature 381: 531–535, 1996. 14. Hart MC, Cooper JA. Vertebrate isoforms of actin capping protein beta have distinct functions In vivo. J Cell Biol 147: 1287–1298, 1999. 15. Hartman TJ, Martin JL, Solaro RJ, Samarel AM, Russell B. CapZ dynamics are altered by endothelin-1 and phenylephrine via PIP2- and PKC-dependent mechanisms. Am J Physiol Cell Physiol 296: C1034 – C1039, 2009. 16. Huang S, Gao L, Blanchoin L, Staiger CJ. Heterodimeric capping protein from Arabidopsis is regulated by phosphatidic acid. Mol Biol Cell 17: 1946 –1958, 2006. 17. Hug C, Miller TM, Torres MA, Casella JF, Cooper JA. Identification and characterization of an actin-binding site of CapZ. J Cell Biol 116: 923–931, 1992. 18. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell 110: 673–687, 2002. 19. Katz AM. The cardiomyopathy of overload: an unnatural growth response in the hypertrophied heart. Ann Intern Med 121: 363–371, 1994. 20. Kim K, McCully ME, Bhattacharya N, Butler B, Sept D, Cooper JA. Structure/function analysis of the interaction of phosphatidylinositol 4,5bisphosphate with actin-capping protein: implications for how capping protein binds the actin filament. J Biol Chem 282: 5871–5879, 2007. 21. Lin YH, Li J, Swanson ER, Russell B. CapZ and actin capping dynamics increase in myocytes after a bout of exercise and abates in hours after stimulation ends. J Appl Physiol 114: 1603–1609, 2013. 22. Lui WY, Lee WM, Cheng CY. Rho GTPases and spermatogenesis. Biochim Biophys Acta 1593: 121–129, 2003. 23. Mao YS, Yin HL. Regulation of the actin cytoskeleton by phosphatidylinositol 4-phosphate 5 kinases. Pflügers Arch 455: 5–18, 2007. 24. Molkentin JD, Lu JR, Antos CL, Markham B, Richardson J, Robbins J, Grant SR, Olson EN. A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell 93: 215–228, 1998. 25. Nachmias VT, Golla R, Casella JF, Barron-Casella E. Cap Z, a calcium insensitive capping protein in resting and activated platelets. FEBS Lett 378: 258 –262, 1996. 26. Oude Weernink PA, Schulte P, Guo Y, Wetzel J, Amano M, Kaibuchi K, Haverland S, Voss M, Schmidt M, Mayr GW, Jacobs KH. Stimu-
AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
STRAIN REGULATES CapZ AND ACTIN DYNAMICS VIA PIP2
27. 28. 29. 30. 31.
32.
33. 34. 35. 36.
lation of phosphatidylinositol-4-phosphate 5-kinase by Rho-kinase. J Biol Chem 275: 10168 –10174, 2000. Pyle WG, Hart MC, Cooper JA, Sumandea MP, de Tombe PP, Solaro RJ. Actin capping protein: an essential element in protein kinase signaling to the myofilaments. Circ Res 90: 1299 –1306, 2002. Pyle WG, La Rotta G, de Tombe PP, Sumandea MP, Solaro RJ. Control of cardiac myofilament activation and PKC-betaII signaling through the actin capping protein, CapZ. J Mol Cell Cardiol 41: 537–543, 2006. Roy P, Rajfur Z, Pomorski P, Jacobson K. Microscope-based techniques to study cell adhesion and migration. Nat Cell Biol 4: E91–E96, 2002. Russell B, Curtis MW, KoshmanYE, Samarel AM. Mechanical stressinduced sarcomere assembly for cardiac muscle growth in length and width. J Mol Cell Cardiol 48: 817–823, 2010. Ruwhof C, van Wamel JT, Noordzij LA, Aydin S, Harper JC, van der Laarse A. Mechanical stress stimulates phospholipase C activity and intracellular calcium ion levels in neonatal rat cardiomyocytes. Cell Calcium 29: 73–83, 2001. Saltel F, Mortier E, Hytonen VP, Jacquier MC, Zimmermann P, Vogel V, Liu W, Wehrle-Haller B. New PI(4,5)P2- and membrane proximal integrin-binding motifs in the talin head control beta3-integrin clustering. J Cell Biol 187: 715–731, 2009. Schafer DA, Hug C, Cooper JA. Inhibition of CapZ during myofibrillogenesis alters assembly of actin filaments. J Cell Biol 128: 61–70, 1995. Schafer DA, Jennings PB, Cooper JA. Dynamics of capping protein and actin assembly in vitro: uncapping barbed ends by polyphosphoinositides. J Cell Biol 135: 169 –179, 1996. Schafer DA, Korshunova YO, Schroer TA, Cooper JA. Differential localization and sequence analysis of capping protein beta-subunit isoforms of vertebrates. J Cell Biol 127: 453–465, 1994. Senyo SE, Koshman YE, Russell B. Stimulus interval, rate and direction differentially regulate phosphorylation for mechanotransduction in neonatal cardiac myocytes. FEBS Lett 581: 4241–4247, 2007.
H1623
37. Smith J, Diez G, Klemm AH, Schewkunow V, Goldmann WH. CapZlipid membrane interactions: a computer analysis. Theor Biol Med Model 3: 30, 2006. 38. Tomar A, George SP, Mathew S, Khurana S. Differential effects of lysophosphatidic acid and phosphatidylinositol 4,5-bisphosphate on actin dynamics by direct association with the actin-binding protein villin. J Biol Chem 284: 35278 –35282, 2009. 39. Torsoni AS, Marin TM, Velloso LA, Franchini KG. RhoA/ROCK signaling is critical to FAK activation by cyclic stretch in cardiac myocytes. Am J Physiol Heart Circ Physiol 289: H1488 –H1496, 2005. 40. Tuominen EK, Holopainen JM, Chen J, Prestwich GD, Bachiller PR, Kinnunen PK, Janmey PA. Fluorescent phosphoinositide derivatives reveal specific binding of gelsolin and other actin regulatory proteins to mixed lipid bilayers. Eur J Biochem 263: 85–92, 1999. 41. Visser MB, Koh A, Glogauer M, Ellen RP. Treponema denticola major outer sheath protein induces actin assembly at free barbed ends by a PIP2-dependent uncapping mechanism in fibroblasts. PLoS One 6: e23736, 2011. 42. Wear MA, Cooper JA. Capping protein: new insights into mechanism and regulation. Trends Biochem Sci 29: 418 –428, 2004. 43. Wear MA, Yamashita A, Kim K, Maeda Y, Cooper JA. How capping protein binds the barbed end of the actin filament. Curr Biol 13: 1531– 1537, 2003. 44. Weernink PA, Meletiadis K, Hommeltenberg S, Hinz M, Ishihara H, Schmidt M, Jakobs KH. Activation of type I phosphatidylinositol 4-phosphate 5-kinase isoforms by the Rho GTPases, RhoA, Rac1, and Cdc42. J Biol Chem 279: 7840 –7849, 2004. 45. Yamashita A, Maeda K, Maeda Y. Crystal structure of CapZ: structural basis for actin filament barbed end capping. EMBO J 22: 1529 –1538, 2003. 46. Yin HL, Janmey PA. Phosphoinositide regulation of the actin cytoskeleton. Annu Rev Physiol 65: 761–789, 2003.
AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00477.2013 • www.ajpheart.org
