NO inhibits stretch-induced MAPK activity by cytoskeletal disruption.

Mesangial cells (MC) grown on extracellular matrix protein-coated plates and exposed to cyclic strain/relaxation proliferate and produce extracellular matrix protein, providing an in vitro model of signaling in stretched MC. Intracellular transduction of mechanical strain involves mitogen-activated protein kinases, and we have shown that p42/44 mitogen-activated protein kinase (extracellular signal-regulated kinase (ERK)) is activated by cyclic strain in MC. In vivo studies show that increased production of nitric oxide (NO) in the remnant kidney limits glomerular injury without reducing glomerular capillary pressure, and we have observed that NO attenuates stretch-induced ERK activity in MC via generation of cyclic guanosine monophosphate (cGMP). Accordingly, we sought to determine whether NO affects strain-induced ERK activity after strain and how this is mediated. Strain-induced ERK activity was dependent on time and magnitude of stretch and was maximal after 10 min at -27 kilopascals. Actin cytoskeleton disruption with cytochalasin D abrogated this. The non-metabolizable cGMP analogue 8-bromo cyclic GMP (8-Br-cGMP) dose-dependently attenuated strain-induced ERK activity. Cytoskeletal stabilization with jasplakinolide prevented this inhibitory effect of 8-Br-cGMP. Cyclic strain increased nuclear translocation of phospho-ERK by immunofluorescent microscopy, again attenuated by 8-Br-cGMP. Jasplakinolide prevented the inhibitory effect of 8-Br-cGMP on activated ERK nuclear translocation after strain. Strain increased ERK-dependent AP-1 nuclear protein binding, which was attenuated by cytochalasin D and 8-Br-cGMP. These data indicate that cGMP can inhibit cyclic strain-induced ERK activity, nuclear translocation, and AP-1 nuclear protein binding. Cytoskeletal disruption leads to the same effect, whereas cytoskeleton stabilization reverses the effect of 8-Br-cGMP. Thus, NO inhibits strain-induced ERK activity by cytoskeletal destabilization.

Glomerular mesangial cells (MC) 1 are positioned as architectural supports for capillary loops and are therefore exposed to pulsatile stretch/relaxation (1). Whereas little resident glomerular cell proliferation or sclerosis is demonstrable in normal animals (2), MC proliferation and matrix production, eventually resulting in sclerosis, can be induced by maneuvers that increase intraglomerular pressure by 10 mmHg (3)(4)(5). Moreover, in these models, preventing the intraglomerular pressure rise attenuates sclerosis (5-7). We and others have shown reduction of sclerosis and MC proliferation in remnant glomeruli by oral L-arginine supplementation to increase NO production (8,9). L-Arginine increases nitric oxide production by the remnant kidney and reduces glomerular endothelin-1 expression but does not lower glomerular capillary pressure (8).
The effects of mechanical forces on MC in vitro can be modelled by culturing cells in wells with deformable bottoms and then applying a vacuum to the wells to generate alternating cycles of strain and relaxation. Initial experiments using this methodology showed induction of mRNA for the proto-oncogene and AP-1 transcription factor component c-fos at 30 min (1). Subsequently, increases in both MC proliferation (2) and collagenous and non-collagenous extracellular matrix protein synthesis were observed by 48 h, the sine qua non of sclerotic injury (10,11).
We and others have studied the link between mechanical stress and c-fos induction in stressed MC (12)(13)(14). We demonstrated increases in all three canonical mitogen-activated protein kinase (MAPK) pathways in response to strain (14). The most well described mammalian MAPK cascade, p42/44 or ERK, is well recognized to lie upstream of AP-1 (15,16). ERKdependent AP-1 induction has been demonstrated in endothelial cells in response to shear (17). In vivo, cyclic strain in the aortic wall activates ERK and AP-1 (18), and glomerular ERK activation and AP-1 nuclear protein binding have been shown in response to angiotensin II infusion (19). AP-1 activation may be important in the pathogenesis of glomerular sclerosis, because AP-1 activation has been shown to mediate transforming growth factor beta-1 induction (20,21).
Recent data indicates that the actin cytoskeleton is important for ERK signaling. Cytochalasin-D prevented strain-induced ERK activity in vascular smooth muscle cells (22). Interestingly, LIM kinase-1 is intimately involved in induction of serum response factor by serum, suggesting a central role for the actin cytoskeleton in intracellular signaling (23). ERK does interact with the actin cytoskeleton (24), and ERK signaling in response to lysophosphatidic acid (25) or epidermal growth factor (26) is dependent on the presence of an intact actin cytoskeleton.
Recent studies indicate a role for NO in inhibition of cytoskeletal organization. Several NO donors and a constitutively active form of cyclic guanosine monophosphate (cGMP) inhibited MC adhesion to extracellular matrix protein via inhibition of focal adhesion kinase phosphorylation and actin cytoskeletal disruption (27). In vascular smooth muscle cells, cGMP-dependent kinase phosphorylated and inactivated RhoA, thereby disrupting the actin cytoskeleton (28). Consequently, we hypothesized that nitric oxide, via cyclic GMP, would limit MC ERK signaling in response to mechanical strain through cytoskeletal disruption.

Application of Strain/Relaxation
Mesangial cells (5 ϫ 10 4 /well) were plated on to 6-well plates with flexible bottoms coated with bovine type I collagen (Flexcell International Corp., McKeesport, PA). Cells were grown to confluence for 72 h and then rendered quiescent by incubation for 24 h in Dulbecco's modified Eagle's medium with 0.5% fetal calf serum. To characterize the time of maximum response, cells were initially exposed to cycles of strain/relaxation for periods of 2, 5, 10, 30, and 60 min generated by a cyclic vacuum produced by a computer-driven system (Flexercell Strain Unit 2000, Flexcell International Corp.). Plates were exposed to continuous cycles of strain/relaxation, with each cycle being 0.5 s of strain and 0.5 s of relaxation, for a total of 60 cycles per min. Initially, the vacuum pressures used were Ϫ10 to Ϫ27 kPa, inducing a 16 -28% elongation in the diameter of the surface. Subsequent experiments were performed at the time and strain level of maximal response, 10 min and Ϫ27 kPa, respectively (average 28% elongation in diameter of the plates).
For the experiments studying NO effects, 8-bromo cyclic guanosine monophosphate (8-bromo-cGMP, Sigma) was added in the indicated concentrations 10 min prior to the initiation of stretch protocols. For experiments studying cytoskeletal manipulation, cytochalasin-D (Sigma) or jasplakinolide (Molecular Probes, Eugene, OR) was added prior to the initiation of stretch protocols at the indicated concentrations.

p42/44 MAPK (ERK) Activity
Protein Isolation and Western Blotting-Initially, the time course and concentration dependence of ERK activity in response to stretch were studied, and subsequent experiments were performed at Ϫ27 kPa at 10 min. Cultures were serum starved overnight prior to stretch protocols. After stretch protocols with or without inhibitors, medium was removed, and the cells were washed once with ice-cold PBS. Protein was isolated as described above, and 40 g of sample was separated on a 12% SDS-polyacrylamide gel electrophoresis gel. After electroblotting to a nitrocellulose membrane, membranes were incubated for 3 h at room temperature with 25 ml of blocking buffer (1ϫ TBS, 0.1% Tween 20 with 5% w/v nonfat dry milk) and then overnight at 4°C with ERK1/2 MAPK (Thr 202 /Tyr 204 ) polyclonal antibody (1:1000)(New England Biolabs, Beverly, MA) in 10 ml of antibody dilution buffer (1ϫ TBS, 0.05% Tween 20 with 5% bovine serum albumin) with gentle rocking at 4°C. Membranes were then washed three times with 1ϫ TBS, 0.05% Tween 20 and then incubated with horseradish peroxidase-conjugated anti-rabbit secondary antibody (1:2000) in 10 ml of blocking buffer for 45 min at room temperature. After three further washes in TBS, the membrane was incubated with LumiGlo reagent and then exposed to x-ray film.
Activity Assays-200 g of total cellular protein was incubated with immobilized phospho-p44/42 MAPK (Thr 202 /Tyr 204 ) monoclonal antibody (15 l, New England Biolabs) with gentle rocking overnight at 4°C. Lysates were microcentrifuged for 30 s at 14,000 rpm to recover the beads, and the pellet was washed twice with 0.5 ml of 1ϫ lysis buffer. For the kinase assay, after immunoprecipitation, pellets were washed twice with 0.5 ml of kinase buffer (25 mM Tris, 5 mM ␤-glycerophosphate, 2 mM dithiothreitol, 0.1 mM sodium orthovanadate, 10 mM MgCl 2 ). ERK activity was then measured by suspending the pellet in 50 l of 1ϫ kinase buffer containing 200 M ATP and 2 g of Elk1 fusion protein as substrate. After incubation for 30 min at 30°C, the reaction was terminated with 25 M 3ϫ SDS sample buffer (187.5 mM Tris-HCl (pH 6.8), 6% w/v SDS, 30% glycerol, 150 mM dithiothreitol, 0.3% w/v bromphenol blue), boiled for 5 min, vortexed, and then microcentrifuged for 2 min. 20 l of sample was then run on a 10% non-reducing SDSpolyacrylamide gel electrophoresis gel. After blotting to nitrocellulose, membranes were incubated for 3 h at room temperature with 25 ml of blocking buffer (1ϫ TBS, 0.1% Tween 20 with 5% w/v nonfat dry milk) and then overnight at 4°C with phospho-specific anti-Elk1 (Ser 383 ) antibody 1:1000 in 10 ml of antibody dilution buffer. Membranes were washed three times with 1ϫ TBS, 0.05% Tween 20 and then incubated with horseradish peroxidase-conjugated anti-rabbit secondary antibody (1:2000) for 1 h at room temperature. After three further washes in TBS, the membrane was processed as above.
Fluorescence Microscopy ERK Nuclear Translocation-After each strain protocol with or without inhibitors, cells were washed three times with PBS and fixed with 3.7% formaldehyde (300 l/well) for 10 min at room temperature. Cells were washed three times with PBS, then permeabilized in 100% methanol for 5 min at Ϫ20°C, washed again with PBS, and incubated with a 1:50 dilution of anti-phospho-ERK (Thr 180 /Tyr 182 ) (New England Biolabs) in PBS for 30 min at room temperature. Cells were washed three times in PBS and incubated with a 1:50 dilution of an Alexa 488 goat anti-rabbit IgG (H ϩ L) conjugate (Molecular Probes) in PBS for 30 min at room temperature in the dark. Cells were washed and then mounted by removing the flexible base from each well and placing it directly on a glass slide using one drop of anti-fade mount medium (Slow Fade, Molecular Probes). A drop of mount medium was then placed on top of the cells and covered with a glass slip. Slides were stored at 4°C in the dark until confocal laser scanning microscopy was performed using a Bio-Rad MRC-600 confocal microscope (Bio-Rad) within 10 days.
F-actin Staining-MC were fixed with formaldehyde exactly as described above and then permeabilized by dipping in acetone for 5 min at Ϫ20°C. Subsequently, Texas Red phalloidin solution (Molecular Probes) was applied for 20 min at room temperature. Cells were then washed, mounted, and analyzed exactly as described above.

Nuclear Protein Binding to AP-1 Consensus Sequences
After each strain protocol, MC were washed in cold PBS, and nuclear extracts were prepared by lysis in hypotonic buffer (20 mM Hepes, pH 7.9, 1 mM EDTA, 1 mM EGTA, 20 mM NaF, 1 mM Na 3 VO 4 , 1 mM Na 4 P 2 O 7 , 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 1 g/ml leupeptin, 1 g/ml pepstatin A, 0.6% Nonidet P-40), homogenized, and sedimented at 16,000 ϫ g for 20 min at 4°C. Pelleted nuclei were resuspended in hypotonic buffer containing 0.42 M NaCl 2 , 20% glycerol and rotated for 30 min at 4°C. After centrifugation for 20 min at 16,000 ϫ g, the supernatant containing nuclear proteins was collected, and the protein concentration was measured with the Bio-Rad assay kit.
Nuclear proteins (3 g) were incubated with 2 g of poly(dI-dC)⅐poly(dI⅐dC) (Amersham Pharmacia Biotech) in binding buffer (20 mM HEPES pH 7.9, 1.8 mM MgCl 2 , 2 mM dithiothreitol, 0.5 mM EDTA, 0.5 mg/ml bovine serum albumin) for 30 min at room temperature and then reacted with radiolabeled consensus oligonucleotides at room temperature for 20 min (50,000 -100,000 cpm). Reaction mixtures were electrophoresed in a 6% polyacrylamide gel and autoradiographed. Competition experiments were performed with 100ϫ excess unlabeled AP-1 consensus oligonucleotides.

Actin Cytoskeletal Disruption Eliminates Strain-induced
ERK Activity-Initial experiments showed time-and strain intensity-dependent ERK activity, as we have previously published (14). Maximal activity was observed at 10 min and Ϫ27 kPa pressure, with decay at 30 min (n ϭ 4, data not shown). To study the effects of cytoskeletal disruption on ERK activity induction, we first sought to determine the dose of cytochala-sin-D that would provide maximal actin cytoskeleton disassembly in MC. Application of a Ϫ27-kPa stretch to MC for 10 min resulted in marked induction of actin stress fibers by phalloidin staining (Fig. 1, compare A and C). Pre-incubation with cytochalasin D for 60 min dose-dependently prevented stress fiber organization, maximally at 1 mg/ml (Fig. 1, compare B and D). Pre-incubation with 1 mg/ml cytochalasin D for 60 min prevented the stretch-induced ERK activity (Fig. 2), indicating actin cytoskeleton-dependent transmission of the stretch signal.
8-Bromo-cGMP Inhibits Strain-induced Actin Cytoskeletal Organization and ERK Activity-To determine whether nitric oxide, via induction of cyclic GMP, could prevent actin cytoskeleton organization and ERK activity in response to strain, MC cultures were incubated with the stable active cGMP analogue 8-bromo-cGMP for 10 min prior to strain. Initially, we sought to determine the effects of NO donors on the cytoskeleton in unstretched MC. Incubation for 10 min with either 70 M S-nitroso-N-acetylpenicillamine (SNAP) or 1 mM 8-bromo-cGMP resulted in marked disruption of actin fibers in unstressed MC (Fig. 3, compare A, B, and C). Again, application of a Ϫ27-kPa stretch for 10 min to MC resulted in marked organization of actin stress fibers by phalloidin staining (Fig. 3,  compare A and D), effectively prevented by 10 min of preincubation with either 70 M SNAP or 1 mM 8-bromo-cGMP (Fig. 3, compare D, E, and F). In addition, pre-incubation with 8-bromo-cGMP for 10 min prior to strain dose-dependently inhibited the stretch-induced ERK activity (data not shown), with complete inhibition at 1 mM (Fig. 4, lane 3).
Cytoskeletal Stabilization Restores ERK Activity in Response to Stretch-The preceding experiments show an association between disruption of the actin cytoskeleton and prevention of stretch-induced ERK activity. To cement this link, we sought to prevent 8-bromo-cGMP-induced actin cytoskeleton disruption with jasplakinolide. Jasplakinolide stabilizes actin stress fibers primarily by decreasing the dissociation rate of actin subunits (29), although at higher concentrations (200 nM) and longer times (24 h), stress fibers disappear and are replaced by F-actin masses (29). Jasplakinolide competes with phalloidin for its actin binding site (30), making actin visualization difficult. Accordingly, we sought to determine whether low concentrations of jasplakinolide could restore stretch-induced ERK signaling in the presence of 8-bromo-cGMP. Fig. 4 shows prevention of stretch-induced ERK activity by 8-bromo-cGMP. Preincubation for 60 min with jasplakinolide at the indicated doses restored ERK1/2 activity, with complete restoration at 50 nM jasplakinolide.

Strain-induced Nuclear Translocation of Phospho-ERK Is Prevented by 8-Bromo-cGMP and Restored by Jasplakinol-
ide-We sought to confirm the above observations by determining nuclear translocation of active (phospho) ERK after stretch using immunofluorescent microscopy. Application of a Ϫ27-kPa cyclic stretch for 10 min led to both induction of phospho-ERK and nuclear translocation (Fig. 5, compare B and C). This was largely prevented by pre-incubation for 10 min with 1 mM 8-bromo-cGMP (Fig. 5D). Addition of 50 nM jasplakinolide with 1 mM 8-bromo-cGMP restored both induction of ERK and nuclear translocation (Fig. 5E). Jasplakinolide alone (Fig. 5F) did not affect induction of phospho-ERK in response to stretch, although more fluorescent label appeared to be retained in the cytoplasm.
Strain-induced AP-1 Nuclear Protein Binding Is Prevented by 8-Bromo-cGMP and Restored by Jasplakinolide-ERK signaling is known to increase AP-1 transactivational activity in several cell lines (15,16) including MC (18). Accordingly, we sought to determine whether strain led to ERK-dependent increases in nuclear protein binding to AP-1 consensus sequences and to determine the effects of 8-bromo-cGMP and jasplakinolide. Strain resulted in a marked increase in binding of nuclear proteins to AP-1 consensus sequences (Fig. 6, compare lanes 1 and 2), which was ERK-dependent, because it was abrogated by pre-treatment with PD98059 (Fig. 6, lane 5). Pre-incubation with 8-bromo-cGMP markedly attenuated the nuclear protein binding induced by stretch (Fig. 6, lane 3). This was again restored by pre-incubation with 50 nM jasplakinolide for 60 min (Fig. 6, lane 4), indicating that the effect of 8-bromo-cGMP was on the actin cytoskeleton. DISCUSSION In the best-characterized animal model of chronic renal failure, the subtotally nephrectomized rat, increased glomerular capillary pressure (as little as 20%) triggers MC responses that ultimately result in glomerulosclerosis (10,11,31). In vitro studies of the application of cyclic mechanical strain to MC have demonstrated that this stimulus results in MC proliferation (2,32,33) and production of collagenous protein (2) and fibronectin (12).
We and others have studied how the mechanical signal is transduced in mesangial cells. The first site of transduction occurs at the cell membrane (34). Initial studies of mechanical strain in MC noted increased proliferation in concert with induction of expression of the proto-oncogene and AP-1 component c-fos (1). Both down-regulation of protein kinase C (32) and calcium chelation (32) were shown to attenuate c-fos ex-pression induced by strain. Studies of the proliferative effects of mechanical strain showed matrix dependence. Cells adherent to fibronectin showed the greatest strain-induced proliferative response, and this was inhibited by coincubation with RGD peptides (35). Integrin-focal adhesion complex interactions have been studied, and tyrosine phosphorylation of the focal adhesion-associated kinase pp125 FAK was seen in stretched MC (36). Integrin binding to extracellular matrix protein leads to clustering and the formation of a signaling complex termed a focal adhesion that associates with actin filaments, leading to their reorganization into filamentous stress fibers (37). Stretched MC elongate and align in the direction of stress, and actin filaments coalesce and orient themselves along this long axis (2). Various tyrosine kinases associate with focal adhesions after their formation, in particular focal adhesion kinase and the Src family kinases (38). In mesangial cells, cyclic strain-induced increases of vascular permeability factor mRNA are Src-dependent (39), providing evidence that the assembly of the focal adhesion complex and its association with kinases are important in the transduction of mechanical signals in MC. Further downstream, Src and focal adhesion kinase activate Ras via the Shc-Grb2-Sos complex (40). Ras is activated in response to cyclic strain in vascular smooth muscle cells (41,42).
Signaling of mechanical stimuli to the cell nucleus after membrane events involves the ubiquitous MAPK cascades. ERK interacts with the actin cytoskeleton (24), which is indispensible for signaling in response to lysophosphatidic acid (25) or epidermal growth factor (26). Each of the MAPK cascades consists of three protein kinases acting sequentially, a mitogenactivated protein kinase kinase activator (MKK), a mitogenactivated protein kinase activator (MEK), and a mitogen-activated protein kinase (15). We have shown activation of all three cascades in MC in response to cyclic strain (14). More recently, we have demonstrated inhibition of strain-induced ERK and stress-activated protein kinase/c-Jun NH 2 -terminal kinase by NO donors (43). In the nucleus, both ERK and stress-activated protein kinase/c-Jun NH 2 -terminal kinase may induce AP-1 (15,16). As noted, cyclic strain in the aortic wall activates ERK and AP-1 (18), and glomerular ERK activation and AP-1 nuclear protein binding occur with angiotensin II infusion (19), a maneuver that would be expected to increase glomerular pressure.
Accordingly, we sought to determine how NO interfered with transmission of the stretch signal to ERK. Given recent data showing that 8-bromo-cGMP inhibited actin cytoskeletal organization via cyclic GMP kinase-mediated phosphorylation of RhoA (28), we elected to study whether the effect of NO donors Application of a Ϫ27-kPa stretch at 60 Hz for 10 min led to a prompt formation of parallel stress fibers (D) that was largely prevented by pre-incubation with either SNAP (E) or 8-bromo-cGMP (F).

FIG. 4. Stretch-induced ERK activity in MC is prevented by 8-bromo-cGMP and restored by jasplakinolide.
To determine whether the inhibitory effect of 8-bromo-cGMP on stretch-induced ERK activity was via cytoskeletal disruption, jasplakinolide was used to stabilize the actin cytoskeleton. MC were exposed to a Ϫ27-kPa stretch at 60 Hz for 10 min, and ERK activity was measured by in vitro phosphorylation of a target Elk-1 protein. The total amount of ERK protein present by Western blot was not affected by stretch, 8-bromo-cGMP, or jasplakinolide (A). Stretch resulted in a marked increase in ERK activity that was inhibited by pre-incubation with 8-bromo-cGMP (1 mM, 10 min). Co-incubation with jasplakinolide (50 nM, 60 min) restored the stretch-induced ERK activity. A lower dose of jasplakinolide was unable to restore stretch-induced ERK activity (B). These data are shown graphically in C (n ϭ 4). 8-Bromo, 8-bromo-cGMP; JAS, jasplakinolide.
on stretch-induced ERK activity was through cytoskeletal disruption.
We first observed that stretch led to the formation of actin stress fibers within 10 min in MC. Prevention of stretch-induced actin stress fiber formation in MC completely eliminated the usual stretch-induced ERK activity. This is consistent with a recent report in vascular smooth muscle cells using the same stretch system (22). Having established cytoskeletal dependence of stretch-induced ERK activity in MC, we sought to characterize the effects of NO on the actin cytoskeleton after stretch. Phalloidin staining of F-actin revealed that either SNAP or 8-bromo-cGMP led to cytoskeletal disassembly after 10 min of incubation. This has been observed in resting aortic smooth muscle cells with 40 min of 100 M 8-bromo-cGMP (28). Pre-incubation with 8-bromo-cGMP also inhibited stretch-induced ERK activation in MC in this study, consistent with its effects on the cytoskeleton. In accord with this, phospho-ERK induction and nuclear translocation, which was easily visualized after 10 min of stretch, was essentially prevented by preincubation with 8-bromo-cGMP.
Whereas these data demonstrate an association between cytoskeletal disruption and prevention of strain-induced ERK activity by 8-bromo-cGMP, we sought to strengthen this link by studying cytoskeletal stabilization. A major finding of this study is that incubation with jasplakinolide in addition to 8-bromo-cGMP prior to and during stretch prevented 8-bromo-cGMP-mediated inhibition of strain-induced ERK1/2 activity. Furthermore, induction and nuclear translocation of phospho-ERK by immunofluorescent microscopy was also preserved by co-incubation with jasplakinolide. The effects of jasplakinolide on the actin cytoskeleton are complex, but jasplakinolide appears to stabilize actin stress fibers by decreasing the dissociation rate of actin subunits (29,30). It binds to actin at or near the same site as phalloidin, at the interface of three actin subunits (44). There it acts to decrease the apparent critical concentration of G-actin, enhancing nucleation and inhibiting subunit dissociation from filaments (29). We did not image the cytoskeleton when jasplakinolide was applied to MC because of competition between jasplakinolide and phalloidin.
Finally, we sought to relate these findings to an intranuclear event and chose to study nuclear protein binding to AP-1 consensus sequences by gel-shift assay, because induction of the AP-1 component c-fos is one of the paradigmatic MC responses to stretch (1). Another major finding of this study was that stretch led to a prompt (10 min) increase in nuclear protein binding to AP-1 consensus sequences, which was prevented by the ERK inhibitor PD98059. Pre-incubation with 8-bromo-cGMP also prevented the stretch-induced increase in binding but not when co-incubated with jasplakinolide. These data indicate that 8-bromo-cGMP-mediated inhibition of ERK signaling to AP-1 is dependent on its ability to induce actin cytoskeletal disassembly.
How might NO affect the cytoskeleton? In cervical epithelia, NO donors were found to disrupt the actin cytoskeleton through action on cyclic GMP kinase (45). Subsequently, it was demonstrated that cyclic GMP kinase phosphorylates RhoA in vitro and that transfection of a non-phosphorylatable RhoA mutant abrogated the ability of 8-bromo-cGMP to induce cytoskeletal disassembly (28). Rho-dependent inhibition of myosin light chain phosphatase activity is important in stimulation of actin cytoskeletal stress fibers in response to carbachol (46) and endothelin-1 (47). Consequently, it is possible that 8-bro-

FIG. 5.
Jasplakinolide restores stretch-induced phospho-ERK nuclear translocation in MC treated with 8-bromo-cGMP. MC on flexible bottom plates were incubated with antiphospho-ERK, washed, incubated with a goat anti-rabbit IgG (H ϩ L) conjugate, and then visualized using confocal microscopy. No staining is seen in the absence of primary antibody (A). Unstretched cells show primarily light cytoplasmic staining (B). Application of a Ϫ27-kPa stretch at 60 Hz for 10 min led to prompt induction and nuclear translocation of phospho-ERK (C). This is completely prevented by pre-incubation with 8-bromo-cGMP (1 mM, 10 min)(D). Co-incubation of jasplakinolide (50 nM for 60 min) and 8-bromo-cGMP (1 mM) prior to stretch restores the stretch-induced nuclear translocation (E). Addition of jasplakinolide (50 nM for 60 min) without 8-bromo-cGMP prior to stretch does not result in any further increase in the intensity of nuclear staining beyond that seen with stretch alone (F).
FIG. 6. Jasplakinolide restores stretch-induced AP-1 nuclear protein binding in MC treated with 8-bromo-cGMP. Extracted MC nuclear proteins were reacted with radiolabeled AP-1 consensus oligonucleotides, electrophoresed, and autoradiographed. Lane 1 shows nuclear protein binding in normal MC. Application of a Ϫ27-kPa stretch at 60 Hz for 10 min led to an increase in nuclear protein binding (lane 2), prevented by pre-incubation with 8-bromo-cGMP (1 mM) for 10 min (lane 3). Co-incubation of jasplakinolide (50 nM) and 8-bromo-cGMP (1 mM) prior to stretch restored the AP-1 nuclear protein binding (lane 4). Induction of AP-1 nuclear protein binding by stretch is ERK-dependent, because pre-incubation with 10 M PD98059 prevents it (lane 5). Competition experiments were performed with 100ϫ excess unlabeled consensus oligonucleotides (data not shown). A representative autoradiograph is shown (n ϭ 4). mo-cGMP inhibits stretch-induced actin cytoskeleton organization in MC through this mechanism.
In conclusion, the data presented here clearly demonstrate that stretch-induced ERK activity and AP-1 nuclear protein binding is dependent on an intact actin cytoskeleton. 8-Bromo-cGMP inhibits stretch-induced ERK activity and AP-1 nuclear protein binding through disruptive effects on the cytoskeleton.