ABSTRACT
When subjected to uni-axial cyclic stretch (120% in length, 1 Hz), fibroblasts (3Y1) aligned perpendicular to the stretch axis in a couple of hours. Concomitantly with this orienting response, protein tyrosine phosphorylation of cellular proteins (molecular masses of approximately 70 kDa and 120-130 kDa) increased and peaked at 30 minutes. Immuno-precipitation experiments revealed that paxillin, pp125FAK, and pp130CAS were included in the 70 kDa, and 120-130 kDa bands, respectively. Treatment of the cells with herbimycin A, a tyrosine kinase inhibitor, suppressed the stretch induced tyrosine phosphorylation and the orienting response suggesting that certain tyrosine kinases are activated by stretch. We focused on pp60src, the most abundant tyrosine kinase in fibroblasts. The kinase activity of pp60src increased and peaked at 20 minutes after the onset of cyclic stretch. Treatment of the cells with an anti-sense S-oligodeoxynucleotide (S-ODN) against pp60src, but not the sense S-ODN, inhibited the stretch induced tyrosine phosphorylation and the orienting response. To further confirm the involvement of pp60src, we performed the same sets of experiments using c-src-transformed 3Y1 (c-src-3Y1) fibroblasts. Cyclic stretch induced a similar orienting response in c-src-3Y1 to that in wild-type 3Y1, but with a significantly faster rate. The time course of the stretch-induced tyrosine phosphorylation was also much faster in c-src-3Y1 than in 3Y1 fibroblasts. These results strongly suggest that cyclic stretch induces the activation of pp60src and that pp60src is indispensable for the tyrosine phosphorylation of pp130CAS, pp125FAK and paxillin followed by the orienting response in 3Y1 fibroblasts.
INTRODUCTION
All tissues in the body are subjected to physical forces originating either from themselves or the environment. In response to these forces, cells exhibit a variety of responses including increased gene expression, protein synthesis, and morphogenesis.
As for fibroblasts, it is known that their shape is dramatically altered in response to mechanical stimulation. Arterial fibroblasts subjected to uni-axial cyclic stretch aligned perpendicularly to the stretch axis (Kanda and Matsuda, 1993). In ligamentous fibroblasts subjected to repetitive stretch, they aligned perpendicularly to the deformation axis of the substrate (Desrosiers et al., 1995). However, signaling mechanisms as well as mechanotransduction processes are largely unknown in mechanically induced shape change in fibroblasts. Similar stretch-induced morphological changes were also observed in different cell types (Dartsch and Haemmerle, 1986; Terracio et al., 1988; Hamasaki et al., 1995) and data on mechanotransduction processes have been accumulated recently. We have been particularly interested in the morphological changes in response to uni-axial cyclic stretch in vascular endothelial cells. A previous study from our laboratory in human vascular endothelial cells (HUVECs) indicated that tyrosine phosphorylation of focal adhesion proteins is critical for stretch induced morphological changes such as orientation and elongation perpendicular to the stretch axis (Naruse et al., 1998a). Treatment with tyrosine kinase inhibitors, such as genistein and herbimycin A, inhibited the stretch induced morphological changes. Moreover, treatment with an antisense S-oligodeoxynucleotide against pp125FAK, which down-regulated the expression of the protein, also suppressed the morphological changes (Naruse et al., 1998b).
These results prompted us to further investigate whether a similar signaling mechanism in HUVECs is involved in the mechanically induced morphological changes in fibroblasts. Compared to HUVECs, fibroblasts have advantages for investigating the intracellular signaling cascade because they have a good proliferation rate, and are suitable for molecular biology because of the availability of transfectants.
The aim of this study was to explore the signaling events downstream of the initial mechanoreception on the plasma membrane of fibroblasts, with the main emphasis on the role of tyrosine phosphorylation. Furthermore, to clarify the role of the tyrosine kinase pp60src in the stretch induced orienting response and protein tyrosine phosphorylation, the alteration in morphology and protein tyrosine phosphorylation were analyzed in 3Y1 fibroblasts treated with an antisense or sense S-ODN against pp60src expression.
MATERIALS AND METHODS
Cell culture
Rat 3Y1 fibroblasts (Kimura et al., 1975) and c-src-transformed 3Y1 (c-src-3Y1) fibroblasts were maintained in D-MEM (Dulbecco’s minimum essential medium) supplemented with 10% fetal calf serum and 100 units/ml penicillin G sodium, 100 μg/ml streptomycin sulfate and 250 μg/mg amphotericin B. Cells were incubated in a tissue culture incubator at 37°C in a humidified atmosphere of 5% CO2 and 95% air.
Stretch apparatus
The cells were stretched as previously described (Naruse et al., 1998a,b; Kato et al., 1998; Sokabe et al., 1997; Suzuki et al., 1997). Briefly, cells were removed from the dish with 0.01% EDTA-0.02% trypsin and transferred onto a silicon chamber coated with 50 μg/ml fibronectin. One end of the chamber is firmly attached to a fixed frame, while the other end is held on a movable frame. The movable frame is connected to a motor driven shaft. The amplitude and the frequency of stretch were controlled by a programmable microcomputer. The silicon membrane was uniformly stretched over the whole membrane area and lateral thinning did not exceed 1% at 20% stretch.
Analysis of cell morphology
The morphology of cells on the silicon chamber was analyzed in photographs of phase contrast images obtained through a ×10 objective (Olympus, Tokyo Japan). The photographs were digitized by a film digitizer (Nikon Cool Scan, Tokyo Japan) and the captured images were analyzed using an image analysis program (Mocha Jandel Scientific) running on Windows.
Immuno-blotting
After the application of cyclic stretch, the cells were washed with ice-cold phosphate-buffered saline (PBS: 137 mM NaCl, 8.10 mM Na2HPO4, 2.68 mM KCl, 1.47 mM KH2PO4, pH 7.40) containing 0.5 mM Na3VO4 and were lysed with a sample buffer (62 mM Tris-HCl, pH 6.8, 2% sodium dodecyl sulfate (SDS), 1% 2-mercaptoethanol, 0.003% Bromophenol Blue, 5% glycerol). Total cell lysate from equivalent cell numbers was separated by 12.5% SDS-PAGE (Laemmli, 1970). The proteins were transferred electrophoretically onto polyvinylidene fluoride (PVDF) membranes, (Immobile P, Millipore Crop, Bedford, MA). The PVDF membranes were blocked with 3% ovalbumin in PBS, the membranes were subsequently probed with monoclonal antibodies in PBS containing 3% ovoalbumin for 1 hour. The antibody-antigen complexes were detected by using horseradish peroxidase-conjugated goat anti-rabbit IgG (1:1000 dilution). Immuno-reactivity was determined using the ECL chemi-luminescence reaction (Amersham Corp., Arlington Heights, IL), according to the instructions provided by the manufacturer and was measured using densitometry.
Immuno-precipitation and in vitro kinase assay
For immuno-precipitation, the cells were washed with ice-cold PBS and were lysed by A buffer (10 mM Tris-HCl, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 1% SDS, pH 7.40). Cell lysates were centrifuged twice at 16,000 g for 20 minutes at 4°C and 50 μl of Protein A-Sepharose anti-mouse IgG were added to pre-clear the lysate. 1 μl of monoclonal anti-pp60src were added and the samples were incubated at 4°C, for 1 hour. Then 30 μl of Protein A-Sepharose anti-mouse IgG were added, and the samples were incubated at 4°C for 1 hour with gentle agitation. The resulting immuno-precipitates were washed five times with A buffer, and three times with kinase buffer (10 mM Tris-HCl, pH 7.40, 5 mM MnCl2). The kinase reaction was started with the addition of 5 μCi of [γ-32 P]ATP, 1 μM ATP and 2.5 μg of enolase. The reaction was terminated after 10 minutes with the addition of an equal volume of SDS-sample buffer and the samples were boiled for 5 minutes. The samples were separated on 12.5% SDS-PAGE and the autoradiogram was analyzed by BAS-1500 radioactivity imaging system (Fuji Photo Film, Tokyo, Japan).
Antisense S-ODN treatment
The cells were removed with 0.01% EDTA, 0.02% trypsin and were suspended at a density of 1×106 cells/ml in PBS. 100 ng of S-ODN was added to 400 μl of cell suspension and the suspension was transferred to an electroporation chamber, and was square electroporated at 75 V, 40 milliseconds, 1 pulse, using an Electro Square Porator-T820 (BTX Inc., San Diego, CA). Then the S-ODN treated cells were transferred into a 4 cm2 silicon chamber, and were incubated for 24 hours. The anti-sense sequence against chicken pp60src (5'-GTC GGG GGC TGC TGT CTT-3') and the sense sequence (3'-CAG CCC CCG ACG ACA GAA-5') were designed using Genetyx (Softwear Development Company, Tokyo, Japan) and were synthesized by Rikaken Co., Ltd, Nagoya Japan.
Materials
Mouse anti-pp130CASmonoclonal antibody, mouse anti-pp125FAK monoclonal antibody, mouse anti-paxillin monoclonal antibody, and PY20 were purchased from Transduction Laboratories (Lexington, KY). Mouse anti-pp60src monoclonal antibody was purchased from Upstate Biotechnology Inc. (Lake Placid, NY). Rabbit anti-mouse IgG and goat anti-rabbit IgG were obtained from Sigma (St Louis, MO). Herbimycin A were purchased from Wako (Tokyo, Japan). Gadolinium (Gd3+) chloride hexahydrate was purchased from Aldrich Chem. Co. Human plasma fibronectin was purified according to the method of Regnault et al. (1988). Other chemicals used were of special grade.
RESULTS
Cyclic stretch induced orienting response and protein tyrosine phosphorylation in 3Y1 fibroblasts
Application of sinusoidal cyclic stretch to the cells induced significant morphological changes particularly in their orientation. Without cyclic stretch, the cells showed no particular orientation (Fig. 1, 0 minutes). The cells began to orient perpendicularly to the stretch axis and the percentage of oriented cells increased with time. After 180 minutes, most of the cells aligned almost perfectly transversely across the stretch axis. A series of histograms in Fig. 1 show a time course of the distribution of cell orientation. Without stretch, cells showed no changes in their morphology at least for 3 hours. In contrast, the cells under cyclic stretch showed a clear indication of alignment, around 90° at 60 minutes after the onset of stretch.
To investigate the involvement of protein tyrosine phosphorylation in the stretch induced orienting response, tyrosine phosphorylation of cellular proteins was examined by immuno-blotting using anti-phosphotyrosine monoclonal antibody. In response to cyclic stretch, a time dependent tyrosine phosphorylation of multiple proteins was observed (Fig. 2A). Those proteins included broad bands of 130 kDa, 125 kDa, and 68 kDa. The elevation of the level of tyrosine phosphorylation was detected from 5 minutes after the onset of cyclic stretch, peaked at 30 minutes and declined gradually thereafter (Fig. 2B). Even without cyclic stretch, a certain level of protein tyrosine phosphorylation was observed, probably because integrin-mediated protein tyrosine phosphorylation occurred as the cells adhered to the extracellular matrix (Burridge et al., 1992).
To ensure if the anti-phosphotyrosine monoclonal antibody used here specifically recognized the tyrosine phosphorylated proteins, immuno-blotting was performed in the presence of an excess amount of phosphorylated tyrosine phosphorylated serine and phosphorylated threonine. Phosphorylated tyrosine inhibited the band, which was detected by anti-phosphotyrosine monoclonal antibody to tyrosine phosphorylated proteins, while phosphoserine and phosphothreonine had no effect on immuno-detection patterns (data not shown). As herbimycin A has been demonstrated to be an effective inhibitor of tyrosine kinase (Satoh et al., 1992), we examined the effect of herbimycin A on the stretch induced tyrosine phosphorylation. When the cells were pre-incubated with 3 M herbimycin A for 10 hours to block tyrosine kinase, the cells did not show any orienting response even 60 minutes after the onset of cyclic stretch (Fig. 3A) and the increase in the amount of tyrosine phosphorylated proteins was not significant (Fig. 3B,C). Those results strongly suggest that protein tyrosine phosphorylation is necessary for the stretch induced orienting response in fibroblasts.
Paxillin, pp125FAK, and pp130CAS were included in tyrosine phosphorylated proteins
Since major tyrosine phosphorylated proteins were detected at 70 kDa and 120-140 kDa, we examined the involvement of tyrosine phosphorylation of pp130CAS (130 kDa; Sakai et al., 1994), pp125FAK (125 kDa; Schaller et al., 1992), and paxillin (68 kDa; Turner et al., 1990) in 3Y1 fibroblasts. Immunoprecipitation using the anti-paxillin monoclonal antibody revealed a single band of tyrosine phosphorylated protein of 68 kDa (Fig. 4A, upper). The level of tyrosine phosphorylation increased (1.7-fold) in response to cyclic stretch without any change in the total amount of paxillin present (Fig. 4A, lower). Pp125FAK was immuno-precipitated from lysates of static and stretched cells, and probed with anti-phosphotyrosine monoclonal antibody. The level of tyrosine phosphorylation of pp125FAK immunoprecipitated from stretched cells was much higher (1.8-fold) compared with that[para-merge-chec in static cells (Fig. 4B). Pp130CAS was examined in the same way and the level of tyrosine phosphorylation of pp130CAS immunoprecipitated from stretched cells was much higher (2.1-fold) compared with that in static cells (Fig. 4C). Fig. 4D summarizes the tyrosine-phosphorylated proteins in 3Y1 fibroblasts.
Pp60src is responsible for the stretch induced tyrosine phosphorylation
As cyclic stretch increased the level of protein tyrosine phosphorylation in 3Y1, we next investigated the involvement of pp60src in the cells, since pp60src is one of the major tyrosine kinases (Golden and Brugge, 1989), and pp125FAK is known as one of the substrates of pp60src. Pp60src was immunoprecipitated from lysates of 3Y1 fibroblasts subjected to uni-axial cyclic stretch. The specific kinase activity of pp60src was measured as the amount of phosphorylation of acid-treated enolase by immuno-complex followed by SDS-PAGE and autoradiography. The autoradiogram of phosphorylated enolase at 40 kDa revealed the increase in kinase activity of pp60src as early as 10 minutes after the stretch onset (Fig. 5A). The immuno-blotting pattern in Fig. 5B indicates that the same amounts of pp60src were loaded in each lane. The activity of pp60src peaked at 20 minutes (3.0±0.2-fold, mean ± s.e., n=5) and decreased gradually (Fig. 5C). The phosphorylation of pp60src was also observed and the time course of the phosphorylation was almost parallel with that of the tyrosine phosphorylation level of acid treated enolase. We also measured the kinase activity of src and the morphological changes in response to a static and constant deformation instead of the oscillatory deformation. However, we did not observe any changes in the kinase activity of pp60src nor in its morphology. To further investigate the role of pp60src in the stretch-dependent orienting response and protein tyrosine phosphorylation, we employed an antisense S-ODN against pp60src to down-regulate the expression of pp60src. The expression level of pp60src protein in the antisense SODN treated cells decreased less than 24% as compared to that in the sense S-ODN treated cells, whereas the expression level of other proteins such as actin was not significantly changed (Fig. 6A). Without stretch stimulation, we could not observe any significant morphological changes in the S-ODN treated cells. The S-ODN treated cells were subjected to uni-axial cyclic stretch, and the levels of tyrosine phosphorylation and pp60src kinase activity were examined. In the sense S-ODN treated cells, the increase in tyrosine phosphorylation was observed in three major protein bands at 68, 125, and 130 kDa, as was observed in the untreated cells and an increase in the kinase activity was also observed (Fig. 6B), whereas in the antisense S-ODN treated cells, the basal tyrosine phosphorylation level was lower compared to the sense S-ODN treated cells, cyclic stretch did not increase the protein tyrosine phosphorylation level significantly, and kinase activity was not observed at all (Fig. 6B). Finally, uni-axial cyclic stretch induced the cell alignment perpendicular to the stretch axis in the sense S-ODN treated cells (Fig. 6C, upper), whereas in the antisense S-ODN treated cells, the orienting response was strongly inhibited (Fig. 6C, lower).
Pp60src transformed 3Y1 fibroblasts showed an accelerated orienting response
We have shown that pp60src was activated by cyclic stretch and that its depletion by antisense S-ODN suppressed the stretch induced orienting response. It is of interest to see whether pp60src transformed cells show the acceleration in the morphological response to uni-axial cyclic stretch. First, we examined the kinase activity of pp60src of c-src transformed 3Y1(c-src-3Y1) fibroblasts during uni-axial cyclic stretch. After the application of uni-axial cyclic stretch, pp60src was immunoprecipitated and the specific kinase activity of pp60src was measured as the phosphorylation of acid-treated enolase by the immuno-complex. In c-src-3Y1 fibroblasts, the tyrosine kinase activity increased and peaked at 10 minutes in response to cyclic stretch (Fig. 7A). Immuno-blotting against pp60src (Fig. 7B) revealed that the same amounts of protein were loaded in each lane, and that in comparison with Fig. 5B, c-src-3Y1 fibroblasts express a 3-fold higher level of pp60src than do wild-type 3Y1 fibroblasts. Fig. 7C shows the comparison of the kinase activity of pp60src in wild-type 3Y1 fibroblasts and c-src-3Y1 fibroblasts during cyclic stretch. We also examined the orienting response of c-src-3Y1 fibroblasts to cyclic stretch. In response to uni-axial cyclic stretch, c-src-3Y1 fibroblasts showed an alignment perpendicular to the stretch axis (Fig. 8). The striking difference between wild type and the transformed fibroblasts was that the orienting response of c-src-3Y1 fibroblasts was much faster than that of wild-type 3Y1 fibroblasts. In order to compare the difference in the time course more precisely, the standard deviations (s.d.) of the orientation distribution in the histograms were plotted against time, because the s.d. may be a good indication of the degree of orienting response. Fig. 9 shows the time course of the s.d. from orientation distributions at each time point for wild-type 3Y1 and c-src-3Y1 fibroblasts, showing that c-src-3Y1 fibroblasts aligned much faster than wild-type 3Y1 fibroblasts. Again, these data confirmed that pp60src plays a crucial role in the stretch dependent orienting response in fibroblasts.
DISCUSSION
Stretch-induced activation of pp60src
In the present study, we have demonstrated that uni-axial cyclic stretch induced the activation of pp60src in fibroblasts and that pp60src is critical for the stretch induced tyrosine phosphorylation of focal proteins and for the orienting response. Several observations are consistent with the notion that phosphotyrosine accumulation is involved in the process of the stretch-induced orienting response. First, the time course of the level of phosphorylation closely resembles that of the orienting response both in 3Y1 and c-src-3Y1. Uni-axial cyclic stretch increased pp60src activity and the activity peaked at 20 minutes (Fig. 5) in 3Y1 fibroblasts. The time course of the tyrosine phosphorylation level of pp125FAK, paxillin, and pp130CAS coincided with that of pp60src activation (Figs 2, 5). Second, treatment with herbimycin A inhibited not only the stretch-induced tyrosine phosphorylation but also the orienting response. Finally, depletion of pp60src by the antisense-SODN suppressed the stretch-induced orienting response (Fig. 3). Moreover, overexpression of c-src in 3Y1 accelerated the stretch induced orienting response (Figs 8, 9). Taken together, it is highly likely that pp60src is involved and is playing a crucial role in the stretch-induced orienting response in fibroblasts.
Involvement of SA channels in the activation of pp60src tyrosine kinase
One probable mechanism for the stretch induced activation of pp60src may be an activation of Ca2+-permeable stretch-activated (SA) channels followed by intracellular Ca2+ concentration ([Ca2+]i) increase. Ca2+-permeable SA channels have been reported in fibroblasts by patch clamp experiments (Stockbridge and French, 1988; Glogauer et al., 1995). The involvement of SA channels in the stretch-induced morphological response was reported in HUVECs. In HUVECs, uni-axial cyclic stretch induced intracellular Ca2+ mobilization, which was blockable by Gd3+, and tyrosine phosphorylation of focal proteins followed by morphological changes (Naruse et al., 1998b). The present study in fibroblasts implies the involvement of the same mechanism as in HUVECs, as a result of the following observations: (1) treatment of the cells with EGTA or BAPTA/AM inhibited the stretch-induced kinase activity of pp60src, suggesting that Ca2+influx followed by an increase in [Ca2+]i contributes to the activation of pp60src. (2) Application of Gd3+, a blocker for the SA channel, inhibited the stretch-induced kinase activity of pp60src as well as the stretch-induced tyrosine phosphorylation of focal proteins and the stretch-induced morphological changes.
At present, there is no evidence that pp60srcis activated directly by the increase in [Ca2+]i. However, a recent observation made by our laboratory suggested the involvement of calcineurin, Ca2+/CaM dependent phosphatase in the stretch-induced pp60src activation in HUVECs (Naruse et al., 1998c), since the stretch induced activation of pp60src was inhibited by FK506, a specific inhibitor for calcineurin. It has been proposed that pp60src is maintained in a low activity state by an interaction between phosphotyrosine 527 and the pp60src homology 2 (SH2) domain and that pp60src is activated upon tyrosine 527 dephosphorylation (Roussel et al., 1991). It is possible that calcineurin dephosphorylates the phosphotyrosine 527 of pp60src. Dephosphorylation of phosphorylated 527 would expose the SH2 domain freely, thus allowing it to bind to specific phosphotyrosine containing proteins (Fukui et al., 1991; Koch et al., 1991).
Role of tyrosine phosphorylation of focal proteins
In fibroblasts, tyrosine phosphorylation of paxillin and pp125FAK has been implicated during actin assembly (Burridge et al., 1992). In bovine endothelial cells, cyclic strain induces the tyrosine phosphorylation of pp125FAK and paxillin and these proteins are suggested to be involved in the morphological changes (Yano et al., 1996; Naruse et al., 1998a). Paxillin has been suggested to be a direct substrate of pp125FAK (Rankin and Rozengurt, 1994) and found to bind in vitro to the SH3 domain of pp60src (Weng et al., 1993). In 3Y1 fibroblasts, morphological alteration and transforming ability were regulated through increasing tyrosine phosphorylation of the p130CAS (Ogawa et al., 1994). It is likely that tyrosine phosphorylation of focal proteins may serve as a proximate signal for actin polymerization (Burridge et al., 1992). Treatment of the cells with herbimycin A suppressed the stretch-induced tyrosine phosphorylation of those proteins and inhibited the orienting response in 3Y1 fibroblasts, suggesting that tyrosine phosphorylation of focal proteins is indispensable for the stretch induced orienting response. How cells sense the direction of stretch remains unclear. One possibility is that cell orientation requires remodeling of focal adhesions; disassemby of existing complexes and formation of new ones to anchor reoriented stress fibers. Support for this idea is provided by the fact that when formation of new focal adhesions is inhibited cyclic stretching induces cell detachment as shown in Fig. 3A.
ACKNOWLEDGEMENTS
We thank Dr M. Hamaguchi for discussion. We also thank M. Takahashi for excellent technical assistance. This work was supported by a Grant-in Aid from the Ministry of Education Science and Culture Japan (09770025 for K.N. and 10480180, 10169227 for M.S.), a research grant (Research for the Future Program) from the Japan Society for the Promotion of Science (K.N. and M.S.), and a grant from Japan Space Forum (M.S.).