Analysis of cyclic-stretching responses using cell-adhesion-patterned cells

Available online at www.sciencedirect.com

Journal of Biotechnology 133 (2008) 82–89

Analysis of cyclic-stretching responses using cell-adhesion-patterned cells Yuki Katanosaka, Jin-Hua Bao, Tomoyo Komatsu, Tomohiko Suemori, Akira Yamada, Satoshi Mohri, Keiji Naruse ∗ Department of Cardiovascular Physiology, Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama University, 2-5-1 Shikata-cho, Okayama 700-8558, Japan Received 24 August 2007; accepted 15 September 2007

Abstract Human vascular endothelial cells form the interface between the bloodstream and vessel walls and are continuously subjected to mechanical stimulation. When endothelial cells are stretched cyclically, along one axis, they align perpendicular to the axis of stretch. We previously reported that applying a cyclic, uni-axial strain to cells induced tyrosine phosphorylation of focal adhesion kinase and stimulated mitogen-activated protein kinase. However, it is difficult to quantify and analyze the spatial distribution of tyrosine phosphorylation in these cells, as they form focal adhesions randomly. In this study, we developed a system to overcome this problem by preparing individual, uniform, patterned cells that could be stretched cyclically and uni-axially. We constructed polydimethylsiloxane stretch chambers and used microcontact printing technology to imprint a pattern of 2 ␮m fibronectin dots (10 lines × 10 columns in a 38 ␮m square) before seeding them with human umbilical vein endothelial cells (HUVEC). We found that most HUVEC attached to the patterned dots after 2 h and were similar in size and morphology, based on phase-contrast microscopy. In this system we were able to statistically analyze tyrosine phosphorylation and actin polymerization in these patterned cells, when subjected to a cyclic, uni-axial strain, using fluorescent microscopy. © 2007 Elsevier B.V. All rights reserved. Keywords: HUVEC; Microcontact printing; PDMS stretch chamber; Tyrosine phosphorylation

1. Introduction It is important for vascular endothelial cells to sense and adapt to mechanical stimuli, arising from, for example, changes in blood flow and blood pressure, in order to respond to, and potentially prevent, the pathological changes involved in diseases, such as arteriosclerosis and hypertension (Davies and Tripathi, 1993; Hudlicka and Brown, 1993; Skalak and Price, 1996). To understand these mechanisms, it is important to clarify the mechanoreceptive properties of endothelial cells. Substantial evidence suggests that many different types of signaling molecule are involved in the biochemical signaling response elicited by mechanical stimuli, including stretchsensitive ion channels, protein kinase C, focal adhesion kinase (FAK), Rho, heterotrimeric G proteins and adenylyl cyclase

∗

Corresponding author. Tel.: +81 86 235 7114; fax: +81 86 235 7430. E-mail address: [email protected] (K. Naruse).

0168-1656/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jbiotec.2007.09.017

(Alenghat and Ingber, 2002; Ingber, 2003). We previously reported that cyclic, uni-axial stretching activated Src and the protein tyrosine phosphorylation of focal proteins, including FAK, p130Cas , and paxillin in human umbilical vein endothelial cells (HUVEC; Suzuki et al., 1997; Naruse et al., 1998; Sai et al., 1999; Wang et al., 2001). Although these studies have provided an analysis of the overall intracellular responses occurring within the cell, there have been no reports concerning the localization of signaling pathway activation to different parts of the cell. Therefore, it has been difficult to discuss intracellular localization, such as the direction of focal adhesion, with respect to the stretch axis, in which tyrosine phosphorylation is induced. However, for vectorial stimulation, such as stretching, it has been predicted that a load will be applied to different regions of a cell relative to the stretch axis. In experiments with cultured cells, it is difficult to standardize their form, axis and size. Therefore, even though the external mechanical stimulation remains constant, not all target cells receive the same amount of stimulation. Moreover, even if the

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89

fixed mechanical stimulation is externally stressed, the localization and quantitative definition of intracellular responses can only be achieved with partial data. Creating patterns of biomolecules on solid substrates can be used not only to control cell-adhesion and growth but also to regulate cell function (Singhvi et al., 1994; Chen et al., 1998; Dike et al., 1999; Ito, 1999; Kane et al., 1999). Microcontact printing is becoming an increasingly popular way to achieve biomolecular patterning (Xia and Whitesides, 1998; Kane et al., 1999), as it is simple, fast and inexpensive, and does not require a clean-room or completely flat surfaces, and allows the creation of complex patterns on a surface. In this study, fibronectin was applied to polydimethylsiloxane (PDMS) stretch chambers by microcontact printing and the focal adhesion induced in HUVECs in this way allowed stretch stimulation to be applied to the cells in a specific pattern. It was therefore possible to analyze the localization of responses to a cyclic-stretching of a fixed size and direction, including the formation of focal adhesions. As this technique allows individual patterned cells to be prepared, without any cell–cell contacts, it could also be used to examine cell-adhesion in response to stretch stimulation.

83

adhering to other parts of the printed chambers, they were blocked with 1% BSA at 37 ◦ C for 30 min, before introducing HUVECs. 2.3. HUVEC isolation and cell cultures

2. Materials and methods

HUVECs were isolated from the umbilical vein of a postpartum woman, collected and handled in accordance with the approval granted by the Ethics Committee of Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Science, Japan. The human umbilical vein was washed in calcium- and magnesium-free PBS (PBS(–)), then subjected to several sequential treatments with 0.2% trypsin (Wako Chemical, Osaka, Japan) in PBS(–) for 10 min at 37 ◦ C. The endothelial cells obtained from the trypsin digests were pooled and passed through a 100 ␮m nylon cell strainer (Falcon, Franlkin Lake, NJ) to remove any undigested matrix. The cells released into the supernatant were collected by centrifugation at 150 × g for 10 min, and resuspended in HuMedia-EG2 (Kurabou, Okayama, Japan) supplemented with 10% fetal calf serum (FCS; Gibco BRL, Grand Island, NY). The cells were incubated at 37 ◦ C in a 5% CO2 atmosphere, and the culture medium was changed every third day.

2.1. Preparation of PDMS rubber stamps

2.4. Preparation of patterned cells

A silicon master was fabricated by a photolithography process on a silicon wafer coated with a positive resist, and the patterns were transferred by deep reactive ion etching to a 1.0 ␮m depth. PDMS rubber stamps were prepared using a 10:1 (v/w) ratio of elastomer TSE3032 (Toshiba, Tokyo, Japan) to hardener TSE3032 (Toshiba). After extensive mixing and degassing under vacuum, unpolymerized PDMS was poured over the silicon wafer mold and allowed to polymerize for 1 h at 65 ◦ C to crosslink the polymer. The stamps formed were then separated from the structured silicon wafers, and examined by light microscopy to assess the quality of the patterns. The rubber stamps were treated with the plasma cleaner/sterilizer PDC-32G (Harrick, Ithaca, NY) to enhance the hydrophilic properties of the surface.

HUVECs were removed from culture vessels with 0.02% trypsin in 0.01% ethylenediaminetetraacetic acid (EDTA) and plated at a density of 4 × 106 cells per well in PDMS stretch chambers printed with fibronectin using the microcontact printing method described above. They were then incubated in HuMedia-EG2 supplemented with 10% FCS, and incubated at 37 ◦ C in a 5% CO2 atmosphere for 30 min. The HUVEC in the silicon chambers were rinsed with medium to remove unattached cells, cultured for 2 h in serum-free medium before they wee used for stretch analysis. To analyze tyrosine phosphorylation in response to stretching, the HUVECs were pre-cultured in serum-free HuMedia-EG2 at 37 ◦ C for 2 h.

2.2. Microcontact printing on PDMS stretch chambers

HUVECs were subjected to stretch using the STRETCH ST140 instrument (Strex, Osaka, Japan). Briefly, one end of the chamber was firmly attached to a fixed frame, while the other was attached to a movable frame connected to a motor-driven shaft. The amplitude and frequency of stretch were controlled by a programmable microcomputer. The silicon membrane was uniformly stretched over the whole membrane area, and the lateral thinning did not exceed 1% at 20% stretch.

To reduce contamination of the printed surface by airborne dust, the stamps were rinsed in phosphate-buffered saline (PBS) and dried in a nitrogen steaming. Fibronectin (Chemicon, Temecula, CA) was labeled with Alexa 568 using a commercial labeling kit (Molecular Probes, Eugene, OR). Each PDMS rubber stamp was coated with a 100-␮l aliquot of 0.1 mg/ml fibronectin for 30 min at 37 ◦ C. After removing excess fibronectin using an air duster (HOZAN, Osaka, Japan), the stamps were rinsed, dried under a nitrogen stream and stamped onto the PDMS stretch chamber. The 38 ␮m arrays of 10 × 10 fibronectin dots, each 2 ␮m across, printed onto the PDMS stretch chambers in this way were examined by light microscopy to check the quality of the fibronectin stamps. To prevent cells

2.5. Stimulation by cyclic uni-axial stretching

2.6. Immunocytochemical analyses Cell arrays on the PDMS stretch chambers were fixed with 4% paraformaldehyde for 15 min at room temperature, permeabilized with 0.1% Triton X-100 and then stained with mouse monoclonal antibodies against phosphotyrosine (pY;

84

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89

Transduction Lab., Flanklin Lake, NJ), FAK (Transduction Lab.), paxillin (Transduction Lab.), or vinculin (Sigma Chemical, St. Louis, MO) diluted 1:500 for 60 min. The samples were then incubated with Alexia 488-conjugated anti-mouse immunoglobulin G (IgG; Invitrogen, Tokyo, Japan) and Alexia 568-conjugated phalloidin (Invitrogen), and examined using a confocal laser-scanning microscope LV1000 (Olympus, Tokyo, Japan) mounted on an Olympus BX50WI epifluorescence microscope with a plan-apochromat ×60 oil-immersion objective lens (Olympus). 2.7. Data analysis Each HUVEC used in these experiments had 100 focal adhesions of 2 ␮m in diameter, determined by the 10 × 10 array of fibronectin dots underlying each cell. For each focal adhesion, we set the same region of interest (ROI) and determined the fluorescent intensity of anti-pY antibody labeling at each one. Results presented are the average fluorescent intensity for several cells (n = 3) for every focal adhesion, as shown in Fig. 5K, L and M.

3. Results and discussion 3.1. Problems in analyzing cellular responses in random cell cultures So far, changes in cellular morphology and intracellular responses to biomechanical forces have been analyzed in randomly cultured cells. It is difficult to standardize the form, axis, and size of adherent cells with a uniform number of receptors or focal adhesions, and to maintain the area and condition of the plasma membrane. The cellular responses to uni-axial cycle stretch applied to these cells have therefore been difficult to analyze. We applied a uni-axial cycle stretch, 20% average strain of 1 Hz (Fig. 1, arrow) to randomly cultured HUVECs for 0, 15 or 60 min, and observed the polymerization of F-actin and local changes in tyrosine phosphorylation. As the molecules which constitute focal adhesions, that is PECAM-1, the negative regulator of Srk family kinase CrK, p130Cas and paxillin, are known to be tyrosine phosphorylated in response to integrin signaling (Suzuki et al., 1997; Naruse et al., 1998; Sai et al., 1999; Wang

Fig. 1. Effect of cyclic uni-axial stretching (1 Hz, 20% stretch) on tyrosine phosphorylation and actin organization in randomly cultured HUVECs. HUVECs at rest (A–C) or subjected to cyclic-stretching for 15 min (D–F) or 60 min (G–I) were labeled with anti-pY antibody (A, D and G) and with phalloidin to label F-actin (B, E and H). The images labeled for pY and F-actin are merged in (C, F and I). White squares show the orientation of anti-pY antibody signals. White arrows show actin patches. Bar: 50 ␮m.

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89

et al., 2001), we assumed that the observed tyrosine phosphorylation reflected the response of focal adhesion molecules to mechanical stimulation. Before stretching, staining the cells with an anti-pY antibody showed a pattern radiating from the inside to the outside of the cells (Fig. 1A, white square). After 60 min of cyclicstretching, however, the pattern of phosphorylation was clearly perpendicular to the stretch axis (Fig. 1G, white square). The fluorescent intensity of anti-pY antibody staining was 843 ± 29 in arbitrary units (AU) (n = 9) in resting cells, 1458 ± 32 AU (n = 9) after stretching for 15 min, and 826 ± 27 AU after stretching for 60 min, indicating an increase in tyrosine phosphorylation after 15 min (Fig. 1D). Most tyrosine phosphorylation developed inside the cells, but, as yet, it has been difficult to discuss the localization and quantitative definition of intracellular responses, such as the direction of focal adhesion, with respect to the stretch axis (Fig. 1D). In resting cells, actin was located in characteristic patches inside the cells, as shown by the white arrow in Fig. 1B. However, after stretch stimulation for 15 or 60 min, actin labeling was found at the cell–cell contacts between individual cells (Fig. 1B, C, E, F, H and I). Actin patches have been shown to move along actin cables and become concentrated at cell tips which are the sites of polarized growth (Pelham and Chang, 2001). Recently, the velocity at which these actin patches move has been shown to be depen-

85

dent on the level of actin polymerization, suggesting that the patches function as centers of dynamic actin-polymerization that contribute to the formation and maintenance of actin cables (Pelham and Chang, 2001). Fig. 1E and H show that the movement of the actin patches to cell tips is also dependent on F-actin polymerization in stretched cells. After cyclic-stretching for 60 min, F-actin lined up within cells perpendicular to the stretch axis (Fig. 1H), as had been observed previously in cyclically stretched, randomly cultured HUVECs, in which F-actin polymerization had been enhanced (Shirinsky et al., 1989). In Fig. 1C, the co-localization of pY and F-actin is not obvious at cell–cell contacts before stretching, but can be clearly seen between most cells after stretch stimulation (Fig. 1F and I). This probably reflected the assembly of focal adhesion complexes and F-actin at the cell–cell contact regions and cell tips in response to stretching. To analyze the region in which the cellular response is induced by vectorial stretching, we standardized the adherence conditions for HUVECs to obtain multiple arrays of cells with a fixed number of focal adhesions. 3.2. Preparation of cell arrays on PDMS stretch chambers Even within the same cell lineage, the size, shape and form of individual cells varied (Fig. 1). We therefore used microcon-

Fig. 2. Preparation of cell-adhesion-patterned cells. (A) Diagram of the microcontact printing technique for stamping fibronectin onto the PDMS stretch chamber. Unpolymerized PDMS was poured into a mold (shown as scanning electron-microscopy images in (B and C)) and heated to set and form a stamp (shown as scanning electron-microscopy images in (D and E)). This stamp was used with fibronectin as “ink”, to stamp patterns onto the stretch chamber (shown as confocal laser-scanning microscope images in (F and G)). (H) Patterned HUVEC in the stretch chamber. Bars: 70 ␮m in (B, D and F); 25 ␮m in (C, E and G); 50 ␮m in (H).

86

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89

Fig. 3. Formation of focal adhesions in patterned HUVEC on a stretch chamber. (A) Alexa 568-labeled fibronectin stamp. Cultured HUVEC stained for FAK (B), vinculin (C) or paxillin (D).

tact printing technology to prepare HUVECs that were uniform in size and shape. Fig. 2A illustrates the microcontact printing technology with a stamp made of PDMS used to form arrays of fibronectin dots on the stretch chambers. Non-polymerized PDMS was poured in a Silicon wafer mold, thermoset, and then removed as a stamp. The stamp was plasma processed using plasma cleaner/sterilizer PDC-32G (Harrick) to enhance the hydrophilic properties of the surface. Then fibronectin used as ink was added to the stamp at 37 ◦ C for 30 min. After 30 min, excess fibronetin was removed using an air duster (HOZAN) and it was stamped onto the stretch chamber to complete printing the stretch chamber with 10 × 10 fibronectin dots of 2 ␮m in 38 ␮m. To prevent the cells from adhering to other parts, the HUVECs were cultured on the printed chamber after being blocked with 1% BSA at 37 ◦ C for 30 min. Fig. 2B and C show SEM images of the mold, Fig. 2D and E show SEM images of the stamp made of PDMS, Fig. 2F and G show Alexia 568-labeled fibronectin printed onto a PDMS stretch chamber and Fig. 2H shows HUVEC cultured on the chamber.

2004). Vinculin is a cytoskeletal protein associated with the cytoplasmic face of both cell–cell and cell–extracellular matrix adherence-type junctions, where it is thought to function as one of several interacting proteins involved in anchoring F-actin and paxillin to the membrane (Weller et al., 1990; Turner, 2000). Fig. 3A shows that the fibronectin dots printed onto the surface beneath the HUVECs were of uniform size and fluorescent intensity, so the individual cells were assumed to have an equal potential to produce fibronectin-containing focal adhesions on

3.3. Localization of HUVEC cell-adhesion molecules on the fibronectin-stamped PDMS stretch chamber The focal adhesion molecules, FAK, vinculin and paxillin, were labeled in individual HUVEC cultured on the stretch chambers (Fig. 3). FAK, which is a non-receptor protein tyrosine kinase, and its substrate paxillin are two important components of the focal adhesion-signaling complex, and are implicated in several fundamental cellular functions including cell survival, motility and invasion (Hanks et al., 2003; Schlaepfer and Mitra,

Fig. 4. Microscopic images of patterned HUVEC cultured on a stretch chamber at rest (A and B) and after 20% cyclic-stretching (C and D). (A and C) Show Alexa 568-labeled fibronectin. (C and D) Are phase-contrast microscopic images of HUVEC. Scale bar: 25 ␮m.

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89

87

Fig. 5. Analysis of tyrosine phosphorylation and actin dynamics in HUVEC at rest (A–C), and after cyclic-stretching for 15 min (D–F), and 60 min (G–I). (A, D and G) Show anti-pY antibody labeling. (B, E and H) Show phalloidin labeling. (C, F and I) Show the merged images. J shows an Alexa 568-labeled fibronectin stamp on a stretch chamber. Block numbers were assigned to each row and column. Scale bar: 25 ␮m. (K, L and M) Are graphs showing fluorescent intensity resulting from anti-pY antibody labeling in each block, representing a focal adhesion.

88

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89

this surface. However, although FAK dots co-localized with fibronectin dots, they varied in fluorescent intensity, indicating that not all focal adhesions contained FAK (Fig. 3B). Strongly positive FAK signals were evident on at least 50% of the 100 fibronectin dots (n = 9). Under static culture conditions, with the cellular dimensions and number of cell scaffolds indicated, we speculated that the number of focal adhesions necessary to maintain cell viability is fixed. Fig. 3C shows that vinculin, which is a paxillin-associated molecule, also localized to the fibronectin dots, but showed variable signal intensities. Unlike FAK, the strongest vinculin signals were observed connecting each dot near the cell membrane. As vinculin also associates with F-actin, the two molecules co-localized near the cell membrane (shown later in Fig. 5B). Fig. 3D shows the localization of paxillin. Labeling for paxillin was both radial from the cell center to the cell membrane, similar to fibronectin, and along the cell membrane, similar to vinculin. 3.4. Change in F-actin organization in response to stretching During stretching, the chamber could be extended by up to 120% laterally, along only one axis, without any change along the perpendicular axis (Fig. 4). Fig. 5 shows the changes in Factin polymerization and tyrosine phosphorylation in HUVEC when cells were subjected to a 20% uni-axial strain at 1 Hz. At rest, F-actin labeling appeared along the circumference of the cells (Fig. 5B). After 15 min of cyclic-stretching, with each scaffold as a base point, F-actin polymerization was active, producing multiple short lengths of F-actin (Fig. 5E). After 60 min, strong labeling of F-actin fibers perpendicular to the stretch axis was visible (Fig. 5H). 3.5. Stretch-induced tyrosine phosphorylation in HUVECs At focal adhesions, formed where cells contact the extracellular matrix, FAK and paxillin form a complex with c-Src tyrosine kinase and participate in integrin-mediated signal transduction as the Src substrate (Burridge and Chrzanowska-Wodnicka, 1996). In endothelial cells, platelet endothelial cell-adhesion molecule-1 (PECAM-1), which is a 130-kDa type I transmembrane glycoprotein, is also localized to focal adhesions (Harada et al., 1995). The tyrosine phosphorylation of PECAM-1 can be induced by mechanical force, applied directly to PECAM-1, or via fibronectin, or by fluid shear stress or osmotic shock (Osawa et al., 1997, 2002; Newman and Newman, 2003). Recently, the Src family substrate p130Cas was shown to act as a primary force sensor, transducing force into mechanical extension (Sawada et al., 2006). In this study, we investigated the position of focal adhesions which were highly responsive to stretch stimulation in individual HUVECs, by measuring the intensity of labeling with anti-pY antibody before and after stretching. Since cells on fibronectin arrays had no cell–cell contacts and received no mechanical stimuli from neighboring cells, it was possible to examine the

effect of mechanical stress caused purely by uni-axial stretching applied to focal adhesions. At rest, anti-pY antibody labeling was observed along the edge of the cells (Fig. 5A). In the absence of an external mechanical force, the focal adhesions are believed to be strongest at the edge of cells. After cyclic-stretching for 15 min, tyrosine phosphorylation occurred at intracellular focal contacts throughout the cells (Fig. 5D). As these signals were restricted to focal contacts and not seen elsewhere in the cell, this suggested that phosphorylation in response to stretching involved the molecules localized in focal adhesions, FAK, PECAM-1, p130Cas and paxillin. After stretching for 60 min, sites of tyrosine phosphorylation were labeled on the edge of the cells (Fig. 5G) and also at the anchor points of actin polymerization (Fig. 5I, white circles). To quantify tyrosine phosphorylation using intracellular labeling with anti-pY antibody, we used ROIs equivalent to the dimensions of each fibronectin dot (Fig. 5J) and measured the fluorescent intensity for each ROI using a confocal laserscanning microscope. Fig. 5K, L and M show the average values obtained from several cells (n = 3) at each ROI. We found the stretch-induced tyrosine phosphorylation of adhesion molecules to be rapid and transitory at intracellular focal contacts. Tyrosine phosphorylation at focal contacts therefore appears to be central to the signal transduction pathways and changes in actin organization in HUVEC that are induced by stretching. In conclusion, we have used microcontact printing technology, for the first time, to prepare many individual cells of the same form and size, and with the same number and pattern of focal adhesions. We were therefore able to apply the same stretch stimuli to focal adhesions at the same coordinates in multiple cells. By quantifying changes in tyrosine phosphorylation in response to cyclic-stretching at each coordinate, we were able to analyze the intracellular pattern of tyrosine phosphorylation statistically and show this method achieved a high signal to noise (S/N) ratio. This technique provides a useful tool for statistically analyzing cellular response to vectorial stimuli, such as uni-axial stretching. Acknowledgments We thank K. Takamura and S. Oka for technical assistance. This work was supported by a grant-in-aid for Scientific Research on Priority Areas “System Cell Engineering by MultiScale Manipulation” (no. 17076006 to K.N.) from the Ministry of Education Science Sports and Culture, Japan. References Alenghat, F.J., Ingber, D.E., 2002. Mechanotransduction: all signals point to cytoskeleton, matrix, and integrins. Sci. STKE 2002, PE6. Burridge, K., Chrzanowska-Wodnicka, M., 1996. Focal adhesions, contractility, and signaling. Annu. Rev. Cell. Dev. Biol. 12, 463–518. Chen, C.S., Marksich, M., Huang, S., Whitesides, G.M., Ingber, D.E., 1998. Micropatterned surfaces for control of cell shape, position, and function. Biotechnol. Prog. 14, 356–363. Davies, P.F., Tripathi, S.C., 1993. Mechanical stress mechanisms and the cell An endothelial paradigm. Circ. Res. 72, 239–245.

Y. Katanosaka et al. / Journal of Biotechnology 133 (2008) 82–89 Dike, L.E., Chen, C.S., Marksich, M., Tien, J., Whitesides, G.M., Ingber, D.E., 1999. Geometric control of switching between growth, apoptosis, and differentiation during angiogenesis using micropatterned substrates. In vitro Cell Dev. Biol. Anim. 35, 441–448. Hanks, S.K., Ryzhova, L., Shin, N.Y., Brabek, J., 2003. Focal adhesion kinase signaling activities and their implications in the control of cell survival and motility. Front. Biosci. 8, d982–d996. Harada, N., Matsuda, N., Fujiwara, K., 1995. Fluid flow and osmotic stress induce tyrosine phosphorylation of an endothelial cell 128 kDa surface glycoprotein. Biochem. Biophys. Res. Commun. 214, 69– 74. Hudlicka, O., Brown, M.D., 1993. Physical forces and angiogenesis. In: Rubanyi, G.M. (Ed.), Mechanoreception by the Vascular Wall. Futura Publishing Co., Mount Kisco, NY, pp. 197–241. Ingber, D.E., 2003. Tensegrity II. How structural networks influence cellular information processing networks. J. Cell. Sci. 116, 1397–1408. Ito, Y., 1999. Surface micropatterning to regulate cell functions. Biomaterials 20, 2333–2342. Kane, R.S., Takayama, S., Ostuni, E., Ingber, D.E., Whitesides, G.M., 1999. Patterning proteins and cells using soft lithography. Biomaterials 20, 2363–2376. Naruse, K., Yamada, T., Sai, X.R., Hamaguchi, M., Sokabe, M., 1998. Involvement of SA channels in orienting response of cultured endothelial cells to cyclic stretch. Am. J. Physiol. 274, H1532–H1538. Newman, P.J., Newman, D.K., 2003. Signal transduction pathways mediated by PECAM-1: new roles for an old molecule in platelet and vascular cell biology. Arterioscler. Thromb. Vasc. Biol. 23, 953–964. Osawa, M., Masuda, M., Harada, N., Lopes, R.B., Fujiwara, K., 1997. Tyrosine phosphorylation of platelet endothelial cell-adhesion molecule-1 (PECAM1, CD31) in mechanically stimulated vascular endothelial cells. Eur. J. Cell. Biol. 72, 229–237. Osawa, M., Masuda, M., Kusano, K., Fujiwara, K., 2002. Evidence for a role of platelet endothelial cell-adhesion molecule-1 in endothelial cell mechanosignal transduction: is it a mechanoresponsive molecule? J. Cell Biol. 158, 773–785.

89

Pelham, R.J., Chang, F., 2001. Role of actin polymerization and actin cables in actin-patch movement in Schizosaccharomyces pombe. Nat. Cell Biol. 3, 235–244. Sai, X., Naruse, K., Sokabe, M., 1999. Activation of pp60(src) is critical for stretch-induced orienting response in fibroblasts. J. Cell. Sci. 112, 1365–1373. Sawada, Y., Tamada, M., Dubin-Thaler, B.J., Cherniavskaya, O., Sakai, R., Tanaka, S., Sheetz, M.P., 2006. Force sensing by mechanical extension of the Src family kinase substrate p130Cas . Cell 127, 1015–1026. Schlaepfer, D.D., Mitra, S.K., 2004. Multiple connections link FAK to cell motility and invasion. Curr. Opin. Genet. Dev. 14, 92–101. Shirinsky, V.P., Antonov, A.S., Birukov, K.G., Sobolevsky, A.V., Romanov, Y.A., Kabaeva, N.V., Antonova, G.N., Smirnov, V.N., 1989. Mechano-chemical control of human endothelium orientation and size. J. Cell. Biol. 109, 331–339. Singhvi, R., Kumar, A., Lopez, G.P., Stephanopoulos, G.N., Wang, D.I., Whitesides, G.M., Ingber, D.E., 1994. Engineering cell shape and function. Science 264, 696–698. Skalak, T.C., Price, R.J., 1996. The role of mechanical stresses in microvascular remodeling. Microcirculation 3, 143–165. Suzuki, M., Naruse, K., Asano, Y., Okamoto, T., Nishikimi, N., Sakurai, T., Nimura, Y., Sokabe, M., 1997. Up-regulation of integrin beta3 expression by cyclic stretch in human umbilical endothelial cells. Biochem. Biophys. Res. Commun. 239, 372–376. Turner, C.E., 2000. Paxillin interactions. J. Cell. Sci. 23, 4139–4140. Wang, J.G., Miyazu, M., Matsushita, E., Sokabe, M., Naruse, K., 2001. Uniaxial cyclic stretch induces focal adhesion kinase (FAK) tyrosine phosphorylation followed by mitogen-activated protein kinase (MAPK) activation. Biochem. Biophys. Res. Comm. 288, 356–361. Weller, P.A., Ogryzko, E.P., Corbern, E.B., Zhidkova, N.I., Patel, B., Price, G.J., Spurr, N.K., Koteliansky, V.E., Critchley, D.R., 1990. Complete sequence of human vinculin and assignment of the gene to chromosome 10. Proc. Nat. Acad. Sci. 87, 5667–5671. Xia, Y.N., Whitesides, G.M., 1998. Soft lithography. Annu. Rev. Mater. Sci. 28, 153–184.