- Email: [email protected]
Cardiac valve interstitial cells secrete fibronectin and form fibrillar adhesions in response to injury
Cardiovascular Pathology 16 (2007) 203 – 211
Original Article
Cardiac valve interstitial cells secrete fibronectin and form fibrillar adhesions in response to injury Cristina Fayeta, Michelle P. Bendeckb, Avrum I. Gotlieba,4 b
a Toronto General Research Institute and Department of Pathology, University Health Network, Toronto, Ontario, Canada M5G 2C4 Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Banting Institute, Toronto, Ontario, Canada M5G 1L5
Received 3 August 2006; received in revised form 21 December 2006; accepted 21 February 2007
Abstract Background: Fibronectin, an extracellular matrix protein, is associated with the general process of tissue repair and is present in heart valves. In order to understand the cellular mechanisms of heart valve repair, we hypothesized that fibronectin is produced and secreted by valvular interstitial cells (VICs), and when up-regulated in VICs involved in active repair, it is associated with prominent fibrillar adhesions composed of tensin and a5h1 integrin. We investigated the interaction of porcine mitral VICs with the underlying fibronectin matrix and the formation and localization of focal and fibrillar adhesion complexes in an in vitro wound model. Methods: Confluent monolayers of VICs were wounded with a 1-mm-wide cell scraper, maintained in standard media and 10% fetal bovine serum, and fixed at various time points after wounding. Immunohistochemistry was used to localize fibronectin, paxillin, tensin, and a5h1 integrin. F-actin was localized with an Alexa-Fluor-568-labeled phalloidin. Cells were examined with a scanning confocal laser microscope. Results: In response to in vitro mechanical wounding, migrating VICs at the wound edge expressed cytoplasmic fibronectin compared to nonwounded confluent monolayers. Over 24 to 48 h, fibrils were deposited into the subcellular space. Coincident with this, staining for a5h1 appeared, and tensin redistributed from focal adhesions to fibrillar adhesions, which colocalized with a5h1. Conclusions: Fibronectin in association with fibrillar adhesions is a component of the matrix that may be secreted by migrating VICs to regulate repair at sites of valve injury. D 2007 Elsevier Inc. All rights reserved. Keywords: Fibronectin; Valve interstitial cells; Fibrillar adhesions; Focal adhesions; Valve repair; Vinculin; Actin cytoskeleton
1. Introduction Valvular interstitial cells (VICs) maintain the integrity and stability of normal valves and regulate repair processes during disease and following valve injury [1–6]. VIC-
Cristina Fayet was a recipient of an Ontario Graduate Scholarship. Michelle P. Bendeck is a Career Investigator of the Heart and Stroke Foundation of Ontario. The studies were supported in part by research funds from the Department of Laboratory Medicine and Pathobiology, University of Toronto (A.I.G.). 4 Corresponding author. Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Banting Institute, Toronto, Ontario, Canada M5G 1L5. Tel.: +1 416 978 2557; fax: +1 416 978 7361. E-mail address: [email protected] (A.I. Gotlieb). 1054-8807/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.carpath.2007.02.008
regulated repair involves the synthesis and remodeling of the valve extracellular matrix [1–4,6–8]. For this reason, many valvular diseases are associated with increased cellularity and changes in matrix composition [9]. Since VICs contract and interact with the surrounding extracellular matrix, dysregulation of matrix metabolism leads to valve dysfunction [4]. The extracellular matrix protein fibronectin is present in many adult tissues and is particularly abundant in the extracellular matrix of injured and regenerating tissues [10–16]. Fibronectin is a multidomain glycoprotein that plays a role in the wound-healing process by providing a suitable matrix for cells to migrate on and by acting as a chemoattractant that induces cell migration toward the site of injury [15,17]. It also stimulates fibroblasts in the healing wound to become myofibroblasts, which are
204
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
important for wound contraction and further fibronectin matrix assembly [14]. Mammalian cells make contact with fibronectin via focal and fibrillar adhesions. These two distinct classes of cell–matrix adhesion complexes differ in morphology, location on the cell membrane, and protein composition. Fibrillar adhesions are small globular or elongated adhesion complexes that consist primarily of the a5h1 integrin and tensin, while lacking other plaque cytoplasmic proteins, such as paxillin and vinculin, which are characteristic of the large, rod-shaped focal adhesions [18,19]. Focal adhesions are composed of focal adhesion kinase; cytoskeletal molecules vinculin, paxillin, talin, and a-actinin; and avh3 and a5h1 integrins [19–22]. The ability of the fibronectin molecule to stretch provides cells with a dynamic and pliable extracellular matrix environment able to accommodate cell activities, such as migration, lamellipodial protrusion, and fibrillar adhesion formation [19,22–24].
The role of fibronectin in heart valve repair has not been studied. Our objective was to understand how VICs interact with fibronectin during wound repair. We studied the formation and localization of focal and fibrillar adhesion complexes that link the cytoskeleton to the extracellular matrix protein fibronectin using an in vitro wound model of a mechanically injured confluent monolayer of porcine mitral VICs [25,26]. The wound edges that are created allow for the study of VIC response to injury, especially cell migration and matrix formation. For comparison, we also studied both low-density VIC cultures in which VICs are migrating and proliferating randomly to form a monolayer and quiescent confluent cultures. The findings support our hypothesis that as VICs undergo a transition from quiescent to activated cells at the wound edge, fibronectin is formed and secreted by VICs at the wound edge and is associated with prominent fibrillar adhesions composed of tensin and a5h1 integrin.
Fig. 1. Micrographs of confocal projections of subconfluent (A and B) and confluent (C and D) cultures of porcine VICs immunostained to detect fibronectin (Fn). F-actin stress fibers were stained with Alexa-568-labeled phalloidin. (A) Cytoplasmic Fn was present around the nuclei of the cells, and Fn fibrils were deposited immediately surrounding the cells and extending between adjacent cells. (B) Fn fibrils were often colocalized along F-actin stress fibers (wide arrow) and at their tips in lamellipodial protrusions (thin arrows). (C) Fn fibrils were present in the extracellular space in the subcellular area and between cells, but no cytoplasmic Fn was detected. (D) Fn fibrils were present in areas surrounding the cells and aligned parallel to the F-actin stress fibers. N=nucleus (green=Fn, red=F-actin stress fibers, yellow=colocalization). Magnification, 600.
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
2. Materials and methods 2.1. Cell culture and in vitro wound model Porcine heart valves were acquired at an abattoir, and the distal third of the anterior mitral valve leaflet was excised for explant culture as previously described [27]. Briefly, endothelial cells were removed using a sterile scalpel blade to scrape the surfaces of the valve. Pieces measuring approximately 45 mm were cut from the tissue, placed in 35-mm tissue culture dishes, and grown in Medium 199 supplemented with 10% fetal bovine serum, 2% penicillin/ streptomycin, and 1% Fungizone in a humidified, 5% CO2/ 95% air incubator at 378C. VICs that grew out of the explants were subcultured for three to four passages. For experiments, VICs were grown on 2222 mm glass coverslips, reaching confluency in approximately 10 days. A sterile P1000 pipette tip was used to make a linear wound across the confluent monolayers that measured 1 mm in width. Cells were subsequently fixed with 4% paraformaldehyde at the time of wounding (0 h) and at 4, 24, or 48 h after wounding. In another set of wounded monolayers, VICs were allowed to nearly close the wound before fixation, which took approximately 4 days. In addition, subconfluent cultures grown for 2–3 days and confluent cultures grown for 10 days were also fixed in their nonwounded states. Experiments were conducted at least in triplicate. 2.2. Immunocytochemistry After fixation, cells were permeabilized for 3.5 min with 0.1% Triton X-100 to allow for intracellular staining. Cells were then washed in PBS (35 min) and incubated at room temperature with one of the following primary antibodies: mouse antichicken tensin (Transduction Laboratories, clone 5, 5 Ag/ml PBS), mouse antichicken paxillin (Transduction Laboratories, clone 349, 1.25 Ag/ml), sheep antihuman
205
fibronectin (Serotec, 1:100), mouse antichicken vinculin (Sigma, clone VIN-11-5, 1:100), mouse antihuman avh3 integrin (Chemicon, clone LM609, 20 Ag/ml), or mouse antihuman a5h1 integrin (Chemicon, clone JBS5, stock concentration range 20–50 Ag/ml ) for 1 h. After three 5-min washes with PBS, coverslips were incubated for 30 min with a species-appropriate secondary antibody, either an Alexa-Fluor-488-conjugated goat antimouse immunoglobulin G (IgG; Molecular Probes, 1:100, excitation/emission=495/519 nm) or an FITC-conjugated AffiniPure donkey antisheep IgG (Jackson ImmunoResearch Labs, 1:100, excitation/emission=492/520 nm). F-actin was localized in double-stained coverslips by incubation with AlexaFluor-568-labeled phalloidin (Molecular Probes, 1:40, excitation/emission=578/600 nm) concomitantly with either the Alexa-Fluor-488- or the FITC-conjugated secondary antibody. Colocalization was identified by the superimposition of red and green to give a yellow color visualized under the confocal microscope. For negative controls, omission of the primary antibody or incubation with a nonspecific mouse IgG (Upstate, 1:200) was used. All controls were negative. The coverslips were washed with PBS (35 min) and mounted on glass slides with Prolong Antifade. Cells were examined with a BioRad MRC 1024ES confocal scanning laser microscope fitted with an argon–krypton mixed-gas laser for simultaneous visualization of multiple fluorescent probes. Images were captured under the 488-nm wavelength excitation line to visualize FITC or Alexa Fluor 488 and the 568-nm wavelength excitation line to visualize Alexa Fluor 568. Fluorescence was acquired at wavelengths N515 nm. A stack of optical sections was taken at 0.4-Am intervals in the Z direction with three Kalman averages, using a 60 oil-immersion objective lens. Projections of the series of optical sections were reconstructed into the same focal plane with LaserSharp software (BioRad, Version 3.2). Three investigators provided qualitative assessments of fibronectin, tensin, a5h1
Fig. 2. Micrographs of confocal projections of porcine VICs at the wound edge immunostained to detect fibronectin (Fn) 4 h (A), 24 h (B), and 48 h (C) after wounding. (A) Cells at the wound edge started migrating into the wound and synthesizing cytoplasmic Fn (arrows) 4 h after wounding. Twenty-four hours (B) and 48 h (C) after wounding, Fn cytoplasmic staining and fibril deposition by the migrating cells had increased. W=wound (green=Fn). Magnification, 600.
206
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
Fig. 3. Micrographs of confocal projections of cultures of porcine VICs immunostained to detect tensin in subconfluent (A and B) and confluent (C and D) monolayers. F-actin stress fibers were stained with Alexa-568-labeled phalloidin. In subconfluent cultures, tensin stained in a long, thin fibrillar pattern running lengthwise throughout the cell center (A) and colocalized along F-actin stress fibers (B; arrow). In the confluent monolayer, tensin was localized both in small weakly stained fibrillar adhesions distributed throughout the cell (C) and in large, rod-shaped focal adhesions at the cell periphery (D; green=tensin, red=F-actin stress fibers, yellow=colocalization). Magnification, 600.
integrin, vinculin, and paxillin staining under all control and experimental conditions. 3. Results 3.1. Fibronectin In subconfluent cultures grown for 2–3 days, cytoplasmic staining for fibronectin was present, especially in areas surrounding the nuclei of the cells. Fibronectin fibrils were also present immediately around the cells and extending between adjacent cells (Fig. 1A). Fibronectin fibrils were deposited in parallel to the F-actin stress fibers, often at the tips of lamellipodial protrusions (Fig. 1B). The cultures became confluent at 7 days, after which cytoplasmic fibronectin was no longer detected in the cells of the monolayer. However, fibronectin fibrils were present in the subcellular area immediately surrounding the cell periphery and between the cells. Furthermore, the intensity of fibronectin staining increased compared with that of subconfluent cultures (Fig. 1C and D).
When the confluent monolayers were wounded, elongated VICs were observed migrating outward from the wound edge as previously described [28]. At the wound edge, there was a loss of fibronectin, but VICs migrating into the wound showed cytoplasmic staining for fibronectin 4 h after wounding (arrow, Fig. 2A). Twenty-four and 48 h after wounding, the migrating VICs showed intense staining of cytoplasmic fibronectin, and fibronectin fibrils were deposited into the matrix (Fig. 2B and C). Forty-eight hours after wounding, cytoplasmic fibronectin stained predominantly the leading half of the cell in front of the nucleus (not shown). Wound closure was nearly complete 4 days after wounding, with only a few regions of the wound still open. Immunostaining for fibronectin increased in the extracellular space of the wound at 4 days when compared to wounds 48 h after wounding (data not shown). The density of the fibronectin fibrils in the closed wound approached that of the confluent monolayer away from the wound area. The fibronectin fibrils appeared longer and thicker than the ones that were first deposited into the extracellular matrix but not yet fully comparable to those of either the
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
207
Fig. 4. Micrographs of confocal projections of porcine VICs at the wound edge immunostained to detect (A) tensin, (B) vinculin, and (C) paxillin. F-actin stress fibers were stained with Alexa-568-labeled phalloidin. Tensin staining increased progressively from (A1) 0 h to (A2) 24 h and (A3) 48 h after wounding. (A3) At 48 h, tensin stained strongly in fibrillar adhesion sites; small, globular staining throughout the cell. (B and C) Paxillin and vinculin showed a constant intensity and pattern of staining at all time points; large, rod-shaped focal adhesions colocalized with the tips of the F-actin stress fibers at the cell periphery. N=nucleus (green=tensin, vinculin, and paxillin; red=F-actin stress fibers; yellow=colocalization). Magnification, 600.
unperturbed confluent monolayer or the monolayer away from the wound area. 3.2. Tensin, paxillin, and vinculin Subconfluent cultures of VICs stained weakly for tensin in an elongated thin fibrillar pattern (Fig. 3A and B). In the confluent monolayer, the staining pattern of tensin was more globular in shape than in the subconfluent cultures (Fig. 3C). In addition, some cells stained for tensin in large, rodshaped contacts located at the cell periphery (Fig. 3D),
which was similar to the staining pattern of the focal adhesion proteins vinculin and paxillin (Fig. 4B and C). After wounding, tensin staining increased progressively in intensity at 24 and 48 h (Fig. 4A1–A3). At 48 h, VICs now showed tensin localized to fibrillar adhesions, characterized by small globular plaques localized throughout the cell (Fig. 4A3). In contrast, staining for the focal adhesion proteins paxillin and vinculin was positive at the time of wounding (0 h; Fig. 4B1 and C1) and remained constant in intensity and pattern 24 and 48 h after wounding (Fig. 4B2 and B3 for paxillin; Fig. 4C2 and C3 for vinculin). These
208
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
Fig. 5. Micrographs of confocal projections of porcine VICs immunostained to detect a5h1 integrin. F-actin stress fibers were stained with Alexa-568-labeled phalloidin. (A) In subconfluent cultures, a5h1 stained strongly in a diffuse punctate pattern in the cytoplasm throughout the cell center and at the cell periphery in large focal adhesions (arrows). (B) Colocalization of the a5h1 integrin to the tips of F-actin stress fibers provided further evidence that a5h1 is present in focal adhesions (arrow). (C) In confluent cultures, a5h1 integrin staining was very faint in the cytoplasm and absent at focal adhesion sites. N=nucleus (green=a5h1 integrin, red=F-actin stress fibers, yellow=colocalization). Magnification, 600.
focal adhesion proteins appeared as large, rod-shaped focal adhesions localized to the tips of F-actin stress fibers along the cell periphery. Once VICs reestablished a confluent monolayer, tensin staining in the closed wounds showed the elongated thin fibrillar pattern running lengthwise along the cell and outlining the cell periphery (not shown). Cells of the monolayer away from the closed wound region also displayed tensin in long fibrillar adhesions throughout the cell center. In addition to this central staining, tensin was localized in rod-shaped structures at the cell periphery, colocalizing with the tips of the F-actin stress fibers. This staining pattern was similar to that observed for the focal adhesion protein paxillin, suggesting that tensin also localized in focal adhesions. 3.3. a5b1 integrin Subconfluent cultures of VICs stained strongly for the fibronectin receptor, the a5b1 integrin. The anti-a5b1 anti-
body stained large, rod-shaped contacts at the cell periphery and in a diffuse punctate pattern throughout the cell center (Fig. 5A). The former pattern likely represents focal adhesions, while the latter likely includes fibrillar adhesions. Cells double stained for a5b1 and F-actin showed that this integrin colocalized with the tips of the F-actin stress fibers (Fig. 5B), as seen with the focal adhesion proteins paxillin and vinculin. Once cultures became confluent, immunostaining for a5b1 integrin in the monolayer decreased dramatically and was barely detectable (Fig. 5C). a5b1 did not localize to the tips of the F-actin stress fibers. These results suggest that a5b1 integrins are not prominent in confluent cultures. There was no staining for a5b1 integrin at the time of wounding (0 h; Fig. 6A). Staining was detected 24 h after wounding (Fig. 6B) and remained constant at 48 h (Fig. 6C). The a5b1 integrin stained in two different patterns. The first was localized to the lamellipodia in migrating cells with some colocalization with the tips of the F-actin stress fibers,
Fig. 6. Micrographs of confocal projections of wounded monolayers double stained for a5h1 integrin and F-actin stress fibers, stained with Alexa-568-labeled phalloidin. (A) VICs of the confluent monolayer did not stain for a5h1 at 0 h. Twenty-four hours (B) and 48 h (C) after wounding, a5h1 stained both in a focal adhesion pattern at the lamellipodia of migrating cells, with colocalization at the tips of the F-actin stress fibers (arrows), and in a punctate pattern throughout the cell, consistent with fibrillar adhesions (green=a5h1 integrin, red=F-actin stress fibers, yellow=colocalization). W=wound. Magnification, 600.
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
consistent with localization at focal adhesions (arrows, Fig. 6B and C). The second stained in a punctate pattern throughout the center of the cell, consistent with localization at fibrillar adhesions. The staining of the a5b1 integrin showed a similar increase in intensity to that of tensin over time in VICs at the wound edge. In the closed wounds, a5b1 integrin stained faintly in a punctate pattern throughout the VIC cytoplasm (not shown). The same staining pattern was observed in the monolayer away from the wound area (not shown). Some regions of the wound had not yet reclosed in 4 days. Interestingly, the cells that were still migrating into these open regions of the wound stained for the a5b1 integrin both in a punctate pattern throughout the cell and in a focal adhesion pattern that localized to the cell periphery. In addition, the a5b1 staining intensity in the cytoplasm of the cells inside open wound regions was brighter than that of the cells both in the closed wounds and in the monolayer away from the wound area. This staining pattern of a5b1 was like that of the cells in the wounded monolayers and in subconfluent cultures.
4. Discussion Fibronectin constitutes part of the valve matrix, which is likely crucial for maintaining valvular structure and function. This is the first experimental study to report on interactions that take place between VICs and fibronectin in response to injury. To examine the dynamics of fibronectin fibril formation, we studied three conditions: subconfluent, confluent, and wounded cultures, which provide the opportunity to study populations of VICs in transition between quiescent and activated states. Cells in subconfluent cultures are initially migrating and proliferating to form the confluent monolayer. Cells in confluent cultures are quiescent and maintain strong attachments to the extracellular matrix via localized sites of firm anchorage called focal adhesions [29]. We showed that, under these three conditions, there are differences in expression and localization of fibronectin and its associated fibrillar adhesion proteins, tensin and a5b1 integrins. We found that the production of a fibronectin extracellular matrix plays a role in VIC repair after wounding. During the initiation of repair as VICs undergo the transition from quiescent to activated cells, fibronectin first appeared in the cell cytoplasm and then fibrils were deposited into the extracellular space to provide a matrix that supports migration. Deposition of fibronectin into the extracellular matrix may facilitate the migration of VICs to the site of injury and enable them to close the wound more efficiently, as has been shown in models of cardiac repair, cardiac cushion cell migration, and epithelial–mesenchymal cell transformation during cardiogenesis [15,30–34]. In fact, cytoplasmic fibronectin, tensin, and the a5b1 integrin were also expressed by VICs in subconfluent cultures, and new
209
fibronectin fibrils were deposited by migrating VICs in association with tensin and a5b1 in fibrillar adhesions. In contrast, cytoplasmic fibronectin disappeared in the quiescent VICs of confluent monolayers, suggesting that fibronectin synthesis was reduced once the cells reached confluence. A decrease in fibronectin protein synthesis has been reported to be associated with a decrease in fibrillar adhesion formation [23,35,36]. This is consistent with our observation that both tensin and a5b1 stained more weakly in the confluent VIC cultures. However, in some cells of the confluent monolayer, tensin staining was found in focal adhesions at the cell periphery, possibly representing tensin that did not translocate from focal to fibrillar adhesion sites. This observation is consistent with the idea that when fibronectin fibrillogenesis is not occurring, as we speculate is happening in the confluent VIC monolayer, fibrillar adhesions disassemble and focal adhesions become the predominant type of cell–matrix attachment. Studies have shown that once cells secrete fibronectin fibrils, they bind to these fibrils via the a5b1 integrin receptor, found within focal adhesions located at the cell periphery, at the tips of F-actin stress fibers [36–38]. Binding initiates a signaling cascade that results in the active transport of a5b1 and tensin along the F-actin stress fiber away from the cell periphery, leaving behind avb3 and paxillin in the focal adhesions. Our observations in VICs showed that fibronectin fibrils often colocalized with the F-actin stress fibers at the tips of lamellipodial protrusions, supporting the idea that fibronectin fibrillogenesis involves the translocation of fibronectin fibrils along the F-actin cytoskeleton. This process results in the formation of tensinrich fibrillar adhesions at the cell center, consisting of the a5b1 integrin in association with extracellular fibronectin fibrils, while lacking the other focal adhesion plaque components [18,19]. It is believed that translocation of this fibronectin–a5b1–tensin complex toward the cell center facilitates additional fibronectin fibril formation by stretching folded fibrils away from the cell periphery and exposing cryptic self-assembly sites that promote the self-association of fibronectin molecules [17,21,35,39]. This is consistent with our observations that porcine VICs at the wound edge deposited fibronectin fibrils into the extracellular matrix, displayed increased a5b1 integrin immunostaining, and formed tensin-rich fibrillar adhesions throughout the cell. In summary, our results suggest that fibronectin fibrillogenesis and the resultant formation of tensin-rich fibrillar adhesions play a role in cell migration in subconfluent and wounded cultures while focal adhesions remain unaffected. A potential mechanism for the closure of an in vitro wound would first involve the production of fibronectin by cells at the wound edge. Once a more pliable fibronectin matrix is deposited, cells are able to migrate more efficiently into the wound and reestablish the monolayer. The increased fibronectin production activates the processes involved in fibronectin fibrillogenesis, resulting in the increased expression of the a5b1 integrin fibronectin receptor. As the
210
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
migrating cells attach to fibronectin via its receptor, tensin gets translocated with this fibronectin–a5b1 complex along F-actin stress fibers toward the cell center and forms tensinrich fibrillar adhesions. These dynamic fibrillar adhesions differ in both composition and function from the more stable focal adhesions. Fibronectin-mediated VIC migration and the formation of fibrillar adhesions have not been previously reported as potential response to valve injury.
5. Summary This is the first experimental study to report on interactions that take place between VICs and the fibronectin matrix in response to injury. Our findings support the hypothesis that fibronectin is formed and secreted by VICs at the wound edge and is associated with the formation of prominent fibrillar adhesions composed of tensin and a5b1 integrin. References [1] Mulholland DL, Gotlieb AI. Cell biology of valvular interstitial cells. Can J Cardiol 1996;12:231 – 6. [2] Mulholland D, Gotlieb AI. Cardiac valve interstitial cells: regulator of valve structure and function. Cardiovasc Pathol 1997;6:167 – 74. [3] Filip DA, Radu A, Simionescu M. Interstitial cells of the heart valves possess characteristics similar to smooth muscle cells. Circ Res 1986; 59:310 – 20. [4] Rabkin E, Aikawa M, Stone JR, Fukumoto Y, Libby P, Schoen FJ. Activated interstitial myofibroblasts express catabolic enzymes and mediate matrix remodeling in myxomatous heart valves. Circulation 2001;104:2525 – 32. [5] Taylor PM, Allen SP, Yacoub MH. Phenotypic and functional characterization of interstitial cells from human heart valves, pericardium and skin. J Heart Valve Dis 2000;9:150 – 8. [6] Messier RH, Bass BL, Aly HM, Jones JL, Domkowski PW, Wallace RB, Hopkins RA. Dual structural and functional phenotypes of the porcine aortic valve interstitial population: characteristics of the leaflet myofibroblast. J Surg Res 1994;57:1 – 21. [7] Roy A, Brand NJ, Yacoub MH. Molecular characterization of interstitial cells isolated from human heart valves. J Heart Valve Dis 2000;9:459 – 64. [8] Tamura K, Jones M, Yamada I, Ferrans VJ. Wound healing in the mitral valve. J Heart Valve Dis 2000;9:53 – 63. [9] Schoen FJ, Edwards WD. In: Silver MD, Gotlieb AI, Schoen FJ, editors. Valvular heart disease in cardiovascular pathology. Philadelphia7 Churchill Livingstone; 2001: p. 402 – 42. [10] Bauters C, Marotte F, Hamon M, Oliviero P, Farhadian F, Robert V, Samuel JL, Rappaport L. Accumulation of fetal fibronectin mRNAs after balloon denudation of rabbit arteries. Circulation 1995;92:904 – 11. [11] Colombi M, Zoppi N, De Petro G, Marchina E, Gardella R, Tavian D, Ferraboli S, Barlati S. Matrix assembly induction and cell migration and invasion inhibition by a 13-amino acid fibronectin peptide. J Biol Chem 2003;278:14346 – 55. [12] Farhadian F, Contard F, Corbier A, Barrieux A, Rappaport L, Samuel JL. Fibronectin expression during physiological and pathological cardiac growth. J Mol Cell Cardiol 1995;27:981 – 90. [13] Ffrench-Constant C, Van de WL, Dvorak HF, Hynes RO. Reappearance of an embryonic pattern of fibronectin splicing during wound healing in the adult rat. J Cell Biol 1989;109:903 – 14.
[14] Magnusson MK, Mosher DF. Fibronectin: structure, assembly, and cardiovascular implications. Arterioscler Thromb Vasc Biol 1998; 18:1363 – 70. [15] Ulrich MM, Janssen AM, Daemen MJ, Rappaport L, Samuel JL, Contard F, Smits JF, Cleutjens JP. Increased expression of fibronectin isoforms after myocardial infarction in rats. J Mol Cell Cardiol 1997;29:2533 – 43. [16] Sechler JL, Corbett SA, Wenk MB, Schwarzbauer JE. Modulation of cell–extracellular matrix interactions. Ann N Y Acad Sci 1998;857: 143 – 54. [17] Pankov R, Yamada KM. Fibronectin at a glance. J Cell Sci 2002; 115:3861 – 3. [18] Zamir E, Katz BZ, Aota S, Yamada KM, Geiger B, Kam Z. Molecular diversity of cell–matrix adhesions. J Cell Sci 1999; 112(Pt 11);1655 – 69. [19] Katz BZ, Zamir E, Bershadsky A, Kam Z, Yamada KM, Geiger B. Physical state of the extracellular matrix regulates the structure and molecular composition of cell–matrix adhesions. Mol Biol Cell 2000;11:1047 – 60. [20] Zamir E, Geiger B. Components of cell–matrix adhesions. J Cell Sci 2001;114:3577 – 9. [21] Geiger B, Bershadsky A, Pankov R, Yamada KM. Transmembrane crosstalk between the extracellular matrix–cytoskeleton crosstalk. Nat Rev Mol Cell Biol 2001;2:793 – 805. [22] Pelham RJ, Wang Y. Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc Natl Acad Sci U S A 1997; 94:13661 – 5. [23] Yamada KM, Pankov R, Cukierman E. Dimensions and dynamics in integrin function. Braz J Med Biol Res 2003;36:959 – 66. [24] Wierzbicka-Patynowski I, Schwarzbauer JE. The ins and outs of fibronectin matrix assembly. J Cell Sci 2003;116:3269 – 76. [25] Gotlieb AI, Rosenthal A, Kazemian P. Fibroblast growth factor 2 regulation of mitral valve interstitial cell repair in vitro. J Thorac Cardiovasc Surg 2002;124:591 – 7. [26] Durbin A, Nadir NA, Rosenthal A, Gotlieb AI. Nitric oxide promotes in vitro interstitial cell heart valve repair. Cardiovasc Pathol 2005; 14:12 – 8. [27] Lester W, Rosenthal A, Granton B, Gotlieb AI. Porcine mitral valve interstitial cells in culture. Lab Invest 1988;59:710 – 9. [28] Durbin AD, Gotlieb AI. Advances towards understanding heart valve response to injury. Cardiovasc Pathol 2002;11:69 – 77. [29] Sastry SK, Burridge K. Focal adhesions: a nexus for intracellular signaling and cytoskeletal dynamics. Exp Cell Res 2000;261:25 – 36. [30] Mjaatvedt CH, Lepera RC, Markwald RR. Myocardial specificity for initiating endothelial–mesenchymal cell transition in embryonic chick heart correlates with a particulate distribution of fibronectin. Dev Biol 1987;119:59 – 67. [31] Clark RA. Potential roles of fibronectin in cutaneous wound repair. Arch Dermatol 1988;124:201 – 6. [32] Clark RA, Wikner NE, Doherty DE, Norris DA. Cryptic chemotactic activity of fibronectin for human monocytes resides in the 120kDa fibroblastic cell-binding fragment. J Biol Chem 1988;263: 12115 – 23. [33] Ffrench-Constant C, Hynes RO. Patterns of fibronectin gene expression and splicing during cell migration in chicken embryos. Development 1988;104:369 – 82. [34] Icardo JM, Nakamura A, Fernandez-Teran MA, Manasek FJ. Effects of injecting fibronectin and antifibronectin antibodies on cushion mesenchyme formation in the chick An in vivo study. Anat Embryol (Berl) 1992;185:239 – 47. [35] Schwarzbauer JE, Sechler JL. Fibronectin fibrillogenesis: a paradigm for extracellular matrix assembly. Curr Opin Cell Biol 1999; 11:622 – 7. [36] Pankov R, Cukierman E, Katz BZ, Matsumoto K, Lin DC, Lin S, Hahn C, Yamada KM. Integrin dynamics and matrix assembly: tensindependent translocation of alpha(5)beta(1) integrins promotes early fibronectin fibrillogenesis. J Cell Biol 2000;148:1075 – 90.
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211 [37] Zamir E, Katz M, Posen Y, Erez N, Yamada KM, Katz BZ, Lin S, Lin A, Bershadsky A, Kam Z, Geiger B. Dynamics and segregation of cell– matrix adhesions in cultured fibroblasts. Nat Cell Biol 2000;2:191 – 6. [38] Ilic D, Kovacic B, Johkura K, Schlaepfer DD, Tomasevic N, Han Q, Kim JB, Howerton K, Baumbusch C, Ogiwara N, Streblow DN, Nelson JA, Dazin P, Shino Y, Sasaki K, Damsky CH. FAK promotes
211
organization of fibronectin matrix and fibrillar adhesions. J Cell Sci 2004;117:177 – 87. [39] Zhong C, Chrzanowska-Wodnicka M, Brown J, Shaub A, Belkin AM, Burridge K. Rho-mediated contractility exposes a cryptic site in fibronectin and induces fibronectin matrix assembly. J Cell Biol 1998;141:539 – 51.
Original Article
Cardiac valve interstitial cells secrete fibronectin and form fibrillar adhesions in response to injury Cristina Fayeta, Michelle P. Bendeckb, Avrum I. Gotlieba,4 b
a Toronto General Research Institute and Department of Pathology, University Health Network, Toronto, Ontario, Canada M5G 2C4 Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Banting Institute, Toronto, Ontario, Canada M5G 1L5
Received 3 August 2006; received in revised form 21 December 2006; accepted 21 February 2007
Abstract Background: Fibronectin, an extracellular matrix protein, is associated with the general process of tissue repair and is present in heart valves. In order to understand the cellular mechanisms of heart valve repair, we hypothesized that fibronectin is produced and secreted by valvular interstitial cells (VICs), and when up-regulated in VICs involved in active repair, it is associated with prominent fibrillar adhesions composed of tensin and a5h1 integrin. We investigated the interaction of porcine mitral VICs with the underlying fibronectin matrix and the formation and localization of focal and fibrillar adhesion complexes in an in vitro wound model. Methods: Confluent monolayers of VICs were wounded with a 1-mm-wide cell scraper, maintained in standard media and 10% fetal bovine serum, and fixed at various time points after wounding. Immunohistochemistry was used to localize fibronectin, paxillin, tensin, and a5h1 integrin. F-actin was localized with an Alexa-Fluor-568-labeled phalloidin. Cells were examined with a scanning confocal laser microscope. Results: In response to in vitro mechanical wounding, migrating VICs at the wound edge expressed cytoplasmic fibronectin compared to nonwounded confluent monolayers. Over 24 to 48 h, fibrils were deposited into the subcellular space. Coincident with this, staining for a5h1 appeared, and tensin redistributed from focal adhesions to fibrillar adhesions, which colocalized with a5h1. Conclusions: Fibronectin in association with fibrillar adhesions is a component of the matrix that may be secreted by migrating VICs to regulate repair at sites of valve injury. D 2007 Elsevier Inc. All rights reserved. Keywords: Fibronectin; Valve interstitial cells; Fibrillar adhesions; Focal adhesions; Valve repair; Vinculin; Actin cytoskeleton
1. Introduction Valvular interstitial cells (VICs) maintain the integrity and stability of normal valves and regulate repair processes during disease and following valve injury [1–6]. VIC-
Cristina Fayet was a recipient of an Ontario Graduate Scholarship. Michelle P. Bendeck is a Career Investigator of the Heart and Stroke Foundation of Ontario. The studies were supported in part by research funds from the Department of Laboratory Medicine and Pathobiology, University of Toronto (A.I.G.). 4 Corresponding author. Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Banting Institute, Toronto, Ontario, Canada M5G 1L5. Tel.: +1 416 978 2557; fax: +1 416 978 7361. E-mail address: [email protected] (A.I. Gotlieb). 1054-8807/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.carpath.2007.02.008
regulated repair involves the synthesis and remodeling of the valve extracellular matrix [1–4,6–8]. For this reason, many valvular diseases are associated with increased cellularity and changes in matrix composition [9]. Since VICs contract and interact with the surrounding extracellular matrix, dysregulation of matrix metabolism leads to valve dysfunction [4]. The extracellular matrix protein fibronectin is present in many adult tissues and is particularly abundant in the extracellular matrix of injured and regenerating tissues [10–16]. Fibronectin is a multidomain glycoprotein that plays a role in the wound-healing process by providing a suitable matrix for cells to migrate on and by acting as a chemoattractant that induces cell migration toward the site of injury [15,17]. It also stimulates fibroblasts in the healing wound to become myofibroblasts, which are
204
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
important for wound contraction and further fibronectin matrix assembly [14]. Mammalian cells make contact with fibronectin via focal and fibrillar adhesions. These two distinct classes of cell–matrix adhesion complexes differ in morphology, location on the cell membrane, and protein composition. Fibrillar adhesions are small globular or elongated adhesion complexes that consist primarily of the a5h1 integrin and tensin, while lacking other plaque cytoplasmic proteins, such as paxillin and vinculin, which are characteristic of the large, rod-shaped focal adhesions [18,19]. Focal adhesions are composed of focal adhesion kinase; cytoskeletal molecules vinculin, paxillin, talin, and a-actinin; and avh3 and a5h1 integrins [19–22]. The ability of the fibronectin molecule to stretch provides cells with a dynamic and pliable extracellular matrix environment able to accommodate cell activities, such as migration, lamellipodial protrusion, and fibrillar adhesion formation [19,22–24].
The role of fibronectin in heart valve repair has not been studied. Our objective was to understand how VICs interact with fibronectin during wound repair. We studied the formation and localization of focal and fibrillar adhesion complexes that link the cytoskeleton to the extracellular matrix protein fibronectin using an in vitro wound model of a mechanically injured confluent monolayer of porcine mitral VICs [25,26]. The wound edges that are created allow for the study of VIC response to injury, especially cell migration and matrix formation. For comparison, we also studied both low-density VIC cultures in which VICs are migrating and proliferating randomly to form a monolayer and quiescent confluent cultures. The findings support our hypothesis that as VICs undergo a transition from quiescent to activated cells at the wound edge, fibronectin is formed and secreted by VICs at the wound edge and is associated with prominent fibrillar adhesions composed of tensin and a5h1 integrin.
Fig. 1. Micrographs of confocal projections of subconfluent (A and B) and confluent (C and D) cultures of porcine VICs immunostained to detect fibronectin (Fn). F-actin stress fibers were stained with Alexa-568-labeled phalloidin. (A) Cytoplasmic Fn was present around the nuclei of the cells, and Fn fibrils were deposited immediately surrounding the cells and extending between adjacent cells. (B) Fn fibrils were often colocalized along F-actin stress fibers (wide arrow) and at their tips in lamellipodial protrusions (thin arrows). (C) Fn fibrils were present in the extracellular space in the subcellular area and between cells, but no cytoplasmic Fn was detected. (D) Fn fibrils were present in areas surrounding the cells and aligned parallel to the F-actin stress fibers. N=nucleus (green=Fn, red=F-actin stress fibers, yellow=colocalization). Magnification, 600.
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
2. Materials and methods 2.1. Cell culture and in vitro wound model Porcine heart valves were acquired at an abattoir, and the distal third of the anterior mitral valve leaflet was excised for explant culture as previously described [27]. Briefly, endothelial cells were removed using a sterile scalpel blade to scrape the surfaces of the valve. Pieces measuring approximately 45 mm were cut from the tissue, placed in 35-mm tissue culture dishes, and grown in Medium 199 supplemented with 10% fetal bovine serum, 2% penicillin/ streptomycin, and 1% Fungizone in a humidified, 5% CO2/ 95% air incubator at 378C. VICs that grew out of the explants were subcultured for three to four passages. For experiments, VICs were grown on 2222 mm glass coverslips, reaching confluency in approximately 10 days. A sterile P1000 pipette tip was used to make a linear wound across the confluent monolayers that measured 1 mm in width. Cells were subsequently fixed with 4% paraformaldehyde at the time of wounding (0 h) and at 4, 24, or 48 h after wounding. In another set of wounded monolayers, VICs were allowed to nearly close the wound before fixation, which took approximately 4 days. In addition, subconfluent cultures grown for 2–3 days and confluent cultures grown for 10 days were also fixed in their nonwounded states. Experiments were conducted at least in triplicate. 2.2. Immunocytochemistry After fixation, cells were permeabilized for 3.5 min with 0.1% Triton X-100 to allow for intracellular staining. Cells were then washed in PBS (35 min) and incubated at room temperature with one of the following primary antibodies: mouse antichicken tensin (Transduction Laboratories, clone 5, 5 Ag/ml PBS), mouse antichicken paxillin (Transduction Laboratories, clone 349, 1.25 Ag/ml), sheep antihuman
205
fibronectin (Serotec, 1:100), mouse antichicken vinculin (Sigma, clone VIN-11-5, 1:100), mouse antihuman avh3 integrin (Chemicon, clone LM609, 20 Ag/ml), or mouse antihuman a5h1 integrin (Chemicon, clone JBS5, stock concentration range 20–50 Ag/ml ) for 1 h. After three 5-min washes with PBS, coverslips were incubated for 30 min with a species-appropriate secondary antibody, either an Alexa-Fluor-488-conjugated goat antimouse immunoglobulin G (IgG; Molecular Probes, 1:100, excitation/emission=495/519 nm) or an FITC-conjugated AffiniPure donkey antisheep IgG (Jackson ImmunoResearch Labs, 1:100, excitation/emission=492/520 nm). F-actin was localized in double-stained coverslips by incubation with AlexaFluor-568-labeled phalloidin (Molecular Probes, 1:40, excitation/emission=578/600 nm) concomitantly with either the Alexa-Fluor-488- or the FITC-conjugated secondary antibody. Colocalization was identified by the superimposition of red and green to give a yellow color visualized under the confocal microscope. For negative controls, omission of the primary antibody or incubation with a nonspecific mouse IgG (Upstate, 1:200) was used. All controls were negative. The coverslips were washed with PBS (35 min) and mounted on glass slides with Prolong Antifade. Cells were examined with a BioRad MRC 1024ES confocal scanning laser microscope fitted with an argon–krypton mixed-gas laser for simultaneous visualization of multiple fluorescent probes. Images were captured under the 488-nm wavelength excitation line to visualize FITC or Alexa Fluor 488 and the 568-nm wavelength excitation line to visualize Alexa Fluor 568. Fluorescence was acquired at wavelengths N515 nm. A stack of optical sections was taken at 0.4-Am intervals in the Z direction with three Kalman averages, using a 60 oil-immersion objective lens. Projections of the series of optical sections were reconstructed into the same focal plane with LaserSharp software (BioRad, Version 3.2). Three investigators provided qualitative assessments of fibronectin, tensin, a5h1
Fig. 2. Micrographs of confocal projections of porcine VICs at the wound edge immunostained to detect fibronectin (Fn) 4 h (A), 24 h (B), and 48 h (C) after wounding. (A) Cells at the wound edge started migrating into the wound and synthesizing cytoplasmic Fn (arrows) 4 h after wounding. Twenty-four hours (B) and 48 h (C) after wounding, Fn cytoplasmic staining and fibril deposition by the migrating cells had increased. W=wound (green=Fn). Magnification, 600.
206
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
Fig. 3. Micrographs of confocal projections of cultures of porcine VICs immunostained to detect tensin in subconfluent (A and B) and confluent (C and D) monolayers. F-actin stress fibers were stained with Alexa-568-labeled phalloidin. In subconfluent cultures, tensin stained in a long, thin fibrillar pattern running lengthwise throughout the cell center (A) and colocalized along F-actin stress fibers (B; arrow). In the confluent monolayer, tensin was localized both in small weakly stained fibrillar adhesions distributed throughout the cell (C) and in large, rod-shaped focal adhesions at the cell periphery (D; green=tensin, red=F-actin stress fibers, yellow=colocalization). Magnification, 600.
integrin, vinculin, and paxillin staining under all control and experimental conditions. 3. Results 3.1. Fibronectin In subconfluent cultures grown for 2–3 days, cytoplasmic staining for fibronectin was present, especially in areas surrounding the nuclei of the cells. Fibronectin fibrils were also present immediately around the cells and extending between adjacent cells (Fig. 1A). Fibronectin fibrils were deposited in parallel to the F-actin stress fibers, often at the tips of lamellipodial protrusions (Fig. 1B). The cultures became confluent at 7 days, after which cytoplasmic fibronectin was no longer detected in the cells of the monolayer. However, fibronectin fibrils were present in the subcellular area immediately surrounding the cell periphery and between the cells. Furthermore, the intensity of fibronectin staining increased compared with that of subconfluent cultures (Fig. 1C and D).
When the confluent monolayers were wounded, elongated VICs were observed migrating outward from the wound edge as previously described [28]. At the wound edge, there was a loss of fibronectin, but VICs migrating into the wound showed cytoplasmic staining for fibronectin 4 h after wounding (arrow, Fig. 2A). Twenty-four and 48 h after wounding, the migrating VICs showed intense staining of cytoplasmic fibronectin, and fibronectin fibrils were deposited into the matrix (Fig. 2B and C). Forty-eight hours after wounding, cytoplasmic fibronectin stained predominantly the leading half of the cell in front of the nucleus (not shown). Wound closure was nearly complete 4 days after wounding, with only a few regions of the wound still open. Immunostaining for fibronectin increased in the extracellular space of the wound at 4 days when compared to wounds 48 h after wounding (data not shown). The density of the fibronectin fibrils in the closed wound approached that of the confluent monolayer away from the wound area. The fibronectin fibrils appeared longer and thicker than the ones that were first deposited into the extracellular matrix but not yet fully comparable to those of either the
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
207
Fig. 4. Micrographs of confocal projections of porcine VICs at the wound edge immunostained to detect (A) tensin, (B) vinculin, and (C) paxillin. F-actin stress fibers were stained with Alexa-568-labeled phalloidin. Tensin staining increased progressively from (A1) 0 h to (A2) 24 h and (A3) 48 h after wounding. (A3) At 48 h, tensin stained strongly in fibrillar adhesion sites; small, globular staining throughout the cell. (B and C) Paxillin and vinculin showed a constant intensity and pattern of staining at all time points; large, rod-shaped focal adhesions colocalized with the tips of the F-actin stress fibers at the cell periphery. N=nucleus (green=tensin, vinculin, and paxillin; red=F-actin stress fibers; yellow=colocalization). Magnification, 600.
unperturbed confluent monolayer or the monolayer away from the wound area. 3.2. Tensin, paxillin, and vinculin Subconfluent cultures of VICs stained weakly for tensin in an elongated thin fibrillar pattern (Fig. 3A and B). In the confluent monolayer, the staining pattern of tensin was more globular in shape than in the subconfluent cultures (Fig. 3C). In addition, some cells stained for tensin in large, rodshaped contacts located at the cell periphery (Fig. 3D),
which was similar to the staining pattern of the focal adhesion proteins vinculin and paxillin (Fig. 4B and C). After wounding, tensin staining increased progressively in intensity at 24 and 48 h (Fig. 4A1–A3). At 48 h, VICs now showed tensin localized to fibrillar adhesions, characterized by small globular plaques localized throughout the cell (Fig. 4A3). In contrast, staining for the focal adhesion proteins paxillin and vinculin was positive at the time of wounding (0 h; Fig. 4B1 and C1) and remained constant in intensity and pattern 24 and 48 h after wounding (Fig. 4B2 and B3 for paxillin; Fig. 4C2 and C3 for vinculin). These
208
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
Fig. 5. Micrographs of confocal projections of porcine VICs immunostained to detect a5h1 integrin. F-actin stress fibers were stained with Alexa-568-labeled phalloidin. (A) In subconfluent cultures, a5h1 stained strongly in a diffuse punctate pattern in the cytoplasm throughout the cell center and at the cell periphery in large focal adhesions (arrows). (B) Colocalization of the a5h1 integrin to the tips of F-actin stress fibers provided further evidence that a5h1 is present in focal adhesions (arrow). (C) In confluent cultures, a5h1 integrin staining was very faint in the cytoplasm and absent at focal adhesion sites. N=nucleus (green=a5h1 integrin, red=F-actin stress fibers, yellow=colocalization). Magnification, 600.
focal adhesion proteins appeared as large, rod-shaped focal adhesions localized to the tips of F-actin stress fibers along the cell periphery. Once VICs reestablished a confluent monolayer, tensin staining in the closed wounds showed the elongated thin fibrillar pattern running lengthwise along the cell and outlining the cell periphery (not shown). Cells of the monolayer away from the closed wound region also displayed tensin in long fibrillar adhesions throughout the cell center. In addition to this central staining, tensin was localized in rod-shaped structures at the cell periphery, colocalizing with the tips of the F-actin stress fibers. This staining pattern was similar to that observed for the focal adhesion protein paxillin, suggesting that tensin also localized in focal adhesions. 3.3. a5b1 integrin Subconfluent cultures of VICs stained strongly for the fibronectin receptor, the a5b1 integrin. The anti-a5b1 anti-
body stained large, rod-shaped contacts at the cell periphery and in a diffuse punctate pattern throughout the cell center (Fig. 5A). The former pattern likely represents focal adhesions, while the latter likely includes fibrillar adhesions. Cells double stained for a5b1 and F-actin showed that this integrin colocalized with the tips of the F-actin stress fibers (Fig. 5B), as seen with the focal adhesion proteins paxillin and vinculin. Once cultures became confluent, immunostaining for a5b1 integrin in the monolayer decreased dramatically and was barely detectable (Fig. 5C). a5b1 did not localize to the tips of the F-actin stress fibers. These results suggest that a5b1 integrins are not prominent in confluent cultures. There was no staining for a5b1 integrin at the time of wounding (0 h; Fig. 6A). Staining was detected 24 h after wounding (Fig. 6B) and remained constant at 48 h (Fig. 6C). The a5b1 integrin stained in two different patterns. The first was localized to the lamellipodia in migrating cells with some colocalization with the tips of the F-actin stress fibers,
Fig. 6. Micrographs of confocal projections of wounded monolayers double stained for a5h1 integrin and F-actin stress fibers, stained with Alexa-568-labeled phalloidin. (A) VICs of the confluent monolayer did not stain for a5h1 at 0 h. Twenty-four hours (B) and 48 h (C) after wounding, a5h1 stained both in a focal adhesion pattern at the lamellipodia of migrating cells, with colocalization at the tips of the F-actin stress fibers (arrows), and in a punctate pattern throughout the cell, consistent with fibrillar adhesions (green=a5h1 integrin, red=F-actin stress fibers, yellow=colocalization). W=wound. Magnification, 600.
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
consistent with localization at focal adhesions (arrows, Fig. 6B and C). The second stained in a punctate pattern throughout the center of the cell, consistent with localization at fibrillar adhesions. The staining of the a5b1 integrin showed a similar increase in intensity to that of tensin over time in VICs at the wound edge. In the closed wounds, a5b1 integrin stained faintly in a punctate pattern throughout the VIC cytoplasm (not shown). The same staining pattern was observed in the monolayer away from the wound area (not shown). Some regions of the wound had not yet reclosed in 4 days. Interestingly, the cells that were still migrating into these open regions of the wound stained for the a5b1 integrin both in a punctate pattern throughout the cell and in a focal adhesion pattern that localized to the cell periphery. In addition, the a5b1 staining intensity in the cytoplasm of the cells inside open wound regions was brighter than that of the cells both in the closed wounds and in the monolayer away from the wound area. This staining pattern of a5b1 was like that of the cells in the wounded monolayers and in subconfluent cultures.
4. Discussion Fibronectin constitutes part of the valve matrix, which is likely crucial for maintaining valvular structure and function. This is the first experimental study to report on interactions that take place between VICs and fibronectin in response to injury. To examine the dynamics of fibronectin fibril formation, we studied three conditions: subconfluent, confluent, and wounded cultures, which provide the opportunity to study populations of VICs in transition between quiescent and activated states. Cells in subconfluent cultures are initially migrating and proliferating to form the confluent monolayer. Cells in confluent cultures are quiescent and maintain strong attachments to the extracellular matrix via localized sites of firm anchorage called focal adhesions [29]. We showed that, under these three conditions, there are differences in expression and localization of fibronectin and its associated fibrillar adhesion proteins, tensin and a5b1 integrins. We found that the production of a fibronectin extracellular matrix plays a role in VIC repair after wounding. During the initiation of repair as VICs undergo the transition from quiescent to activated cells, fibronectin first appeared in the cell cytoplasm and then fibrils were deposited into the extracellular space to provide a matrix that supports migration. Deposition of fibronectin into the extracellular matrix may facilitate the migration of VICs to the site of injury and enable them to close the wound more efficiently, as has been shown in models of cardiac repair, cardiac cushion cell migration, and epithelial–mesenchymal cell transformation during cardiogenesis [15,30–34]. In fact, cytoplasmic fibronectin, tensin, and the a5b1 integrin were also expressed by VICs in subconfluent cultures, and new
209
fibronectin fibrils were deposited by migrating VICs in association with tensin and a5b1 in fibrillar adhesions. In contrast, cytoplasmic fibronectin disappeared in the quiescent VICs of confluent monolayers, suggesting that fibronectin synthesis was reduced once the cells reached confluence. A decrease in fibronectin protein synthesis has been reported to be associated with a decrease in fibrillar adhesion formation [23,35,36]. This is consistent with our observation that both tensin and a5b1 stained more weakly in the confluent VIC cultures. However, in some cells of the confluent monolayer, tensin staining was found in focal adhesions at the cell periphery, possibly representing tensin that did not translocate from focal to fibrillar adhesion sites. This observation is consistent with the idea that when fibronectin fibrillogenesis is not occurring, as we speculate is happening in the confluent VIC monolayer, fibrillar adhesions disassemble and focal adhesions become the predominant type of cell–matrix attachment. Studies have shown that once cells secrete fibronectin fibrils, they bind to these fibrils via the a5b1 integrin receptor, found within focal adhesions located at the cell periphery, at the tips of F-actin stress fibers [36–38]. Binding initiates a signaling cascade that results in the active transport of a5b1 and tensin along the F-actin stress fiber away from the cell periphery, leaving behind avb3 and paxillin in the focal adhesions. Our observations in VICs showed that fibronectin fibrils often colocalized with the F-actin stress fibers at the tips of lamellipodial protrusions, supporting the idea that fibronectin fibrillogenesis involves the translocation of fibronectin fibrils along the F-actin cytoskeleton. This process results in the formation of tensinrich fibrillar adhesions at the cell center, consisting of the a5b1 integrin in association with extracellular fibronectin fibrils, while lacking the other focal adhesion plaque components [18,19]. It is believed that translocation of this fibronectin–a5b1–tensin complex toward the cell center facilitates additional fibronectin fibril formation by stretching folded fibrils away from the cell periphery and exposing cryptic self-assembly sites that promote the self-association of fibronectin molecules [17,21,35,39]. This is consistent with our observations that porcine VICs at the wound edge deposited fibronectin fibrils into the extracellular matrix, displayed increased a5b1 integrin immunostaining, and formed tensin-rich fibrillar adhesions throughout the cell. In summary, our results suggest that fibronectin fibrillogenesis and the resultant formation of tensin-rich fibrillar adhesions play a role in cell migration in subconfluent and wounded cultures while focal adhesions remain unaffected. A potential mechanism for the closure of an in vitro wound would first involve the production of fibronectin by cells at the wound edge. Once a more pliable fibronectin matrix is deposited, cells are able to migrate more efficiently into the wound and reestablish the monolayer. The increased fibronectin production activates the processes involved in fibronectin fibrillogenesis, resulting in the increased expression of the a5b1 integrin fibronectin receptor. As the
210
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211
migrating cells attach to fibronectin via its receptor, tensin gets translocated with this fibronectin–a5b1 complex along F-actin stress fibers toward the cell center and forms tensinrich fibrillar adhesions. These dynamic fibrillar adhesions differ in both composition and function from the more stable focal adhesions. Fibronectin-mediated VIC migration and the formation of fibrillar adhesions have not been previously reported as potential response to valve injury.
5. Summary This is the first experimental study to report on interactions that take place between VICs and the fibronectin matrix in response to injury. Our findings support the hypothesis that fibronectin is formed and secreted by VICs at the wound edge and is associated with the formation of prominent fibrillar adhesions composed of tensin and a5b1 integrin. References [1] Mulholland DL, Gotlieb AI. Cell biology of valvular interstitial cells. Can J Cardiol 1996;12:231 – 6. [2] Mulholland D, Gotlieb AI. Cardiac valve interstitial cells: regulator of valve structure and function. Cardiovasc Pathol 1997;6:167 – 74. [3] Filip DA, Radu A, Simionescu M. Interstitial cells of the heart valves possess characteristics similar to smooth muscle cells. Circ Res 1986; 59:310 – 20. [4] Rabkin E, Aikawa M, Stone JR, Fukumoto Y, Libby P, Schoen FJ. Activated interstitial myofibroblasts express catabolic enzymes and mediate matrix remodeling in myxomatous heart valves. Circulation 2001;104:2525 – 32. [5] Taylor PM, Allen SP, Yacoub MH. Phenotypic and functional characterization of interstitial cells from human heart valves, pericardium and skin. J Heart Valve Dis 2000;9:150 – 8. [6] Messier RH, Bass BL, Aly HM, Jones JL, Domkowski PW, Wallace RB, Hopkins RA. Dual structural and functional phenotypes of the porcine aortic valve interstitial population: characteristics of the leaflet myofibroblast. J Surg Res 1994;57:1 – 21. [7] Roy A, Brand NJ, Yacoub MH. Molecular characterization of interstitial cells isolated from human heart valves. J Heart Valve Dis 2000;9:459 – 64. [8] Tamura K, Jones M, Yamada I, Ferrans VJ. Wound healing in the mitral valve. J Heart Valve Dis 2000;9:53 – 63. [9] Schoen FJ, Edwards WD. In: Silver MD, Gotlieb AI, Schoen FJ, editors. Valvular heart disease in cardiovascular pathology. Philadelphia7 Churchill Livingstone; 2001: p. 402 – 42. [10] Bauters C, Marotte F, Hamon M, Oliviero P, Farhadian F, Robert V, Samuel JL, Rappaport L. Accumulation of fetal fibronectin mRNAs after balloon denudation of rabbit arteries. Circulation 1995;92:904 – 11. [11] Colombi M, Zoppi N, De Petro G, Marchina E, Gardella R, Tavian D, Ferraboli S, Barlati S. Matrix assembly induction and cell migration and invasion inhibition by a 13-amino acid fibronectin peptide. J Biol Chem 2003;278:14346 – 55. [12] Farhadian F, Contard F, Corbier A, Barrieux A, Rappaport L, Samuel JL. Fibronectin expression during physiological and pathological cardiac growth. J Mol Cell Cardiol 1995;27:981 – 90. [13] Ffrench-Constant C, Van de WL, Dvorak HF, Hynes RO. Reappearance of an embryonic pattern of fibronectin splicing during wound healing in the adult rat. J Cell Biol 1989;109:903 – 14.
[14] Magnusson MK, Mosher DF. Fibronectin: structure, assembly, and cardiovascular implications. Arterioscler Thromb Vasc Biol 1998; 18:1363 – 70. [15] Ulrich MM, Janssen AM, Daemen MJ, Rappaport L, Samuel JL, Contard F, Smits JF, Cleutjens JP. Increased expression of fibronectin isoforms after myocardial infarction in rats. J Mol Cell Cardiol 1997;29:2533 – 43. [16] Sechler JL, Corbett SA, Wenk MB, Schwarzbauer JE. Modulation of cell–extracellular matrix interactions. Ann N Y Acad Sci 1998;857: 143 – 54. [17] Pankov R, Yamada KM. Fibronectin at a glance. J Cell Sci 2002; 115:3861 – 3. [18] Zamir E, Katz BZ, Aota S, Yamada KM, Geiger B, Kam Z. Molecular diversity of cell–matrix adhesions. J Cell Sci 1999; 112(Pt 11);1655 – 69. [19] Katz BZ, Zamir E, Bershadsky A, Kam Z, Yamada KM, Geiger B. Physical state of the extracellular matrix regulates the structure and molecular composition of cell–matrix adhesions. Mol Biol Cell 2000;11:1047 – 60. [20] Zamir E, Geiger B. Components of cell–matrix adhesions. J Cell Sci 2001;114:3577 – 9. [21] Geiger B, Bershadsky A, Pankov R, Yamada KM. Transmembrane crosstalk between the extracellular matrix–cytoskeleton crosstalk. Nat Rev Mol Cell Biol 2001;2:793 – 805. [22] Pelham RJ, Wang Y. Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc Natl Acad Sci U S A 1997; 94:13661 – 5. [23] Yamada KM, Pankov R, Cukierman E. Dimensions and dynamics in integrin function. Braz J Med Biol Res 2003;36:959 – 66. [24] Wierzbicka-Patynowski I, Schwarzbauer JE. The ins and outs of fibronectin matrix assembly. J Cell Sci 2003;116:3269 – 76. [25] Gotlieb AI, Rosenthal A, Kazemian P. Fibroblast growth factor 2 regulation of mitral valve interstitial cell repair in vitro. J Thorac Cardiovasc Surg 2002;124:591 – 7. [26] Durbin A, Nadir NA, Rosenthal A, Gotlieb AI. Nitric oxide promotes in vitro interstitial cell heart valve repair. Cardiovasc Pathol 2005; 14:12 – 8. [27] Lester W, Rosenthal A, Granton B, Gotlieb AI. Porcine mitral valve interstitial cells in culture. Lab Invest 1988;59:710 – 9. [28] Durbin AD, Gotlieb AI. Advances towards understanding heart valve response to injury. Cardiovasc Pathol 2002;11:69 – 77. [29] Sastry SK, Burridge K. Focal adhesions: a nexus for intracellular signaling and cytoskeletal dynamics. Exp Cell Res 2000;261:25 – 36. [30] Mjaatvedt CH, Lepera RC, Markwald RR. Myocardial specificity for initiating endothelial–mesenchymal cell transition in embryonic chick heart correlates with a particulate distribution of fibronectin. Dev Biol 1987;119:59 – 67. [31] Clark RA. Potential roles of fibronectin in cutaneous wound repair. Arch Dermatol 1988;124:201 – 6. [32] Clark RA, Wikner NE, Doherty DE, Norris DA. Cryptic chemotactic activity of fibronectin for human monocytes resides in the 120kDa fibroblastic cell-binding fragment. J Biol Chem 1988;263: 12115 – 23. [33] Ffrench-Constant C, Hynes RO. Patterns of fibronectin gene expression and splicing during cell migration in chicken embryos. Development 1988;104:369 – 82. [34] Icardo JM, Nakamura A, Fernandez-Teran MA, Manasek FJ. Effects of injecting fibronectin and antifibronectin antibodies on cushion mesenchyme formation in the chick An in vivo study. Anat Embryol (Berl) 1992;185:239 – 47. [35] Schwarzbauer JE, Sechler JL. Fibronectin fibrillogenesis: a paradigm for extracellular matrix assembly. Curr Opin Cell Biol 1999; 11:622 – 7. [36] Pankov R, Cukierman E, Katz BZ, Matsumoto K, Lin DC, Lin S, Hahn C, Yamada KM. Integrin dynamics and matrix assembly: tensindependent translocation of alpha(5)beta(1) integrins promotes early fibronectin fibrillogenesis. J Cell Biol 2000;148:1075 – 90.
C. Fayet et al. / Cardiovascular Pathology 16 (2007) 203 – 211 [37] Zamir E, Katz M, Posen Y, Erez N, Yamada KM, Katz BZ, Lin S, Lin A, Bershadsky A, Kam Z, Geiger B. Dynamics and segregation of cell– matrix adhesions in cultured fibroblasts. Nat Cell Biol 2000;2:191 – 6. [38] Ilic D, Kovacic B, Johkura K, Schlaepfer DD, Tomasevic N, Han Q, Kim JB, Howerton K, Baumbusch C, Ogiwara N, Streblow DN, Nelson JA, Dazin P, Shino Y, Sasaki K, Damsky CH. FAK promotes
211
organization of fibronectin matrix and fibrillar adhesions. J Cell Sci 2004;117:177 – 87. [39] Zhong C, Chrzanowska-Wodnicka M, Brown J, Shaub A, Belkin AM, Burridge K. Rho-mediated contractility exposes a cryptic site in fibronectin and induces fibronectin matrix assembly. J Cell Biol 1998;141:539 – 51.