Jararhagin, a snake venom metalloproteinase-disintegrin, stimulates epithelial cell migration in an in vitro restitution model

Toxicon 44 (2004) 861–870 www.elsevier.com/locate/toxicon

Jararhagin, a snake venom metalloproteinase-disintegrin, stimulates epithelial cell migration in an in vitro restitution model E´rica Pereira Costa, Marinilce Fagundes Santos* Department of Developmental and Cell Biology, Institute of Biomedical Sciences, University of Sa˜o Paulo, ICB/USP, Av. Prof. Lineu Prestes 1524, Sa˜o Paulo, SP CEP 05508-000, Brazil Received 8 March 2004; accepted 19 August 2004 Available online 14 October 2004

Abstract The snake venom metalloproteinase-disintegrin jararhagin (JG) has no chemotactic activity but stimulates the migration of neutrophils in vivo through a mechanism still unclear. In this study we investigated the effects of jararhagin on epithelial cell adhesion and migration in vitro. F-actin arrangement and the distribution of laminin, fibronectin, several integrins and phosphorylated Focal Adhesion Kinase (FAK) were studied using rhodamine–phalloidin and immunofluorescence. Maximum stimulation of migration (about 100%) was obtained with 5 mg/ml JG, with about 38% inhibition of cellular adhesion. In migratory cells the toxin stimulated the formation of filopodia, lamellipodia and stress fibers. The pericellular fibronectin matrix was lost in migrating cells, while laminin was less affected. The toxin stimulated FAK phosphorylation and the recruitment of av-containing integrins to focal contacts, whereas integrins containing the a2 subunit were reduced in these junctions. Inactivation of the toxin with 1,10 phenanthroline showed that the catalytic activity is important for the effect of jararhagin on cell migration, FAK phosphorylation and for the recruitment of av, but not as much for the anti-adhesive effect. In conclusion, jararhagin stimulates the migration of epithelial cells in vitro through a mechanism that involves its proteolytic activity, qualitative changes in cellular adhesion and the formation of actin-rich cellular processes. q 2004 Elsevier Ltd. All rights reserved. Keywords: Snake venom metalloproteinase; Jararhagin; Disintegrin; Cell migration; Cytoskeleton; Integrins

1. Introduction Snake Venom Metalloproteinases (SVMPs) are responsible for most of the local and systemic effects observed during envenoming by snakes from the Viperidae family: haemorrhage, myonecrosis, skin lesions, inflammation, cytokine-dependent inflammatory cellular influx, complement activation and activation of endogenous matrix metalloproteinases (MMPs) (Bjarnason and Fox, 1994; Bode et al., 1993; Gutie´rrez and Rucavado, 2000). * Corresponding author. Tel.: C55 11 3091 7371; fax: C55 11 3091 7219. E-mail address: [email protected] (M.F. Santos). 0041-0101/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.toxicon.2004.08.009

Jararhagin, a hemorrhagic toxin isolated from the venom of Bothrops jararaca, is a SVMP of the Reprolysin family of zinc metalloproteinases, containing a metalloproteinase domain followed by an ECD disintegrin-like domain and a cysteine-rich domain (Paine et al., 1992). This toxin shows structural homology with other proteins such as the cell surface A Disintegrin And Metalloproteinases (ADAMs), RGD disintegrins from venoms and several MMPs (Bjarnason and Fox, 1994; Bode et al., 1993; GomisRuth et al., 1993; Niewiarowski et al., 1994; Usami et al., 1994). In platelets and fibroblasts jararhagin binds the a2b1 integrin, inhibiting platelet aggregation and collagen adhesion in vitro (Moura-da-Silva et al., 2001; Zigrino et al., 2002). In mice, using a dorsal air-pouch model,

862

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

the toxin induced the influx of inflammatory cells, mostly neutrophils, without having direct chemotactic activity (Costa et al., 2002). The mechanisms involved in this latter effect are still unclear, as well as the possible effects of jararhagin in other migratory cell types. Several factors affect cell adhesion and migration, including the concentration of adhesive molecules in the extracellular matrix (ECM), the number and affinity of integrins and interactions between these receptors and the cytoskeleton (Cox and Huttenlocher, 1998; Schwartz and Shattil, 2000). Integrins are heterodimeric transmembrane proteins composed by alpha and beta subunits, and from this combination depends their ligand specificity. While the extracellular portion of most integrins binds specific regions of ECM proteins, the intracellular portion is often linked to the cytoskeleton through several structural/ signaling proteins such as talin, vinculin and paxillin, which participate in large protein complexes known as focal adhesions (Giancotti, 2000). Regulation of the cell-ECM interactions during migration can also be effected by proteolysis, usually requiring strict spatial and temporal targeting of proteinase activity. The versatile use of different proteinase systems, with a variety of localization mechanisms and cleavage targets, is being revealed by several studies (Ellis and Murphy, 2001). MMPs, for example, were believed to function primarily as regulators of ECM composition and to facilitate cell migration simply by removing barriers such as collagen. Recent studies have expanded their roles well beyond ECM degradation they also cleave many growth factors, cytokines and cell adhesion molecules in the extracellular milieu, modulating their functions irreversibly. In particular, some MMPs that associate with the cell surface have arisen as intriguing regulators of cellular functions, including migration (Seiki, 2002; Stamenkovic, 2003). Besides, ECM preexisting modules or cryptic sites revealed by partial enzymatic hydrolysis positively or negatively regulate MMP expression and activation, further influencing cell migration (Hornebeck et al., 2002). Due to SVMPs similarity to several MMPs and ADAMs, their effects on adhesion and migration of different cell types have also been subject of intense investigation, particularly as potential therapeutic tools for cancer treatment, inhibiting metastasis and angiogenesis (Beviglia et al., 1995; Cominetti et al., 2003; Correa et al., 2002; Schmitmeier et al., 2003; Sheu et al., 1997; Soszka et al., 1991; Tomczuk et al., 2003). The aim of this study was to verify the effects of jararhagin on cell adhesion and migration, using a wellestablished in vitro model for epithelial restitution, the initial phase of healing of superficial wounds in skin and mucosae. The ability of cells to migrate plays a pivotal role during this process (McComarck et al., 1992; Santos et al., 1997a,b).

2. Materials and methods 2.1. Materials Culture ware was purchased from Corning Glass Works (Corning, NY). Medium, fetal bovine serum (FBS) and other cell culture reagents were obtained from Invitrogen (Carlsbad, CA). Other chemicals and biochemicals were obtained from Sigma (St. Louis, MO) and Mallinkrodt (Phillipsburg, NJ). Matrigel was obtained from BD Biosciences (Bedford, MA). Polyclonal antibodies to antiintegrin subunits a2, av and fibronectin were obtained from Chemicon International, Inc. (Temecula, CA). The polyclonal antibody against laminin, the monoclonal antibody against phosphorylated FAK and DAPI (4,6diamino-2-phenylindole) were obtained from Sigma (St. Louis, MO), while the Rhodamine-phalloidin was obtained from Molecular Probes (Eugene, OR). Secondary antibodies were obtained from Jackson Immunolabs (West Grove, Pennsylvania). 2.2. Purification of jararhagin from the venom and inactivation The venom from Bothrops jararaca snakes was obtained from the Laboratory of Herpetology, Butantan Institute, Sa˜o Paulo, kindly donated by Dr. Ana Maria Moura-da-Silva. Jararhagin was purified as previously described (Paine et al., 1992). Briefly, the venom was fractionated by hydrophobic interaction using FPLC Phenyl-Superose columns. The most hydrophobic fraction (containing the hemorrhagic and proteolytic activities) was applied to Mono-Q chromatography columns (FPLC). After elution, the purity of the toxin was checked by SDS-PAGE, always showing a single band of 52-kDa, corresponding to jararhagin. The protein concentration was estimated by the Bradford method (1976). Jararhagin proteolytic activity inactivation was obtained by incubation of 100 mg/ml of the toxin in the presence of 10 mM 1,10 phenanthroline at 37 8C for 30 min. 2.3. Cell culture The stock of IEC-6 cells (ATCC # CRL 1592) was maintained in a humidified, 37 8C incubator in an atmosphere of 95% air-5% CO2. This non transformed cell lineage was derived from undifferentiated rat intestinal crypt cells (Quaroni et al., 1979). The medium consisted of Dulbecco’s Modified Eagle’s Medium with 5% FBS plus 10 mg insulin, 100 units of penicillin, 100 mg of streptomycin sulfate and 0.29 mg of L-glutamine/ml (DMEM-FBS). The stock was passaged weekly. 2.4. Cell migration This assay was carried out as previously described (McComarck et al., 1992). Approximately 0.5!106 cells

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

863

Fig. 1. Dose–response curve showing the effects of jararhagin on directional migration (A) and adhesion of IEC-6 cells to treated Petri dishes (B). *P!0.05, according to ANOVA and Tukey post-test.

were plated in 35 mm dishes, fed on day 2 and tested on day 4. For assays using exogenously provided ECM, 35 mmdishes were coated with Matrigel, according to the manufacturer’s instructions. To initiate migration, the cell layer was scratched with a razor blade, beginning at the diameter of the dish and extending over an area 30 mm wide. After the scratch the medium was changed, and the dishes were returned to the incubator. For the dose– response experiments, jararhagin was added to the medium after the scratch at several concentrations ranging from of 0.31 to 15 mg/ml. After 6 h, the dishes were washed with 0.1 M phosphate buffer pH 7.4 (PB), fixed with 4% formaldehyde/0.2% Triton X-100 for 10 min and washed with PB. The areas with the highest migration rate were photographed using a CCD camera coupled to an inverted Zeiss Axiovert 135 microscope. Cells were counted, and migration was expressed as percent of controlGstandard deviation (SD) of at least 2–3 different experiments with triplicates.

2.5. Attachment assay This method was already described (Santos et al., 1997b). Cells were taken up with trypsin plus 1 mM EDTA in Hank’s balanced salt solution (HBSS). A final suspension was prepared to a density of 2.5!105 cells/ml in DMEM-FBS. After treatment with trypsin, cells were allowed to recover for one hour at room temperature (RT) in suspension, in DMEM-FBS. For the assay, cells were plated at 3!104 cells/ cm2 and incubated for 20 min in DMEM-FBS with or without the toxin at 37 8C in an atmosphere of 95% air-5% CO2. Following incubation the dishes were placed on ice, and cells were washed three times with ice-cold PB, fixed with 4% formaldehyde/0.2% Triton X-100 in PB for 10 min at RT, and washed three times with PB. The attached cells were immediately photographed using an inverted-phase microscope with attached CCD camera. Three pictures were randomly taken from each dish at 100!magnification and the total number of cells/area was counted. Attachment was

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

864

2.8. Immunofluorescence

Fig. 2. Migration (A) and adhesion (B) of control IEC-6 cells (C), cells treated with 5 mg/ml Jararhagin (J) or 5 mg/ml jararhagin inactivated with 1,10 phenanthroline (IJG). *P!0.05, according to ANOVA and Tukey post-test.

expressed as percent of controlGSD of at least two independent experiments with triplicates.

Approximately 0.2!106 cells were plated per 35 mm dish (each containing a non-coated or a Matrigel-coated glass cover slip), fed on day 2 and, on day 4, the migration assay was performed as already described. After 3 h, the cells were treated with jararhagin in the doses of 5 mg/ml (glass) or 1.25 mg/ml (Matrigel) and the plates were incubated for additional 20 min. Alternatively, cells received the toxin in the beginning of the migration assay and migrated for 3 h. Afterwards, they were fixed and permeabilized as already described for the F-actin staining. All proteins were stained by indirect immunofluorescence. The non specific binding sites were blocked with 0.5% non immune goat or rabbit serum in PB/0.2% Triton X-100 during 1 h and 30 min. Incubation with the primary antibodies (anti-laminin and anti-fibronectin 1:100, antiphosphorylated FAK, anti-av, and anti-a2 1:50) diluted in PB with or without 0.2% Triton X-100 (this reagent was not used for laminin and fibronectin staining) was done in a humid chamber at RT overnight. Incubation with the secondary antibody labelled with FITC (diluted 1:200 in PB/0.2% Triton X-100) was done for 1 h and 30 min at RT. After washing in PB, nuclear staining was performed with 5 mg/ml DAPI (Sigma) during 15 min. After washing, the slides were mounted using a glycerol–carbonate solution, and analyzed in a Nikon PCM2000 confocal microscope. Figures were mounted with Adobe Photoshop 3.0 (Deneba), and printed on a Kodak XLS 8600 PS printer. Only brightness and contrast were adjusted. 2.9. Statistical methods

2.6. Cell viability assays IEC-6 cells were incubated during 20 min at RT in DMEM-FBS with or without jararhagin in different concentrations (time chosen for the majority of the experiments). The cell viability was measured using the Trypan blue exclusion method.

All experiments consisted of at least duplicate samples. Analysis of variance followed by Tukey’s procedure were used. Results were considered significantly different at P!0.05.

3. Results 2.7. F-actin staining

3.1. Jararhagin increases the migration of IEC-6 cells 6

Approximately 0.2!10 cells were plated per 35 mm dish (each containing a glass cover slip), fed on day 2 and, after 4 days, the migration assay was performed as described previously. After 3 h, jararhagin was added at a concentration of 5 mg/ml to the experimental dishes, and cells were fixed and permeabilized with 2% formaldehyde/0.2% Triton X-100 in PEM buffer (10 mM PIPES, 5 mM EGTA, 2 mM MgCl2, pH 6.8) for 10 min at RT and post-fixed with 95% ethanol for 5 min at K20 8C after different periods of time (5, 10, 15 or 30 min). F actin was detected by rhodamine–phalloidin staining according to the manufacturer’s instructions. Samples were imaged with a Nikon PCM 2000 Laser Scanning Confocal microscope.

Jararhagin present in the medium during 6 h increased the migration of IEC-6 cells up to the concentration of 10 mg/ml, whereas higher concentrations had no effect (Fig. 1A). The maximum increase in migration (100%) was obtained with 5 mg/ml jararhagin. Other concentrations of jararhagin increased migration in 38% (1.25 mg/ml), 59% (2.5 mg/ml), and 42% (10 mg/ml). In the presence of Matrigel, only the concentration of 1.25 mg/ml was effective, increasing cellular migration in about 53% (data not shown). The catalytic activity of jararhagin was important for this effect on cell migration, because inactivated jararhagin had no effect (Fig. 2A).

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

865

Fig. 3. Effects of jararhagin on F-actin arrangement in migrating IEC-6 cells. Migrating cells were treated with 5 mg/ml jararhagin for different periods of time. A, control migrating cells; B-D, treatment with jararhagin for 10, 15 and 30 min, respectively. There is an increase in actin polymerization and formation of processes such as filopodia (arrowheads), lamellipodia (arrows) and stress fibers.

3.2. Jararhagin decreases the adhesion of IEC-6 cells to the substrate For the adhesion experiments, IEC-6 cells were seeded upon coated or uncoated (treated) dishes in the presence of jararhagin in several concentrations, during 20 min. This period of time was chosen after preliminary experiments testing different periods for adhesion (20 min, 1 and 3 h, data not shown). Adhesion is a fast phenomenon, depending on the substrate. Even with poor substrates, after 20 min the cells were adherent, but not spread out yet. Fig. 1B shows that jararhagin reduced cellular adhesion in an apparent dosedependent manner, being significantly different from the control at the concentrations of 5, 10 and 15 mg/ml (inhibition of 38, 67 and 73%, respectively). In the presence of Matrigel

the toxin reduced cellular adhesion in 48% (1.25 mg/ml), 56% (2.5 mg/ml) and 81% (5 mg/ml) (data not shown). Actually, high concentrations (above 5 mg/ml with uncoated dishes and above 1.25 mg/ml with Matrigel-coated dishes) also promoted some de-adhesion in the cellular monolayer. This effect was not related to cytotoxicity, since viability tests showed absolutely no cytotoxic effect of the toxin (99% viability after treatment with 15 mg/ml jararhagin for 20 min when compared to the control cells). For this effect upon cell adhesion the catalytic activity of jararhagin was not important, considering that inactivated jararhagin had the same effect observed with the native toxin (Fig. 2B). We have to consider, however, the possibility that some residual 1,10 phenanthroline could also affect adhesion.

866

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

Fig. 4. Effects of jararhagin and inactivated jararhagin on FAK phosphorylation (A–C), and distribution of the integrin subunits av (D–F) and a2 (G–I) in IEC-6 cells. Migrating cells were treated with 5 mg/ml jararhagin (B, E, H) or inactivated jararhagin (C, F, I) during 20 min, fixed and processed for immunofluorescence. A, D and G, control cells. Jararhagin increases FAK phosphorylation, augments av and reduces a2 in focal contacts. These effects depend on the proteolytic activity of jarahagin. Arrows point to focal contacts.

3.3. Jararhagin promotes F-actin rearrangement in migrating IEC-6 cells, increasing the amount of phosphorylated FAK in focal contacts Although there was apparently a gradual increase in the F-actin content in migratory cells in the presence of 5 mg/ml jararhagin over time, the most striking effects were observed after 15 min of incubation, with numerous long filopodia irradiating from the cells located in the migrating edge(Fig. 3). This effect was still observed after 20 min (data not shown). After 30 min of incubation with

the toxin, cells showed extensive lamellipodia and an increase in stress fibers (Fig. 3). Immunohistochemistry to phosphorylated FAK (PFAK), a 125 kDa-kinase associated with focal contacts, was used in order to verify the distribution of these junctions in migrating cells in the presence of the toxin (5 mg/ml) for 20 min. The amount of P-FAK also indicates the activity of these junctions, since FAK exerts a central role in integrin signaling. Fig. 4(A-C) shows that the toxin promoted an increase in FAK phosphorylation in migrating cells.

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

867

Fig. 5. Effects of jararhagin and inactivated jararhagin on fibronectin (A–C) and laminin (D–F) distribution in IEC-6 cells. Migrating cells were treated with 5 mg/ml jararhagin (B, E) or inactivated jararhagin (C, F) during 20 min, fixed and processed for immunofluorescence. A and D, control cells. Due to its proteolytic activity, jararhagin rapidly changes the pericellular matrix.

Using several photographs and the Image Pro Plus software (Media Cybernetics, LP) we quantified this labeling measuring the number of P-FAK-containing focal contacts and the area occupied by this labeling in the photographs (mean area of P-FAK labeling in each contact and total area). The results showed that jararhagin significantly increased the number of P-FAK-containing focal contacts in 100%, while the total area of FAK labeling was significantly increased in 160%, due to an additional increase in the amount of labeling in each focal contact. Inactivation of jararhagin with 1,10 phenanthroline abolished this effect. 3.4. Jararhagin promotes the recruitment of av- and reduces the amount of a2-containing integrins in focal contacts Immunohistochemistry to the av integrin subunit showed that jararhagin-treated cells had a higher amount of this protein concentrated in focal contacts, in comparison to control cells. As shown in Fig. 4(D-F), there was a recruitment of these integrins to focal contacts in the presence of the toxin, while the amount of a2-containing integrins (probably integrin a2b1) in these junctions was decreased (Fig. 4 (G-I)). In control cells the av subunit was often observed in focal contacts, and this distribution was accentuated in the presence of jararhagin,

independently of the substrate (uncoated dishes or Matrigel). Inactivation of the metalloproteinase activity of jararhagin abolished its effects on the distribution of avand a2-containing integrins. 3.5. Jararhagin alters fibronectin and laminin in the pericellular matrix Due to its metalloproteinase activity, jararhagin at a concentration of 5 mg/ml for 20 min reduced the fibronectin and laminin contents around the cells. This effect was particularly evident for the fibronectin matrix, which was more abundant and lost its fibrillar aspect around migratory cells after treatment with the toxin (Fig. 5A-C). Although jararhagin was also effective reducing laminin, in migrating cells the content of this protein was lower than that observed for fibronectin (Fig. 5D-F). This proteolytic effect of jararhagin was also observed on Matrigel alone (data not shown). As expected, inactivated jararhagin had no effect on fibronectin and laminin present in the pericellular matrix.

4. Discussion In order to study the effects of jararhagin or other similar SVMP on cell migration, one has to consider, at least, its

868

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

metalloproteinase activity and other domains such as disintegrin-like and cysteine-rich. Most studies involving disintegrins and cell migration focused on angiogenesis and tumor cell invasion, with variable results. The venom RGD disintegrin Albolabrin inhibited the attachment of B16-F10 mouse melanoma cells to ECM and also inhibited mouse lung colonization by IV-injected cells in an in vivo experimental metastasis system (Soszka et al., 1991). It was shown recently that jararhagin also inhibited tumorigenicity of Skmel-28 human melanoma cells injected SC in mice, reducing lung metastasis within 180 days (Correa et al., 2002). In vitro assays showed that jararhagin (active or inactivated by 1,10-phenanthroline) inhibited adhesion and invasion, but also inhibited cellular proliferation. The direct effect of the toxin on cell migration, however, was not evaluated. In this study, using a well established model for epithelial restitution and migration, we demonstrate for the first time that jararhagin enhances the migration of epithelial cells up to 100%—as much as growth factors that stimulate migration of those cells (Santos et al., 1997a)—while inhibiting cellular adhesion to the substrate in a dosedependent manner. The effect of jararhagin on epithelial cell migration was dependent on its proteolytic activity, because it was abolished after the inactivation of the toxin with 1,10 phenanthroline. Several metalloproteinases increase cell migration and invasion through the cleavage or shedding of cell surface proteins, including ECM proteins in the pericellular matrix, surface receptors, and surface metalloproteinases (Nabeshima et al., 2002). Jararhagin clearly altered the pericellular matrix of migrating cells, particularly its fibronectin content. This disorganization would be expected to have a negative effect on cell migration, considering that the fibronectin synthesized by IEC-6 cells is important for fast migration (Goke et al., 1996). On the other hand, fibronectin fragments with biological activity generated by the proteolysis might be signaling through integrins in the cell surface, for example integrins containing the av subunit. Several authors reported that the proteolysis of compounds of the basal lamina expose cryptic sites capable of promoting cellular migration (Bower-sox and Sorgenete, 1982; Gianelli et al., 1997; Legrand et al., 1999; Xu et al., 2001). Besides, these ECM preexisting modules or cryptic sites positively or negatively regulate MMP expression and activation, further influencing matrix invasion by cancer cells (Hornebeck et al., 2002). Additionally, certain MMPs can bind to integrins or other receptors on the cell surface, thereby providing a mechanism for localized matrix degradation (Steffensen et al., 2001). Curiously, in the presence of exogenous ECM (Matrigel) the effect of jararhagin on cell migration was still significant but less pronounced. Maybe this result was related to a higher competition for cell surface receptors, important for a localized proteolysis promoted by jararhagin. Another

possibility is that in the presence of Matrigel the ECM fragments generated by jararhagin activity might be different, since the amount of laminin and collagen IV are much higher in Matrigel when compared to the extracellular matrix produced by IEC-6 cells. The inhibition of adhesion promoted by jararhagin was similar to other studies reporting inhibition of cellular adhesion to ECM by metalloproteinases containing disintegrin-like domains, or disintegrins (Beviglia et al., 1995; Cominetti et al., 2003; Correa et al., 2002; Schmitmeier et al., 2003; Sheu et al., 1997; Soszka et al., 1991; Tomczuk et al., 2003). Considering that the inhibition of adhesion was dose-dependent and not affected by the toxin inactivation, it is possible that this effect was due to the specific binding of jararhagin to integrins. Correa et al. (2002) had already shown that active and inactivated jararhagin had the same inhibitory effect on adhesion of melanoma cells to ECM. Interestingly, the inhibition of cellular adhesion under conditions of maximum migration was about 38–48%, in the absence or presence of exogenous matrix. An interesting thought is that the enhanced migration could be directly related to the inhibition of cellular adhesion. DiMilla et al. (1991) suggested, through a mathematical model, that an intermediate level of adhesion to the substrate is optimum for cell migration, because a too high or too low adhesiveness would be prejudicial. In IEC-6 cells, however, the effect on cell migration was abolished after the treatment of jararhagin with 1,10 phenanthroline, while adhesiveness inhibition was still present. The treatment of IEC-6 cells with jararhagin induced profound alterations in the F-actin cytoskeleton, in a timerelated manner. These alterations consisted of the appearance of long filopodia after 15 min, and the increase in lamellipodia and stress fibers afterwards. Within 1 h, these effects had disappeared (data not shown). All of these morphological changes were compatible with an enhanced migratory behavior of these cells and activation of signaling proteins belonging to the Rho family of GTPases (Santos et al., 1997a). Cell locomotion involves the activation of these GTPases and the rearrangement of the actin cytoskeleton to form cellular processes and focal contacts, providing the necessary traction for migration (Hall, 1998; Ridley and Hall, 1992). FAK exerts a central role in integrin signaling in focal contacts, and our results showed an increase in FAK phosphorylation after treatment with jararhagin. This effect was abolished after proteolytic inactivation of the toxin, suggesting that the binding of disintegrin-domain of jararhagin was not directly involved. This result suggests higher integrin signaling in adhesion contacts, probably related to the enhanced migratory behavior. Integrins containing the av subunit, which bind fibronectin and other RGD-containing proteins, were specifically recruited to focal adhesions after treatment of the cells with jararhagin. The av subunit, which can form receptors with

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

several b subunits, is involved in the adhesion and migration of keratinocytes and PKC activation (Huang et al., 1998; Thomas et al., 2001) and FAK phosphorylation in tumoral cells stimulated by RGD-disintegrins (Ritter et al., 2000, 2001). We believe that the recruitment of av-containing integrins was related to the enhanced migratory behavior but not due to direct jararhagin binding, since this toxin has an ECD disintegrin-like domain. Control migrating cells showed the presence of a2 in focal contacts, but the treatment with jararhagin in solution significantly reduced this labeling. Since jararhagin binds the a2b1 integrin in other cell types, it was expected that the same occurred in IEC-6 cells, which constitutively express this integrin. The binding could explain the absence of a2 in focal contacts, similar to the observation made by Eble et al. (2002) with the venom disintegrin rhodocetin, an inhibitor of a2b1 which failed to cluster this integrin in the cell surface. Inactivation of the toxin abolished this effect in IEC-6 cells, however, suggesting that the proteolytic activity, rather than the disintegrin-like domain, was important. Interestingly, Kamiguti et al. (1996) showed that the expression of a2b1 in platelets following jararhagin pretreatment was markedly reduced due to the proteolysis of the b1 subunit by the toxin. While the disintegrin-like domain was important for ligand specificity, cleavage was dependent of proteolytic activity. We do not know yet if a2b1 proteolysis happens in IEC-6 cells after jararhagin treatment, but our results are compatible with it. In summary, our results have demonstrated that the SVMP jararhagin directly stimulates epithelial cell migration in a manner that involves its metalloproteinase domain. Its activity leads to an increase in FAK phosphorylation and the recruitment of integrins containing the av subunit to focal contacts, leading to the formation of several actin-rich processes related to cell motility, regulated by GTPases from the Rho family of GTPases. Given the homology between the metalloproteinase, disintegrin-like and cysteine-rich domains of jararhagin and those of the members of the ADAMs family of proteins and other MMPs, this study demonstrates the potential of jararhagin as a useful tool to elucidate the action of other cell surface and matrix proteins containing similar domains in cell adhesion and migration.

Acknowledgements The authors would like to thank Dr Ana Maria Moura-da-Silva for the venom and purification of the toxin, Marley Janua´rio da Silva and Leandro Mantovani de Castro for the excellent technical assistance, and FAPESP for the scholarship for E´rica Costa (grant # 00/12769-7) and financial support (grants 97/09507-6 and 01/09047-2).

869

References Beviglia, L., Stewart, G.J., Niewiarowski, S., 1995. Effect of four disintegrins on the adhesive and metastatic properties of B16F10 melanoma cells in a murine model. Oncol. Res. 7, 7–20. Bjarnason, J.B., Fox, J.W., 1994. Hemorrhagic metalloproteinases from snake venoms. Pharmacol. Ther. 62, 325–372. Bode, W., Gomis-Ru¨th, F.S., Sto¨ckler, W., 1993. Astacin, serralysins, snake venom and matrix metalloproteinases exhibit identical zinc-binding environments (HESSHSSGSSG and Metturn) and topologies and should be grouped into a common family, the metzincins. FEBS Lett. 331, 134–140. Bower-sox, J.C., Sorgenete, N., 1982. Chemotaxis of aortic endothelial cells in response to fibronectin. Cancer Res. 42, 2547–2551. Bradford, M.A., 1976. A rapid and sensitive method for the quantitation of microgram amounts of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248– 254. Cominetti, M.R., Ribeiro, J.U., Fox, J.W., Selistre-deAraujo, H.S., 2003. BaG, a new dimeric metalloproteinase/disintegrin from the Bothrops alternatus snake venom that interacts with alpha5beta1 integrin. Arch. Biochem. Biophys. 416, 171–179. Correa Jr., M.C., Maria, D.A., Moura-da-Silva, A.M., Pizzocaro, K.F., Ruiz, I.R., 2002. Inhibition of melanoma cells tumorigenicity by the snake venom toxin jararhagin. Toxicon 40, 739–748. Costa, E.P., Clissa, P.B., Teixeira, C.F., Moura-da-Silva, A.M., 2002. Importance of metalloproteinases and macrophages in viper snake envenomation-induced local inflammation. Inflammation 26, 13–17. Cox, E.A., Huttenlocher, A., 1998. Regulation of integrin-mediated adhesion during cell migration. Microsc. Res. Tech. 43, 412–419. DiMilla, P.A., Barbee, K., Lauffenburger, D.A., 1991. Mathematical model for the effects of adhesion and mechanics on cell migration speed. Biophys. J. 60, 15–37. Eble, J.A., Niland, S., Dennes, A., Schmidt-Hederich, A., Bruckner, P., Brunner, G., 2002. Rhodocetin antagonizes stromal tumor invasion in vitro and other alpha2beta1 integrin-mediated cell functions. Matrix Biol. 21, 547–558. Ellis, V., Murphy, G., 2001. Cellular strategies for proteolytic targeting during migration and invasion. FEBS Lett. 506, 1–5. Giancotti, F.G., 2000. Complexity and specificity of integrin signaling. Nat. Cell Biol. 2, E13–E14. Gianelli, G., Falk-Marzillier, J., Schiraldi, O., StetlerStevenson, W.G., Quaranta, V., 1997. Induction of cell migration by matrix metalloproteinase-2 cleavage of laminin5. Science 277, 225–228. Gomis-Ruth, F.X., Kress, L.F., Bode, W., 1993. First structure of a snake venom metalloproteinase: a prototype for matrix metalloproteinases/collagenases. EMBO J. 12, 4151–4157. Goke, M., Zuk, A., Podolsky, D.K., 1996. Regulation and function of extracellular matrix in intestinal epithelial restitution in vitro. Am. J. Physiol. 271, G729–G740. Gutie´rrez, J.M., Rucavado, A., 2000. Snake venom metalloproteinases: their role in the pathogenesis of local tissue damage. Biochimie 82, 841–850.

870

E´.P. Costa, M.F. Santos / Toxicon 44 (2004) 861–870

Hall, A., 1998. Rho GTPases and the actin cytoskeleton. Science 279, 509–514. Hornebeck, W., Emonard, H., Monboisse, J.C., Bellon, G., 2002. Matrix-directed regulation of pericellular proteolysis and tumor progression. Semin. Cancer Biol. 12, 231–241. Huang, X., Wu, J., Spong, S., Sheppard, D., 1998. The integrin avb6 is critical for keratinocyte migration on both its known ligand, fibronectin and on vitronectin. J. Cell Sci. 111, 2189–2195. Kamiguti, A.S., Hay, C.R., Zuzel, M., 1996. Inhibition of collageninduced platelet aggregation as the result of cleavage of alpha 2 beta 1-integrin by the snake venom metalloproteinase jararhagin. Biochem. J. 320, 635–641. Legrand, C., Gilles, C., Zahm, J.M., Polette, M., Buisson, A.C., Kaplan, H., Birembaut, P., Tournier, J.M., 1999. Airway epithelial cell migration dynamics: MMP-9 in cell—ECM remodeling. J. Cell. Biol. 146, 517–529. McComarck, S.A., Viar, M.J., Johnson, L.R., 1992. Migration of IEC-6 cells: a model for mucosal healing. Am. J. Physiol. 263, G494–G501. Moura-da-Silva, A.M., Marcinkiewicz, C., Marcinkiewicz, M., Niewiarowski, S., 2001. Selective recognition of alpha (2) beta (1) integrin by jararhagin, a metalloproteinase/disintegrin from Bothrops jararaca venom. Thromb. Res. 102, 153–159. Nabeshima, K., Inoue, T., Shimao, Y., Sameshima, T., 2002. Matrix metalloproteinases in tumor invasion: role for cell migration. Pathol. Int. 52, 255–264. Niewiarowski, S., McLane, M.A., Kloczewiak, M., Stewart, G.J., 1994. Disintegrins and other naturally occurring antagonists of platelet fibrinogen receptors. Semin. Hematol. 31, 289–300. Paine, M.J.I., Desmond, H.P., Theaskton, R.G.D., Crapton, J.M., 1992. Purification, cloning and molecular characterization of a high molecular weight hemorrhagic metalloproteinase, jararhagin, from Bothrops jararaca venom. J. Biol. Chem. 267, 22869–22876. Quaroni, A., Wands, J., Trelstad, R.L., Isselbacher, K.J., 1979. Epithelioid cell cultures from rat small intestine. Characterization by morphologic and immunologic criteria. J. Cell Biol. 80, 248–265. Ridley, A.J., Hall, A., 1992. The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 401–410. Ritter, M.R., Zhou, Q., Markland Jr., F.S., 2000. Contortrostatin, a snake venom disintegrin, induces avb6-mediated tyrosine phosphorylation of CAS and FAK in tumor cells. J. Cell Biochem. 79, 28–37. Ritter, M.R., Zhou, Q., Markland Jr., F.S., 2001. Contortrostatin, a homodimeric disintegrin, actively disrupts focal adhesion and cytoskeletal structure and inhibits cell motility through a novel mechanism. Cell Commun. Adhes. 8, 71–86. Santos, M.F., McComarck, S.A., Guo, Z., Okolicany, J., Zheng, Y., Johnson, L.R., Tigyi, G., 1997a. Rho proteins play a critical role in cell migration during the early phase of mucosal restitution. J. Clin. Invest. 100, 216–225.

Santos, M.F., Viar, M.J., McComarck, S.A., Johnson, L.R., 1997b. Polyamines are important for attachment of IEC-6 cell to extracellular matrix. Am. J. Physiol. 273, G175–G183. Schmitmeier, S., Markland, F.S., Ritter, M.R., Sawcer, D.E., Chen, T.C., 2003. Functional effect of contortrostatin, a snake venom disintegrin, on human glioma cell invasion in vitro. Cell Commun. Adhes. 10, 1–16. Schwartz, M.A., Shattil, S.J., 2000. Signaling networks linking integrins and Rho family GTPases. Trends Biochem. Sci. 25, 388–391. Seiki, M., 2002. The cell surface: the stage for matrix metalloproteinase regulation of migration. Curr. Opin. Cell Biol. 14, 624–632. Sheu, J.R., Yen, M.H., Kan, Y.C., Hung, W.C., Chang, P.T., Luk, H.N., 1997. Inhibition of angiogenesis in vitro and in vivo: comparison of the relative activities of triflavin, an Arg-Gly-Asp-containing peptide and anti-alpha(v)beta3 integrin monoclonal antibody. Biochim. Biophys. Acta. 1336, 445–454. Soszka, T., Knudsen, K.A., Beviglia, L., Rossi, C., Poggi, A., Niewiarowski, S., 1991. Inhibition of murine melanoma cellmatrix adhesion and experimental metastasis by albolabrin, an RGD-containing peptide isolated from the venom of Trimeresurusalbolabris. Exp. Cell Res. 196, 6–12. Stamenkovic, I., 2003. Extracellular matrix remodeling: the role of matrix metalloproteinases. J. Pathol. 200, 448–464. Steffensen, B., Hakkinen, L., Larjava, H., 2001. Proteolytic events of wound-healing-coordinated interactions among matrix metalloproteinases (MMPs), integrins, and extracellular matrix molecules. Crit. Rev. Oral Biol. Med. 12, 373– 398. Thomas, G.J., Poomsawat, S., Lewis, M.P., Hart, I.R., Speight, P.M., Marshall, J.F., 2001. avb6 integrin upregulates matrix metalloproteinase 9 and promotes migration of normal oral keratinocytes. J. Invest. Dermatol. 116, 898–904. Tomczuk, M., Takahashi, Y., Huang, J., Murase, S., Mistretta, M., Klaffky, E., Sutherland, A., Bolling, L., Coonrod, S., Marcinkiewicz, C., Sheppard, D., Stepp, M.A., White, J.M., 2003. Role of multiple beta1 integrins in cell adhesion to the disintegrin domains of ADAMs 2 and 3. Exp. Cell Res. 290, 68–81. Usami, Y., Fujimura, Y., Miura, S., Shima, H., Yoshida, E., Yoshioka, A., Hirano, K., Susuki, M., Tita´n, K., 1994. A 28 kDa-protein with disintegrin-like structure (Jararhagin-C) purified from Bothrops jararaca venom inhibits collagen- and ADP-induced platelet aggregation. Biochem. Biophys. Res. Commun. 201, 331–339. Xu, J., Rodriguez, D., Petitclerc, E., Kim, J.J., Hangai, M., Yuen, S.M., Davis, G.E., Brooks, P.C., 2001. Proteolytic exposure of a cryptic site within collagen type IV is required for angiogenesis and tumor growth in vivo. J. Cell Biol. 154, 1069–1079. Zigrino, P., Kamiguti, A.S., Eble, J., Drescher, C., Nischt, R., Fox, J.W., Mauch, C., 2002. The reprolysin jararhagin, a snake venom metalloproteinase, functions as a fibrillar collagen agonist involved in fibroblast cell adhesion and signaling. J. Biol. Chem. 277, 40528–40535.