ilable at ScienceDirect

Toxicon 79 (2014) 45–54

Contents lists ava

Toxicon

journal homepage: www.elsevier .com/locate/ toxicon

Inhibitory effects of recombinant RTS-jerdostatin on integrina1b1 function during adhesion, migration and proliferation ofrat aortic smooth muscle cells and angiogenesis

Gema Bolás a, Flávia Figueiredo de Rezende b, Carolina Lorente a, Libia Sanz a,Johannes A. Eble b, Juan J. Calvete a,*

a Instituto de Biomedicina de Valencia, CSIC, Valencia, Spainb Institute for Physiological Chemistry and Pathobiochemistry, University of Muenster, Waldeyerstr. 15, 48149 Muenster, Germany

a r t i c l e i n f o

Article history:Received 17 September 2013Received in revised form 22 November 2013Accepted 18 December 2013Available online 11 January 2014

Keywords:Recombinant RTS disintegrinJerdostatina1b1 integrin inhibitorCell adhesionAngiogenesis

* Corresponding author. Instituto de BiomedicinaJaime Roig 11, 46010 Valencia, Spain. Tel.: þ34 96 3369 0800.

E-mail address: jcalvete@ibv.csic.es (J.J. Calvete).

0041-0101/$ – see front matter � 2013 Elsevier Ltdhttp://dx.doi.org/10.1016/j.toxicon.2013.12.006

a b s t r a c t

Jerdostatin, a short RTS-disintegrin cloned from venom gland mRNA of Protobothrops jer-donii, selectively blocks the adhesion of a1b1 integrin to collagen IV. Integrin a1b1 is highlyexpressed in smooth muscle cells (SMC) surrounding small blood vessels and vascularendothelial cells. Vascular SMC adhesion, migration and proliferation are important pro-cesses during normal vascular development. Using recombinant jerdostatin we haveinvestigated the role of the a1b1 integrin on the adhesion of vascular SMC to collagen IV,and the potential relevance of blocking this crucial component of focal adhesions as ananti-angiogenic strategy. Our results show that jerdostatin does not interact with canonicalcollagen-binding site on the isolated A-domain of the a1 integrin subunit. r-Jerdostatininhibited the adhesion of RASMCs to immobilized CB3 fragment in a dose-dependentmanner, triggering to round-up, retraction, and finally detachment of the cells. r-Jerdos-tatin did not affect the adhesion of human SMCs to CB3, presumably because the highexpression of a2b1 integrin compensated for a1b1 integrin blockage by jerdostatin. r-Jerdostatin dose-dependently inhibited a1b1 integrin-dependent HUVEC tube formation.However, VEGF-driven tube formation in the matrigel assay was only completely abolishedwhen binding of integrin a2b1 to collagen was also inhibited by the C-type lectin-likerhodocetin. As a whole, our work emphasizes the relevance of using specific inhibitorsfor dissecting the role of a1b1 integrin in physiological and pathological conditions.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Venoms of Viperidae snakes contain b1 and b3 integrinantagonists, thedisintegrins, that represent a familyof small(40–84 amino acids), cysteine-rich polypeptides (Calvete,2010, 2013). The proper pairing of cysteines determinesthe conformation of amobile loop,which protrudes 14–17Åfrom the globular protein core and which harbors an

de Valencia, C.S.I.C.,39 1778; fax: þ34 96

. All rights reserved.

integrin-recognition tripeptide motif at its apex. The activeloop and the C-terminal tail exhibit concerted motions andform a conformational epitope involved in extensive in-teractions with both subunits of integrin receptors that ac-count for the specificity and selectivity of disintegrinstowards b1 and b3 integrin receptors (Monleon et al., 2003;Calvete, 2010; Carbajo et al., 2011). Short (R/K)TS-dis-integrins that selectively target the a1b1 integrin (Calveteet al., 2007), such as obtustatin (Marcinkiewicz et al.,2003; Brown et al., 2008), viperistatin (Staniszewska et al.,2009; Momic et al., 2011), lebestatin (Olfa et al., 2005) andjerdostatin (Sanz et al., 2005), cluster into a distinct cladewithin the disintegrin family (Sanz-Soler et al., 2012).

G. Bolás et al. / Toxicon 79 (2014) 45–5446

Jerdostatin, an RTS-disintegrin cloned from venom glandmRNA of Protobothrops jerdonii, selectively and effectivelyblocks the adhesion of a1b1 integrin-expressing-K562 cellto collagen IV in vitro (Sanz et al., 2005). Type IV collagen isan exclusive constituent of thebasementmembraneswhereit forms complex supramacromolecular networks that in-fluence cell adhesion, migration, and differentiation(Khoshnoodi et al., 2008). Jerdostatin, recombinantly pro-duced in a mammalian cell system, also inhibited thebinding of soluble a1b1 integrin to the CB3 fragment ofcollagen IV in a dose-dependent manner. Moreover, it dis-rupted the adhesion of glioblastoma (RuGli) cells to collagenIV (Juarez et al., 2010).

Cell proliferation, cytoskeletal reorganization, andmigration mediated via the Ras-Shc-MAPK pathway, aretriggered through a1b1 integrin-dependent signalingmechanisms (Pozzi et al., 1998), and are central processesfor normal vascular development and arterial repair afterinjury (Schapira et al., 2005). Integrin a1b1 is also highlyexpressed in vascular and visceral smooth muscle cells(SMC) surrounding small blood vessels, as well as inmicrovascular endothelial cells during vascular endothelialgrowth factor (VEGF)-induced tumor angiogenesis (Sengeret al., 1997, 2002). Compelling evidence supports the viewthat the ability of a primary tumor to grow beyond a fewmillimeters in size and tometastasize depends on its abilityto stimulate angiogenesis for proper nourishment andremoval of metabolic waste products (Folkman, 2006). Thedependency of tumor cell growth and metastasis on neo-vascularization holds promises in the development oftherapeutic strategies to treat aggressive cancers(Twardowski and Gradishar, 1997). Previous studies haveshown that naturally-occurring KTS-disintegrins, leb-estatin and obtustatin, inhibit endothelial cell proliferationand angiogenesis by specifically blocking the interaction ofa1b1 integrin with collagen IV (Olfa et al., 2005; Brownet al., 2008). However, the effects of a1b1-blocking dis-integrins on vascular smooth muscle cells (VSMC) have notbeen investigated. The aims of this study were to charac-terize how the RTS-disintegrin, jerdostatin, recombinantlyproduced in Escherichia coli (Sanz et al., 2005), inhibitsa1b1 integrin functions in adhesion, migration and prolif-eration of vascular smooth muscle cells, and to assesswhether it is able to block tubulogenesis of endothelialcells.

2. Materials and methods

2.1. Materials

Soluble human recombinant integrin a1b1 ectodomainwas produced in transfected Drosophila Schneider cells andpurified according to a previously reported protocol (Ebleet al., 2006). The collagen IV fragment CB3 was generatedas described (Kern et al., 1993). Antibodies used in thisstudy included: polyclonal anti-PEP160 against the C-ter-minal tail of jerdostatin (CKPSYPGNG) raised in rabbit byAbintek Biopharma S.L. (Parque Tecnológico de Bizkaia,Derio, Bizkaia, Spain) using a standard immunization pro-tocol; monoclonal antibody MAB1973 against the humanintegrin a1b1 (Merck Millipore, Darmstadt, Germany);

JA218 (anti integrin a2 subunit, purified according toTuckwell et al., 2000); and anti-human integrin b1 subunitraised in rabbit (Eble et al.,1993). KTS-disintegrin lebestatin[Q3BK14] was isolated from the venom of Macroviperalebetina as described (Bazaa et al., 2005). RecombinantRGD-disintegrin ocellatusin, was produced from an Echisocellatus venom gland cDNA library (Sanz-Soler et al.,2012). The GST-tagged A-domain of the integrin a1 sub-unit, produced according to Calderwood et al. (1997), wasdetected with a rabbit antiserum against glutathione-S-transferase (Invitrogen, Karlsruhe, Germany). C-typelectin-like rhodocetin was purified as described (Eble et al.,2009).

2.2. Cell lines

Rat aortic smooth muscle A7r5 cells (RASMC) werepurchased from ATCC (Manassas, VA, USA) and cultured inDulbecco’s modified Eagle’s medium (DMEM high glucosecontent; PAA, Germany) supplemented with 10% fetal calfserum (FCS, Invitrogen) containing 100 U/mL penicillin and0.1 mg/mL streptomycin. Primary human arterial smoothmuscle cells (HASMC) (PromoCell, Heidelberg, Germany)were grown in smooth muscle cell growth medium (Pro-mocell) supplemented with growth factors (Promocell), 5%fetal calf serum, 100 U/mL of penicillin and 0.1 mg/mLstreptomycin. Both cell lines were grown at 37 �C in a hu-midified incubator containing 5% CO2. Human fibrosarcomaHT 1080 cells were cultivated in DMEM supplemented with10% FCS, 100 U/mL penicillin and 0.1 mg/mL streptomycin.Human umbilical vein endothelial cells (HUVECs) were agift from Dr. Herminia González (Fundación Hospital Clín-ico, Valencia, Spain). Cells were cultured in endothelialmedium (GIBCO, Life technologies) supplemented with 10%FBS, growth factors (5 mg/mL EGF, 1.5 mg/mL bFGF), 0.1 mg/mL of gentamicin and amphotericin, and 1 mg/mL of hy-drocortisone (Gibco, Life technologies). The experimentswere carried out at the 4–6 cell passage.

2.3. Cloning, expression and purification of r-jerdostatin

Recombinant jerdostatin was produced in an E. coliexpression system as described (Sanz-Soler et al., 2012).The purity of the isolated protein was assessed by SDS-PAGE (see below) followed by Western blot andelectrospray-ionization mass spectrometry using a QTrap2000 instrument (Applied Biosystems) equipped with ananoelectrospray source (Proxeon, Denmark). Proteinconcentrationwas determined using the bicinchoninic acidprotein method (BCA� Protein Assay, Pierce).

2.4. SDS-PAGE and Western blot analysis

Recombinant jerdostatin was analyzed by Tris-Tricine-SDS-10%-PAGE under reducing conditions and visualizedby Coomassie blue staining. For immunoblot analysis, theSDS-PAGE gel was electro-transferred to PVDF (Hybond-P,GE Healthcare) membrane. Remaining binding sites on thePVDF membrane were blocked by incubationwith 5% (w/v)non-fat dried milk in PBS (20 mMNa2HPO4 pH 7.5, 150 mMNaCl) (blocking buffer) for 1 h at room temperature. The

G. Bolás et al. / Toxicon 79 (2014) 45–54 47

membrane was incubated overnight with a 1:500 (v/v)dilution of anti-PEP160 polyclonal antibody in blockingbuffer, washed three times (5 min each) with PBS con-taining 0.1% (v/v) Tween 20, and incubated with anti-rabbitIgG peroxidase-conjugated (Sigma), 1:10,000 (v/v) dilutionin blocking buffer, washed three times as above, andchemiluminescence developed using the ECL Plus kit (GEHealthcare).

2.5. Binding of r-jerdostatin to soluble a1b1 integrin

Integrin interaction with recombinant jerdostatin (r-jerdostatin) was assayed using an ELISA (Juarez et al.,2010). Briefly, the collagen type IV fragment CB3 wasimmobilized on a 96-well plate overnight at 4 �C in 20mMTris, 150 mM NaCl, pH 7.5 containing 2 mM MgCl2 (TBS/Mg2þ) at a concentration of 5 mg/mL. The plate waswashed three times with TBS/Mg2þ and blocked with 1%(v/v) BSA in TBS/Mg2þ at room temperature for 1 h. Sol-uble a1b1 integrin (at 3.5 mg/mL final concentration) or30 mg/mL of a1 A-domain, in the same buffer, were mixedwith different concentrations of recombinant jerdostatinand incubated for 2 h at room temperature. The plate waswashed twice with 50 mM Hepes (pH 7.5) containing150 mM NaCl, 2 mM MgCl2 and 1 mM MnCl2, and boundintegrin was fixed with 2.5% (v/v) glutaraldehyde in thesame buffer for 10 min. The a1b1 integrin was detectedusing a rabbit anti-b1 antiserum, 1:2000 (v/v) and a1 A-domain with a 1:1000 dilution of anti-GST (Glutation S-transferase rabbit Ig antiserum) for 1 h at room temper-ature. After washing, goat anti-rabbit IgG conjugated withalkaline phosphatase (AP) (1:2000) was added. Finally theAP-catalyzed conversion of pNpp was measured at405 nm.

2.6. Inhibition of cell adhesion on CB3 by r-jerdostatin

CB3 (1 mg/mL) and type I collagen (10 mg/mL) in PBS/Mg2þ were immobilized overnight at 4 �C on 96-wellplates. Wells were washed three times with PBS (pH 7.4)and blocked for 1 h with 0.1% BSA in PBS. RASMC, HASMCand HT 1080 cells were seeded at 0.2 � 106 cells/mL for45 min in the presence of different concentrations of r-jerdostatin and rhodocetin tetramer (abgd) at 37 �C, in a5% CO2 atmosphere. Non-adherent cells were washed-offwith warm PBS, and adherent cells were labeled with0.5 mg/mL 2N,7N-bis-2-carboxyethyl-5-(and-6)-carboxyfluorescein (BCECF, Invitrogen) in DMEM phenol red-freemedium (GIBCO) for 15 min at 37 �C under light-protection (De Santana Evangelista et al., 2009). After awashing step, cells were either fixed and analyzed underthe microscope, or lysed in order to release and quantifythe fluorescence of the incorporated BCECF. Fixation wasperformed with 4% formaldehyde in PBS for 10 min fol-lowed by washing with PBS. Images were acquired using20 times magnification with an Olympus IX71 microscopecontrolled by Metamorph (version 7.7.4.0, Molecular De-vices, Inc. Downingtown, PA, USA) software. Fluorescencemeasurement was obtained after cell lysis with 30 mMTris–HCl, pH 8.8 and 0.1% SDS for 10 min. Fluorescentsignals were detected with a microplate reader (BioTek;

excitation: 485 nm, emission: 535 nm). Non-inhibited and10 mM EDTA treated cells were used as positive andnegative controls, respectively.

2.7. Real time cell adhesion and migration of RASMC on CB3

Adhesion and migration of rat ASMC were carried outusing the xCELLigence RT-CA� system (Roche Diagnostics,Penzberg, Germany), following the manufacturer’s in-structions. Briefly, E- (electrode) and CIM- (Cell Invasionand Migration) plates were coated with 1 mg/mL of CB3overnight at 4 �C. The plates were washed three times withPBS and blocked with 0.1% BSA in PBS for 1 h at roomtemperature. 0.2 � 106 cells/mL were added and incubatedwith different concentrations of snake venom proteins. TheCIM-Plate contained growth medium supplemented with10% fibronectin–free FBS as a cell attractant. Cells treatedwith and without 10 mM EDTA were used as negative andpositive controls, respectively. Impedance values wererecorded as cell index every 3 min for 24 h. They representcell adhesion and migration on the E- and CIM-plate,respectively. Data analysis was performed using the RTCAsoftware 1.2 (Roche Diagnostics). The increase of the signalover time, determined as the slope (with a unit of h�1)indicates the migration speed of the cells.

2.8. Effect of r-jerdostatin on proliferation of RASMC

To quantify cell proliferation, the BrdU ProliferationAssay (Calbiochem�) was employed, following the manu-facturer’s instructions. Briefly, a 96 well-plate was coatedovernight with 1 mg/mL of CB3 in TBS/Mg2þ. The plate wasblockedwith 0.1% of BSA in PBS for 1 h. After blocking, wellswere washed with PBS three times and 10,000 cells/wellwere seeded with different concentrations of r-jerdostatinfor 1 h. A working solution of BrdU (1:2000) was added toeach well into the culture media. After a 24 h-incubation at37 �C and 5% CO2. cells were fixed, permeabilized and theDNA was denatured. This treatment allowed inmunode-tection with anti-BrdU monoclonal antibody, which wasquantified with a secondary horseradish peroxidase-conjugated goat anti-mouse antibody. Peroxidase-catalyzed conversion of the fluorogenic substrate wasquantified by fluorescence (excitation: 320 nm; emission:460 nm).

2.8.1. Endothelial tube formation assayA 48-well-plate was coated with 150 ml of Matrigel (BD

Bioscience) and allowed to polymerize for 30–60 min at37 �C and 5% CO2 a humidified atmosphere. BD Bioscience’sMatrigel matrix contains 5.0–7.5 ng/mL of VEGF. A sus-pension of 3.5 � 104 HUVECs on endothelial growth me-dium were seeded per well on top of the gel and differentconcentrations of jerdostatin, rhodocetin and a mixture ofboth were added. Cells were incubated for 12–16 h at 37 �C.The assay was performed in duplicate. The effect of thesnake venom components was viewed using an invertedmicroscope Leica DM IRE2 with a digital camera Leica DFC280 incorporated. Ten times magnification images wereacquired and analyzed with the Image J software.

Fig. 1. Inhibitory activity of r-jerdostatin on the binding of soluble a1b1 integrin to CB3 fragment. (A) 96-well plates coated with 5 mg/mL of CB3 wereincubated with a fixed concentration of a1b1 integrin (3.5 mg/mL) previously incubated with increasing concentrations of r-jerdostatin. Residual integrin bindingwas detected by ELISA. KTS-disintegrin lebestatin and RGD-ocellatusin were used as positive and negative inhibition controls, respectively. (B) Representation ofthe efficacy of jerdostatin to inhibit the binding of a1b1 integrin to immobilized CB3 fragment of collagen type IV. The calculated half maximal inhibitoryconcentration (IC50) of r-jerdostatin was 30 nM. (C) To assess the inhibitory activity of r-jerdostatin towards soluble a1 A-Domain binding to immobilized CB3fragment, 30 mg/mL of soluble a1 A-Domain were mixed with increasing concentrations of r-jerdostatin and the reaction mixture added to the wells of a mi-crotiter plate coated with 5 mg/mL of CB3. A colorimetric method was used to detect residual integrin binding.

G. Bolás et al. / Toxicon 79 (2014) 45–5448

2.9. Flow cytometry

To profile integrin expression on HASMC, flow cyto-metric analyses were carried out using the anti-collagen-binding integrin antibodies MAB1973 and JA-218. To thisend, cells were harvested with trypsin, adjusted to5�105 cells/mL per condition and blockedwith 1% BSA and1% horse serum in PBS for 1 h on ice. Cells were thenincubated with the primary antibodies (5 mg/mL), washedtwice with PBS and incubated with a 1:2000 (v/v) dilutionof goat anti-mouse antibody conjugated to R-phycoerythrin(Invitrogen) for one hour on ice. IgG from mouse serum(dilution 1:1000 (v/v)) (Sigma) was used as an isotypecontrol. Fluorescence was detected using a CyFlow�

fluo-rometer (Partec, Görlitz, Germany). Data were analyzed

Fig. 2. Integrin expression profile on HASMC. The cell surface expression of the coanalysis (CyFlow�). 20 � 103 events were counted per sample were gated using the(FL1) was determined with the WinMDI 2.9 software. The histograms represent thcontrol (CTR).

with the software Windows Multiple Document Interface(WinMDI) 2.9 (Miscellaneous software, The ScrippsResearch Institute, CA, USA), and plotted as histogramsusing Microsoft Excel.

2.10. Statistical analysis

Data were analysed using GraphPad Prism 5.0 program.Data in Figs. 2 and 3 were approximated with a non-linearregression curve. Differences in adhesion and migrationand tube formation were assessed by using One-WayANOVA statistic and Bonferoni test; and Two-WayANOVA. All data are shown as mean � SEM of at leastthree independent measurements. P values �0.05 wereconsidered significant.

llagen-binding integrins a1b1 and a2b1 was determined by flow cytometricparameters of relative size (FSC) and granularity (SSC). Fluorescence intensitye number of fluorescent HASMCs (A) in comparison to the negative isotype

Fig. 3. Jerdostatin inhibits the adhesion of RASMC on CB3 fragment. Cells were seeded on a 96-well plate coated with CB3 (1 mg/mL) and with increasingconcentrations of r-jerdostatin. In (A) cell adhesion was determined by ELISA. Results represents mean � standard error, N ¼ 4 for each condition; ns: notsignificant *p < 0.05; **p < 0.01; ***p < 0.001. (B) For adhesion assays, cells were incubated with r-jerdostatin at 1500 nM (c), 150 nM (d), 15 nM (e) and 1.5 nM (f).Positive (a) and negative (10 mM EDTA) (b) controls were performed in parallel.

G. Bolás et al. / Toxicon 79 (2014) 45–54 49

3. Results and discussion

3.1. Expression and purification of recombinant jerdostatin

E. coli BL21 strain transformed with pET32a/TEV-jerdostatin plasmids overexpressed soluble jerdostatin-thioredoxin-His6 fusion protein (Sanz-Soler et al., 2012).Western blotting was used to follow the presence of jer-dostatin through the different purification steps.Electrospray-ionization mass spectrometry confirmed thecorrect primary sequence of the purified recombinant jer-dostatin molecule (calculatedmonoisotopic mass including4 disulphide bonds: 4765.4 Da).

3.2. Recombinant jerdostatin blocks the binding of solublea1b1 integrin to CB3

The interaction of the recombinant jerdostatin witha1b1 integrin was tested in an inhibition ELISA assay, in

which the binding of soluble a1b1 integrin to immobilizedcollagen fragment CB3 was challenged by addingincreasing concentrations of r-jerdostatin. The recombi-nant disintegrin dose-dependently blocked the binding ofa1b1 to CB3 with an IC50 of 30 nM (Fig. 1A and B). However,r-jerdostatin had no effect on the binding to immobilizedCB3-fragment of the isolated a1 A-domain (Fig. 1C), indi-cating that the isolated collagen-binding domain of thea1b1 integrin does not harbor the high-affinity r-jerdos-tatin recognition site. These results indicate that, contraryto RGD-disintegrins and a2b1 integrin-blocking C-typelectin-like proteins, all of which target the ligand bindingsite of their cognate integrin receptor, jerdostatin does notinteract with the isolated A-domain of the a1 integrinsubunit. This is in concordance with a previous reportshowing that the short KTS-disintegrin obtustatin did notimpaired the interaction of recombinant a1 subunit A-domain to collagen IV (Marcinkiewicz et al., 2003). Simi-larly, arresten (the C-terminal NC1 domain of collagen IV)

Fig. 4. RASMC attachment in the presence of increasing concentrations of jerdostatin. (A) Time-dependency of c RASMC attachment as a function of the r-jerdostatin concentration specified at the right. RASMC, positive control: no r-jerdostatin added; 10 mM EDTA, negative control: binding of RASMC aftertreatment with 10 mM EDTA. (B) Signals of RASMC attachment on CB3 fragment after incubation with r-jerdostatin for 1 h. (C) Dose-dependency of jerdostatin’sinhibitory potential after 45 min of treatment (IC50 ¼ 25.6 nM). ns: not significant; * denotes p < 0.05; **, p < 0.01; ***, p < 0.001.

G. Bolás et al. / Toxicon 79 (2014) 45–5450

another specific antagonist of integrin a1b1 (Sudhakaret al., 2005), does not comprise the canonical collagen tri-ple helical integrin-binding motif recognized by integrinsa1b1 (Eble et al., 1993) and a2b1 (Knight et al. 1998, 2000),clearly indicating the occurrence of yet unrevealed integrinbinding sites in collagen IV.

3.3. Integrin expression profile of ASMC

The expression of a1b1 and a2b1 integrins changesduring the transition from the contractile to the syntheticVSMC phenotype (Hedin et al., 1988; Gotwals et al., 1996;Bunni et al., 2011). The predominant phenotype in normalblood vessels is the quiescent or differentiated contractilephenotype, which functions as the regulator of vasodila-tation, vasoconstriction and blood flow, whereas dediffer-entiated SMCs aremotile and exhibit a synthetic phenotypeused for vascular remodeling. Flow-cytometric analysis offreshly isolated human aortic (hA) SMCs showed highexpression of the a1b1 integrin, while a2b1 integrin wasabsent, indicating that these cells bear a resting, contractilephenotype (Clyman et al., 1990). On the other hand, and inline with previous report (Skinner et al., 1994; Belkin et al.,1990), cultured human VSMCs mainly expressed a2, a3, a5,av, and b1 integrin subunits, whereas the al subunit wassignificantly downregulated after the first few passages.Although Flow Cytometry is not a quantitative technique,the differences observed between the FACS profiles dis-played in Fig. 2 suggest that there is a higher expression ofa2 than a1 integrin subunit, and that the lack of a1b1integrin can be compensated by overexpression of anothercollagen-binding integrin, presumably a2b1.

3.4. Inhibition of the adhesion of RASMCs

Rat ASMCs utilize predominantly b1 integrins to attachto surfaces coated with fibronectin, laminin, and collagentypes I and IV (Clyman et al., 1990). The a1b1 integrin re-ceptor has higher affinity for type IV than for type I collagen(Syfrig et al., 1991; Kern et al., 1993). Our results show thatr-jerdostatin inhibited the adhesion of RASMCs to immo-bilized CB3 fragment in a dose-dependent manner(Fig. 3A). In addition, treatment of RASMCs with r-jerdos-tatin led to round-up, retraction, and finally detachment ofthe cells (Fig. 3B). Time-course analysis corroborated that r-jerdostatin induced detachment of proliferating RASMCs(Fig. 4). On the other hand, r-jerdostatin did not affect theadhesion of human SMCs to CB3 (Fig. 5A), presumablybecause the high expression of a2b1 integrin compensatedfor a1b1 integrin blockage by jerdostatin (Skinner et al.,1994). To check this possibility, we used rhodocetin, a C-type lectin-like protein from Calloselasma rhodostomavenom that selectively blocks the binding of integrin a2b1to collagens type I and IV (Eble et al., 2001; Eble et al.,2002). As positive control, rhodocetin completelyimpaired the a2b1 integrin-mediated attachment of fibro-sarcoma HT1080 cells to collagen. Rhodocetin also inhibi-ted HASMC adhesion to CB3 by about 54%, whereas r-jerdostatin hardly had any effect (less than 6% inhibition).In contrast, when r-jerdostatin and rhodocetin weresimultaneously added, HASMC attachment to CB3 wasentirely prevented (Fig. 5B). The finding that a completeinhibition of the adhesion of HASMC to CB3 requires an-tagonists of both collagen-binding integrins, a1b1 anda2b1, may be relevant for the prevention of vascular

Fig. 5. Effect of r-jerdostatin on HASMC adhesion. (A) A 96-well plate coated with CB3 (1 mg/mL) was seeded with HASMC in the presence of incremental serialdilutions (1:2) of r-jerdostatin. (B) HT1080 and HASM cells were seed on a 96 well plate coated with rhodocetin (5 mg/mL), r-jerdostatin (3 mg/mL), and a mixtureof both. Fluorescence was determined with an ELISA plate reader. Addition of 10 mM of EDTA was used as negative control in all experiments. N ¼ 4 for eachcondition; *p < 0.05; **p < 0.01; ***p < 0.001.

G. Bolás et al. / Toxicon 79 (2014) 45–54 51

pathology such as neointimal hyperplasia during reste-nosis, as proliferating VSMCs likely express both collagen-binding integrins, a1b1 and a2b1.

3.5. Inhibition of RASMC migration

VSM cell migration is involved in vessel wall formationduring neovascularization, atherosclerosis or vascularinjury. Pro-migratory effects have been ascribed to themajor constituents of the vascular extracellular matrix,collagens I and IV (Nelson et al., 1996). Hence, to testwhether a1b1 plays a role in the migration of rat ASMCs oncollagen, we monitored cell migration through CB3-coatedfilters by impedance measurement in real-time. Even at thehighest concentrations of r-jerdostatin, only a weak inhi-bition of not more than 40% was detected. Rhodocetinexerted an even lower inhibitory effect than r-jerdostatin,

Fig. 6. RASMC migration on CB3 was not affected by blocking collagen-binding integrins. Graphic representation of the effect of r-jerdostatinand rhodocetin on the migration of RASMC on CB3 after one hour treatment.A CIM-plate was coated with 1 mg/mL of CB3, RASMC were seeded andtreated with different concentrations of r-jerdostatin and rhodocetin. Areduction in slope (migration speed) indicates an increased inhibitory po-tential of the antagonist(s).

and the combined action of the two venom proteins wassimilar as that of r-jerdostatin alone (Fig. 6).

The rate of cell migration is determined by the adhesionstrength in a non-linear manner (Palecek et al., 1997). If cellattachment is too weak, cells cannot migrate due to aslippery substratum, an effect which is overcome by anincrease in adhesion strength. However, if adhesionstrength is too strong, cells stick firmly to their substratum,thereby preventing cell movement. Thus, an optimalintegrin-mediated adhesion strength is required for high-est migration speeds (Louis and Zahradka, 2010). Our re-sults suggest that migration of rat ASMC on CB3 is partlydepended on a1b1 but not on a2b1 integrin. In addition,the failure to block cell migration employing specific an-tagonists of integrins, a1b1 and a2b1 indicates thatRASMCs use an alternative migratory mechanism, which isthought to be mediated via b3 integrin (Smyth et al., 2001).During the early phase of SMC migration, b1 and b3integrins exhibit a different cellular distribution. b1 integ-rins rapidly cluster into focal plaques thereby anchoring the

Fig. 7. Proliferation of RASMC on CB3. A 96 well-plate was coated with CB3(1 mg/mL). After blocking with 0.1% of BSA in PBS, RASMCs were seeded inthe presence of decreasing concentrations of r-jerdostatin. BrdU was added24 h later. For inmunodetection anti-BrdU signal was quantified at 320 nmexcitation wavelength and 460 nm emission wavelength. The relative fluo-rescence signal (RFU) was calculated as the ratio of the signal obtained froma labeled sample (þBrdU) to an unlabeled sample (�BrdU) after subtractionof the endogenous fluorescence of buffer controls.

Fig. 8. Inhibitory effect of r-jerdostatin on endothelial tube formation.A) HUVECs plated on a 48 well-plate covered with matrigel matrix. Controlcells without matrigel (a), HUVECs on matrigel (b); on matrigel and humancollagen IV (c). HUVEC tube formation evaluated after 16 h in the presence ofdifferent concentrations of jerdostatin, 2.5 mg/mL (d), 5 mg/mL(e,f), 10 mg/mL(g); 5 mg/mL of rhodocetin (h) and 5 mg/mL of rhodocetin and 10 mg/mL ofjerdostatin (i). B) Graphic representation of the average number of tubesformed.

G. Bolás et al. / Toxicon 79 (2014) 45–5452

cells, while avb3 integrin plays a still elusive role in rataortic SMC cell migration (Clyman et al., 1990,1992; Li et al.,2010).

3.6. Effect of r-jerdostatin on the proliferation of rat ASMCs

Proliferation of SMCs is an important response tovascular injury and neointima formation during athero-sclerosis. Cell survival and proliferation are tightly regu-lated by integrin-dependent adhesion to the extracellularmatrix. Although integrin a1b1 has been reported to play arole in the regulation of cell proliferation by triggering theRas-MAPK pathway via the adapter protein Shc (Coloradoet al., 2000; Brown et al., 2008), r-jerdostatin failed toinhibit the proliferation of VSMCs (Fig. 7). Anti-proliferativeactivity of another selective antagonist of integrin a1b1, theKTS-disintegrin obtustatin, on endothelial cells was relatedto induction of apoptosis (Brown et al., 2008). We have notanalyzed possible apoptotic effects of jerdostatin. However,our result are in concordance with those reported by Bunniet al. (2011) showing that a KTS-containing peptide did notcompletely inhibit proliferation of AngII-induced VSMC.

These pieces of evidence suggest that in addition to a1b1integrin other receptors may be involved in the prolifera-tion of VSMCs.

3.7. Inhibition of HUVEC tube formation

Angiogenesis, the formation of new vessels from exist-ing vasculature, requires the migration and proliferation ofendothelial cells, which reorganize to form a tubularnetwork of vessel-like structures. In order to investigate r-jerdostatin effect on HUVEC angiogenesis, a tube formationassay on matrigel was performed. At concentrations of 2.5–10 mg/mL r-jerdostatin dose-dependently inhibited a1b1integrin-dependent HUVEC tube formation. At 10 mg/mL,the number of branching points decreased substantially(Fig. 8). Again, binding of integrin a2b1 to collagen wasspecifically inhibited by the C-type lectin-like rhodocetin,and in the presence of both integrin antagonists, VEGF-driven tube formation in the matrigel assay was almostcompletely abolished (Fig. 8).

4. Concluding remarks

A conditional b1-integrin knockout mouse modelrevealed the importance of b1-integrins for vasculogenesisduring development (Abraham et al., 2008) and in adultmice (Liu and Leask, 2012). On the other hand, theknockout mice (a1-/-) showed a normal development of thevascular system (Gardner et al., 1996; Louis et al., 2007) buta reduction of tumor angiogenesis was observed (Pozziet al., 2000). Hence, a1b1 seemly plays different rolesdepending on which cell type within the vessel wall ex-presses this integrin receptor. Adhesion, migration andproliferation of vascular smooth muscle cells are ruled bydistinct integrin-mediated mechanisms. Our studies on theeffects of r-jerdostatin on integrin a1b1 functions duringadhesion, migration and proliferation of rat aortic smoothmuscle cells and angiogenesis highlight both the relevanceof a1b1 integrin on VSMC focal adhesions, and emphasizesthe usefulness of a specific a1b1 integrin inhibitor as a toolfor studying the role of this collagen receptor in physio-logical and pathological conditions.

5. Ethical statement

International ethical guidelines for scientific paperswere followed in the preparation of this manuscript.

Acknowledgments

This work is part of GB’s PhD Thesis (JAE-Predoctoralprogram) and has been partly financed by grants BFU2007-61563 and BFU2010-17273 from the Spanish Ministries ofScience and Innovation and Economy and Competitiveness(Madrid, Spain), and PROMETEO/2010/005 from the Gen-eralitat Valenciana (to JJC). J.A.E receives financial supportfrom the Deutsche Forschungsgemeinschaft (SFB/TR23project A8). GBwas recipient of a JAE short-term fellowshipin JA Eble’s laboratory.

G. Bolás et al. / Toxicon 79 (2014) 45–54 53

Appendix A. Supplementary data

Supplementary data related to this article can be foundat http://dx.doi.org/10.1016/j.toxicon.2013.12.006.

Conflict of interest

The authors declare that there are no conflicts ofinterests.

References

Abraham, S., Kogata, N., Fässler, R., Adams, R.H., 2008. Integrin b1 subunitcontrols mural cell adhesion, spreading, and blood vessel wall sta-bility. Circ. Res. 102, 562–570.

Bazaa, A., Marrakchi, N., El Ayeb, M., Sanz, L., Calvete, J.J., 2005. Snakevenomics: comparative analysis of the venom proteomes of theTunisian snakes Cerastes cerastes, Cerastes vipera and Macroviperalebetina. Proteomics 5, 4223–4235.

Belkin, V.M., Belkin, A.M., Koteliansky, V.E., 1990. Human smooth muscleVLA-1 integrin: purification, substrate specificity, localization inaorta, and expression during development. J. Cell Biol. 111, 2159–2170.

Brown, M.C., Staniszewska, I., Del Valle, L., Tuszynski, G.P.,Marcinkiewicz, C., 2008. Angiostatic activity of obtustatin as a1b1integrin inhibitor in experimental melanoma growth. Int. J. Cancer123, 2195–2203.

Bunni, M.A., Kramarenko, I.I., Walker, L., Raymond, J.R.,Garnovskaya, M.N., 2011. Role of integrins in angiotensin II-inducedproliferation of vascular smooth muscle cells. Am. J. Physiol. CellPhysiol. 300, C647–C656.

Calderwood, D.A., Tuckwell, D.S., Eble, J., Kühn, K., Humphries, M.J., 1997.The integrin a1 A-domain is a ligand binding site for collagens andlaminin. J. Biol. Chem. 272, 12311–12317.

Calvete, J.J., 2010. Brief history and molecular determinants of snakevenom disintegrin evolution. In: Kini, R.M., Markland, F.,McLane, M.A., Morita, T. (Eds.), Toxins and Hemostasis: from Bench toBedside. Springer, Amsterdam, pp. 285–300 (Chapter 18).

Calvete, J.J., 2013. The continuing saga of snake venom disintegrins.Toxicon 62, 40–49.

Calvete, J.J., Marcinkiewicz, C., Sanz, L., 2007. KTS and RTS-disintegrins:anti-angiogenic viper venom peptides specifically targeting the a1b1integrin. Curr. Pharm. Des. 13, 2853–2859.

Carbajo, R.J., Sanz, L., Mosulen, S., Perez, A., Marcinkiewicz, C., Pineda-Lucena, A., Calvete, J.J., 2011. NMR structure and dynamics of re-combinant wild type and mutated jerdostatin, a selective inhibitor ofintegrin a1b1. Proteins 79, 2530–2542.

Clyman, R.I., Turner, D.C., Kramer, R.H., 1990. An a1b1-like integrin re-ceptor on rat aortic smooth muscle cells mediates adhesion to lami-nin and collagen types I and IV. Arteriosclerosis 10, 402–409.

Clyman, R.I., Mauray, F., Kramer, R.H., 1992. b1 and b3 integrinshave different roles in the adhesion and migration of vascularsmooth muscle cells on extracellular matrix. Exp. Cell Res. 200,272–284.

Colorado, P.C., Torre, A., Kamphaus, G., Maeshima, Y., Hopfer, H.,Takahashi, K., Volk, R., Zamborsky, E.D., Herman, S., Sarkar, P.K.,Ericksen, M.B., Dhanabal, M., Simons, M., Post, M., Kufe, D.W.,Weichselbaum, R.R., Sukhatme, V.P., Kalluri, R., 2000. Anti-angiogeniccues from vascular basement membrane collagen. Cancer Res. 60,2520–2526.

De Santana Evangelista, K., Andrich, F., Figueiredo de Rezende, F.,Niland, S., Cordeiro, M.N., Horlacher, T., Castelli, R., Schmidt-Hederich, A., Seeberger, P.H., Sanchez, E.F., Richardson, M., Gomes deFigueiredo, S., Eble, J.A., 2009. Plumieribetin, a fish lectin homologousto mannose-binding B-type lectins, inhibits the collagen-binding a1b1integrin. J. Biol. Chem. 284, 34747–34759.

Eble, J.A., Golbik, R., Mann, K., Kühn, K., 1993. The a1b1 integrin recogni-tion site of the basement membrane collagen molecule [a1(IV)]2a2(IV). EMBO J. 12, 4795–4802.

Eble, J.A., Beermann, B., Hinz, H.J., Schmidt-Hederich, A., 2001. a2b1integrin is not recognized by rhodocytin but is the specific, high af-finity target of rhodocetin, an RGD-independent disintegrin andpotent inhibitor of cell adhesion to collagen. J. Biol. Chem. 276, 12274–12284.

Eble, J.A., Niland, S., Dennes, A., Schmidt-Hederich, A., Bruckner, P.,Brunner, G., 2002. Rhodocetin antagonizes stromal tumor invasion

in vitro and other a2b1 integrin-mediated cell functions. Matrix Biol.21, 547–558.

Eble, J.A., Kassner, A., Niland, S., Morgelin, M., Grifka, J., Grassel, S., 2006.Collagen XVI harbors an integrin a1b1 recognition site in its C-ter-minal domains. J. Biol. Chem. 281, 25745–25756.

Eble, J.A., Niland, S., Bracht, T., Mormann, M., Peter-Katalinic, J.,Pohlentz, G., Stetefeld, J., 2009. The a2b1 integrin-specific antagonistrhodocetin is a cruciform, heterotetrameric molecule. FASEB J. 23,2917–2927.

Folkman, J., 2006. Angiogenesis. Annu. Rev. Med. 57, 1–18.Gardner, H., Kreidberg, J., Koteliansky, V., Jaenisch, R., 1996. Deletion of

integrin a1 by homologous recombination permits normal murinedevelopment but gives rise to a specific deficit in cell adhesion. Dev.Biol. 175, 301–313.

Gotwals, P.J., Chi-Rosso, G., Lindner, V., Yang, J., Ling, L., Fawell, S.E.,Koteliansky, V.E., 1996. The a1b1 integrin is expressed during neo-intima formation in rat arteries and mediates collagen matrix reor-ganization. J. Clin. Invest 97, 2469–2477.

Hedin, U., Bottger, B.A., Forsberg, E., Johansson, S., Thyberg, J., 1988.Diverse effects of fibronectin and laminin on phenotypic properties ofcultured arterial smooth muscle cells. J. Cell Biol. 107, 307–319.

Juarez, P., Bolás, G., de Rezende, F.F., Calvete, J.J., Eble, J.A., 2010. Recom-binant expression in human cells of active integrin a1b1-blockingRTS-disintegrin jerdostatin. Toxicon 56, 1052–1058.

Kern, A., Eble, J., Golbik, R., Kühn, K., 1993. Interaction of type IV collagenwith the isolated integrinsa1b1 and a2b1. Eur. J. Biochem215,151–159.

Khoshnoodi, J., Pedchenko, V., Hudson, B.G., 2008. Mammalian collagenIV. Microsc. Res. Tech. 71, 357–370.

Knight, C.G., Morton, L.F., Onley, D.J., Peachey, A.R., Messent, A.J.,Smethurst, P.A., Tuckwell, D.S., Farndale, R.W., Barnes, M.J., 1998.Identification in collagen type I of an integrin a2b1-binding sitecontaining an essential GER sequence. J. Biol. Chem. 273, 33287–33294.

Knight, C.G., Morton, L.F., Peachey, A.R., Tuckwell, D.S., Farndale, R.W.,Barnes, M.J., 2000. The collagen-binding A-domains of integrins a1b1and a2b1 recognize the same specific amino acid sequence, GFOGER,in native (triple-helical) collagens. J. Biol. Chem. 275, 35–40.

Li, G., Jin, R., Norris, R.A., Zhang, L., Yu, S., Wu, F., Markwald, R.R.,Nanda, A., Conway, S.J., Smyth, S.S., Granger, D.N., 2010. Periostinmediates vascular smooth muscle cell migration through the integ-rins avb3 and avb5 and focal adhesion kinase (FAK) pathway.Atherosclerosis 208, 358–365.

Liu, S., Leask, A., 2012. Integrin b1 is required for maintenance of vasculartone in postnatal mice. J. Cell Commun. Signal 6, 175–180.

Louis, S.F., Zahradka, P., 2010. Vascular smooth muscle cell motility: frommigration to invasion. Exp. Clin. Cardiol. 15, e75–e85.

Louis, H., Kakou, A., Regnault, V., Labat, C., Bressenot, A., Gao-Li, J.,Gardner, H., Thornton, S.N., Challande, P., Li, Z., Lacolley, P., 2007. Role ofa1b1-integrin in arterial stiffness and angiotensin-induced arterialwallhypertrophy in mice. Am. J. Physiol. Heart Circ. Physiol. 293, 597–604.

Marcinkiewicz, C., Weinreb, P.H., Calvete, J.J., Kisiel, D.G., Mousa, S.A.,Tuszynski, G.P., Lobb, R.R., 2003. Obtustatin: a potent selective in-hibitor of a1b1 integrin in vitro and angiogenesis in vivo. Cancer Res.63, 2020–2023.

Momic, T., Arlinghaus, F.T., Arien-Zakay, H., Katzhendler, J., Eble, J.A.,Marcinkiewicz, C., Lazarovici, P., 2011. Pharmacological aspects ofVipera xantina palestinae venom. Toxins 3, 1420–1432.

Monleon, D., Moreno-Murciano, M.P., Kovacs, H., Marcinkiewicz, C.,Calvete, J.J., Celda, B., 2003. Concerted motions of the integrin-bindingloop and the C-terminal tail of the non-RGD disintegrin obtustatin. J.Biol. Chem. 278, 45570–45576.

Nelson, P.R., Yamamura, S., Kent, K.C., 1996. Extracellular matrix proteinsare potent agonists of human smooth muscle cell migration. J. Vasc.Surg. 24, 25–33.

Olfa, K.Z., José, L., Salma, D., Amine, B., Najet, S.A., Nicolas, A., Maxime, L.,Raoudha, Z., Kamel, M., Jacques, M., Jean-Marc, S., Mohamed, E.A.,Naziha, M., 2005. Lebestatin, a disintegrin from Macrovipera venom,inhibits integrin-mediated cell adhesion, migration and angiogenesis.Lab. Investig. 85, 1507–1516.

Palecek, S.P., Loftus, J.C., Ginsberg, M.H., Lauffenburger, D.A., Horwitz, A.F.,1997. Integrin-ligand binding properties govern cell migration speedthrough cell-substratum adhesiveness. Nature 385, 537–540.

Pozzi, A., Moberg, P.E., Miles, L.A., Wagner, S., Soloway, P., Gardner, H.A.,2000. Elevated matrix metalloprotease and angiostatin levels inintegrin a1 knockout mice cause reduced tumor vascularization. Proc.Natl. Acad. Sci. U. S. A. 97, 2202–2207.

Pozzi, A., Wary, K.K., Giancotti, F.G., Gardner, H.A., 1998. Integrinalpha1beta1 mediates a unique collagen-dependent proliferationpathway in vivo. J. Cell Biol. 142, 587–594.

G. Bolás et al. / Toxicon 79 (2014) 45–5454

Sanz, L., Chen, R.Q., Pérez, A., Hilario, R., Juárez, P., Marcinkiewicz, C.,Monleón, D., Celda, B., Xiong, Y.L., Pérez-Payá, E., Calvete, J.J., 2005.cDNA cloning and functional expression of jerdostatin, a novel RTS-disintegrin from Trimeresurus jerdonii and a specific antagonist ofthe a1b1 integrin. J. Biol. Chem. 280, 40714–40722.

Sanz-Soler, R., Lorente, C., Company, B., Sanz, L., Juárez, P., Pérez, A.,Zhang, Y., Jin, Y., Chen, R., Eble, J.A., Calvete, J.J., Bolás, G., 2012. Re-combinant expression of mutants of the Frankenstein disintegrin,RTS-ocellatusin. Evidence for the independent origin of RGD and KTS/RTS disintegrins. Toxicon 60, 665–675.

Schapira,K., Lutgens,E., deFougerolles,A., Sprague,A., Roemen,A.,Gardner,H.,Koteliansky, V., Daemen, M., Heeneman, S., 2005. Genetic deletion orantibody blockade of a1b1 integrin induces a stable plaque phenotype inApoE-/- mice. Arterioscler. Thromb. Vasc. Biol. 25, 1917–1924.

Senger, D.R., Claffey, K.P., Benes, J.E., Perruzzi, C.A., Sergiou, A.P.,Detmar, M., 1997. Angiogenesis promoted by vascular endothelialgrowth factor: regulation through a1b1 and a2b1 integrins. Proc. Natl.Acad. Sci. U. S. A. 94, 13612–13617.

Senger, D.R., Perruzzi, C.A., Streit, M., Koteliansky, V.E., deFougerolles, A.R., Detmar, M., 2002. The a1b1 and a2b1 integrinsprovide critical support for vascular endothelial growth factorsignaling, endothelial cell migration, and tumor angiogenesis. Am. J.Pathol. 160, 195–204.

Skinner, M.P., Raines, E.W., Ross, R., 1994. Dynamic expression of a1b1 anda2b1 integrin receptors by human vascular smooth muscle cells. a2b1

integrin is required for chemotaxis across type I collagen-coatedmembranes. Am. J. Pathol. 145, 1070–1081.

Smyth, S.S., Reis, E.D., Zhang, W., Fallon, J.T., Gordon, R.E., Coller, B.S., 2001.b3-integrin-deficient mice but not P-selectin-deficient mice developintimal hyperplasia after vascular injury: correlation with leukocyterecruitment to adherent platelets 1 hour after injury. Circulation 103,2501–2507.

Staniszewska, I., Walsh, E.M., Rothman, V.L., Gaathon, A., Tuszynski, G.P.,Calvete, J.J., Lazarovici, P., Marcinkiewicz, C., 2009. Effect of VP12 andviperistatin on inhibition of collagen-receptor-dependent melanomametastasis. Cancer Biol. Ther. 8, 1507–1516.

Sudhakar, A., Nyberg, P., Keshamouni, V.G., Mannam, A.P., Li, J.,Sugimoto, H., Cosgrove, D., Kalluri, R., 2005. Human alpha1 type IVcollagen NC1 domain exhibits distinct antiangiogenic activity medi-ated by a1b1 integrin. J. Clin. Investig. 115, 2801–2810.

Syfrig, J., Mann, K., Paulsson, M., 1991. An abundant chick gizzard integrinis the avian a1b1 integrin heterodimer and functions as a divalentcation-dependent collagen IV receptor. Exp. Cell Res. 194, 165–173.

Tuckwell, D.S., Smith, L., Korda, M., Askari, J.A., Santoso, S., Barnes, M.J.,Farndale, R.W., Humphries, M.J., 2000. Monoclonal antibodies identifyresidues 199–216 of the integrin a2 vWFA domain as a functionallyimportant region within a2b1. Biochem. J. 350, 485–493.

Twardowski, P., Gradishar, W.J., 1997. Clinical trials of antiangiogenicagents. Curr. Opin. Oncol. 9, 584–589.