ilable at ScienceDirect

Toxicon 58 (2011) 398–409

Contents lists ava

Toxicon

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

Osteopontin, a chemotactic protein with cytokine-like properties,is up-regulated in muscle injury caused by Bothrops lanceolatus(fer-de-lance) snake venom

Valéria Barbosa-Souza a,b, Daniel Kiss Contin a,b, Waldemar Bonventi Filho c,Albetiza Lôbo de Araújo a, Silvia Pierre Irazusta b,c,d, Maria Alice da Cruz-Höfling b,*

aDepartamento de Farmacologia, Faculdade de Ciências Médicas, Universidade Estadual de Campinas (UNICAMP), CP 6111, 13081-970 Campinas, SP, BrazilbDepartamento de Histologia e Embriologia, Instituto de Biologia, Universidade Estadual de Campinas (UNICAMP), CP 6109, 13083-970, Campinas, SP, Brazilc Faculdade de Tecnologia de Sorocaba, Centro Estadual de Educação Tecnológica Paula Souza (CEETPS), 18013-280 Sorocaba, SP, Brazild Programa de Pós-Graduação, CEETPS, 01124-010 São Paulo, SP, Brazil

a r t i c l e i n f o

Article history:Received 15 March 2011Received in revised form 18 July 2011Accepted 19 July 2011Available online 5 August 2011

Keywords:Bothropic venomMyonecrosisMuscle regenerationOsteopontinMyogeninCD68 macrophage

* Corresponding author. Departamento de HistoInstituto de Biologia, Universidade Estadual de Cam6109, 13081-970 Campinas, SP, Brazil. Tel.: þ55 19 353289 3124.

E-mail address: hofling@unicamp.br (M.A. da Cr

0041-0101� 2011 Elsevier Ltd.doi:10.1016/j.toxicon.2011.07.011

Open access under the E

a b s t r a c t

Osteopontin (OPN) is a chemotactic, adhesive protein whose receptors include someintegrins and matrix proteins known to have role in inflammatory and repair processes.We examined the time course of OPN expression at acute and chronic stages after intra-muscular injection of Bothrops lanceolatus venom in rats. Additionally, we examined theexpression of CD68 (a marker for phagocytic macrophages) and the myogenic factors,myoD and myogenin. There was a biphasic upregulation of OPN (6–48 h and 3–14 dayspost-venom), i.e., during acute inflammation and myogenic cell proliferation and differ-entiation phases. OPN was detected in CD68 þ macrophages, fibroblasts, normal anddamaged myofibers, myoblasts and myotubes. Myogenin was expressed in the cytoplasm(atypical pattern) and nucleus of myoblasts and myotubes from 18 h to 7 days, after whichit was expressed only in nuclei. Macrophage numbers, OPN and myogenin expression werestill elevated at 7, 14 and 7 days. At 3 days, when OPN achieved the peak, some clusters ofmyoblasts were within regions of intense collagen deposition. Fibrosis may representlimitation for repairing processes and may explain the small diameter of regenerated fibersat 21 days post-venom. The expression of OPN in the course of venom-induced damageand regeneration suggests stages-specific mediation role along the whole process.

� 2011 Elsevier Ltd. Open access under the Elsevier OA license.

1. Introduction

Skeletal muscle regeneration is possible because of theexistence of undifferentiated satellite cells located betweenthe basal membrane and fiber sarcolemma (Mauro, 1961).In response to mechanical, nutritional, hormonal, chemicaland toxic stimuli, satellite cells proliferate and differentiate

logia e Embriologia,pinas (UNICAMP), CP21 6224; fax: þ55 19

uz-Höfling).

lsevier OA license.

to repair damaged fibers, or fuse to pre-existing myofibersto increase their size or form new fibers; the overall resultof these processes is an increase in muscle mass. Under-standing the mechanisms that regulate these myogenicprocesses is central in the development of therapeuticstrategies to deal with muscle disabilities.

Envenoming by Bothrops snakes (family Viperidae) ischaracterized by prominent local damage mediated bymyotoxins, hemorrhagins and hemolytic enzymes of thevenom (Gutiérrez and Ownby, 2003; Gutiérrez et al., 2005);such damage can result in important morbidity at the bitesite and impairment of muscle function (Rosenfeld, 1971;França and Málaque, 2003).

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409 399

Bothrops lanceolatus (fer-de-lance) is responsible formost snakebites on the Caribbean island of Martinica(Thomas et al., 1995). Compared to other Bothrops species,B. lanceolatus venom is less myotoxic (Bogarín et al., 1999;Gutiérrez et al., 2008), but induces thrombosis in humans(Thomas et al., 1995; Malbranque et al., 2008); the latterresponse is not seen in mice (Gutiérrez et al., 2008). B.lanceolatus venom contains L-amino acid oxidase, serineproteases, phospholipase A2 (PLA2) (Lôbo de Araújo et al.,1994, 1998) and zinc-containing metalloproteinases(MMPs) (Stroka et al., 2005; Gutiérrez et al., 2008). Studiesin vitro have also shown that the venom contains thrombin-like activity but no coagulant or defibrinogenating activi-ties (Stocker et al., 1974; Lôbo de Araújo et al., 2001). B.lanceolatus venom stimulates leucocyte migration andedema formation (increase in vascular permeability) that ismediated by arachidonic acid metabolites (lipoxygenaseand cyclooxygenase products), bradykinin, histamine andserotonin (Lôbo de Araújo et al., 2000; Guimarães et al.,2004).

In this work, we examined the expression of osteo-pontin (OPN) during muscle damage and regenerationfollowing the intramuscular injection of B. lanceolatusvenom. In addition, we assessed changes in myoD, myo-genin and CD68. OPN is an O-glycosylated phosphoproteinexpressed by a variety of cells and tissues involved ina range of physiological processes, including the synthesisof collagen fibrils, angiogenesis, cell migration, woundhealing and immunomodulation (Wang and Denhardt,2008). MyoD and myogenin, which belong to themyogenic regulatory factors (MRF) family of proteins, havea key role in the early and late stages of myogenesis duringdevelopment and repair (Chargé and Rudnicki, 2004). CD68is a transmembrane receptor of M1 (resident) macro-phages, a pro-inflammatory population of phagocytic cellsthat respond to acute muscle injury after neutrophil inva-sion (reviewed in Tidball and Villalta (2010)). The resultsdescribed here contribute to our understanding of the localeffects induced by B. lanceolatus venom, and the biology ofmuscle regeneration in general.

2. Materials and methods

2.1. Snake venom

Lyophilized B. lanceolatus venom (supplied by the Unitédes Venins, Institute Pasteur, Paris, France) was recon-stituted in 0.05 M phosphate-buffered saline (PBS), pH 7.4.

2.2. Animals

Six to eight-week-old male Wistar rats (Rattus norvegi-cus; 200–300 g) were provided by the MultidisciplinaryCenter for Biological Investigation at the State University ofCampinas (CEMIB/UNICAMP). This study was approved bythe institutional Committee for Ethics in Animal Use(CEUA/UNICAMP, protocol no. 941-1) and the experimentswere done according to the guidelines of the BrazilianSociety for Laboratory Animal Science (SBCAL; formerlyBrazilian College for Animal Experimentation - COBEA).

2.3. Production of muscle injury

Ten groups of 12 Wistar rats (6 controls and 6 treated)received an intramuscular (i.m.) injection of 100 ml of0.05M phosphate-buffered saline (PBS), pH 7.4, or 100 mg ofvenom/100 ml, respectively, in the right gastrocnemiusmuscle. This amount of venom was chosen based onpreliminary experiments with 50, 100 and 150 mg of venomwhich showed that 100 mg of venom/100 mL gave the bestresponse. After injection, the rats were maintained in cages(five/cage) at 22 �C, on a 12 h light–dark cycle, with foodand water ad libitum. At various times after treatment (1, 3,6 or 18 h, or 1, 2, 3, 7, 14 or 21 days) the rats were anes-thetized with halothane (Cristália, Itapira, SP) and killed bycervical dislocation and the injected muscle was immedi-ately dissected. Samples were taken from themedial aspectaround the injection site and processed for histological andimmunohistochemical analysis.

2.4. Histology

Muscle samples were fixed in 4% paraformaldehyde andembedded in paraffin. Serial 5 mm thick cross-sections andlongitudinal sections were mounted on silanized glassslides. One section from each sectioning plane (cross-section or longitudinal section) per animal was stained byhematoxylin and eosin (H&E) or Masson’s trichrome (MT)(n¼ 6/time interval/staining method), the latter being usedto stain connective tissue.

2.5. Measurement of muscle fiber diameter

To detect acute changes in muscle fiber size, the smalldiameter (to avoid error if fibers were not perfectly cross-sectioned) of muscle cells in the damaged area wasmeasured 1 h post-venom and compared with the corre-sponding time-matchedPBS controls. The damaged areawasdefined as that presenting hemorrhage, edema and/oralteredmuscle fibers. For each rat, the small diameter of 200fibers (total of 1200 fibers per group since each group con-tained six rats) was measured using a photomicroscope(BX51, Olympus, Japan), a 200� magnification, and ImagePro-Plus software. To evaluate regeneration, the smalldiameter of regenerated fibers with centrally-located nuclei(n ¼ 205 fibers) at 21 days post-venomwas compared withthatof apparentlynormalfibers (peripherally-locatednuclei;n ¼ 205) in the same rats (n ¼ 6). To ensure that the fiberswith peripherally-located nuclei in envenomed muscleswere indeed normal measurements were also taken of 205fibers in control rats (21 days after the injection of PBS).

2.6. Immunohistochemistry

Sections (5 mm thick) of gastrocnemius muscle fromeach interval (1, 3, 6 or 18 h, or 1, 2, 3, 7, 14 or 21 days) weredeparaffinized with xylene and hydrated with decreasingconcentrations of ethanol and distilled water. Endogenousperoxidase activity was blocked by immersing the slides ina 3% H2O2 solution. Antigen retrieval was done by incu-bating the sections with 10 mM sodium citrate buffer, pH6.0, in a steamer (95–99 �C) for 30 min. The slides were

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409400

subsequently incubated with reconstituted milk powder toblock nonspecific antigenic sites. Subsequently, the slides(n ¼ 6 sections/time-point; one section per rat) wereincubated overnight (16–18 h) at 4 �C with monoclonalprimary antibodies: anti-CD68 (Dako� M0718; 1:200),anti-OPN (Sigma O7264, 1 mg/ml, anti-myoD 1 (Dako�

M3512; 1:50,1:75,1:100,1:150) and anti-myogenin (Dako�

M3559; 1:100) according to manufacturer’s recommenda-tions (modified when necessary). After washing, thesections were incubated with biotinylated anti-mouse oranti-rabbit secondary antibody (Advanced� HRP link,Dako�) for 30 min at room temperature, rinsed with PBSand incubated with Advanced� HRP Enzyme for 30 min atroom temperature. Horseradish peroxidase (HRP) activitywas detected using the chromogenic substrate dia-minobenzidine (DAB-Advanced�, Dako�). All incubationswere done in a humidified chamber. The antibody reactionswere stopped by washing the slides with distilled water.The sections were counterstained with Ehrlich’s hematox-ylin and then mounted in Canada balsam. For each anti-body, a negative control was done by replacing the primaryantibody with 1% PBS-BSA. Macrophages expressing OPNwas identified by double labeling the cells for CD68 andOPN. For this, the sections were incubated sequentiallywith both antibodies and then stainedwith the Envision Kitdouble stain system (Dako�).

2.7. Image analysis: quantification of immunoreactivity

2.7.1. Anti-osteopontinTo quantify the immunoreactivity to OPN, two fields in

the damaged area per section were evaluated (n ¼ 6animals per time-point, total of 12 fields/time-point). Allfield images were photographed with an Olympus BX51photomicroscope using fixed parameters for light intensity,magnification (200�) and color (24 bits). The images werecaptured with a computer-aided image analysis system(Image Pro-Plus 4.0, Media Cybernetics) connected to thephotomicroscope. Image analysis (quantification of theoptical density of immunoreactivity) was done using GIMP2.6.4 software (GNU Image Manipulation Program, CNETNetworks Inc.) that segmented the images by color. Thissegmentation by color made it possible to determine thepercentage of pixels for staining by a given antibody.

2.7.2. Anti-CD68 and anti-myogeninCD68-positive macrophages were quantified by count-

ing the number of positive cells in ten fields in the damagedarea per section (one section/rat and six rats/time interval,i.e., 60 fields per time interval) in the control and enve-nomed groups. Myogenin was quantified by counting thenumber of nuclei positive for the protein in muscle fibers.

2.8. Statistical analysis

The immunohistochemical data were expressed as themean � S.D. Multiple comparisons were done using one-way analysis of variance (ANOVA) followed by Bonferro-ni’s multiple comparisons test (GraphPad Prism 4.0 soft-ware, San Diego, CA, USA). Cell diameters (1 h post-venomvs. control group, regenerated fibers vs. intact fibers at 21

days post-venom and intact fibers in envenomedmuscle vs.normal fibers in control PBS-injected muscle after 21 days)were compared using Student’s t-test. A value of p < 0.05indicated significance.

3. Results

3.1. Histological analysis

Control muscles injected with 100 ml PBS showed fiberswith normal morphology and no detectable abnormality inblood vessels or connective tissue at all times intervals(Fig.1A). In contrast,1hafter venominjection (100mg/100ml)the muscles showed intense interstitial hemorrhage, edema(fiber enlargement and an increase in the tissue inter-fascicular space), and emaciated myofibers; neutrophilswere very few or absent. At 3 h, some of the swollen edem-atous cells showed “delta” lesions and, more rarely, denseclumps of hypercontracted myofibrils; an inflammatoryneutrophilic infiltratewasobservedalongwithextravascularredbloodcells (Fig.1B,C). At6h, hemorrhagediminishedandwas replaced by foci of a dense polymorphonuclear-richinflammatory infiltrate (not shown), while at 18 h macro-phages gradually appeared in the interstitial space and innecrotic fibers (Fig. 1D). At 3 days post-venom, clusters ofproliferative myoblasts were observed in the damaged area(Fig. 1E) and at 7 days post-venom basophilic myoblasts andmyotubes of different sizes were abundant (Fig. 1F). By 21days post-venom, the former damaged region was nowoccupied by discrete foci of small centrally nucleated myo-fibers (shown in Fig. 3B). Masson’s trichrome, which stainscollagen tissue blue, showed that in control muscle thecollagenous matrix was clearly visible only in the perimy-sium around venous and arterial branches; the endomysialmatrix was barely seen (Fig. 2A). At 1 h post-venom, thecollagen bundles appeared ragged (Fig. 2B), while at 3 and 7days post-venom there was a noticeable deposition in someregions, with proliferative myoblasts and differentiatingmyotubes giving these regions a fibrotic aspect (Fig. 2C,D,respectively).

3.2. Muscle cell morphometry

Onehour after venom injection, themean small diameterof fibers in the affected muscle region was significantlygreater (59.8 � 8.2 mm; n ¼ 1200 fibers) than in controlsinjectedwithPBS (39.9�6.7mm;n¼1200fibers) (Fig. 3A).At21 days post-venom, only 205 regenerated cells (those withcentral nuclei) were detected in cross-sections of gastroc-nemius (n ¼ 6 rats). The mean diameter of these fibers wassignificantly smaller (21.42 � 5.65 mm) than that of appar-ently normalfiberswithperipheral nuclei invenom-injectedmuscle (41.16 � 7.96 mm; Fig. 3B). The diameter of the latterfibers was not significantly different from that of normalfibers in rats injected with PBS (42.0 � 6.03 mm).

3.3. Immunohistochemistry

3.3.1. Anti-CD68The activation of resident macrophages (M1 phagocytic

population) in the damaged region was evaluated by

Fig. 1. Cross-sections of rat gastrocnemius at various times after injection of PBS (A), or BLV (B–F). (A) Control group (1 h): the fibers and fascicles had a normalmorphology. (B) V1h - One hour post-BLV, showing disorganized fascicles and extensive hemorrhage (H). There were a few ruptured fibers and the myofibrilsshowed clumping; most myofibers were edematous (*). (C) V3h - Three hours post-venom: the most prominent features at this time were interstitial edema anda neutrophil infiltrate (n), mainly in contact with necrotic fibers. (D) V18h - Eighteen hours post-venom: note the presence of numerous macrophages and limitednecrosis. (E) V3d - Three days post-venom: at this time small foci of proliferative myoblasts were surrounded by normal muscle fibers (nf). (F) V7d - Seven dayspost-venom: detail of the regenerative region (RR) showing basophilic myoblasts and myotubes (distinguished by their difference in size) and damagedacidophilic fibers (df). n/m ¼ neutrophil/macrophage, P ¼ perimysium, T ¼ tendon. Hematoxylin-eosin staining. Bar: 40 mm.

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409 401

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409402

counting the number of cells expressing CD68 protein inenvenomed and control muscles; the number of CD68-positive macrophages increased significantly up to 24 h(461.67� 144.67), but decreased gradually thereafter. From7 days post-venom onwards the number of CD68-positivecells was lower than observed at 1 h (Fig. 4A,B).

3.3.2. Anti-OPNOsteopontin was expressed in the cytoplasm of intact

and venom-damaged fibers (Fig. 5A,B), macrophages,myoblasts and myotubes and fibroblast (Figs. 6 and 7). OPNimmunolabeling inmusclefiberswasquantifiedusingGIMP2.6.4 software (Fig. 5C,D), with the number of pixelsreflecting the intensity of immunolabeling; this quantifica-tion allowed the comparison of OPN expression (Fig. 5E).Basal OPN labeling in controls did not vary significantly overtime. In envenomed muscle, OPN expression was signifi-cantly increased from 3 h to 14 days post-venom; maximalexpression occurred at 3 days (31 � 3.1%), and was slightlylower at 7 days (27 � 1.2%) and 14 days (24.2 � 3.2%) post-venom. At 21 days post-venom, the pixel density did notdiffer from the PBS control or envenomed muscle after 1 h.

3.3.3. Dual labeling for OPN/CD68 in macrophagesImage analyses of venom-treated muscles at 3 days

post-venom showed double-labeled macrophages next to

Fig. 2. Masson’s Trichrome staining of collagen fibrils in gastrocnemius muscle inperimysial collagenous matrix around venous and arterial blood vessels. (B) V1h -bundles. (C) V3d - Three days post-venom, showing fibrotic regeneration region (RR(dr) were observed. (D) V7d - Seven days post-venom: note the proliferation of thregeneration (dr). bv ¼ blood vessels; mnt ¼ motor nerve trunk. Bars: A,B ¼ 40 mm

the endomysial space (alkaline phosphatase reaction in redplus peroxidase-based reaction in brown for CD68 andOPN, respectively) and in close contact with OPN-labeledmuscle fibers (Fig. 6). The 3 day post-venom interval waschosen for double labeling because it corresponded to peakof OPN expression in muscle fibers.

3.3.4. Myoblasts and myotubesOPN reactivity was strong in the regenerating region of

envenomed muscle, but was rare or absent in regions notaffected by venom. Fig. 7A–C shows regenerating fibers at 7days post-venom. The muscle proliferative region con-tained mainly myotubes, with myoblasts being rarer. Bothmyoblasts (proliferative cells) and myotubes (differenti-ating cells) were strongly positive for OPN; mature fiberswere also OPNþ (Fig. 7A,B). OPN-positive fibroblasts wereobserved in the interstitium (Fig. 7C). Although the numberof macrophages was highly reduced, their reactivity was asstrong as in the previous time intervals (Fig. 7D).

3.3.5. Myogenin and myoDAt day 7 post-venom, whenmyogenin expressionwas at

its peak, this protein was detected in the nucleus andcytoplasm of myoblasts and myotubes (Fig. 8A,B) whereasat subsequent intervals it was expressed only in thenucleus. Myogenin expression in envenomed muscle was

jected with PBS (A) or venom (B–D). (A) Control (1 h), showing a normalOne hour post-venom, showing disorganization of the perimysial collagen); abnormal central nucleated myofibers indicative of defective regeneratione collagen matrix in the fascicle interstitium and the presence of defective; C,D ¼ 25 mm.

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409 403

significantly greater than in control muscle from 18 h to 14days post-venom, with a peak at 7 days (152.63 � 60.45)followed by a decrease thereafter (Fig. 8C). No immuno-labeling for anti-myoD was observed at any time interval,despite several attempts using different dilutions andincubation protocols.

4. Discussion

B. lanceolatus venom produced local tissue damagecompatible with disturbances in hemostasis. At 3–6 h post-venom there was extensive hemorrhage, with inflamma-tory neutrophils and macrophages disseminated amongstthe swollen or disintegrated muscle fibers. Class P-I (Strokaet al., 2005) and P-III (Gutiérrez et al., 2008) Zn2þ-depen-dent metalloproteinases present in this venom probablycontributes to the observed muscle damage, and inflam-matory response, as also reported for other Bothropsvenoms (Gutiérrez, 1995; Rucavado et al., 1998, 2002; Lainget al., 2003). Class P-III snake venommetalloproteinases (P-III SVMPs) possess disintegrin-like and cysteine-richdomains that are critical for their hemorrhagic activitysince endothelial cell integrins and specific ECM proteinswith a role in matrix/endothelium stabilization are themain targets for these proteinases (Fox and Serrano, 2008,2009). Experimentally, the local inflammatory response

Fig. 3. Small diameter in muscle fibers 1 h after injection of PBS or venom. (A) Onemuscles was significantly greater than in control fibers (*p < 0.05). (B) V1h - One ho(C) By 21 days post-venom, the small diameter of intact fibers (unaffected by venomof regenerated fibers (affected by venom and showing central nuclei: arrows) (*p < 0diameters of intact and regenerated fibers in venom-injected muscles. Hematoxymean � SD of the number of fibers indicated.

induced by B. lanceolatus venom includes leucocytemigration, enhanced vascular permeability and the partic-ipation of metabolites of the lipoxygenase and cyclo-oxygenase (COX) pathways (Lôbo de Araújo et al., 2000;Rucavado et al., 2002; Guimarães et al., 2004). Oxygenand nitrogen free radical formation and the release ofcytotoxic mediators such as TNF-a, TGF-b, IGF and IL-6(Tidball, 2005) are also associated with venom-inducedinflammation (Rucavado et al., 2002).

In skeletal muscle, inflammatory mediators can activatequiescent satellite cells, increase myoD and myogeninexpression, and improvemuscle differentiation and growth(Tidball, 2005; Chazaud et al., 2009; Chen and Li, 2009;Tidball and Villalta, 2010). Here, we found that CD68-positive macrophages (M1 population) reached theirhighest number from 18 h to 48 h after venom injection,when necrotic fragments of destroyed muscle fibers occu-pied the foci of injury. Interestingly, CD68-positive macro-phages also expressed OPN; this expression was biphasic,with the first one peak occurring from 6 to 48 h post-venom.

OPN (also knownas Eta1 or T-lymphocyte activation 1) issynthesized in a variety of tissues and cells, includinginflammatory cells and myoblasts (Pereira et al., 2006;Uaesoontrachoon et al., 2008). This protein is a member ofthe small integrin-binding ligand N-linked glycoprotein

hour after venom administration, the small diameter of fibers in envenomedur post-venom: fibers in the envenomed region were edematous. Bar: 25 mm.and hence showing peripheral nuclei) were significantly greater than those.05). (D) V21d - Twenty-one days post-venom: there was no difference in thelin-eosin staining. Bar: 40 mm. In (A) and (C), the columns represent the

Fig. 4. Light microscopy of macrophages positive for CD68 membrane receptor in rat gastrocnemius muscle. One day (24 h) post-venom, phagocytic M1macrophages (m) expressing CD68 protein are located in tendon connective tissue (T) (panel A) and in the endomysium of edematous muscle fibers (mf) (B–D).Insert ¼ high magnification of CD68 þ macrophages; mnt ¼ motor nerve trunk. Counterstaining with Ehrlich’s hematoxylin. Bars: A,B,C ¼ 40 mm, D ¼ 25 mm. (E)Histogram showing the number of CD68-labeled macrophages at different times after injection of PBS (100 ml) or venom (100 mg/100 ml) in rat gastrocnemiusmuscle. The columns represent the number of positive cells in ten separate fields per section of injured muscle (n ¼ 6 animals/per time interval). The columnsrepresent the mean � SD. *p < 0.05 compared to the PBS control at each time-point. #p < 0.05 compared to 1, 3 and 6 h and 3, 7, 14 and 21 days post-venom.

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409404

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409 405

(SIBLING) family of proteins whose receptors include someintegrins (avb3, avb1, avb5, a8b1, a9b1, a4b1, a4b7) andCD44 variants (hyaluran receptor) (Mylona et al., 2006;Wang and Denhardt, 2008). The receptors for OPNmediate adhesion, migration and survival in a variety of cell

Fig. 5. Immunohistochemistry for osteopontin (OPN) in rat gastrocnemius muscle tfibers (mf) labeled for OPN. (B) Three days post-venom: intact and damaged OPN(C,D): explanation of how the OPN immunoreactive regions in panels (A) and (B) wemnt ¼ motor nerve trunk. Bar: 40 mm. (E) Histogram based on the optical densities oThe software GIMP 2.6.4 was used to compare two random images per animal (n ¼ 6compared to control (PBS-treated) muscle at each interval. There were significantcompared to 1, 3, 6, 18, 24 h and 21 days post-venom.

types, including neutrophils and myogenic cells(Uaesoontrachoon et al., 2008). The OPN adhesive domainscontain Arg-Gly-Asp (RGD) and serine-valine-valine-tyrosine-glutamate-leucine-arginine (SVVYGLR) sequencesthat allow OPN adhesion to integrins and matrix proteins,

hree days post-venom. (A) C3d - Control, three days: note the intact muscle-positive fibers were observed. Counterstaining with Ehrlich’s hematoxylin.re selected and analyzed in order to determine the optical density of labeling.f OPN immunoreactive regions in PBS-(control) and venom-treated muscles.animals per time interval). The columns represent the mean � SD. *p < 0.05differences between envenomed groups in various time-points. #p < 0.05

Fig. 6. A,B. CD68/osteopontin (OPN) double labeling in rat gastrocnemius three days after venom injection (V3d). Note CD68-labeled macrophages (m) in contactwith fiber debris in (A) and in the endomysium (arrows) in (B). Muscle fibers in both regions were immunolabeled for OPN. hf ¼ hypercontracted fiber.Counterstaining with Ehrlich’s hematoxylin. Bar: 25 mm.

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409406

such as fibronectin, thereby facilitating myogenesis(Yokosaki et al., 1999, 2005; Scatena et al., 2007).

Integrins are anchorage proteins which mediates cell(myoblasts, immune cells and others) to fibronectin, lam-inins and collagens fibrils of the ECM (Heino and Käpylä,2009). Integrins also anchor endothelial cells to theunderlying basal lamina and therefore have a role in bloodvessel structure and organization. During developmentaland reparative myogenesis, integrins promote the fusion of

Fig. 7. Osteopontin (OPN) immunolabeling in myofibers, myoblasts, myotubes, macOPN labeling in muscle fibers (mf), myoblasts (Mb), myotubes (Mt) and fibroblasts (fstrongly positive for OPN seven days post-venom. Counterstaining with Ehrlich’s h

myogenic cells and their interactionwith ECM proteins (seeSanchez et al., 2010; Vetrone et al., 2009). Interestingly,OPN molecule contains thrombin cleavage site and sitessusceptible to cleavage by MMPs, then allowing moleculeto bind integrins and specific CD44 variants. (Wang andDenhardt, 2008). The fragments generated maintain andenhance the adhesive properties of full-length OPN byexposing the cryptic RGD (avb3, avb1, avb5, a8b1) andSVVYGLR (a9b1, a4b1, a4b7) domains for integrin-binding

rophages and fibroblasts seven days post-venom (V7d). Panels A and C show). Bars: A ¼ 25 mm; B ¼ 10 mm; C ¼ 30 mm. Panels D and E show macrophagesematoxylin. Bars ¼ 10 mm.

Fig. 8. Anti-myogenin immunolabeling in regenerative region of rat gastrocnemius seven days post-venom (V7d). Cross-section (A) and longitudinal section (B)of myotubes (Mt) showed cytoplasmic and nuclear labeling for OPN; the cells varied in size, with larger cells (*) showing weaker reactivity for myogenin thansmaller ones. Mb ¼ myoblasts. Panel C shows that myogenin expression was significantly higher 18–14 days post-venom and was maximal seven days post-venom. *p < 0.05 when compared with the paired PBS control group. #p < 0.05 compared to all other post-venom intervals. Bars: 25 mm.

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409 407

(Yokosaki et al., 1999, 2005; Scatena et al., 2007). Thebiphasic upregulation of OPN expression (6–48 h and 3–14days post-venom) correlated with two distinct phasesfollowing B. lanceolatus venom-induced muscle injury. Thefirst of these, which corresponded to the early acuteinflammatory degenerative phase, was critical for thesecond stage that was characterized by muscle repair andremodeling subsequent to satellite cell activation. Whetherthe OPN expressed by macrophages and muscle fibers6–48 h post-venom at sites of acute inflammation acted asa chemotactic cytokine and adhesive molecule is notknown. However, OPN mediates activities such as phago-cytosis and cytokine production by macrophages and otherimmune cells (Wang and Denhardt, 2008). These activitiesare necessary to activate dormant satellite cells, migrationand proliferation. Similarly, the second phase of OPNupregulation seen at 3–14 days post-venom correlatedtemporally with the beginning of myoblasts proliferation,their migration and the subsequent transformation into

myotubes (differentiation) and the growth of regeneratingmyofibers with centrally-located nuclei (maturationphase). In this second phase, which was more pronouncedthan the first, OPN was produced mainly by myogenic cells,including differentiated cells, and also by fibroblasts andM1 macrophages, as shown by double immunolabeling forCD68/OPN. In a study of muscle regeneration after theinjection of cardiotoxin (CTX) from Naja naja atra snakevenom, Hirata et al. (2003) showed that the gene expres-sion for OPN was notably increased at 2 and 4 days afterenvenoming. These authors suggested that OPN might bean important mediator in the early phase of muscleregeneration following intoxication with CTX.

In this study, we also examined the pattern of myoD andmyogenin expression during regeneration correlated thiswith OPN expression. MyoD is a myogenic transcriptionfactor associated with the early stage of differentiation,whereas myogenin occurs in the late stage (Dedieu et al.,2002; Holterman and Rudnicki, 2005). However, no myoD

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409408

immunolabeling was observed in this work; in contrast,myogenin expression increased steadily from 18 h to 7 dayspost-venom and decreased from 14 days onwards,although its levels were still higher than in the time-matched controls (Fig. 8). That OPN expressed bymyoblasts and myotubes at this stage would representa role in the adhesion of regenerating myotubes toelements of the ECM in order to promote the appropriateconditions for regenerative myogenesis is unknown.However, Pereira et al. (2006) suggested that OPN expres-sion at a time interval characterized by myoblast prolifer-ation, migration and fusion probably reflected an importantrole of the protein in these events. In studies in vitro,Uaesoontrachoon et al. (2008) reported that OPN releasedby myoblasts served as a link between the inflammatoryresponse and myogenesis during the early phase of muscleregeneration and repair. Our findings corroborate the closerelationship in timing between the second phase of OPNupregulation and the significant increase in myogeninexpression initiated at 18 h, with peaks at 3 days and 7 dayspost-venom.

Our results showed that B. lanceolatus venom promotedconnective tissue disorganization in the acute stage ofenvenoming followed by patches of intense collagendeposition 3–7 days post-venom. Fibrotic processes mayrepresent a barrier for tissue revascularization and limit theaccess of important molecules or cells involved in tissueregeneration. The finding that the small diameter ofregenerated fibers at 21 days post-venomwas significantlylower than in time-matched controls suggests that fibrosismay have impaired complete regeneration. It is worthmentioning that OPN has been pointed out as a pro-fibroticpromoter in hepatic and renal diseases (Lorena et al., 2006;Irita et al., 2008). In cardiac muscle dysfunction (Singhet al., 2010) and skeletal and cardiac muscles of mdx mice(Vetrone et al., 2009) the upregulation of OPN has beencorrelated with enhanced collagen synthesis and accumu-lation, whereas deletion of the OPN gene reduced fibrosisand improved regeneration.

Our findings also showed two other interesting data:the expression of myogenin in the cytoplasm of myoblastsandmyotubes instead of its usual expression in the nucleus,and the population of CD68 þ macrophages significantlyelevated in the proliferative stage of myoblasts (3 dayspost-venom), and in the acute inflammatory phase (3–6 hpost-venom). Nuclear myogenin is needed for regulation ofthe transcription of specific myogenic promoters whereasits retention in the cytoplasm may regulate the biologicalactivity of proteins and prevent differentiation; the transferof myogenin into the nucleus occurs when proliferativesignals cease and the protein level increases significantly(Ferri et al., 2009). On the other hand, macrophages canrelease products that inhibit the transition of myogeniccells from proliferative to differentiating stages (Merlyet al., 1999). Whether this significant presence of phago-cytic M1 macrophages on day 3 post-venom has a role inthe atypical retention of myogenin in the cytoplasm and indelayed muscle repair is unknown. This is an interestingpossibility since it was only from day 14 post-venomonwards that myogenin labeling was no longer observedin the cytoplasm and that CD68macrophage numbers were

as low as in control muscle. Together, the expression ofmyogenin in the cytoplasm of myoblasts andmyotubes, thepatches of interstitial fibrosis and the high number ofmacrophages seen up to the seventh day post-venom inenvenomed muscles could explain the small size ofregenerating fibers 21 days post-envenoming.

In conclusion, osteopontin, a chemotactic protein withcytokine-like properties was found to be up-regulated inmuscle injury caused by B. lanceolatus (fer-de-lance) snake.The upregulation of OPN occurred during the acute stage ofinflammation and during myogenic cell proliferation anddifferentiation. The expression of OPN by cells ofa myogenic lineage, macrophages and fibroblasts agreeswith its role as an adhesive chemotactic matricellularproteinwith cytokine-like properties that canmodulate theexpression of myogenic transcriptional factors and, hence,muscle regeneration. In our experimental model, threeweeks after envenoming, the regenerating fibers weresmall, indicating delayed regeneration. Since OPN has beenalso described as pro-fibrotic protein in adverse conditions,its possible mediation in collagen deposition in the regionof myoblast proliferation needs to be investigated. As far aswe know, this is the only report to associate OPN expres-sion in a rat model of muscle regeneration after the intra-muscular injection of Bothrops snake venom.

5. Conflict of interest

The authors have no conflict of interest related to thiswork.

Acknowledgments

The authors thank Marta B. Leonardo, MSc, and GlauceAparecida Pinto, PhD, for excellent technical assistance andDr. Stephen Hyslop for criticism and revising the language.This work was supported by a grant from Fundação deAmparo à Pesquisa do Estado de São Paulo (FAPESP, grantno. 2005/60929-7). V.B.S. was supported by an MScstudentship from Coordenação de Aperfeiçoamento dePessoal de Nível Superior (CAPES), Brazil. S.P.I. was a post-doctoral researcher in the Venom and Toxin Laboratory ofM.A.C.H. M.A.C.H. is supported by a research fellowshipfrom Conselho Nacional de Desenvolvimento Científico eTecnológico (CNPq), Brazil.

References

Bogarín, G., Romero, M., Rojas, G., Lutsch, C., Casadamont, M., Lang, J.,Otero, R., Gutiérrez, J.M., 1999. Neutralization by a monospecificBothrops lanceolatus antivenom of toxic activities induced byhomologous and heterologous Bothrops snake venoms. Toxicon 37,551–557.

Chargé, S.B., Rudnicki, M.A., 2004. Cellular and molecular regulation ofmuscle regeneration. Physiol. Rev. 84, 209–238.

Chazaud, B., Brigitte, M., Yacoub-Youssef, H., Arnold, L., Gherardi, R.,Sonnet, C., Lafuste, P., Chrétien, F., 2009. Dual and beneficial roles ofmacrophages during skeletal muscle regeneration. Exerc. Sport Sci.Rev. 37, 18–22.

Chen, X., Li, Y., 2009. Role of MMP in skeletal muscle: migration, differ-entiation, regeneration and fibrosis. Cell Adh. Migr. 3, 337–341.

Dedieu, S., Mazères, G., Cottin, P., Brustis, J.J., 2002. Involvement ofmyogenic regulator factors during fusion in the cell line C2C12. Int. J.Dev. Biol. 46, 235–241.

V. Barbosa-Souza et al. / Toxicon 58 (2011) 398–409 409

Ferri, P., Barbieri, E., Burattini, S., Guescini, M., D’Emilio, A., Biagiotti, L.,Del Grande, P., De Luca, A., Stocchi, V., Falcieri, E., 2009. Expressionand subcellular localization of myogenic regulatory factors during thedifferentiation of skeletal muscle C2C12 myoblasts. J. Cell Biochem.108, 1302–1317.

Fox, J.W., Serrano, S.M., 2008. Insights into and speculations about snakevenom metalloproteinases (SVMP) synthesis, folding and disulfidebond formation and their contribution to venom complexity. FEBS J.275, 3016–3030.

Fox, J.W., Serrano, S.M., 2009. Timeline of key events in snake venommetalloproteinase research. J. Proteomics 72, 200–208.

França, F.O.S., Málaque, C.M.S., et al., 2003. Acidente botrópico. In:Cardoso, J.L. (Ed.), Animais Peçonhentos no Brasil: Biologia, Clínica eTerapêutica dos Acidentes, pp. 72–86. Sarvier, São Paulo.

Guimarães, A.Q., Cruz-Höfling, M.A., Araújo, P.M.F., Bon, C., Lôbo deAraújo, A., 2004. Pharmacological and histopathological character-ization of Bothrops lanceolatus (Fer de lance) venom-induced edema.Inflamm. Res. 53, 284–291.

Gutiérrez, J.M., Ownby, C.L., 2003. Skeletal muscle degeneration inducedby venom phospholipases A2: insights into the mechanisms of localand systemic myotoxicity. Toxicon 42, 914–931.

Gutiérrez, J.M., Rucavado, A., Escalante, T., Díaz, C., 2005. Hemorrhageinduced by snake venom metalloproteinases: biochemical andbiophysical mechanisms involved in microvessel damage. Toxicon 45,997–1011.

Gutiérrez, J.M., Sanz, L., Escolano, J., Fernández, J., Lomonte, B., Angulo, Y.,Rucavado, A., Warrell, D.A., Calvete, J.J., 2008. Snake venomics of thelesser Antillean pit vipers Bothrops caribbaeus and Bothrops lanceo-latus: correlation with toxicological activities and immunoreactivityof heterologous antivenom. J. Proteome Res. 7, 4396–4408.

Gutiérrez, J.M., 1995. Clinical toxicology of snake bites in Central América.In: Meier, White (Eds.), Handbook of Clinical Toxicology of AnimalVenoms and Poisons. CRC Press, Boca Ratón, pp. 645–665.

Heino, J., Käpylä, J., 2009. Cellular receptors of extracellular matrixmolecules. Curr. Pharm. Des. 15, 1309–1317.

Hirata, A., Masuda, S., Tamura, T., Kai, K., Ojima, K., Fukase, A.,Motoyoshi, K., Kamakura, K., Miyagoe-Suzuki, Y., Takeda, S., 2003.Expression profiling of cytokines and related genes in regeneratingskeletal muscle after cardiotoxin injection: a role for osteopontin. Am.J. Pathol. 163, 203–215.

Holterman, C.E., Rudnicki, M.A., 2005. Molecular regulation of satellitecell function. Semin. Cell Dev. Biol. 16, 575–584.

Irita, J., Okura, T., Kurata, M., Miyoshi, K., Fukuoka, T., Higaki, J., 2008.Osteopontin in rat renal fibroblasts: functional properties and tran-scriptional regulation by aldosterone. Hypertension 51, 507–513.

Laing, G.D., Clissa, P.B., Theakston, R.D., Moura-da-Silva, A.M., Taylor, M.J.,2003. Inflammatory pathogenesis of snake venommetalloproteinase-induced skin necrosis. Eur. J. Immunol. 33, 3458–3463.

Lôbo de Araújo, A., Donato, J.L., Bon, C., 1998. Purification from Bothropslanceolatus (fer de lance) venom of a fibrino(geno)lytic enzyme withesterolytic activity. Toxicon 36, 745–758.

Lôbo de Araújo, A., Kamiguti, A., Bon, C., 2001. Coagulant and anticoag-ulant activities of Bothrops lanceolatus (Fer de lance) venom. Toxicon39, 371–375.

Lôbo de Araújo, A., Radvanyi, F., Bon, C., 1994. Purification of an acidicphospholipase A2 from Bothrops lanceolatus (fer de lance) venom:molecular and enzymatic properties. Toxicon 32, 1069–1081.

Lôbo de Araújo, A., Souza, A.O., Cruz-Höfling, M.A., Flores, C.A., Bon, C.,2000. Bothrops lanceolatus (Fer de lance) venom induces oedemaformation and increases vascular permeability in mouse hind paw.Toxicon 38, 209–221.

Lorena, D., Darby, I.A., Gadeau, A.P., Leen, L.L., Rittling, S., Porto, L.C.,Rosenbaum, J., Desmoulière, A., 2006. Osteopontin expression innormal and fibrotic liver. Altered liver healing in osteopontin-deficient mice. J. Hepatol. 44, 383–390.

Malbranque, S., Piercecchi-Marti, M.D., Thomas, L., Barbey, C., Courcier, D.,Bücher, B., Ridarch, A., Smadja, D., Warrell, D.A., 2008. Fatal diffusethrombotic microangiopathy after a bite by the "Fer-de-Lance" pitviper (Bothrops lanceolatus) of Martinique. Am. J. Trop. Med. Hyg. 78,856–861.

Mauro, A., 1961. Satellite cell of skeletal muscle fibers. J. Biophys. Biochem.Cytol. 9, 493–495.

Merly, F., Lescaudron, L., Rouaud, T., Crossin, F., Gardahaut, M.F., 1999.Macrophages enhance muscle satellite cells proliferation and delaytheir differentiation. Muscle Nerve 22, 724–732.

Mylona, E., Jones, K.A., Mills, S.T., Pavlath, G.K., 2006. CD44 regulatesmyoblast migration and differentiation. J. Cell Physiol. 209, 314–321.

Pereira, R.O., Carvalho, S.N., Stumbo, A.C., Rodrigues, C.A.B., Porto, L.C.,Moura, A.S., Carvalho, L., 2006. Osteopontin expression in coculture ofdifferentiating rat fetal skeletal fibroblasts and myoblasts. In VitroCell Dev. Biol. Anim. 42, 4–7.

Rosenfeld, G., 1971. Symptomatology, pathology, and treatment of snakebites in South America. In: Bücherl, W., Buckley, E.E., Delofeu, V.(Eds.), Venomous Animals and their venoms. Academic Press, NewYork, pp. 345–841.

Rucavado, A., Escalante, T., Teixeira, C.F., Fernándes, C.M., Díaz, C.,Gutiérrez, J.M., 2002. Increments in cytokines and matrix metal-loproteinases in skeletal muscle after injection of tissue-damagingtoxins from the venom of the snake Bothrops asper. MediatorsInflamm. 11, 121–128.

Rucavado, A., Nuñes, J., Gutiérrez, J.M., 1998. Blister formation and skindamage induced by BaP1, a haemorrhagic metalloptotease from thevenom of the snake Bothrops asper. Int. J. Exp. Pathol 79, 245–254.

Sanchez, E.F., Schneider, F.S., Yarleque, A., Borges, M.H., Richardson, M.,Figueiredo, S.G., Evangelista, K.S., Eble, J., 2010. The novel metal-loproteinase atroxlysin-I from Peruvian Bothrops atrox (Jergón) snakevenom acts both on blood vessel ECM and platelets. Arch. Biochem.Biophys. 496, 9–20.

Scatena, M., Liaw, L., Giachelli, C.M., 2007. Osteopontin: a multifunctionalmolecule regulating chronic inflammation and vascular disease.Arterioscler. Thromb. Vasc. Biol. 27, 2302–2309.

Singh, M., Foster, C.R., Dalal, S., Singh, K., 2010. Osteopontin: role inextracellular matrix deposition and myocardial remodeling post-MI. J.Mol. Cell Cardiol. 48, 538–543.

Stocker, K., Christ, W., Leloup, P., 1974. Characterization of the venoms ofvarious Bothrops species by immunoelectrophoresis and reactionwith fibrinogen agarose. Toxicon 12, 415–417.

Stroka, A., Donato, J.L., Bon, C., Hyslop, S., Lôbo de Araújo, A., 2005.Purification and characterization of a hemorrhagic metalloproteinasefrom Bothrops lanceolatus (Fer-de-lance) snake venom. Toxicon 45,411–420.

Thomas, L., Tyburn, B., Bucher, B., Pecout, F., Ketterlé, J., Rieux, D.,Smadja, D., Garnier, D., Plumelle, Y., 1995. Prevention of thrombosesin human patients with Bothrops lanceolatus envenoming inMartinique: failure of anticoagulants and efficacy of a monospecificantivenom. Research Group on Snake Bites in Martinique. Am. J. Trop.Med. Hyg. 52, 419–426.

Tidball, J.G., 2005. Inflammatory processes in the muscle injury andrepair. Am. J. Physiol. Regul. Integr. Comp. Physiol. 288, R345–R353.

Tidball, J.G., Villalta, S.A., 2010. Regulatory interactions between muscleand the immune system during muscle regeneration. Am. J. Physiol.Regul. Integr. Comp. Physiol. 298, R1173–R1187.

Uaesoontrachoon, K., Yoo, H.J., Tudor, E.M., Pike, R.N., Mackie, E.J., Pagel, C.N., 2008. Osteopontin and skeletal muscle myoblast: association withmuscle regeneration and regulation of myoblast function in vitro. Int.J. Biochem. Cell Biol. 10, 2303–2314.

Vetrone, S.A., Montecino-Rodriguez, E., Kudryashova, E., Kramerova, I.,Hoffman, E.P., Liu, S.D., Miceli, M.C., Spencer, M.J., 2009. Osteopontinpromotes fibrosis in dystrophic mousemuscle bymodulating immunecell subsets and intramuscular TGF-b. J. Clin. Invest. 119, 1583–1594.

Wang, K.X., Denhardt, D.T., 2008. Osteopontin: role in immune regulationand stress responses. Cytokine Growth Factor Rev. 19, 333–345.

Yokosaki, Y., Matsuura, N., Sasaki, T., Murakami, I., Schneider, H.,Higashiyama, S., Saitoh, Y., Yamakido, M., Taooka, Y., Sheppard, D.,1999. The integrin a9b1 binds to a novel recognition sequence(SVVYGLR) in the thrombin-cleaved amino-terminal fragment ofosteopontin. J. Biol. Chem. 274, 36328–36334.

Yokosaki, Y., Tanaka, K., Higashikawa, F., Yamashita, K., Eboshida, A., 2005.Distinct structural requirements for binding of the integrins avb6,avb3, avb5, a5b1 and a9b1 to osteopontin. Matrix Biol. 24, 418–427.