Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part D journal homepage: www.elsevier.com/locate/cbpd

Molecular characterization of metalloproteases from Bothrops alternatus snake venom Fernando Fonseca Pereira de Paula a,e, Juliana Uema Ribeiro b, Livia Mara Santos b, Dulce Helena Ferreira de Souza c, Eduardo Leonardecz a,d, Flávio Henrique-Silva a, Heloisa Sobreiro Selistre-de-Araújo b,⁎ a

Departamento de Genética e Evolução, Universidade Federal de São Carlos, São Carlos, SP, Brazil Departamento de Ciências Fisiológicas, Universidade Federal de São Carlos, São Carlos, SP, Brazil c Departamento de Química, Universidade Federal de São Carlos, São Carlos, SP, Brazil d EMBRAPA Genetic Resources and Biotechnology, Brazil e Universidade de Brasília, Campus Planaltina — UnB/FUP Brazil, Brazil b

a r t i c l e

i n f o

Article history: Received 3 June 2014 Received in revised form 9 September 2014 Accepted 12 September 2014 Available online 27 October 2014 Keywords: Alternagin-C Transcriptome Venom gland Bothrops alternatus Disintegrin

a b s t r a c t We have previously demonstrated that alternagin-C (ALT-C), a disintegrin-like, Cys-rich protein isolated from Bothrops alternatus snake venom, induces human vascular endothelial cell (HUVEC) proliferation and angiogenesis in in vitro and in vivo assays. Therefore this protein could be interesting as a new approach for tissue regeneration studies. However, its primary sequence was not completely determined since the protein isolated from crude venom is usually a mixture of isoforms. Here we describe the transcriptome analysis of B. alternatus from the venom glands of a single male specimen. About 800 good-quality contigs were screened for snake venom metalloproteases/disintegrins, resulting in the following expression profile for these enzymes: 4% for P-I, 7% for P-II and 89% for P-III SVMPs. The PII-SVMP sequence code for RGD-disintegrins and all the expressed PIIIsequences have the ECD adhesive motif. A cDNA sequence coding for an ALT-C homolog was completely sequenced and characterized. Comparative sequence and structural analyses suggested new features that distinguish SVMP classes such as two prolyl endopetidase cleavage sites. All these data add new information on the expression pattern of metalloproteases of B. alternatus venom and may have practical applications for the production of recombinant disintegrins for cell adhesion studies. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Snake venom metalloproteases (SVMPs) comprise a family of highly complex but conserved proteins found in most Viperidae venom. The SVMPs are classified as members of the adamalysin family of the metzincin clan of the metalloendopetidase super family (Takeda et al., 2012). The SVMPs are Zn2 +-dependent enzymes which play central roles in venom toxicity by degrading the extracellular matrix (ECM) components such as collagen, laminin and fibronectin, resulting in hemorrhage, edema and necrosis in affected victims (Bjarnason and Fox, 1994). The SVMPs can also activate endothelial cells, and inhibit both blood coagulation and platelet aggregation (Moura-da-Silva et al., 2007). The SVMPs are synthesized as inactive precursor forms due to the presence of a highly conserved motif PKMCGVT known as cysteine switch in the proenzyme domain (Grams et al., 1993). The SVMPs are multimodular proteins that are divided into three classes (PI to PIII) ⁎ Corresponding author. Tel./fax: +55 2133518333/8401. E-mail address: [email protected] (H.S. Selistre-de-Araújo).

http://dx.doi.org/10.1016/j.cbd.2014.09.001 1744-117X/© 2014 Elsevier Inc. All rights reserved.

accordingly to their molecular mass and domain organization (Fox and Serrano, 2008). Members of the PI class are represented by the catalytic domain only, while the PII and PIII classes have additional C-terminal adhesive domains linked to the proteolytic domain. The catalytic domain is characterized by the presence of a conserved zincbinding consensus sequence HEXXHXXGXXHD (Hooper, 1994). The members of the PII class are characterized by the presence of a disintegrin domain, where an adhesive tripeptide motif such as RGD, VGD, or KGD, among others, is found (Calvete, 2013). The SVMPs of PIII class have instead a disintegrin-like domain with E/DCD adhesive motif, followed by a Cys-rich domain. After enzyme activation during venom secretion, some SVMPs undergo proteolysis of such domains, releasing the free disintegrin domain or the disintegrin-like domain linked to the Cys-rich domain (DC domain). However, there are also SVMPs that do not release the DC domain and may be purified from the venom as full P-III SVMPs (Moura-da-Silva et al., 2003). Additional complexity is also given by post-translational modifications such as domain dimerization resulting in at least eleven subclasses of SVMPs (Fox and Serrano, 2008). Such diversity has been suggested to be derived by accelerated evolution (Moura-da-Silva et al., 1996), post-

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

translational modifications (Fox and Serrano, 2008), and more recently, by genetic recombination between the SVMP classes (Moura-da-Silva et al., 2011). However, the first description of the genomic organization of a PIII SVMP gene suggesting a complex exon–intron organization gave additional evidence of the complexity of the SVMPs (Sanz et al., 2012). The processed disintegrin domains bind to integrin receptors in platelet or cell surfaces with high affinity and specificity therefore inhibiting platelet aggregation or cell adhesion to the ECM (Gould et al., 1990). Inhibition of cell adhesion is particularly relevant in metastasis therapy since tumor cells need to bind to many substrates such as the endothelium or ECM components in order to achieve a secondary site. In this way, the integrins were considered as important targets for metastasis prevention since integrin inhibition blocks cell adhesion to the ECM (Felding-Habermann et al., 2001). A few integrin blockers have entered in clinical trials for cancer therapy with encouraging results such as cilengitide, an inhibitor of the vitronectin receptors, the αvβ3 and αvβ5 integrins (Desgrosellier and Cheresh, 2010). Alternagin-C (ALT-C), a disintegrin-like-Cys-rich protein isolated from the venom of the Brazilian snake Bothrops alternatus (Souza et al., 2000) binds specifically to the integrin α2β1, one of the major collagen receptors in several cell types. ALT-C induces in vitro migration of human neutrophils by triggering classical integrin-mediated intracellular signaling including FAK and IP3K phosphorylation, Erk translocation to the nucleus and actin polymerization, crucial steps for the cell migration (Mariano-Oliveira et al., 2003). In addition, ALT-C is a strong inducer of HUVEC proliferation in vitro, by up-regulating the expression of VEGF (vascular endothelial growth factor) and of its receptor VEGFR 2 (vascular endothelial growth factor receptor 2) (Cominetti et al., 2004). DisBa-01 is a recombinant RGD-disintegrin from B. alternatus venom that inhibits αvβ3 integrin binding (Ramos et al., 2008) as well as cell migration (Selistre-de-Araujo et al., 2010). DisBa-01 has potent antithrombotic activity (Kauskot et al., 2008) and it was shown to inhibit experimental metastasis in nude mice (Ramos et al., 2008). To contribute for a better understanding of SVMP complexity, we constructed a cDNA library from the venom glands of a single specimen of B. alternatus, characterized its tissue expression pattern and identified the sequence of several metalloproteases. This data may have practical

75

Table 1 Comparative analysis of the two transcriptomes from B. alternatus.

Sequenced ESTs (tn*) Singlets (%) Contigs (%) No hits ESTs (%**) Cell processs ESts (%) Toxic ESTs (%) SVMPs, total (%***) PI SVMP (%****) PII SVMP (%****) PIII SVMP (%****) BBPs (%***) C-type lectins (%***) PLA2s (%***) Serine proteases (%***) Other transcripts (%***)

Cardoso et al. (2010)

This work

5350

1920 33 67 21 18 61 59 4 7 89 12 16 1.5 5 6.5

70 7 23 81 0 nd nd 8 1.5 5.6 2 2

*tn = total number; **% from the total sequenced ESTs; *** % from toxic ESTs; **** from SVMP ESTs; nd = not determined.

applications in the design of new integrin-binding drugs for modulation of integrin activity. 2. Materials and methods 2.1. Venom gland extraction One male specimen of B. alternatus (family Viperidae, subfamily Crotalinae) was kindly provided by Prof. Dr. Augusto Shinya Abe (Department of Zoology, UNESP — Rio Claro, Brazil). Three days before the removal of the glands the venom was collected for increasing mRNA synthesis. The glands were extracted and kept on liquid nitrogen in TRIzol reagent (Invitrogen®) until use. 2.2. RNA extraction and mRNA isolation Total RNA from the venom gland was extracted by the method previously described (Chomczynski and Sacchi, 1987) slightly modified;

Fig. 1. Transcripts in Bothrops alternatus venom gland. A) General distribution; B) transcripts related to cellular physiology; C) transcripts related to toxins; and D) percentage of each SVMP class among the metalloprotease transcripts.

76

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

briefly, the tissue was homogenized and nucleic acids and proteins were separated by the addition of phenol/chloroform, the upper phase was precipitated with isopropanol, washed with ethanol 70% and eluted

in RNAse free water. The RNA integrity was verified in a formaldehyde agarose gel electrophoresis. The mRNA was isolated using PolyATtract® mRNA Isolation Systems kit (Promega®) following fabricant instructions.

Table 2 Identification of metalloproteases/disintegrins-encoding sequences from B. alternatus venom gland. Sequencea

Access

Blastx

Motif

a1.10 a1.5 a2.10 a2.2 a2.3 a2.4 a2.5 a2.7 b1.10 b1.4 b1.6 b1.7 b1.9 b2.5 b2.7 b2.9 c1.1 c1.10 c1.11 c1.9 c2.12 c2.5 c2.8 d1.10 d1.11 d1.12 d1.2 d2.1 d2.12 d2.2 d2.3 d2.7 d2.9 e1.10 e1.12 e1.2 e1.4 e1.5 e1.7 e2.1 e2.11 e2.2 e2.6 f1.1 f1.10 f1.12 f1.3 f1.7 f1.9 f2.1 f2.10 f2.11 f2.12 f2.2 f2.3 f2.5 f2.6 g1.3 g1.4 g1.5 g1.9 g2.2 g2.3 g2.4 h1.1 h1.12 h1.2 h1.4 h2.9

gb|AAM09692.1| emb|CAA48323.1| gb|ABD34831.1| sp|P85314.1| gb|AAM09692.1| gb|AAK15542.1| gb|AAG48931.5| gb|AAK15542.1| gb|AAK15542.1| gb|ABD34835.1| gb|AAF28364.1| gb|AAK15542.1| gb|AAK15542.1| gb|AAP78951.1| gb|AAM09692.1| gb|AAC61986.2| emb|CAA48323.1| gb|AAL47169.1| gb|AAL47169.1| gb|ABB76282.1| gb|AAG48931.5| gb|AAK15542.1| gb|AAP78951.1| gb|AAC61986.2| sp|P84035.1| gb|AAG48931.5| gb|AAC61986.2| gb|AAC61986.2| gb|ABD34832.1| sp|P85420.1| gb|AAK15542.1| gb|AAK15542.1| gb|ABD34830.1| gb|AAL47169.1| gb|AAC61986.2| gb|ABD73129.1| gb|AAC61986.2| gb|AAK15542.1| gb|AAM09692.1| gb|AAM09693.1| gb|AAM09692.1| gb|ABD73129.1| gb|ABB76281.1| gb|AAC61986.2| gb|AAK15542.1| gb|AAK15542.1| gb|AAL47169.1| gb|AAC61986.2| gb|AAK15542.1| sp|P0C7B1.1| gb|AAK15542.1| gb|ABD34832.1| gb|AAC61986.2| gb|AAC61986.2| gb|AAK15542.1| gb|ABD34830.1| gb|AAM09692.1| gb|AAM09693.1| gb|AAK15542.1| gb|AAK15542.1| gb|AAG48931.5| gb|AAM09692.1| gb|AAG48931.5| gb|AAC61986.2| gb|AAK15542.1| gb|AAL47169.1| gb|AAM09692.1| gb|AAL47169.1| sp|P84907.1|

Metalloproteinase precursor [Bothrops insularis] Jararhagin [Bothrops jararaca] Venom metalloproteinase bothrojarin2 [Bothrops jararaca] Zinc metalloproteinase BmooMPalfa-I Metalloproteinase precursor [Bothrops insularis] Bothrostatin precursor [Bothrops jararaca] Hemorrhagic metalloproteinase HF3 [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Venom metalloproteinase and disintegrin jararacin [Bothrops jararaca] Non-hemorrhagic fibrin(ogen)olytic metalloprotease [Bothrops neuwiedi] Bothrostatin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Metalloprotease BOJUMET II [Bothrops jararacussu] Metalloproteinase precursor [Bothrops insularis] Bothropasin precursor [Bothrops jararaca] Jararhagin [Bothrops jararaca] Berythractivase [Bothrops erythromelas] Berythractivase [Bothrops erythromelas] Type I metalloproteinase [Bothrops asper] Hemorrhagic metalloproteinase HF3 [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Metalloprotease BOJUMET II [Bothrops jararacussu] Bothropasin precursor [Bothrops jararaca] Zinc metalloproteinase basparin-A Hemorrhagic metalloproteinase HF3 [Bothrops jararaca] Bothropasin precursor [Bothrops jararaca] Bothropasin precursor [Bothrops jararaca] Venom metalloproteinase bothrojarin2 [Bothrops jararaca] Zinc metalloproteinase atroxlysin-1 (Atroxlysin-I) Bothrostatin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Venom metalloproteinase bothrojarin2 [Bothrops jararaca] Berythractivase [Bothrops erythromelas] Bothropasin precursor [Bothrops jararaca] RGD-P-III class hemorrhagic metalloprotease I [Bothrops jararacussu] Bothropasin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Metalloproteinase precursor [Bothrops insularis] Metalloproteinase precursor [Bothrops insularis] Metalloproteinase precursor [Bothrops insularis] RGD-P-III class hemorrhagic metalloprotease I [Bothrops jararacussu] Type II metalloproteinase [Bothrops asper] Bothropasin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Berythractivase [Bothrops erythromelas] Bothropasin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Zinc metalloproteinase-disintegrin BaG Bothrostatin precursor [Bothrops jararaca] Venom metalloproteinase bothrojarin2 [Bothrops jararaca] Bothropasin precursor [Bothrops jararaca] Bothropasin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Venom metalloproteinase bothrojarin2 [Bothrops jararaca] Metalloproteinase precursor [Bothrops insularis] Metalloproteinase precursor [Bothrops insularis] Bothrostatin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Hemorrhagic metalloproteinase HF3 [Bothrops jararaca] Metalloproteinase precursor [Bothrops insularis] Hemorrhagic metalloproteinase HF3 [Bothrops jararaca] Bothropasin precursor [Bothrops jararaca] Bothrostatin precursor [Bothrops jararaca] Berythractivase [Bothrops erythromelas] Metalloproteinase precursor [Bothrops insularis] Berythractivase [Bothrops erythromelas] Zinc metalloproteinase leucurolysin-a (leuc-a)

Reprolysin family propeptide Reprolysin family propeptide

a

Codes for B. alternatus SVMPs singlets and contigs.

Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide

Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Disintegrin; Zinc-dep. metalloprot. Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide snake disintegrins; Reprolysin family propeptide Zinc-dependent metalloprotease Reprolysin family propeptide Reprolysin family propeptide

Reprolysin family propeptide Reprolysin family propeptide x Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide

Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide related Zn-dependent metalloprot related Zn-dependent metalloprot Reprolysin family propeptide Reprolysin family propeptide Reprolysin family propeptide

Reprolysin family propeptide Reprolysin family propeptide

Hit

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

77

Fig. 2. Sequence comparison of the active catalytic domain of metalloproteases predicted from fully sequenced clones from B. alternatus. Each sequence is coded by a number (left) which represents individual clones. Identical residues in all sequences are represented in red; identical residues in the majority of sequences are represented in blue, and black residues mean no consensus in those positions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

2.3. cDNA library The cDNA library was constructed using 5 μg of mRNA with CloneMiner™ cDNA Library Construction Kit (Invitrogen®), which combines the SuperScript™ II Reverse Transcriptase with the Gateway® system according to manufacturer's instructions. The vector used was pDONR™222 (Invitrogen®) transformed by electroporation in Escherichia coli ElectroMAX DH10B T1 and selected in kanamicinsupplemented media. Plasmidial DNAs were isolated from clones and used as template for the sequencing reaction. Sequencing was carried out with DYEnamic™ ET dye terminator kit (GE Healthcare®) using M13 universal primer (GTAAAACGACGGCCAG) and several internal ones designed for complete sequencing of the coding transcripts. The sequencing reactions were run in a MegaBACE™ DNA analysis system (GE Healthcare®).

2.4. Bioinformatics The chromatograms were submitted to the dCAS — Desktop cDNA annotation system (Guo et al., 2009). The first step of this pipeline process is the low quality and vector contamination sequences trimming using PHRED (Ewing and Green, 1998 Ewing et al., 1998) and the Univec database. Only sequences with Phred N15 and longer than 80 bp were used in the assembly and generation of a consensus sequence using BLAST (Altschul et al., 1990) and Cap3 (Huang, 1999). Finally the sequences were faced up to the major databases available (NR — non-redundant database; EBML-EBI European Molecular Biology Laboratory; DDBJ — DNA Data Bank of Japan; PDB — protein data bank; GO — gene ontology; Mit-pla mitochondrial and plasmidial database of NCBI; RRNA — NCBI; Pfam (protein family sequence database); SMART (simple modular architecture research tool) and KOG (Eukaryotic Orthologous Groups)) which generated a plan with the search results in hyperlinks to allow the manual annotation. For the annotation only those with e-value lower than 10− 15 were considered. The results were deposited in the expressed sequence tag (ESTs) databank at the National Center for Biotechnology Information (NCBI), USA. Protein sequence alignments were done using MultAlin (Corpet, 1988) and Praline (Simossis and Heringa, 2003) and the structural superposition of catalytic domain of the BaP1, a P-I class from Bothrops asper, and bothropasin, a P-III class from Bothrops jararaca was with Pymol (The PyMOL Molecular Graphics System, Version 1.5.0.4, Schrödinger, LLC). PDB code for BaP1 is 2W14 (http://www.rcsb.org/pdb/explore/explore.do?

structureId=2W14) (Lingott et al., 2009) and for bothropasin is 3DSL (http://www.rcsb.org/pdb/explore/explore.do?structureId= 3DSL) (Muniz et al., 2008).

3. Results and discussion 3.1. Library quality About 1920 clones were sequenced and 812 were used for the generation of contigs after the quality cut off (NCBI EST databank GO787666–GO788432). The average size of cloned inserts was 487 bp and the shortest and longest sequences were 101 and 2313 bp, respectively. The sequences composing uniques were 268 (33%) and the other 544 (67%) formed 57 contigs. The contigs and singlets were classified into three main groups: (i) hypothetical proteins or mismatched in databases (unknown) corresponding to 20.9% (170 clones); (ii) products related to cellular processes or functions 17.8% (145 clones), and (iii) transcripts coding for toxic compounds 61.2% (497 clones) as shown in Fig. 1A. The expression pattern was similar to those found in other cDNA libraries for Bothrops snakes where the toxic transcripts were predominant such as in the libraries of B. jararaca (Cidade et al., 2006) and Bothrops insularis (Junqueira-de-Azevedo and Ho, 2002). Only Bothrops atrox (Neiva et al., 2009) presented more unknown transcripts. The transcripts related to cellular processes or functions were divided into 6 subcategories (Fig. 1B): (i) metabolism (39%), those related to anabolism and catabolism of nutrients; (ii) components associated to transcription and translation processes (27%), such as binding factors and enzymes; (iii) processing (17%), those related to post-translational modifications; (iv) homeostasis (7%), the transcripts related to cellular regulation; (v) degradation (6%), those related to breakdown of polypeptides; and (vi) structural function (4%). The most expressed subcategories were those related to metabolism, processing, transcription and translation (Fig. 1B). This is indeed expected in a highly active and specialized tissue in the synthesis and secretion of toxic compounds. A similar pattern was found for other Bothrops snakes as well as in other genera (Junqueira-de-Azevedo and Ho, 2002; Kashima et al., 2004; Cidade et al., 2006). The toxic coding transcripts were divided into 8 subcategories (Fig. 1C): metalloproteases (58.5%), C-type lectins (16.5%), serine proteases (5.1%), BPP (bradykinin potentiating peptides, 11.8%), CRISP (Cys-rich secretory proteins, 2.2%), VEGF (vascular endothelial growth factor)-like proteins (2.5%), LAO (L-aminoacid oxidase, 1.8%) and

78

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

PLA2 (phospholipases A2, 1.4%). Corroborating with other cDNA libraries of Bothrops genera, the most abundant transcripts were those related to the metalloprotease protein family. However, the pattern of expression of other toxic transcripts is more variable among Bothrops species and this is known to be due not only to speciation, but seasonality, sex, age and food availability can also interfere (Antunes et al., 2010; de Morais-Zani et al., 2013; Dias et al., 2013). Cardoso et al. (2010) reported a complete transcriptome analysis of the venom gland of Bothrops alternatus but their results are significantly different from the ones described here. About 5000 expressed sequence tags (ESTs) were assembled into 838 contigs and 4512 singletons, which were mostly related to toxin transcripts. The major differences between the work done by Cardoso et al. and ours can be compared in Table 1. One possible explanation for the differences is the fact that cDNA library from the other group was made by mixing the two venom glands from three distinct snake individuals, which, despite enriched in general information, may lose individual variability. We could not find any information on sex, age or geographic origin of the animals used in that study. It is largely known by the literature that venom composition can be strongly affected by those variables (de Morais-Zani et al., 2013; Dias et al., 2013). In our work, we used one single venom gland from an adult male specimen of B. alternatus from São Paulo State. Compared to the work from Cardoso et al. (2010), our data demonstrated that the number of individual snake venom glands used in library generation also impacts the transcriptome analyzed. 3.2. Analysis of SVMP sequences The metalloprotease group was formed by 28 contigs from which 13 full-length sequences were subjected to rounds of extension using internal primers for the distinct domains. Those sequences were

Fig. 4. Predicted amino acid sequences of transcripts coding for the signal peptide only. Each number represents a distinct transcript.

classified into three groups according to the presence of stop codons or additional domains C-terminally to the catalytic domain as P-I (4%), P-II (7%) and P-III (89%) (Fig. 1D and Table 2, Supl. 1). The PIII-group comprises the major expressed class of SVMP, as also found for other Bothrops species (Cidade et al., 2006; Cardoso et al., 2010). In addition to the transcriptome studies, proteomic techniques also demonstrated that P-III SVMPs were the majority of expressed proteins in the venom of B. alternatus (Ohler et al., 2010). These authors found about 46% of P-III SVMPs but other discrepancies from our work were observed such as higher expression levels of serine proteases (24%) and PLA2s (7%) and lower levels of lectin type-C (1.7%). Once again, the venom used in this work was a mixture from different animals from both sexes, which can explain the differences in the results between distinct groups. In addition to geographical or ontogenic differences in the snakes, the role of post-transcriptional and posttranslational events is likely to be equally relevant, as recently discussed by Casewell et al. (2014). By using a combination of integrated multispecies approach, coupling molecular, proteomic, and evolutionary methodologies, these authors demonstrated that events occurring after RNA transcription largely determine the final protein composition of the venom.

Fig. 3. Sequence comparison of the disintegrin, disintegrin-like and Cys-rich domains of metalloproteases predicted from fully sequenced clones from B. alternatus. Each sequence is coded by a number (left) which represents individual clones. Underlined residues represent the adhesive motifs in P-II (RGD) and in P-III (ECD) SVMPs. The sequences of previously described DisBa-01 and ALT-C disintegrins were also included for comparison. X represents non-determined amino acid residues in ALT-C by protein sequencing.

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

79

Fig. 5. DNA and predicted protein sequences of a homolog of alternagin and ALT-C (ALT-C-h). The presence of signal peptide, prodomain, metalloprotease and disintegrin, Cys-rich domains is indicated. The 5′ and 3′ UTRs are represented in italics. ALT-C amino acid sequenced is also included for comparison (shaded); blank spaces indicate regions that were not sequenced. Differences between ALT-C and ALT-C-h are underlined.

80

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

In the search for dissimilarities among SVMP classes, we compared the deduced amino acid sequences of the completed sequences from the transcripts (Figs. 2 and 3). Regarding the catalytic domain, all the sequences show high level of identity within the same class but significant differences are observed comparing different classes (Fig. 2). We also compared the deduced sequences from the disintegrin and DC domains (Fig. 3). In all PIII complete sequences, the presence of the conserved ECD motif was observed (Fig. 3). Studies with synthetic peptides related to an extended ECD motif have demonstrated to reproduce the effects of the whole DC domain, although significant higher values for effective doses were needed (Ramos et al., 2007). Five clones of P-II class SVMPs were fully sequenced and all of them were demonstrated to be homologs to DisBa-01, an RGDdisintegrin previously described by our group (Ramos et al., 2008). Interestingly, two clones had extended C-terminus and another one had a C13W substitution (numbering as in Fig. 3). We could not find disintegrins from the PII SVMPs with other adhesive motif than RGD. Therefore we conclude that RGD-disintegrins are the types of disintegrin mostly expressed in the B. alternatus venom. However, extensive sequencing must be accomplished in order to detect non-RGD disintegrins that may be under represented in the venom gland library. Three clones were found coding for a signal peptide and a small part of the prodomain only, comprising about predicted peptide of 105 amino acid residues (Fig. 4). As far as we know, this is the first description of such pattern of expression. However, the functional significance of these sequences or if it is just the side products of a highly complex mRNA processing, remains to be elucidated. One clone (10f06) belonging to the P-III group was named ALT-C-h (for ALT-C homolog) has 2343 bp and an open reading frame of 1905 bp, predicting a preprometalloprotease with disintegrin-like/ Cys-rich domains as usually found for this class of proteins (Fig. 5). The presence of signal peptide and of the conserved motif PKMCGV related to the cysteine switch mechanism (Grams et al., 1993) is observed. The sequence codes for a mature protein of 421 amino acid residues with an estimated mass of 47.1 kDa, value that differs from the one reported previously for ALT-C (55 kDa) (Souza et al., 2000). This difference could be explained by glycosylation of the native protein since two putative glycosylation sites were observed at N264 (NYTL) and N372 (NCSY). The mature protein has the highly conserved zincbinding motif HEXXHXXGXXH of the reprolysin metalloprotease family (Gomis-Rüth et al., 1993). The metalloprotease domain is followed by the disintegrin-like, the Cys-rich domains and the 3′ untranslated region. The DNA sequence presented in this work is about 99% similar to bothropasin, a PIII SVMP from B. jararaca venom and has more than 90% of similarity when compared with other SVMPs deposited in GenBank (Table 2, Supl. 1). The partial amino acid sequence of ALT-C determined by protein sequencing (Souza et al., 2000) was aligned to the deduced sequence of ALT-C-h as with other sequences from disintegrin, Cys-rich domains from P-III SVMPs (Figs. 3 and 5). The sequence of ALT-C-h matches almost 100% with the sequence determined by Edman sequencing, with the exception of two residues, G567 and A593. Interestingly, these two residues are highly conserved in other P-III SVMPs and therefore may represent mistakes in the protein sequencing of ALT-C. This may be probably due to the high amount of expressed isoforms in the venom gland, since the PIII-SVMP represents more than 50% of expressed toxin genes. It is also possible that ALT-C preparation represents a mixture of isoforms but this remains to be elucidated. It is generally believed that P-III SVMPs have stronger hemorrhagic activity than the P-II and P-I; in addition, the disintegrin domain is easily processed from its precursor form of the P-II class, in contrast with the disintegrin-like, Cys-rich domains, which can be isolated from the venom as a full protein having catalytic and disintegrin-like, Cys-rich domains (Fox and Serrano, 2008). Therefore we decided to search for

possible explanations by comparing the sequences from representative members of the three classes from different venoms. The sequence analysis of the catalytic domain of the 13 complete sequences from this work and of some representatives from other snake venoms demonstrated significant differences between the three classes of SVMPs (Fig. 6 and Suppl. 1), in addition to the well-described differences in the disintegrin and disintegrin-like domains (Ramos and Selistre-deAraujo, 2006; Fox and Serrano, 2008). Sequence alignment of the catalytic domain from these sequences showed 67% of identity for P-I, 70% for PII and 72% for P-III class, and the following residues are extremely conserved in the P-III sequences but not in P-I or P-II: K72, P73, H122, Y133, P135, V139, P167, E206, and P215. We speculate if some of the conserved residues could have a role in metalloprotease function or domain processing. For that, we superpose the threedimensional structures of the catalytic domain of one P-I SVMP (BaP1, PDB code 2W14) and of the one P-III SVMP (bothropasin, PDB code 3DSL) and we analyzed the residues cited above (Fig. 7). It is possible to see that the side chain of residues Y133, P135, V139, P167 are closely exposed and it could be a site for molecular recognition. It is important to note that small differences in residue numbering is due to the gaps introduced (or not) by the alignment softwares, such as in Fig. 7. In addition, K72 is changed to Q/T and H122 is replaced by an aspartate residue in the majority of other P-I and P-II proteins. On the other hand, E206 and P215 are located in the processing region of the disintegrin domain and may have a role in the release of this domain from the precursor. We also performed an in silico analysis of peptidase cleavage sites (Gasteiger et al., 2005) of bothropasin and BaP1. Interestingly, bothropasin has two cleavage sites for prolyl-endopeptidase at P73 and P123, which are not found in P-I or P-II proteins. Prolineendopeptidases (EC 3.4.21.26) are serine proteases that preferentially cleave at Pro in position P1 (Kay, 1993) and in most cases a basic amino acid (Lys, His, Arg) is found in position P2. This is just the case of P-III SVMPs, where P73 and P123 are preceded by a conserved K72 and H122. These data means that P-III SVMPs have a prolylendopeptidase cleavage site that is not found in most P-I or P-II; however, the functional relevance of this finding remains to be investigated. Prolyl-endopetidase activity is known to be blocked when another Pro is at position P1′. Interestingly, P-III SVMPs have a double PP at positions 216–217, while P217 is highly conserved in all SVMP classes, P216 is conserved only in P-III proteases. The presence of prolyl peptidases in snake venom was first described in Elapidae venom (Vonk et al., 2013) and more recently related to a P-I metalloprotease (Okamoto et al., 2014) but its function in the venom is still not well understood. According to our results, the P-III SVMPs could act as substrates for venom prolyl endopeptidases but this hypothesis remains to be experimentally confirmed. In conclusion, our data contribute to the understanding of venom complexity and provide new clues to achieve a better comprehension of the molecular mechanism of action of SVMPs and domain processing.

Acknowledgments This work was supported by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo, SP, Brazil, Grant 1998/14138-2) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil, Grant 303249/2009-9).

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.cbd.2014.09.001.

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

81

82

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83

Fig. 7. Sequential alignment and superposition of three-dimensional structures of the catalytic domain of BaP1 (magenta) and bothropasin (green). On the left an overview of the superposition of the catalytic domain showing the residues discussed in the text. On the right a closer view of the region. Catalytic zinc is represented as a gray ball. Numbering is slightly different from the text due to gaps introduced (or not) by the alignment softwares. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

References Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403–410. http://dx.doi.org/10.1016/S0022-2836(05) 80360-2. Antunes, T.C., Yamashita, K.M., Barbaro, K.C., Saiki, M., Santoro, M.L., 2010. Comparative analysis of newborn and adult Bothrops jararaca snake venoms. Toxicon 56, 1443–1458. http://dx.doi.org/10.1016/j.toxicon.2010.08.011. Bjarnason, J.B., Fox, J.W., 1994. Hemorrhagic metalloproteinases from snake venoms. Pharmacol. Ther. 62, 325–372. http://dx.doi.org/10.1016/0163-7258(94)90049-3. Calvete, J.J., 2013. The continuing saga of snake venom disintegrins. Toxicon 62, 40–49. http://dx.doi.org/10.1016/j.toxicon.2012.09.005. Cardoso, K.C., Da Silva, M.J., Costa, G.G.L., Torres, T.T., Del Bem, L.E.V., Vidal, R.O., Menossi, M., Hyslop, S., 2010. A transcriptomic analysis of gene expression in the venom gland of the snake Bothrops alternatus (urutu). BMC Genomics 11, 605. http://dx.doi.org/10. 1186/1471-2164-11-605. Casewell, N.R., Wagstaff, S.C., Wüster, W., Cook, D.A.N., Bolton, F.M.S., King, S.I., Pla, D., Sanz, L., Calvete, J.J., Harrison, R.A., 2014. Medically important differences in snake venom composition are dictated by distinct postgenomic mechanisms. PNAS 111, 9205–9210. http://dx.doi.org/10.1073/pnas.1405484111. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156–159. http://dx.doi.org/10.1006/abio.1987.9999. Cidade, D. a P., Wermelinger, L.S., Lôbo-Hajdu, G., Dávila, A.M.R., Bon, C., Zingali, R.B., Albano, R.M., 2006. Molecular diversity of disintegrin-like domains within metalloproteinase precursors of Bothrops jararaca. Toxicon 48, 590–599. http://dx.doi.org/ 10.1016/j.toxicon.2006.07.010. Cominetti, M.R., Terruggi, C.H.B., Ramos, O.H.P., Fox, J.W., Mariano-Oliveira, A., De Freitas, M.S., Figueiredo, C.C., Morandi, V., Selistre-de-Araujo, H.S., 2004. Alternagin-C, a disintegrin-like protein, induces vascular endothelial cell growth factor (VEGF) expression and endothelial cell proliferation in vitro. J. Biol. Chem. 279, 18247–18255. http://dx.doi.org/10.1074/jbc.M311771200.

Corpet, F., 1988. Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res. 16, 10881–10890. http://dx.doi.org/10.1093/nar/16.22.10881. De Morais-Zani, K., Grego, K.F., Tanaka, A.S., Tanaka-Azevedo, A.M., 2013. Proteomic analysis of the ontogenetic variability in plasma composition of juvenile and adult Bothrops jararaca snakes. Int. J. Prot. 2013, 135709. http://dx.doi.org/10.1155/2013/ 135709. Desgrosellier, J.S., Cheresh, D.A., 2010. Integrins in cancer: biological implications and therapeutic opportunities. Nat. Rev. Cancer 10, 9–22. http://dx.doi.org/10.1038/ nrc2748. Dias, G.S., Kitano, E.S., Pagotto, A.H., Sant'anna, S.S., Rocha, M.M.T., Zelanis, A., Serrano, S.M.T., 2013. Individual variability in the venom proteome of juvenile Bothrops jararaca specimens. J. Proteome Res. 12, 4585–4598. http://dx.doi.org/10.1021/pr4007393. Ewing, B., Green, P., 1998. Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8, 186–194. http://dx.doi.org/10.1101/gr.8.3.186. Ewing, B., Hillier, L., Wendl, M.C., Green, P., 1998. Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8, 175–185. http://dx.doi. org/10.1101/gr.8.3.175. Felding-Habermann, B., O'Toole, T.E., Smith, J.W., Fransvea, E., Ruggeri, Z.M., Ginsberg, M.H., Hughes, P.E., Pampori, N., Shattil, S.J., Saven, A., Mueller, B.M., 2001. Integrin activation controls metastasis in human breast cancer. Proc. Natl. Acad. Sci. U. S. A. 98, 1853–1858. http://dx.doi.org/10.1073/pnas.98.4.1853. Fox, J.W., Serrano, S.M.T., 2008. Insights into and speculations about snake venom metalloproteinase (SVMP) synthesis, folding and disulfide bond formation and their contribution to venom complexity. FEBS J. 275, 3016–3030. http://dx.doi.org/10.1111/j. 1742-4658.2008.06466.x. Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M.R., Appel, R.D., Bairoch, A., 2005. Protein identification and analysis tools on the ExPASy server. In: Walker, J.M. (Ed.), The Proteomics Protocols Handbook. Humana Press, pp. 571–607 http:// dx.doi.org/10.1385/1-59259-890-0:571. Gomis-Rüth, 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.

Fig. 6. Sequence comparison of P-I, P-II and P-III SVMPs from different venoms using the Multalin software. P-I sequences are: BaP1 (PDB 2W14), VM1B_BOTIN (Q8QG89), ACLH (U18234), ACLF (U18233), fibrolase (AAB26922), Gloydius (AY204247), VM1B1 (P83512); P-II sequences are: VM2J2 (Q98SP2), metalloprotease 2 (FJ429180), MP_IIx1 (ADO21508), MP_IIx2 (ADO21509), MP_IIx3 (ADO2510), MP_IIb1 (ADO21506); P-III sequences are: jararhagin (P30431), bothropasin (O93523) catrocollastatin (AAC59672), acurhagin (AAS57937), acutolysin-E (Q9W6M5), ACLD (AAC18911). Accession numbers to sequences in the GenBank (www.ncbi.nlm.nih.gov) are presented inside brackets. All other sequences are from this work. Residues in red are identical in all sequences, in blue, residues conserved in the majority of sequences, and residues in black mean no consensus for that position. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

F.F.P. de Paula et al. / Comparative Biochemistry and Physiology, Part D 12 (2014) 74–83 Gould, R.J., Polokoff, M.A., Friedman, P.A., Huang, T.-F., Holt, J.C., Cook, J.J., Niewiarowski, S., 1990. Disintegrins: a family of integrin inhibitory proteins from viper venoms. Exp. Biol. Med. 195, 168–171. http://dx.doi.org/10.3181/00379727-195-43129B. Grams, F., Huber, R., Kress, L.F., Moroder, L., Bode, W., 1993. Activation of snake venom metalloproteinases by a cysteine switch-like mechanism. FEBS Lett. 335, 76–80. http://dx.doi.org/10.1016/0014-5793(93)80443-X. Guo, Y., Ribeiro, J.M.C., Anderson, J.M., Bour, S., 2009. dCAS: a desktop application for cDNA sequence annotation. Bioinformatics 25, 1195–1196. http://dx.doi.org/10. 1093/bioinformatics/btp129. Hooper, N.M., 1994. Families of zinc metalloproteases. FEBS Lett. 354, 1–6. http://dx.doi. org/10.1016/0014-5793(94)01079-X. Huang, X., 1999. CAP3: a DNA sequence assembly program. Genome Res. 9, 868–877. http://dx.doi.org/10.1101/gr.9.9.868. Junqueira-de-Azevedo,, I. de L., Ho, P.L., 2002. A survey of gene expression and diversity in the venom glands of the pit viper snake Bothrops insularis through the generation of expressed sequence tags (ESTs). Gene 299, 279–291. http://dx.doi.org/10.1016/ S0378-1119(02)01080-6. Kashima, S., Roberto, P.G., Soares, A.M., Astolfi-Filho, S., Pereira, J.O., Giuliati, S., Faria, M., Xavier, M.A.S., Fontes, M.R.M., Giglio, J.R., França, S.C., 2004. Analysis of Bothrops jararacussu venomous gland transcriptome focusing on structural and functional aspects: I—gene expression profile of highly expressed phospholipases A2. Biochimie 86, 211–219. http://dx.doi.org/10.1016/j.biochi.2004.02.002. Kauskot, A., Cominetti, M.R., Ramos, O.H.P., Bechyne, I., Renard, J.-M., Hoylaerts, M.F., Crepin, M., Legrand, C., Selistre-de-Araujo, H.S., Bonnefoy, A., 2008. Hemostatic effects of the recombinant DisBa-01, a novel RGD-disintegrin from Bothrops alternatus. Front. Biosci. 13, 6604–6616 (http://dx.doi.org/10.2741/3176). Kay, J., 1993. Specificity of proteolysis. FEBS Lett. 331, 201. http://dx.doi.org/10.1016/ 0014-5793(93)80331-N. Lingott, T.J., Schleberger, C., Gutierrez, J.M., Merfort, I., 2009. High-resolution crystal structure of the snake venom metalloproteinase bap1 complexed with a peptidomimetic: insight into inhibitor binding. Biochemistry 48, 6166. http://dx.doi.org/10.1021/bi9002315. Mariano-Oliveira, A., Coelho, A.L.J., Terruggi, C.H.B., Selistre-de-Araújo, H.S., Barja-Fidalgo, C., De Freitas, M.S., 2003. Alternagin-C, a nonRGD-disintegrin, induces neutrophil migration via integrin signaling. Eur. J. Biochem. 270, 4799–4808. http://dx.doi.org/10. 1046/j.1432-1033.2003.03867.x. Moura-da-Silva, A.M., Theakston, R.D., Crampton, J.M., 1996. Evolution of disintegrin cysteine-rich and mammalian matrix-degrading metalloproteinases: gene duplication and divergence of a common ancestor rather than convergent evolution. J. Mol. Evol. 43, 263–269. http://dx.doi.org/10.1007/BF02338834. Moura-da-Silva, A.M., Della-Casa, M.S., David, A.S., Assakura, M.T., Butera, D., Lebrun, I., Shannon, J.D., Serrano, S.M., Fox, J.W., 2003. Evidence for heterogeneous forms of the snake venom metalloproteinase jararhagin: a factor contributing to snake venom variability. Arch. Biochem. Biophys. 409, 395–401. Moura-da-Silva, A.M., Butera, D., Tanjoni, I., 2007. Importance of snake venom metalloproteinases in cell biology: effects on platelets, inflammatory and endothelial cells. Curr. Pharm. Des. 13, 2893–2905. Moura-da-Silva, A.M., Furlan, M.S., Caporrino, M.C., Grego, K.F., Portes-Junior, J.A., Clissa, P.B., Valente, R.H., Magalhães, G.S., 2011. Diversity of metalloproteinases in Bothrops neuwiedi snake venom transcripts: evidences for recombination between different classes of SVMPs. BMC Genet. 12, 94. http://dx.doi.org/10.1186/1471-2156-12-94. Muniz, J.R.C., Ambrosio, A.L.B., Selistre-de-Araujo, H.S., Cominetti, M.R., Moura-da-Silva, A.M., Oliva, G., Garratt, R.C., Souza, D.H.F., 2008. The three-dimensional structure of bothropasin, the main hemorrhagic factor from Bothrops jararaca venom: insights for a new classification of snake venom metalloprotease subgroups. Toxicon 52, 807–816. http://dx.doi.org/10.1016/j.toxicon.2008.08.021.

83

Neiva, M., Arraes, F.B.M., de Souza, J.V., Rádis-Baptista, G., Prieto da Silva, A.R.B., Walter, M.E.M.T., Brigido, M., de M., Yamane, T., López-Lozano, J.L., Astolfi-Filho, S., 2009. Transcriptome analysis of the Amazonian viper Bothrops atrox venom gland using expressed sequence tags (ESTs). Toxicon 53, 427–436. http://dx.doi.org/10.1016/j. toxicon.2009.01.006. Ohler, M., Georgieva, D., Seifert, J., von Bergen, M., Arni, R.K., Genov, N., Betzel, C., 2010. The venomics of Bothrops alternatus is a pool of acidic proteins with predominant hemorrhagic and coagulopathic activities. J. Proteome Res. 9, 2422–2437. http://dx. doi.org/10.1021/pr901128x. Okamoto, D.N., Kondo, M.Y., Oliveira, L.C.G., Honorato, R.V., Zanphorlin, L.M., Coronado, M.A., Araújo, M.S., da Motta, G., Veronez, C.L., Andrade, S.S., Oliveira, P.S.L., Arni, R.K., Cintra, A.C.O., Sampaio, S.V., Juliano, M.A., Juliano, L., Murakami, M.T., Gouvea, I.E., 2014. P-I class metalloproteinase from Bothrops moojeni venom is a post-proline cleaving peptidase with kininogenase activity: insights into substrate selectivity and kinetic behavior. Biochim. Biophys. Acta 1844, 545–552. http://dx.doi.org/10.1016/j. bbapap.2013.12.014. Ramos, O.H.P., Selistre-de-Araujo, H.S., 2006. Snake venom metalloproteases — structure and function of catalytic and disintegrin domains. Comp. Biochem. Physiol. C Vol. 142, 328–346. Ramos, O.H.P., Terruggi, C.H.B., Ribeiro, J.U., Cominetti, M.R., Figueiredo, C.C., Bérard, M., Crepin, M., Morandi, V., Selistre-de-Araujo, H.S., 2007. Modulation of in vitro and in vivo angiogenesis by alternagin-C, a disintegrin-like protein from Bothrops alternatus snake venom and by a peptide derived from its sequence. Arch. Biochem. Biophys. 461, 1–6. http://dx.doi.org/10.1016/j.abb.2007.02.021. Ramos, O.H.P., Kauskot, A., Cominetti, M.R., Bechyne, I., Pontes, C.L.P., Chareyre, F., Manent, J., Vassy, R., Giovannini, M., Legrand, C., Selistre-de-Araujo, H.S., Crépin, M., Bonnefoy, A., 2008. A novel avb3-blocking RGD-disintegrin, DisBa-01, Inhibits bFGF-induced angiogenesis and melanoma metastasis. Clin. Exp. Metast. 25, 53–64. http://dx.doi.org/10.1007/s10585-007-9101-y. Sanz, L., Harrison, R.A., Calvete, J.J., 2012. First draft of the genomic organization of a PIII-SVMP gene. Toxicon 60, 455–469. http://dx.doi.org/10.1016/j.toxicon.2012.04. 331. Selistre-de-Araujo, H.S., Pontes, C.L.S., Montenegro, C.F., Martin, A.C.B.M., 2010. Snake venom disintegrins and cell migration. Toxins 2, 2606–2621. http://dx.doi.org/10. 3390/toxins2112606. Simossis, V.A., Heringa, J., 2003 Octt. The PRALINE online server: optimising progressive multiple alignment on the web. Comput. Biol. Chem. 27 (4–5), 511–519. http://dx. doi.org/10.1016/j.compbiolchem.2003.09.002. Souza, D.H., Iemma, M.R., Ferreira, L.L., Faria, J.P., Oliva, M.L., Zingali, R.B., Niewiarowski, S., Selistre-de-Araujo, H.S., 2000. The disintegrin-like domain of the snake venom metalloprotease alternagin inhibits alpha2beta1 integrin-mediated cell adhesion. Arch. Biochem. Biophys. 384, 341–350. http://dx.doi.org/10.1006/abbi.2000.2120. Takeda, S., Takeya, H., Iwanaga, S., 2012. Snake venom metalloproteinases: structure, function and relevance to the mammalian ADAM/ADAMTS family proteins. Biochim. Biophys. Acta 1824, 164–176. http://dx.doi.org/10.1016/j.bbapap.2011.04.009. Vonk, F.J., Casewell, N.R., Henkel, C.V., Heimberg, A.M., Jansen, H.J., McCleary, R.J.R., Kerkkamp, H.M.E., Vos, R.A., Guerreiro, I., Calvete, J.J., Wüster, W., Woods, A.E., Logan, J.M., Harrison, R.A., Castoe, T.A., de Koning, A.P.J., Pollock, D.D., Yandell, M., Calderon, D., Renjifo, C., Currier, R.B., Salgado, D., Pla, D., Sanz, L., Hyder, A.S., Ribeiro, J.M.C., Arntzen, J.W., van den Thillart, G.E.E.J.M., Boetzer, M., Pirovano, W., Dirks, R.P., Spaink, H.P., Duboule, D., McGlinn, E., Kini, R.M., Richardson, M.K., 2013. The king cobra genome reveals dynamic gene evolution and adaptation in the snake venom system. Proc. Natl. Acad. Sci. U. S. A. 110, 20651–20656. http://dx.doi. org/10.1073/pnas.1314702110.