Expression of a Recombinant Phoneutria Toxin Active in Calcium Channels

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Toxicon 60 (2012) 907–918

Contents lists available at SciVerse ScienceDirect

Toxicon

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

Expression of a recombinant Phoneutria toxin active in calcium channels

I.A. Souza a,c, E.A. Cino d, W.Y. Choy d, M.N. Cordeiro e, M. Richardson e, C. Chavez-Olortegui f, M.V. Gomez c,g, M.A.M. Prado a,b,c, V.F. Prado a,b,c,*

a Molecular Brain Research Group, Robarts Research Institute, Department of Anatomy & Cell Biology, Schulich School of Medicine & Dentistry, University of Western Ontario, Canada b Department of Physiology and Pharmacology, University of Western Ontario, Canada c Program in Molecular Pharmacology, School of Medicine, Federal University of Minas Gerais, Brazil d Department of Biochemistry, University of Western Ontario, Canada e Ezequiel Dias Foundation, Brazil f Department of Biochemistry, Federal University of Minas Gerais, Brazil g Program in Biomedicine, I.E.P. Santa Casa, Belo Horizonte, Minas Gerais, Brazil

article info abstract

Article history: PnTx3-4 is a toxin isolated from the venom of the spider Phoneutria nigriventer that blocks Received 27 February 2012 N-, P/Q-, and R-type voltage-gated calcium channels and has great potential for clinical Received in revised form 24 April 2012 applications. In this report we used the SUMO system to express large amounts of Accepted 24 May 2012 recombinant PnTx3-4 peptide, which was found in both soluble and insoluble fractions of Available online 31 May 2012 bacterial extracts. We puriﬁed the recombinant toxin from both fractions and showed that the recombinant peptide showed biological activity similar to the native PnTx3-4. In silico Keywords: analysis of the primary sequence of PnTx3-4 indicated that the peptide conforms to all the Phoneutria nigriventer Spider toxin criteria of a knottin scaffold. Additionally, circular dichroism spectrum analysis of the Calcium channel blocker recombinant PnTx3-4 predicted that the toxin structure is composed of approximately 53% Bacterial expression turns/unordered, 31% a-helix and 16% b-strand, which is consistent with predicted model Circular dichroism of the PnTx3-4 knottin scaffold available at the knottin database (http://knottin.cbs.cnrs. Knottin scaffold fr). These studies provide the basis for future large scale production and structure- function investigation of PnTx3-4. Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.

1. Introduction calcium, potassium, sodium and ligand-gated channels (Doering and Zamponi, 2003; Li and Tomaselli, 2004; Peptide toxins obtained from animal venoms are Castellino and Prorok, 2000; Lewis et al., 2000; Favreau resourceful compounds to investigate ion channels, et al., 1999). Peptide toxins have also been used as poten- contributing to our understanding of key channels regu- tial lead compounds for the development of novel thera- lating excitability of neurons and cardiomyocytes. Toxins peutic drugs (Alonso et al., 2003; Heading, 2002; Jones and obtained from the venom of different spiders and sea snails Bulaj, 2000; Livett et al., 2004; Lewis, 2009). Importantly, have provided the framework to understand the structure– a synthetic neuroactive peptide equivalent to the u-con- function relationship of a variety of channels including otoxin MVIIA, one of the toxins that target voltage-gated calcium channels, has been approved for the treatment of pain (Williams et al., 2008).

Abbreviations: Gnd-HCl, Guanidine hydrochloride; IPTG, Isopropyl b-D- Calcium is essential in many physiological mechanisms 1-thiogalactopyranoside; RP-HPLC, Reverse phase-high performance including hormone and neurotransmitter release, muscle liquid chromatography. contraction and gene transcription; however, excess * Corresponding author. Robarts Research Institute, P.O. Box 5015, 100 calcium inﬂux can generate a cascade of events that cause Perth Dr., London, ON N6A 5K8, Canada. Tel.: þ1 519 663 5777x24889; fax: þ1 519 931 5789. cytotoxicity and cell death, making calcium a key player E-mail addresses: [email protected], [email protected] (V.F. Prado). in ischemic neuronal death (Lau and Tymianski, 2010;

Arundine and Tymianski, 2003; Sattler and Tymianski, 2. Materials and methods 2000). After an ischemic injury, calcium floods into neurons through different channels including voltage-gated Oligonucleotides used for PCR reactions were synthe- calcium channels, ionotropic glutamate receptors such as sized by Sigma. Restriction endonucleases were purchased N-methyl-D-aspartate (NMDA) and a-amino-3-hydroxy-5- from New England Biolabs. The pE-SUMO LIC vector and metil-4-isoxiazole propionate (AMPA) receptors (Lau and SUMO protease I were obtained from LifeSensors Inc. Tymianski, 2010). Therefore, there is an intensive search (Malvern, USA). ORIGAMI (DE3) competent cells were for calcium channel blockers and glutamate receptors supplied by Novagen Inc. (Madison, USA). Acetonitrile, antagonists in the attempt to develop novel neuroprotective Fura-2AM, glutamate dehydrogenase and Percoll were drugs (Domin et al., 2010; Lipton, 2007, 2006). obtained from Sigma Chemical Co. (MO, USA). The venom of the Brazilian ‘armed’ spider Phoneutria nigriventer has a number of peptides that are effective 2.1. Plasmid construction blockers of distinct calcium, potassium and sodium channels (de Castro Junior et al., 2008; Vieira et al., 2007, 2005; \Four oligonucleotides named Tx34A53, Tx34A35, Cardoso et al., 2003; Carneiro et al., 2003; Vieira et al., 2003; Tx34B53 and Tx34B35 were used as template for the PCR Reis et al., 2000; Penaforte et al., 2000; Reis et al., 1999; reaction that produced the coding region for the PnTx3-4 Kushmerick et al., 1999; Mesquita et al., 1998; Kalapothakis toxin (Table 1). Another two oligonucleotides, Tx34SUMOF et al., 1998b, 1998a; Moura et al., 1998; Miranda et al., 1998; and Tx34SUMOR (Table 1) were used as primers of the same Guatimosim et al., 1997; Prado et al., 1996). Three of these reaction. Bsa I and Xba I restriction sites (underlined) were toxins, named PnTx3-3, PnTx3-4 and PnTx3-6 are voltage- added at the 50 and 30 ends of the DNA, respectively, and gated calcium channel blockers that interfere with the a stop codon (TCA) was also added at the 30 end in order to release of glutamate from isolated nerve terminals stop translation after the last amino acid of the toxin. The (Carneiro et al., 2010; Prado et al., 1996; Souza et al., 2008; amplification reaction contained the template oligonucleo- de Castro Junior et al., 2008; Vieira et al., 2007, 2005; Reis tides in a 0.1 mM concentration, the amplification primers in et al., 1999). PnTx3-4 irreversibly inhibits P/Q and N-type a1mM concentration, 250 mM of each deoxynucleotide channels, whereas its action against R-type channels is triphosphate and 2 U of the thermostable recombinant Taq incomplete and reversible (Dos Santos et al., 2002). PnTx3-3 polymerase. The reactions were subjected to initial dena- and PnTx3-6 reversibly and non-specifically inhibit a broad turation of 4 min at 95 C and subsequent 40 cycles that spectrum of high-voltage-activated Ca2þ channels, namely consisted of denaturing the DNA at 95 C for 1 min, annealing L-, N-, P/Q-, and R-type, with varying potency (Vieira et al., at 55 C for 1 min and elongation at 72 C for 1 min. 2005, 2003; Leao et al., 2000). Recent studies have sug- The PCR product was purified using the QIAquick TM gested that these peptides can interfere with processes Gel Extraction kit (Qiagen, USA), digested with Bsa I and related to ischemia-induced glutamate release and Xba I and cloned into pE-SUMO LIC Vector. The plasmid responses to pain (Dalmolin et al., 2011; Agostini et al., 2011; obtained (6xHis-SUMO-PnTx3-4) encodes a fusion protein Pinheiro et al., 2009; Souza et al., 2008). These three composed of an 18.0 kDa Yeast SUMO protein (Smt3) fused peptides decrease glutamate release as well as neuronal cell at the N-terminal with a 6xHis tag and at the C-terminal death in retina slices submitted to ischemic injury (Agostini et al., 2011). Additionally, PnTx3-3 and PnTx3-6 have been Table 1 Oligonucleotide sequences used to generate the synthetic PnTx3-4 cDNA. shown to be effective for the control of neuropathic pain in animal models with no adverse motor effect (Dalmolin et al., Name Sequence Size (bp) 2011; Souza et al., 2008); PnTx3-4 attenuates neuronal Tx34SUMOF 50-AATTGGTCTCAAGGTTCTTGTATTAAC 36 0 death and electrophysiological consequences of oxygen and GTTGGTGAT-3 Tx3-4SUMOR 50-AATTTCTAGATCAATGGTTTTTTTTAC 34 glucose deprivation in brain slices (Pinheiro et al., 2009); 0 AACGTTT-3 and PnTx3-6 has analgesic effects in rodent models of Tx34A53 50-TGTATTAACGTTGGTGATTTTTGTGAT 120 chronic and acute pain (de Souza et al., 2011; Souza et al., GGTAAAAAAGATGATTGTCAGT 2008). Therefore, these peptides have the potential to be GTTGTCGTGATAACGCCTTTTGTTCTTGT used in the therapeutic management of pain and/or as TCTGTTATTTTTGGTTATAAAACCAACTG TCGTTGTGAAGTT-30 neuroprotective drugs. Tx34A35 50-AACTTCACAACGACAGTTGGTTTTATA 120 Purification of toxins from P. nigriventer’s venom is an ACCAAAAATAACAGAACAAGAACAAAAG expensive, inefficient and time-consuming process. More- GCGTTATCACGACAACACTGACAATCATC over, the yield for most toxins present in the venom is very TTTTTTACCATCACAAAAATCACCAACGTT AATACA-30 low (Cordeiro et al., 1993), making it difficult to complete Tx34B53 50-TGTCGTTGTGAAGTTGGTACCACCGCC 120 characterize these peptides. Furthermore, pharmacological ACCTCTTATGGTATTTGTATGGCCAAACA use of these peptides will only be feasible if they can be TAAATGTGGTCGTCAGACCACCTGTACCA produced in large scale. Generation of recombinant toxins AACCATGTCTGTCTAAACGTTGTAAAAAA 0 using Escherichia coli is an alternative approach and has been AACCAT-3 Tx34B35 50-ATGGTTTTTTTTACAACGTTTAGA 120 used previously to obtain functional recombinant toxins from CAGACATGGTTTGGTACAGGTG the P. nigriventer spider (Souza et al., 2008; Carneiro et al., GTCTGACGACCACATTTATGTT 2003). In this study, we demonstrate for the first time the TGGCCATACAAATACC ATAAGAGGIGGCGGTGGTACC functional expression of the toxin PnTx3-4, a valuable scaf- 0 fold for the development of new neuroprotective drugs. AACTTCACAACGACA-3 I.A. Souza et al. / Toxicon 60 (2012) 907–918 909 with the PnTx3-4 toxin (8.0 kDa). The sequence of the reduced with 10 mM DTT for 4 h at RT. Before the refolding, recombinant plasmid was confirmed by automatic the DTT was removed from the sample by filtration using sequencing using the dideoxynucleotide chain-termination VIVASPIN 6 (3 kDa MWCO). The sample was diluted 20 reaction (Sanger et al., 1977). times into the refolding buffer (550 mM Gnd-HCl, 440 mM L-arginine, 55 mM Tris, 21 mM NaCl, 0.88 mM KCl, 1 mM 2.2. Expression and purification of the fusion protein EDTA, 1 mM GSH and 1 mM GSSG, pH 8.2) to a final protein concentration of 0.1–0.2 mg/mL. The recombinant toxin E. coli ORIGAMI (DE3) cells containing pGro7 chaperone was added in 5 aliquots with a 2 min interval between each plasmid were transformed with 6xHis-SUMO-PnTx3-4. one to minimize the precipitation of folding intermediates. An isolated colony was inoculated in 10 mL of LB medium The reaction was performed at 4 C for 24 h. For desalting supplemented with Ampicillin (100 mgmL1) and and to check the refolded recombinant toxin HPLC profile, Chloramphenicol (20 mgmL1) for cultivation at 30 C the sample was submitted to a C18 Reverse Phase Chro- overnight. The culture was then diluted 100-fold in 1 L of LB matography and the elution samples were lyophilized and 1 medium (with antibiotics) containing 0.5 mg mL of L- kept at 20 C until needed. arabinose (for chaperones expression) and cultured to – an OD600 of 0.5 0.8. The expression of the recombinant 2.5. Protein analysis fusion protein was induced with 0.6 mM Isopropyl b-D-1- thiogalactopyranoside (IPTG), overnight, at 18 C with All purification steps were followed by SDS-PAGE and shaking. Cells were harvested by centrifugation at 5000 g Western blotting. Proteins were resolved on 4–20% for 15 min, suspended in 30 mL of Resuspension buffer gradient gels (Lonza) and stained with RAPIDstain Reagent – (20 mM Tris HCl, 150 mM NaCl, pH 7.5) containing (CALBIOCHEM) or transferred to a PVDF membrane (Milli- protease inhibitor cocktail (CALBIOCHEM), lysed by passing pore). The membrane was incubated overnight at 4 C with – twice through a French Press (16,000 18,000 psi) and cell anti-P. nigriventer total venom peroxidase antibodies debris were removed by centrifugation. (1:1250) and developed with ECL Plus Western blotting The recombinant toxin expressed in the supernatant Detection System (Amersham). was purified by affinity chromatography using a Ni-NTA agarose resin (Qiagen). After purification, elutions were pooled and dialyzed against 20 mM Tris–HCl, 150 mM NaCl, 2.6. Preparation of synaptosomes pH 7.5 for 24 h to remove imidazole. To purify the recombinant toxin present in inclusion bodies, the pellet All experiments were carried out in incompliance obtained after cell lysis was solubilised in a denaturing with the Canadian Council of Animal Care (CCAC) guidelines for the care and use of animals. The protocol was buffer (6 M Guanidine-HCl, 100 mM NaH2PO4, 10 mM Tris, 20 mM Imidazole pH 8.0) and incubated for 1 h at RT. The approved by the University of Western Ontario Institutional sample was then centrifuged at 32,000 g for 30 min and the Animal Care and Use Committee (protocol # 2008-127). All supernatant was purified by affinity chromatography using efforts were made to minimize the suffering of animals. a Ni-NTA agarose resin. The elutions were dialyzed first Cerebral cortices from adult mice (C57BL/6) were isolated against 1 M Guanidine-HCl, 0.05 M Tris, 0.15 M NaCl, pH 8.0 and homogenized in 0.32 M sucrose solution containing 1 mM EDTA and 0.25 mM DTT. The homogenate was and later against 0.1 M Gnd-HCl, 0.05 M Tris, 0.15 M NaCl, 1 mM DTT, pH 8.0. centrifuged at 1000 g for 10 min at 4 C and the supernatant was purified by discontinuous Percoll gradient centrifuga- 2.3. Cleavage of 6xHis-SUMO-PnTx3-4 fusion protein and tion as described by (Dunkley et al., 2008) with minor fi purification of the recombinant PnTx3-4 protein modi cations. The synaptosomal fraction was resuspended in Krebs-Ringer-Hepes (KRH) buffer (124 mM NaCl, Samples of 6xHis-SUMO-PnTx3-4 were incubated with 4 mM KCl, 1.2 mM MgSO4, 25 mM HEPES and 10 mM fi SUMO protease I (1 U:100 mg) for 1 h at 30 C and then Glucose and adjusted to pH 7.4) to a nal concentration of – incubated overnight at 4 C to allow complete cleavage. 0.5 1.0 mg/mL for each sample. DTT was added to the reaction to a final concentration of 2þ 2 mM. The recombinant PnTx3-4 was then purified from 2.7. Measurement of [Ca ]i the 6xHis-SUMO-tag and protease by C8 Reverse Phase- HPLC using a CH3CN discontinuous gradient in 0.1% TFA. Synaptosomes were loaded with 5 mM fura-2AM (stock The absorbance was detected at 214 nm. For the PnTx3-4 solution 1 mM in DMSO) for measurements of intra- 2þ isolated from the supernatant, peaks corresponding to the synaptosomal free calcium concentration [Ca ]i. The recombinant toxin were pooled, freeze-dried and stored quantification was performed in a PTI spectrofluorimeter at 20 C until needed. For the PnTx3-4 isolated from the with the fluorescence emission recorded at 510 nm using pellet, peaks corresponding to the recombinant toxin 340/380 excitation ratio. Synaptosomes were stirred were pooled and treated for refolding. throughout the experiment and maintained at 35 C. Native and recombinant toxins were added to the synaptosomal 2.4. Refolding suspension 6 min prior to membrane depolarization with 33 mM KCl. Calibration was performed as described by The pure PnTx3-4 lyophilized was resuspended in 6 M (Prado et al., 1996) using SDS and EGTA for maximum and Gnd-HCl, 50 mM Tris, pH 8.0 and the disulfide bonds were minimum fluorescence values. 910 I.A. Souza et al. / Toxicon 60 (2012) 907–918

2.8. Measurement of continuous glutamate release Interestingly, the sequence similarity is observed essen- tially in the amino-terminal end of the proteins (first 51 Glutamate release was monitored by measuring the amino acid residues) while the carboxy-terminal end does increase of fluorescence caused by NADPH being produced not show either similarity in amino acid sequence or þ in the presence of NADP and glutamate dehydrogenase. At conserved localization of cysteine residues. the beginning of each fluorimetric assay, 1 mM of CaCl2, þ 1 mM of NADP , and 50 U of glutamate dehydrogenase were 4.2. Cloning and expression of the fusion protein added to the suspension. The excitation wavelength was set at 360 nm and the emission wavelength was monitored at We used the amino acid sequence of PnTx3-4 (Fig. 1), 450 nm. Native and recombinant toxins were incubated also named u-Phonetoxin-IIA (Dos Santos et al., 2002; with the synaptosomes for 30 min prior to each assay. Cassola et al., 1998), to design a synthetic cDNA. The Calcium independent glutamate release was measured by nucleotide sequence was chosen following the E. coli codon removing CaCl2 and adding 2 mM EGTA to the preparation. usage (Sharp and Li, 1987) to improve expression of the transcript in prokaryotic cells. The designed PnTx3-4 cDNA 2.9. Statistical analysis (Fig. 2A) was generated by PCR using six overlapping oligonucleotides (Table 1; Fig. 2B) and cloned into the pE- The results were expressed as mean SEM. The data SUMO vector (LifeSensors Inc.). The fusion protein gener- were analyzed by one-way analysis of variance (ANOVA) ated after expression of the pE-SUMO-PnTx3-4 plasmid followed by Tukey test (SigmaSTAT) and Kruskal-Wallis included (i) a (His)6 tag followed by (ii) the yeast SUMO ANOVA followed by Dun’s multiple comparison test. protein (Smt3) and, (iii) the PnTx3-4 peptide (Fig. 2C). Fusion protein expression was induced in the E. coli 3. Circular dichroism (CD) strain ORIGAMI (DE3) by addition of 0.6 mM IPTG. As shown in Fig. 3A, the 6xHis-Smt3-PnTx3-4 fusion protein To get information about the secondary structure of the was highly expressed. Although a large amount of the toxin PnTx3-4, the CD spectrum of the functional refolded recombinant protein was present in the soluble fraction toxin was collected using a spectropolarimeter Jasco-810 (Fig. 3A, lane 3), considerable amount of the protein was (Jasco Corp.) in water. The temperature was kept at 25 C also present in inclusion bodies (Fig. 3A, lane 2). Because and the spectrum was measured from 260 nm to 190 nm soluble proteins have the potential to be in their native using a 1 mm path length cell. A minimum of 10 scans were conformation, while proteins found in inclusion bodies done at a time. need to be denatured and refolded to assume their native To get an estimation of secondary structures present in structure, we chose to purify the recombinant PnTx3-4 the toxin, the data obtained were analyzed using three from supernatant and pellet separately. different algorithms; CDSSTR, CONTIN and SELCON and two reference sets for each (Sreerama and Woody, 2000; 4.3. Purification of PnTx3-4 from the supernatant Sreerama et al., 1999; Van Stokkum et al., 1990). The soluble recombinant 6xHis-SUMO-PnTx3-4 was 4. Results purified from the supernatant by affinity chromatography using a Ni-NTA agarose resin. Purified fraction was dialyzed 4.1. PnTx3-4 amino-acid sequence analysis for 24 h to remove the imidazole. This purification step isolated the 6xHis-SUMO-PnTx3-4 from most of the bacte- Fig. 1 shows the amino-acid sequence of the P. nig- rial proteins (Fig. 3B, lane 5). To specifically separate the riventer PnTx3-4, toxin and its alignment to two related PnTx3-4 from its N-terminal 6xHis-SUMO tag we digested peptides from the spider Agelenopsis aperta that, as PnTx3- the peptide with SUMO Protease I, which recognizes the 4, block N-, P/Q-, and R-type calcium channels. These three SUMO structure at the N-terminus of the fusion protein and peptides share the same number of amino acid residues cleaves the junction (peptide bond linking the last amino (76-residues) and are highly conserved in their primary acid residue of SUMO to the first amino acid residue of the sequence, showing w70% similarity and w50% identity. recombinant peptide). After digestion, PnTx3-4 was then

Fig. 1. Alignment of PnTx3-4 to u-AgaIIIA and u-AgaIIIB, spider peptides that block calcium ion channels. Percentage of identity is shown based on PnTx3-4. The sequence alignment was obtained from http://www.ch.embnet.org/software/ClustalW.html. Gaps () have been introduced to enhance similarities. Cysteine residues are colored. Predicted arrangement of disulfide bonds for PnTx3-4 and u-AgaIIIA according to database of structural models (http://knottin.cbs.cnrs.fr; see Fig. 7) is shown in colored brackets. Disulfide bridges in the cystine knot motif are shown in green and red. Additional disulfide bonds are shown in blue and orange. PnTx3-4 predicted arrangement of disulfide bonds correspond to: (C2–C19); (C9–C25); (C16–C51); (C18–C39); (C27–C37); (C57–C63); (C67–C72). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) I.A. Souza et al. / Toxicon 60 (2012) 907–918 911

Fig. 2. A) Sequence of the synthetic PnTx3-4 cDNA and the codified product. The restriction sites Bsa I and Xba I (underlined) were added to facilitate cloning. B) Schematic representation of each oligonucleotide used for the PCR reaction that generated the PnTx3-4 cDNA. C) Representation of the recombinant fusion protein, the digested fragments and purified PnTx3-4. purified by reverse phase chromatography in a Waters HPLC measured in mouse cortical synaptosomes depolarized with system using a discontinuous CH3CN gradient. Two main 33 mM KCl in the presence of 1 mM CaCl2 (Fig. 4A and B). peaks were observed, one with retention time of about Addition of 16 nM of native PhTx3-4 (purified from the spider þ 31 min and the other with retention time of about 41 min venom) decreased glutamate release by 36%. Ca2 removal (Fig. 3C). Fractions from each peak were pooled and from the medium, by adding 2.0 mM EGTA before depolar- analyzed by SDS-PAGE and Western blot. We determined ization with KCl, decreased glutamate release by the same that the recombinant PnTx3-4 eluted in peak 1, which pre- magnitude (Fig. 4A and B), corroborating previous findings þ sented a band of approximately 8 kDa (Fig. 3D, lane 1) that that native PnTx3-4 effect is mainly restricted to the Ca2 - could be recognized by a polyclonal antibody raised against dependent (exocytotic) glutamate release (de Castro Junior the spider venom (Fig. 3E, lane 2). Approximately 0.6 mg of et al., 2008). Addition of 16 nM of recombinant PnTx3-4 pure recombinant PnTx3-4 was reproducibly obtained per peptide was able to block glutamate release from cortical litter of bacterial culture (Table 2). synaptosomes as efficiently as the native toxin (Fig. 4A and B).

2þ 4.4. Effect of soluble recombinant PnTx3-4 on glutamate 4.5. Effect of soluble recombinant PnTx3-4 on [Ca ]i release To further test the biological activity of the recombinant þ To investigate whether the recombinant peptide pre- PnTx3-4 we investigated its effect on blocking Ca2 chan- sented biological activity analogous to native PnTx3-4, we nels involved in glutamate release from cortical synapto- þ tested it initially in a glutamate release assay (de Castro Junior somes. To do that, we measured changes in cytosolic Ca2 et al., 2008; Prado et al., 1996). Total glutamate release was in fura-2-loaded synaptosomes (Prado et al., 1996). 912 I.A. Souza et al. / Toxicon 60 (2012) 907–918

Fig. 3. Expression and purification of the soluble recombinant PnTx3-4. A) Analysis by Coomassie-stained SDS-PAGE 4–20%. MW: molecular weight standard (Dual Color Precision Plus). Lane 1: bacterial extract before induction. Lane 2: insoluble proteins extracted from inclusion bodies. Lane 3: soluble proteins from the supernatant B) Affinity purification of the soluble recombinant PnTx3-4. Lane 1: cell extract before induction. Lane 2: cell extract after induction. Lane 3: soluble proteins recovered after cell breakage. Lane 4: proteins that did not bind to the resin. Lane 5: purified fusion protein. SDS-PAGE 4–20% gel was stained with the RAPIDStain reagent (Calbiochem). C) Purification of the digested recombinant PnTx3-4 by RP-HPLC using acetonitrile gradient to separate the toxin from its N-terminal 6xHis-SUMO tag. D) SDS-PAGE 4–20% showing the peaks from the chromatography. Gel was stained using the Silver Stain Plus kit from Bio-Rad. E) Immunoblot analysis of the recombinant fusion protein. Lane 1: purified fusion protein. Lane 2: purified fusion protein digested with SUMO protease I. Lane 3: Phoneutria nigriventer’s total venom. Antibody: anti-P. nigriventer total venom peroxidase. Arrows indicate bands corresponding to 6xHis-SUMO-PnTx3-4 (26 kDa), 6xHis-SUMO tag (18 kDa) and toxin PnTx3-4 (8 kDa).

Synaptosomes depolarized with 33 mM KCl in the presence lane 2), we chose to improve our purification yield by of 1 mM CaCl2 showed a fast increase in internal calcium purifying it from the pellet. To do that, recombinant 6xHis- concentration (Fig. 4C). Addition of 16 nM of native PnTx3- SUMO-PnTx3-4 present in the pellet was first solubilised in þ 4 6 min before KCl depolarization inhibited internal Ca2 6 M of Guanidine-HCl (Fig. 5A) and then purified by affinity increase by approximately 30%. Addition of similar chromatography using a Ni-NTA agarose resin. After concentration of the recombinant PnTx3-4 peptide to the removal of the imidazole by dialysis, the N-terminal tag þ preparation also blocked Ca2 channels, however, the was cleaved off by digestion with SUMO protease I (Fig. 5B, þ inhibition of internal Ca2 increase observed was smaller lane 2). The recombinant toxin was purified by RP-HPLC (approximately 20% inhibition). and two peaks with retention times of about 32 and 41 min respectively were observed (Fig. 5D and E). The 4.6. Purification of PnTx3-4 from inclusion bodies peak with 32 min retention time presented one band of 8 kDa that could be recognized by a polyclonal antibody Because the 6xHis-SUMO-PnTx3-4 fusion protein raised against the spider venom (Fig. 5C, lane 1 and 2). This showed to be highly expressed as inclusion bodies (Fig. 3, peptide presented no biological activity when tested in the I.A. Souza et al. / Toxicon 60 (2012) 907–918 913

Table 2 0.1–0.2 mg/mL. Nine different refolding buffers were tested Purification of recombinant PnTx3-4 from E. coli. (Table 3), ranging from strong to weak denaturing condi- Fraction Total protein tions. Refolding was allowed to proceed for 24 h at 4 C, (per L of culture) samples were submitted to RP-HPLC and tested. We Supernatant (soluble proteins)a 350 mg estimated refolding yields by measuring biological activity Soluble fusion protein (IMAC purification)a 5.3 mg using the glutamate release assay as described for experi- fi c – Digested and puri ed (HPLC) 0.5 0.8 mg ments in Fig. 4; that is, 16 nM of each refolded peptide was Mass of Gnd-HCI extractb 580 mg Insoluble fusion protein (IMAC purification)a 19.5 mg added to mouse cortical synaptosomes prior to depolar- c Digested and purified (HPLC) 3.0 mg ization with 33 mM KCl in the presence of 1 mM CaCl2 and Refoldedc 1.5–2.0 mg total glutamate release was measured (Fig. 5F). As our a Determined using D Protein assay from Bio-Rad. experiments consistently showed that 16 nM of native c þ b Calculated as weigh difference between pellet of French Press and PnTx3-4 or Ca2 removal from the medium (by adding pellet of 6 M Guanidine-HCI extraction. 2.0 mM EGTA before depolarization with KCl) decreased c Determined by A280 nm. glutamate release by the same magnitude (Fig. 4), for these experiments we compared inhibition of glutamate release glutamate release assay (Fig. 4D and E) indicating that the by each refolded peptide to that of EGTA containing buffer. peptide was not properly folded. A refolded sample that presented decrease of glutamate þ release similar to that of Ca2 free medium would 4.7. Refolding of PnTx3-4 isolated from inclusion bodies be considered to have 100% of the peptides properly refolded. As can be seen in Table 3 and Fig. 5F, refolding of Our next step was to determine the optimized condition PnTx3-4 was suppressed at the lowest and highest dena- necessary to obtain reliably refolded, biologically active turant concentrations (buffers 1–4, 8 and 9). Highest PnTx3-4. To do that, we incubated the recombinant PnTx3-4 PnTx3-4 activity was observed in trial five, which contained in a strong denaturing buffer (6 M Gnd-HCl, 50 mM Tris, 0.5 M Gnd-HCl, 0.4 M L-arginine, 1 mM GSH and 1 mM 10 mM DTT, pH 8.0) to completely unfold the protein. After GSSG. Under these conditions, more than 80% of the solu- 4 h of incubation at RT, DTT was removed by filtration bilised PnTx3-4 was refolded. Approximately 1.5–2mg (VIVASPIN 6 column; 3 kDa MWCO). The toxin was then of refolded PnTx3-4 peptide was obtained by using trial five diluted into a refolding buffer to a final concentration of conditions (Table 2).

þ Fig. 4. Biological effects of the recombinant PnTx3-4. A and B) Effect of the soluble recombinant PnTx3-4 compared to the native toxin in K -induced glutamate þ release in synaptosomes. Synaptosomes were incubated with the recombinant and native toxins for 30 min and then depolarized with 33 mM of KCl. Ca2 - þ independent glutamate release was measured in the absence of Ca2 and in the presence of 2 mM EGTA. The results are shown by mean SEM of glutamate release (nmol/mg of protein). C) Inhibition of calcium influx in depolarized synaptosomes. Synaptosomes loaded with fura-2AM were depolarized with 33 mM of 2þ KCl after 6 min of incubation with the recombinant and native toxins. The [Ca ]i was calculated by fluorimetric assay and the results are expressed as þ mean SEM of intrasynaptosomal calcium concentration (nM). D and E) The effect of the recombinant PnTx3-4 purified from inclusion bodies in K -induced glutamate release in synaptosomes. ANOVA followed by Tukey test were used to determine statistical significance. *P < 0.001, #P < 0.004. 914 I.A. Souza et al. / Toxicon 60 (2012) 907–918

Fig. 5. Purification of the recombinant toxin PnTx3-4 obtained from inclusion bodies. Arrows indicate bands corresponding to 6xHis-SUMO-PnTx3-4 (26 kDa), 6xHis-SUMO tag (18 kDa) and toxin PnTx3-4 (8 kDa). A) SDS-PAGE 4–20% of solubilized inclusion bodies. MW: Molecular weight standard (Dual Color Precision Plus). B) SDS-PAGE 4–20% of 6xHis-SUMO-PnTx3-4 digested with SUMO protease I. Lane 1: 6xHis-SUMO-PnTx3-4 purified by affinity chromatography. Lane 2: 6xHis-SUMO-PnTx3-4 digested. C) Immunoblot analysis of the recombinant fusion protein before and after digestion. Lane 1: purified fusion protein. Lane 2: purified fusion protein digested with SUMO protease I. Antibody: anti-P. nigriventer total venom peroxidase. D) Purification of the recombinant PnTx3-4. After digestion with SUMO protease I, the recombinant toxin was purified by RP-HPLC using acetonitrile gradient to separate it from the N-terminal 6xHis-SUMO tag. E) SDS-PAGE 4–20% showing the peaks from the chromatography presented in D. Gel was stained with the RAPIDStain reagent (Calbiochem). F) Effects of refolded þ PnTx3-4 in K -induced glutamate release in synaptosomes. Kruskal–Wallis ANOVA followed by Dun’s multiple comparison tests were used to determine statistical significance. *P < 0.05.

4.8. Circular dichroism spectrum and estimated secondary structures for PnTx3-4

To gather information about the secondary structure of the toxin, we obtained the circular dichroism spectrum of the functional, refolded, recombinant PnTx3-4 (Fig. 6). Analysis of the spectrum using the CDSSTR, CONTIN and SELCON algorithms (Van Stokkum et al., 1990; Sreerama and Woody, 2000; Sreerama et al., 1999) predicted that the toxin structure is composed of approximately 53% turns/unordered, 31% a-helix and 16% b-strand.

5. Discussion

In this report we provide a method for expression and puriﬁcation of recombinant PnTx3-4 with native bioactivity. Identifying ideal conditions for heterologous expression of Fig. 6. Circular dichroism spectrum of the refolded recombinant PnTx3-4. functional PnTx3-4 was rather challenging, even more The concentration of the peptide was 0.5 mg/mL. Data is the average of 3 challenging than ﬁnding the conditions to express other P. separate recordings with a minimum of 10 scans each. I.A. Souza et al. / Toxicon 60 (2012) 907–918 915

Fig. 7. Predicted spatial structure models for the spider toxins PnTx3-4 and Omega-Agatoxin-IIIA. Ribbon models were drawn using the PyMOL program according to structural model calculations obtained from the knottin database (http://knottin.cbs.cnrs.fr). nigriventer toxins (Souza et al., 2008; Carneiro et al., 2003; Only part of the recombinant PnTx3-4 was expressed as Kushmerick et al., 1999; Torres et al., 2010; Diniz et al., a soluble protein. The yield of soluble PnTx3-4 after all the 2006). This difficulty was probably due to the fact that purification steps ranged from 0.5 to 0.8 mg/L, which is in PnTx3-4 requires the formation of a larger number of the same range to what has been reported for other animal disulfide bonds than the other peptides present in the P. toxins successfully expressed in E. coli (Johnson et al., 2000; nigriventer’s venom (Penaforte et al., 2000; Gomez et al., Meng et al., 2011; Che et al., 2009; Souza et al., 2008; 2002). That is, seven disulfide bonds are necessary to Carneiro et al., 2003). More importantly, the soluble properly fold PnTx3-4 into its native conformation (Figs. 1 recombinant protein showed biological activity very and 7). Initial attempts using expression systems that similar to the native PnTx3-4, both in the glutamate release generate His-tag-fusion proteins under the control of the assay as well as in the measurement of intrasynaptosomal strong T7 promoter (Studier et al., 1990), or the tightly free calcium concentration. These results indicate that, regulated araBAD promoter (pBAD) (Guzman et al., 1995) similar to the native peptide, soluble recombinant PnTx3-4 þ were not successful. These trials either did not generate is able to block Ca2 channels involved in glutamate release fusion proteins in soluble form or the induction of the from cortical synaptosomes. protein expression was very low (data not shown). Only the Because most of the recombinant PnTx3-4 aggregated SUMO system was suitable to express large amounts of the as inclusion bodies we also searched for conditions to protein, which was found in both soluble and insoluble provide efficient refolding of the insoluble recombinant form. The SUMO system uses the SUMO protein (Small PnTx3-4. Finding the exact conditions to renature proteins Ubiquitin-like Modifier) as a fusion partner, improving the is usually time-consuming as refolding conditions for solubility of the expressed protein (Marblestone et al., 2006; individual proteins vary considerably (Singh and Panda, Malakhov et al., 2004; Butt et al., 2005). In addition, we co- 2005; Lilie et al., 1998). The basic protocol requires that expressed the chaperones GroEL and GroES to improve the purified inclusion bodies are first solubilised with a strong protein folding process (Thomas et al.,1997), and carried out denaturant, such as guanidine hydrochloride (GdnHCl), to the expression in the ORIGAMI (DE3) strain, which has an produce a completely unfolded protein. DTT is also added oxidizing cytoplasm that provides a better environment for to allow reduction of disulfide bridges (Fahnert et al., 2004). the proper folding of the toxin (Stewart et al., 1998). It has The solubilised protein is then diluted or dialyzed into been shown that expression of disulfide rich peptides in a refolding buffer to reduce the denaturant concentration, ORIGAMI (DE3) strain substantially improve the yield of allowing the protein to refold based on the information active proteins purified (Prinz et al., 1997). contained in its primary sequence. As the denaturant is removed, protein aggregation tends to compete with

Table 3 renaturation therefore, it is crucial to identify the ideal Refolding conditions. milieu to recover maximal amounts of native protein. fl a Several factors in uence renaturation/aggregation during Buffer Gnd-HCI L-arginine Redox environment Percent (M) (M) refolded refolding including protein concentration, concentration of strong and weak denaturants, pH, temperature, and the 1 0 0 2 mM GSH 0.2 mM GSSG NA 2 0 0.4 2 mM GSH 0.4 mM GSSG NA redox environment (Fahnert, 2004; Lilie et al., 1998). Out of 3 0 0.8 1 mM GSH 1 mM GSSG NA 9 different buffer conditions (Table 3) that we tried, only 4 0.55 0 2 mM GSH 0.4 mM GSSG NA buffer 5, which contained 0.5 M Gnd-HCl, 0.4 M L-arginine, 5 0.55 0.4 1 mM GSH 1 mM GSSG 82.5% 1 mM GSH and 1 mM GSSG, allowed proper refolding of 6 0.55 0.8 2 mM GSH 0.2 mM GSSG 42.5% – 7 1.1 0 1 mM GSH 1 mM GSSG 62.5% PnTx3-4. Using buffer 5 we managed to obtain 1.5 2.0 mg/L 8 1.1 0.4 2 mM GSH 0.2 mM GSSG 12.5% of PnTx3-4 refolded after purification from inclusion 9 1.1 0.8 2 mM GSH 0.4 mM GSSG 17.5% bodies. Importantly, the refolded peptide also showed a Each buffer contains 55 mM Tris, 21 mM NaCI, 0.88 mM KCI, 1.0 mM biological activity very similar to the native peptide. These EDTA, pH 8.2. results indicate that a balanced molar ratio of reduced to 916 I.A. Souza et al. / Toxicon 60 (2012) 907–918 oxidized thiol reagents (glutathione) was essential to sequence identity of 70%. These data suggest that the provide the appropriate redox potential to allow formation disulfide bonding patterns of the two molecules are likely and reshuffling of disulfide bonds (Misawa and Kumagai, to be very similar; however, there has been no NMR study 1999; Wetlaufer et al., 1987). Proper renaturation of on either PnTx3-4 or u-Aga-IIIA to define their three- PnTx3-4 also relied on the presence of non-denaturing dimensional structure. concentrations of the chaotroph GdnHCl. Low concentra- Recently Kozlov and Grishin (2005), based on the fact tions of GdnHCl or urea have been suggested to contribute that the majority of spider toxins share similarity in to refolding of proteins by slowing down the refolding cysteine arrangement and disulfide bridge pattern, devel- kinetics and as a consequence, shifting the competition oped a new algorithm that reliably predicts the three- between renaturation and aggregation toward the rena- dimensional structure of the cysteine knot motif based on turation reaction (Fahnert et al., 2004; Lilie et al., 1998). primary sequence analysis. Interestingly, these authors Additionally, the presence of L-arginine also contributed to showed that PnTx3-4 and u-Aga-IIIA primary structure efficient renaturation of PnTx3-4. Although the mechanism conform to all the criteria of a knottin scaffold (Kozlov and by which L-arginine facilitates renaturation is still not Grishin, 2005). An automated modeling procedure is now completely understood, it has been hypothesized that available for predicting the three-dimensional structure of increased solubilization of folding intermediates might be knottins (Gracy and Chiche, 2010, 2011) and a database of involved (Lilie et al., 1998). It is important to note that, structural models for all known knottin sequences is freely although biological assays indirectly suggest that accessible from the web site http://knottin.cbs.cnrs.fr. Fig. 7 recombinant PnTx3-4 and the native PnTx3-4 share similar shows the comparison between the knottin database pre- properties, we cannot rule out the possibility that minor dicted three-dimensional structures of PnTx3-4 and u-Aga- structural differences might exist. Future studies including IIIA toxins. The two peptides show remarkable structural investigating whether recombinant peptides co-migrate similarity (Fig. 7C), not only at the N-terminal end, where with the native toxin on HPLC and comparative mass they show high sequence similarity, but also at the C- spectroscopy analysis will be necessary to clarify this issue. terminus, where the peptides do not show amino acid Analysis of the peptide masses of different spider sequence similarity or conserved localization of cysteine venoms revealed a bimodal molecular weight distribution, residues (Fig. 1). Based on the fact that the different steps of with 60–70% of the peptides showing 30–50 amino-acids, the homology modeling were carefully optimized using and a secondary grouping (less than 10%) showing a large set of knottins with known structures and the peptides 60–80 amino acids long (Escoubas, 2006). Struc- accuracy of predicted models was shown to lie between tural data, although limited, come mainly from the more 1.50 and 1.96 Å (Gracy and Chiche, 2010), it is tempting to abundant short peptides. These studies indicate that short propose that the predicted model for PnTx3-4 is a close spider peptides show mainly two different structural representation of the actual structure of the toxin. In fact, motifs characterized by different cysteine arrangements our CD spectrum analysis of the refolded toxin indicated and structural features. The most common motif is the that PnTx3-4 contains predominantly random coil forma- “inhibitor cystine knot” (ICK), also named knottin, with tion, which corroborates the predicted model proposed. a consensus sequence of C1X3–7–C2X3–8–C3X0–7–C4X1–4– The functional expression of recombinant PnTx3-4 and the C5X4–13–C6, where C represents cysteine residues and X is structural analysis reported here provide the basis for any amino acid residue. Disulfide bond pairing observed in future large scale production and structure-function all molecules of this type follow the arrangement: C1–C4, investigation of this peptide. C2–C5, C3–C6. Spatial structure of peptides with ICK motif is characterized by the presence of a b-hairpin and a pecu- Acknowledgments liar “knot” (origin of its name) (Escoubas, 2006; Vassilevski et al., 2009). The other less prominent structural scaffold This work was supported by the “Milenium Institute for for short spider toxins is the DDH (disulfide-directed beta- development of drugs based on toxins” (Milenio-2005; hairpin) motif, with a consensus sequence C1 X5–19 C2 X2 Brazil; V.F.P., M.A.M. P and M.V.G.), Capes Toxinology (G/P) X2 C3 X6–19 C4, and arrangement of disulfide bonds Program 1444/2011 and PRONEX-2005 (ABORDAGEM C1–C3, C2–C5 (Vassilevski et al., 2009; Escoubas, 2006). It GENETICO-MOLECULAR PARA O ESTUDO DO SISTEMA has been proposed that the DDH motif came earlier in COLINERGICO; Brazil; V.F.P; M.A.M. P; M.V.G.). Canadian evolution and the ICK scaffold should be considered to be Foundation for Innovation (CFI, V.F.P & M.A.M.P), the a molecular evolution of the DDH motif (Shu et al., 2002; Ontario Research Fund (ORF, V.F.P & M.A.M.P) and the Wen et al., 2005). University of Western Ontario (V.F.P. & M.A.M.P.). I. A. Souza Very few of the longer polypeptides present in spider received a PhD fellowship from CAPES (Brazil) and an venoms have been isolated and sequenced to date. In award from the Foreign Affairs and International Trade addition, the three-dimensional structure of the few long Canada (DFAIT) – Grant Agreement for Emerging Leaders in spider peptides that have been described in the literature the Americas (ELAP). remains undetermined (Vassilevski et al., 2009). PnTx3-4 and the closely related peptide u-Aga-IIIA belong to this Ethical statement class of peptides (Fig. 1)(Goncaves et al., 2011; de Castro Junior et al., 2008; Reis et al., 2000; Yan and Adams, All experiments were carried out in compliance with the 2000). They both are 76 amino acids long, show similar Canadian Council of Animal Care (CCAC) guidelines for the placement of the cysteine residues, and have overall care and use of animals. The protocol was approved by the I.A. Souza et al. / Toxicon 60 (2012) 907–918 917

University of Western Ontario Institutional Animal Care antagonist MTEP: effects on kainate-induced excitotoxicity in the rat – and Use Committee (protocol # 2008-127). All efforts were hippocampus. Pharmacol. Rep. 62, 1051 1061. Dos Santos, R.G., Van, R.C., Martin-Moutot, N., Mansuelle, P., Cordeiro, M. made to minimize the suffering of animals. N., Diniz, C.R., Mori, Y., De Lima, M.E., Seagar, M., 2002. Phoneutria nigriventer omega-phonetoxin IIA blocks the Cav2 family of calcium Conflict of interest statement channels and interacts with omega-conotoxin-binding sites. J. Biol. Chem. 277, 13856–13862. Dunkley, P.R., Jarvie, P.E., Robinson, P.J., 2008. A rapid Percoll gradient The authors declare that there are no conflicts of procedure for preparation of synaptosomes. Nat. Protoc. 3, 1718–1728. fi interest. Escoubas, P., 2006. Molecular diversi cation in spider venoms: a web of combinatorial peptide libraries. Mol. Divers. 10, 545–554. Fahnert, B., 2004. Folding-promoting agents in recombinant protein production. Methods Mol. Biol. 267, 53–74. References Fahnert, B., Lilie, H., Neubauer, P., 2004. Inclusion bodies: formation and utilisation. Adv. Biochem. Eng. Biotechnol. 89, 93–142. Agostini, R.M., do Nascimento Pinheiro, A.C., Binda, N.S., Romano Silva, M. Favreau, P.,Le, G.F., Benoit, E., Molgo, J.,1999. A review on conotoxins targeting A., Cordeiro, M.N., Richardson, M., Sena Guimaraes, A.L., Gomez, M.V., ion channels and acetylcholine receptors of the vertebrate neuromus- 2011. Phoneutria spider toxins block ischemia-induced glutamate cular junction. Acta Physiol. Pharmacol. Ther. Latinoam. 49, 257–267. release and neuronal death of cell layers of the retina. Retina 31, Gomez, M.V., Kalapothakis, E., Guatimosim, C., Prado, M.A., 2002. Pho- 1392–1399. neutria nigriventer venom: a cocktail of toxins that affect ion chan- Alonso, D., Khalil, Z., Satkunanthan, N., Livett, B.G., 2003. Drugs from the nels. Cell Mol. Neurobiol. 22, 579–588. sea: conotoxins as drug leads for neuropathic pain and other Goncaves, J.M., Ferreira, J., Prado, M.A., Cordeiro, M.N., Richardson, M., neurological conditions. Mini. Rev. Med. Chem. 3, 785–787. Pinheiro, A.C., Silva, M.A., Junior, C.J., Souza, A.H., Gomez, M.V., 2011. Arundine, M., Tymianski, M., 2003. Molecular mechanisms of calcium- The effect of spider toxin PhTx3-4, omega-conotoxins MVIIA and dependent neurodegeneration in excitotoxicity. Cell Calcium 34, MVIIC on glutamate uptake and on capsaicin-induced glutamate 325–337. release and [Ca2þ]i in spinal cord synaptosomes. Cell Mol. Neurobiol. Butt, T.R., Edavettal, S.C., Hall, J.P., Mattern, M.R., 2005. SUMO fusion 31, 277–283. technology for difficult-to-express proteins. Protein Expr. Purif. 43, 1– Gracy, J., Chiche, L., 2010. Optimizing structural modeling for a specificprotein 9. scaffold: knottins or inhibitor cystine knots. BMC. Bioinform. 11, 535. Cardoso, F.C., Pacifico, L.G., Carvalho, D.C., Victoria, J.M., Neves, A.L., Cha- Gracy, J., Chiche, L., 2011. Structure and modeling of knottins, a promising vez-Olortegui, C., Gomez, M.V., Kalapothakis, E., 2003. Molecular molecular scaffold for drug discovery. Curr. Pharm. Des. 17, 4337–4350. cloning and characterization of Phoneutria nigriventer toxins active Guatimosim, C., Romano-Silva, M.A., Cruz, J.S., Beirao, P.S., Kalapothakis, E., on calcium channels. Toxicon 41, 755–763. Moraes-Santos, T., Cordeiro, M.N., Diniz, C.R., Gomez, M.V., Prado, M.A., Carneiro, A.M., Kushmerick, C., Koenen, J., Arndt, M.H., Cordeiro, M.N., 1997. A toxin from the spider Phoneutria nigriventer that blocks calcium Chavez-Olortegui, C., Diniz, C.R., Gomez, M.V., Kalapothakis, E., channels coupled to exocytosis. Br. J. Pharmacol. 122, 591–597. Prado, M.A., Prado, V.F., 2003. Expression of a functional recombinant Guzman, L.M., Belin, D., Carson, M.J., Beckwith, J., 1995. Tight regulation, þ Phoneutria nigriventer toxin active on K channels. Toxicon 41, 305– modulation, and high-level expression by vectors containing the 313. arabinose PBAD promoter. J. Bacteriol. 177, 4121–4130. Carneiro, D.S., Vieira, L.B., Cordeiro, M.N., Richardson, M., Castro-Junior, C. Heading, C.E., 2002. Conus peptides and neuroprotection. Curr. Opin. J., Gomez, M.V., Reis, H.J., 2010. Effects of new Phoneutria spider Investig. Drugs 3, 915–920. þ toxins on glutamate release and [Ca2 ]i in rat cortical synaptosomes. Johnson, T.M., Quick, M.W., Sakai, T.T., Krishna, N.R., 2000. Expression of Cell Mol. Biol. (Noisy. -le-grand) 56 (Suppl.), OL1223–OL1230. functional recombinant scorpion beta-neurotoxin Css II in E. coli. Cassola, A.C., Jaffe, H., Fales, H.M., Afeche, S.C., Magnoli, F., Cipolla-Neto, J., Peptides 21, 767–772. 1998. Omega-Phonetoxin-IIA: a calcium channel blocker from the Jones, R.M., Bulaj, G., 2000. Conotoxins – new vistas for peptide thera- spider Phoneutria nigriventer. Pflugers Arch. 436, 545–552. peutics. Curr. Pharm. Des. 6, 1249–1285. Castellino, F.J., Prorok, M., 2000. Conantokins: inhibitors of ion flow Kalapothakis, E., Penaforte, C.L., Beirao, P.S., Romano-Silva, M.A., Cruz, J.S., through the N-methyl-D-aspartate receptor channels. Curr. Drug Prado, M.A., Guimaraes, P.E., Gomez, M.V., Prado, V.F., 1998a. Cloning Targets 1, 219–235. of cDNAS encoding neurotoxic peptides from the spider Phoneutria Che, N., Wang, L., Gao, Y., An, C., 2009. Soluble expression and one-step nigriventer. Toxicon 36, 1843–1850. purification of a neurotoxin Huwentoxin-I in Escherichia coli. Kalapothakis, E., Penaforte, C.L., Leao, R.M., Cruz, J.S., Prado, V.F., Protein Expr. Purif. 65, 154–159. Cordeiro, M.N., Diniz, C.R., Romano-Silva, M.A., Prado, M.A., Cordeiro, M.N., de Figueiredo, S.G., Valentim, A.C., Diniz, C.R., von, E.V., Gomez, M.V., Beirao, P.S., 1998b. Cloning, cDNA sequence analysis and Gilroy, J., Richardson, M., 1993. Purification and amino acid sequences patch clamp studies of a toxin from the venom of the armed spider of six Tx3 type neurotoxins from the venom of the Brazilian ‘armed’ (Phoneutria nigriventer). Toxicon 36, 1971–1980. spider Phoneutria nigriventer (Keys). Toxicon 31, 35–42. Kozlov, S., Grishin, E., 2005. Classification of spider neurotoxins using Dalmolin, G.D., Silva, C.R., Rigo, F.K., Gomes, G.M., Cordeiro, M.N., structural motifs by primary structure features. Single residue Richardson, M., Silva, M.A., Prado, M.A., Gomez, M.V., Ferreira, J., 2011. distribution analysis and pattern analysis techniques. Toxicon 46, Antinociceptive effect of Brazilian armed spider venom toxin Tx3-3 in 672–686. animal models of neuropathic pain. Pain 152, 2224–2232. Kushmerick, C., Kalapothakis, E., Beirao, P.S., Penaforte, C.L., Prado, V.F., de Castro Junior, C.J., Pinheiro, A.C., Guatimosim, C., Cordeiro, M.N., Cruz, J.S., Diniz, C.R., Cordeiro, M.N., Gomez, M.V., Romano-Silva, M.A., Souza, A.H., Richardson, M., Romano-Silva, M.A., Prado, M.A., Prado, M.A., 1999. Phoneutria nigriventer toxin Tx3-1 blocks A-type þ þ Gomez, M.V., 2008. Tx3-4 a toxin from the venom of spider Pho- K currents controlling Ca2 oscillation frequency in GH3 cells. J. neutria nigriventer blocks calcium channels associated with exocy- Neurochem. 72, 1472–1481. tosis. Neurosci. Lett. 439, 170–172. Lau, A., Tymianski, M., 2010. Glutamate receptors, neurotoxicity and de Souza, A.H., Lima, M.C., Drewes, C.C., da Silva, J.F., Torres, K.C., neurodegeneration. Pflugers Arch. 460, 525–542. Pereira, E.M., de Castro, C.J.J., Vieira, L.B., Cordeiro, M.N., Leao, R.M., Cruz, J.S., Diniz, C.R., Cordeiro, M.N., Beirao, P.S., 2000. Inhi- Richardson, M., Gomez, R.S., Romano-Silva, M.A., Ferreira, J., bition of neuronal high-voltage activated calcium channels by the Gomez, M.V., 2011. Antiallodynic effect and side effects of Phal- omega-phoneutria nigriventer Tx3-3 peptide toxin. Neuropharma- pha1beta, a neurotoxin from the spider Phoneutria nigriventer: cology 39, 1756–1767. comparison with omega-conotoxin MVIIA and morphine. Toxicon. Lewis, R.J., 2009. Conotoxin venom peptide therapeutics. Adv. Exp. Med. Diniz, M.R., Theakston, R.D., Crampton, J.M., Cordeiro, M.N., Pimenta, A.M., Biol. 655, 44–48. De Lima, M.E., Diniz, C.R., 2006. Functional expression and purifica- Lewis, R.J., Nielsen, K.J., Craik, D.J., Loughnan, M.L., Adams, D.A., Sharpe, I. tion of recombinant Tx1, a sodium channel blocker neurotoxin from A., Luchian, T., Adams, D.J., Bond, T., Thomas, L., Jones, A., Matheson, J. the venom of the Brazilian “armed” spider, Phoneutria nigriventer. L., Drinkwater, R., Andrews, P.R., Alewood, P.F., 2000. Novel omega- Protein Expr. Purif. 50, 18–24. conotoxins from Conus catus discriminate among neuronal calcium Doering, C.J., Zamponi, G.W., 2003. Molecular pharmacology of high channel subtypes. J. Biol. Chem. 275, 35335–35344. voltage-activated calcium channels. J. Bioenerg. Biomembr. 35, 491– Li, R.A., Tomaselli, G.F., 2004. Using the deadly mu-conotoxins as probes 505. of voltage-gated sodium channels. Toxicon 44, 117–122. Domin, H., Zieba, B., Golembiowska, K., Kowalska, M., Dziubina, A., Lilie, H., Schwarz, E., Rudolph, R., 1998. Advances in refolding of proteins Smialowska, M., 2010. Neuroprotective potential of mGluR5 produced in E. coli. Curr. Opin. Biotechnol. 9, 497–501. 918 I.A. Souza et al. / Toxicon 60 (2012) 907–918

Lipton, S.A., 2006. Paradigm shift in neuroprotection by NMDA receptor Sharp, P.M., Li, W.H., 1987. The codon Adaptation Index–a measure of blockade: memantine and beyond. Nat. Rev. Drug Discov. 5, 160–170. directional synonymous codon usage bias, and its potential applica- Lipton, S.A., 2007. Pathologically-activated therapeutics for neuro- tions. Nucleic Acids Res. 15, 1281–1295. protection: mechanism of NMDA receptor block by memantine and S- Shu, Q., Lu, S.Y., Gu, X.C., Liang, S.P., 2002. The structure of spider toxin nitrosylation. Curr. Drug Targets 8, 621–632. huwentoxin-II with unique disulfide linkage: evidence for structural Livett, B.G., Gayler, K.R., Khalil, Z., 2004. Drugs from the sea: conopeptides evolution. Protein Sci. 11, 245–252. as potential therapeutics. Curr. Med. Chem. 11, 1715–1723. Singh, S.M., Panda, A.K., 2005. Solubilization and refolding of bacterial Malakhov, M.P., Mattern, M.R., Malakhova, O.A., Drinker, M., Weeks, S.D., inclusion body proteins. J. Biosci. Bioeng. 99, 303–310. Butt, T.R., 2004. SUMO fusions and SUMO-specific protease for effi- Souza, A.H., Ferreira, J., Cordeiro, M.N., Vieira, L.B., De Castro, C.J., cient expression and purification of proteins. J. Struct. Funct. Geno- Trevisan, G., Reis, H., Souza, I.A., Richardson, M., Prado, M.A., Prado, V.F., mics 5, 75–86. Gomez, M.V., 2008. Analgesic effect in rodents of native and recombi- Marblestone, J.G., Edavettal, S.C., Lim, Y., Lim, P., Zuo, X., Butt, T.R., 2006. nant Ph alpha 1beta toxin, a high-voltage-activated calcium channel Comparison of SUMO fusion technology with traditional gene fusion blocker isolated from armed spider venom. Pain 140, 115–126. systems: enhanced expression and solubility with SUMO. Protein Sci. Sreerama, N., Venyaminov, S.Y., Woody, R.W., 1999. Estimation of the 15, 182–189. number of helical and strand segments in proteins using CD spec- Meng, E., Cai, T.F., Li, W.Y., Zhang, H., Liu, Y.B., Peng, K., Liang, S., Zhang, D. troscopy. Protein Sci. 8, 370–380. Y., 2011. Functional expression of spider neurotoxic peptide Sreerama, N., Woody, R.W., 2000. Estimation of protein secondary huwentoxin-I in E. coli. PLoS. One 6, e21608. structure from CD spectra: comparison of CONTIN, SELCON and Mesquita Jr., F., Prado, M.A., Gomez, R.S., Romano-Silva, M.A., Gomez, M.V. CDSSTR methods with an expanded reference set. Anal. Biochem. 287, þ , 1998. The effect of calcium channels blockers in the K -evoked 252–260. release of [3H]adenine nucleotides from rat brain cortical synapto- Stewart, C., Bailey, J., Manoil, C., 1998. Mutant membrane protein toxicity. somes. Neurosci. Lett. 258, 57–59. J. Biol. Chem. 273, 28078–28084. Miranda, D.M., Romano-Silva, M.A., Kalapothakis, E., Diniz, C.R., Studier, F.W., Rosenberg, A.H., Dunn, J.J., Dubendorff, J.W., 1990. Use of T7 Cordeiro, M.N., Santos, T.M., Prado, M.A., Gomez, M.V., 1998. Pho- RNA polymerase to direct expression of cloned genes. Methods neutria nigriventer toxins block tityustoxin-induced calcium influx in Enzymol. 185, 60–89. synaptosomes. Neuroreport 9, 1371–1373. Thomas, J.G., Ayling, A., Baneyx, F., 1997. Molecular chaperones, folding Misawa, S., Kumagai, I., 1999. Refolding of therapeutic proteins produced catalysts, and the recovery of active recombinant proteins from E. coli. in Escherichia coli as inclusion bodies. Biopolymers 51, 297–307. To fold or to refold. Appl. Biochem. Biotechnol. 66, 197–238. Moura, J.R., Prado, M.A., Gomez, M.V., Kalapothakis, E., Diniz, C.R., Torres, F.S., Silva, C.N., Lanza, L.F., Santos, A.V., Pimenta, A.M., De Lima, M. Cordeiro, M.N., Romano-Silva, M.A., 1998. Investigation of the effect of E., Diniz, M.R., 2010. Functional expression of a recombinant toxin – PhTx2, from the venom of the spider Phoneutria nigriventer, on the rPnTx2-6-active in erectile function in rat. Toxicon 56, 1172–1180. release of [3H]-acetylcholine from rat cerebrocortical synaptosomes. Van Stokkum, I.H.M., Spoelder, H.J.W., Bloemendal, M., Van Grondelle, R., Toxicon 36, 1189–1192. Groen, F.C.A., 1990. Estimation of protein secondary structure and Penaforte, C.L., Prado, V.F., Prado, M.A., Romano-Silva, M.A., Guimaraes, P. error analysis from CD spectra. Anal. Biochem. 191, 110–118. E., De, M.L., Gomez, M.V., Kalapothakis, E., 2000. Molecular cloning of Vassilevski, A.A., Kozlov, S.A., Grishin, E.V., 2009. Molecular diversity of cDNAs encoding insecticidal neurotoxic peptides from the spider spider venom. Biochem. (Mosc. ) 74, 1505–1534. Phoneutria nigriventer. Toxicon 38, 1443–1449. Vieira, L.B., Kushmerick, C., Hildebrand, M.E., Garcia, E., Stea, A., Pinheiro, A.C., da Silva, A.J., Prado, M.A., Cordeiro, M.N., Richardson, M., Cordeiro, M.N., Richardson, M., Gomez, M.V., Snutch, T.P., 2005. Batista, M.C., de Castro Junior, C.J., Massensini, A.R., Guatimosim, C., Inhibition of high voltage-activated calcium channels by spider toxin Romano-Silva, M.A., Kushmerick, C., Gomez, M.V., 2009. Phoneutria PnTx3-6. J. Pharmacol. Exp. Ther. 314, 1370–1377. spider toxins block ischemia-induced glutamate release, neuronal Vieira, L.B., Kushmerick, C., Reis, H.J., Diniz, C.R., Cordeiro, M.N., Prado, M. death, and loss of neurotransmission in hippocampus. Hippocampus A., Kalapothakis, E., Romano-Silva, M.A., Gomez, M.V., 2003. PnTx3-6 þ 19, 1123–1129. a spider neurotoxin inhibits K -evoked increase in [Ca2þ](i) and þ Prado, M.A., Guatimosim, C., Gomez, M.V., Diniz, C.R., Cordeiro, M.N., Ca2 -dependent glutamate release in synaptosomes. Neurochem. Int. Romano-Silva, M.A., 1996. A novel tool for the investigation of 42, 277–282. glutamate release from rat cerebrocortical synaptosomes: the toxin Vieira, L.B., Pimenta, A.M., Richardson, M., Bemquerer, M.P., Reis, H.J., Tx3-3 from the venom of the spider Phoneutria nigriventer. Biochem. J. Cruz, J.S., Gomez, M.V., Santoro, M.M., Ferreira-de-Oliveira, R., 314 (Pt 1), 145–150. Figueiredo, S.G., Snutch, T.P., Cordeiro, M.N., 2007. Leftward shift in þ Prinz, W.A., Aslund, F., Holmgren, A., Beckwith, J., 1997. The role of the the voltage-dependence for Ca2 currents activation induced by thioredoxin and glutaredoxin pathways in reducing protein disulfide a new toxin from Phoneutria reidyi (Aranae, Ctenidae) venom. Cell bonds in the Escherichia coli cytoplasm. J. Biol. Chem. 272, 15661– Mol. Neurobiol. 27, 129–146. 15667. Wen, S., Wilson, D.T., Kuruppu, S., Korsinczky, M.L., Hedrick, J., Pang, L., Reis, H.J., Gomez, M.V., Kalapothakis, E., Diniz, C.R., Cordeiro, M.N., Szeto, T., Hodgson, W.C., Alewood, P.F., Nicholson, G.M., 2005. Prado, M.A., Romano-Silva, M.A., 2000. Inhibition of glutamate uptake Discovery of an MIT-like atracotoxin family: spider venom peptides by Tx3-4 is dependent on the redox state of cysteine residues. Neu- that share sequence homology but not pharmacological properties roreport 11, 2191–2194. with AVIT family proteins. Peptides 26, 2412–2426. Reis, H.J., Prado, M.A., Kalapothakis, E., Cordeiro, M.N., Diniz, C.R., De Wetlaufer, D.B., Branca, P.A., Chen, G.X., 1987. The oxidative folding of Marco, L.A., Gomez, M.V., Romano-Silva, M.A., 1999. Inhibition of proteins by disulfide plus thiol does not correlate with redox poten- glutamate uptake by a polypeptide toxin (phoneutriatoxin 3-4) from tial. Protein Eng. 1, 141–146. the spider Phoneutria nigriventer. Biochem. J. 343 (Pt 2), 413–418. Williams, J.A., Day, M., Heavner, J.E., 2008. Ziconotide: an update and Sanger, F., Nicklen, S., Coulson, A.R., 1977. DNA sequencing with chain- review. Expert Opin. Pharmacother. 9, 1575–1583. terminating inhibitors. Proc. Natl. Acad. Sci. U. S. A. 74, 5463–5467. Yan, L., Adams, M.E., 2000. The spider toxin omega-Aga IIIA defines a high Sattler, R., Tymianski, M., 2000. Molecular mechanisms of calcium- affinity site on neuronal high voltage-activated calcium channels. J. dependent excitotoxicity. J. Mol. Med. (Berl) 78, 3–13. Biol. Chem. 275, 21309–21316.