Title: Pore Formation by Equinatoxin, a Eukaryotic Pore-forming Toxin, Requires a Flexible N-terminal Region and a Stable β-Sandwich
Abstract: Actinoporins are eukaryotic pore-forming proteins that create 2-nm pores in natural and model lipid membranes by the self-association of four monomers. The regions that undergo conformational change and form part of the transmembrane pore are currently being defined. It was shown recently that the N-terminal region (residues 10–28) of equinatoxin, an actinoporin from Actinia equina, participates in building of the final pore wall. Assuming that the pore is formed solely by a polypeptide chain, other parts of the toxin should constitute the conductive channel and here we searched for these regions by disulfide scanning mutagenesis. Only double cysteine mutants where the N-terminal segment 1–30 was attached to the β-sandwich exhibited reduced hemolytic activity upon disulfide formation, showing that other parts of equinatoxin, particularly the β-sandwich and importantly the C-terminal α-helix, do not undergo large conformational rearrangements during the pore formation. The role of the β-sandwich stability was independently assessed via destabilization of a part of its hydrophobic core by mutations of the buried Trp117. These mutants were considerably less stable than the wild-type but exhibited similar or slightly lower permeabilizing activity. Collectively these results show that a flexible N-terminal region and stable β-sandwich are pre-requisite for proper pore formation by the actinoporin family. Actinoporins are eukaryotic pore-forming proteins that create 2-nm pores in natural and model lipid membranes by the self-association of four monomers. The regions that undergo conformational change and form part of the transmembrane pore are currently being defined. It was shown recently that the N-terminal region (residues 10–28) of equinatoxin, an actinoporin from Actinia equina, participates in building of the final pore wall. Assuming that the pore is formed solely by a polypeptide chain, other parts of the toxin should constitute the conductive channel and here we searched for these regions by disulfide scanning mutagenesis. Only double cysteine mutants where the N-terminal segment 1–30 was attached to the β-sandwich exhibited reduced hemolytic activity upon disulfide formation, showing that other parts of equinatoxin, particularly the β-sandwich and importantly the C-terminal α-helix, do not undergo large conformational rearrangements during the pore formation. The role of the β-sandwich stability was independently assessed via destabilization of a part of its hydrophobic core by mutations of the buried Trp117. These mutants were considerably less stable than the wild-type but exhibited similar or slightly lower permeabilizing activity. Collectively these results show that a flexible N-terminal region and stable β-sandwich are pre-requisite for proper pore formation by the actinoporin family. Interaction of proteins with lipid membranes is fundamental for many biological processes. Proteins have to bind transiently or permanently to the membranes to elicit biological functions crucial for cells (1Goñi F.M. Mol. Membr. Biol. 2002; 19: 237-245Crossref PubMed Scopus (43) Google Scholar). Many examples of protein-membrane interactions are known, most notably those involved in cell signaling events (2Cho W. J. Biol. Chem. 2001; 276: 32407-32410Abstract Full Text Full Text PDF PubMed Scopus (192) Google Scholar, 3Hurley J.H. Misra S. Annu. Rev. Biophys. Biomol. Struct. 2000; 29: 49-79Crossref PubMed Scopus (226) Google Scholar, 4Lemmon M.A. Ferguson K.M. Schlessinger J. Cell. 1996; 85: 621-624Abstract Full Text Full Text PDF PubMed Scopus (431) Google Scholar). Furthermore, a large number of protein toxins attach to membranes as the first step in creating transmembrane pores that are permeable for various compounds (5Gouaux E. Curr. Opin. Struct. Biol. 1997; 7: 566-573Crossref PubMed Scopus (151) Google Scholar). The lipid membrane is the first obstacle that needs to be overcome by toxins and in fact creation of transmembrane pores is a very efficient way of killing cells. Pore-forming toxins (PFTs) 1The abbreviations used are: PFT, pore-forming toxin; ANS, 1-anilinonaphthalene-8-sulfonic acid; BRBC, bovine red blood cells; CD, circular dichroism; DTT, dithiothreitol; EqtII, equinatoxin II; PC, 1,2-dioleoyl-sn-glycero-3-phosphorylcholine; POC, phosphorylcholine; SM, sphingomyelin; StII, sticholysin II; mN, millinewton(s). are thus a very important group of natural toxins (6Menestrina G. Dalla Serra M. Lazarovici P. Pore-forming Peptides and Protein Toxins. Taylor & Francis, London2003Google Scholar), and the best studied examples are bacterial PFTs, because they are also important virulence factors in human and animal bacterial disease. In recent years they have been used to study fundamental biological processes such as protein-membrane and protein-protein interactions within the lipid membrane milieu and conformational changes associated with the change of environment from polar to hydrophobic as encountered within the core of lipid membranes. In general all PFTs intoxicate cells by a multistep mechanism that involves binding to the membrane, oligomerization, and transfer of a region of polypeptide across the lipid bilayer to form a functional pore (5Gouaux E. Curr. Opin. Struct. Biol. 1997; 7: 566-573Crossref PubMed Scopus (151) Google Scholar). Heterogeneous PFTs can be divided into two groups according to the structural element present in the final pore. So-called β-PFTs form pores by creation of transmembrane β-barrels. Examples include α-toxin from Staphylococcus aureus, cholesterol-dependent cytolysins and anthrax protective antigen (7Heuck A.P. Tweten R.K. Johnson A.E. Biochemistry. 2001; 40: 9065-9073Crossref PubMed Scopus (128) Google Scholar). In contrast, α-PFTs, like colicins from Escherichia coli, form pores by bundles of α-helices (8Lakey J.H. Slatin S.L. Van Der Goot F.G. Pore-Forming Toxins. 257. Springer Verlag, Heidelberg, Germany2001: 131-161Google Scholar). Although pores of β-PFTs are stable structures that resist heat or detergents, α-PFT pores are not stable and have provided little structural information. Actinoporins are a group of eukaryotic PFTs that are very important in our understanding of how α-PFTs function (9Anderluh G. Maček P. Structure. 2003; 11: 1312-1313Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar, 10Anderluh G. Maček P. Toxicon. 2002; 40: 111-124Crossref PubMed Scopus (348) Google Scholar). These 20-kDa, mainly basic, proteins can efficiently lyse various cells and permeabilize model lipid membranes by using an α-helix in a mechanism that is slowly being defined in vitro (11Athanasiadis A. Anderluh G. Maček P. Turk D. Structure. 2001; 9: 341-346Abstract Full Text Full Text PDF PubMed Scopus (188) Google Scholar, 12Hinds M.G. Zhang W. Anderluh G. Hansen P.E. Norton R.S. J. Mol. Biol. 2002; 315: 1219-1229Crossref PubMed Scopus (126) Google Scholar, 13Hong Q. Gutiérrez-Aguirre I. Barlič A. Malovrh P. Kristan K. Podlesek Z. Maček P. Turk D. González-Mañas J.M. Lakey J.H. Anderluh G. J. Biol. Chem. 2002; 277: 41916-41924Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar, 14Bonev B.B. Lam Y.H. Anderluh G. Watts A. Norton R.S. Separovic F. Biophys. J. 2003; 84: 2382-2392Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar, 15Malovrh P. Viero G. Dalla Serra M. Podlesek Z. Lakey J.H. Maček P. Menestrina G. Anderluh G. J. Biol. Chem. 2003; 278: 22678-22685Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar, 16Anderluh G. Dalla Serra M. Viero G. Guella G. Maček P. Menestrina G. J. Biol. Chem. 2003; 278: 45216-45223Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, 17Mancheño J.M. Martin-Benito J. Martinez-Ripoll M. Gavilanes J.G. Hermoso J.A. Structure. 2003; 11: 1319-1328Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar, 18Caaveiro J.M.M. Echabe I. Gutiérrez-Aguirre I. Nieva J.L. Arrondo J.L.R. González-Mañas J.M. Biophys. J. 2001; 80: 1343-1353Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar). Actinoporins represent a unique family of cysteineless proteins with high sequence similarity (10Anderluh G. Maček P. Toxicon. 2002; 40: 111-124Crossref PubMed Scopus (348) Google Scholar). The two most studied examples are equinatoxin II (EqtII) from the sea anemone Actinia equina and sticholysin II (StII) from Stichodactyla helianthus. They are both globular proteins composed of a tightly folded β-sandwich flanked on two sides by α-helices (Fig. 1, A and B) (11Athanasiadis A. Anderluh G. Maček P. Turk D. Structure. 2001; 9: 341-346Abstract Full Text Full Text PDF PubMed Scopus (188) Google Scholar, 12Hinds M.G. Zhang W. Anderluh G. Hansen P.E. Norton R.S. J. Mol. Biol. 2002; 315: 1219-1229Crossref PubMed Scopus (126) Google Scholar, 17Mancheño J.M. Martin-Benito J. Martinez-Ripoll M. Gavilanes J.G. Hermoso J.A. Structure. 2003; 11: 1319-1328Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). The N-terminal segment containing an amphipathic α-helix (helix A) is the only part of the molecule that can be displaced without disrupting the β-sandwich fold. Actinoporins strictly follow the multistep scheme of pore formation. The initial attachment to the membrane is achieved by a cluster of aromatic amino acid residues from one of the broad loops at the bottom of the β-sandwich and helix B (13Hong Q. Gutiérrez-Aguirre I. Barlič A. Malovrh P. Kristan K. Podlesek Z. Maček P. Turk D. González-Mañas J.M. Lakey J.H. Anderluh G. J. Biol. Chem. 2002; 277: 41916-41924Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar, 19Maček P. Zecchini M. Pederzolli C. Dalla Serra M. Menestrina G. Eur. J. Biochem. 1995; 234: 329-335Crossref PubMed Scopus (55) Google Scholar, 20Malovrh P. Barlič A. Podlesek Z. Maček P. Menestrina G. Anderluh G. Biochem. J. 2000; 346: 223-232Crossref PubMed Scopus (81) Google Scholar) and by a recently defined phosphorylcholine (POC) binding site (17Mancheño J.M. Martin-Benito J. Martinez-Ripoll M. Gavilanes J.G. Hermoso J.A. Structure. 2003; 11: 1319-1328Abstract Full Text Full Text PDF PubMed Scopus (205) Google Scholar). In the next step the segment 10–28 is transferred to the lipid-water interface (13Hong Q. Gutiérrez-Aguirre I. Barlič A. Malovrh P. Kristan K. Podlesek Z. Maček P. Turk D. González-Mañas J.M. Lakey J.H. Anderluh G. J. Biol. Chem. 2002; 277: 41916-41924Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar, 15Malovrh P. Viero G. Dalla Serra M. Podlesek Z. Lakey J.H. Maček P. Menestrina G. Anderluh G. J. Biol. Chem. 2003; 278: 22678-22685Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar), where it lies flat on the membrane (15Malovrh P. Viero G. Dalla Serra M. Podlesek Z. Lakey J.H. Maček P. Menestrina G. Anderluh G. J. Biol. Chem. 2003; 278: 22678-22685Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar) and, finally, in the last step forms the walls of the final pore (15Malovrh P. Viero G. Dalla Serra M. Podlesek Z. Lakey J.H. Maček P. Menestrina G. Anderluh G. J. Biol. Chem. 2003; 278: 22678-22685Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). The final oligomeric pore of actinoporins has not yet been directly visualized. Instead, the number of monomers was deduced only from the cross-linking studies and kinetic experiments. Both, EqtII and StII, form 2-nm pores in diameter composed of four monomers (21Belmonte G. Pederzolli C. Maček P. Menestrina G. J. Membrane Biol. 1993; 131: 11-22Crossref PubMed Scopus (192) Google Scholar, 22Tejuca M. Dalla Serra M. Ferreras M. Lanio M.E. Menestrina G. Biochemistry. 1996; 35: 14947-14957Crossref PubMed Scopus (157) Google Scholar, 23Tejuca M. Dalla Serra M. Potrich C. Alvarez C. Menestrina G. J. Membr. Biol. 2001; 183: 125-135Crossref PubMed Scopus (98) Google Scholar, 24De los Rios V. Mancheño J.M. Lanio M.E. Oñaderra M. Gavilanes J.G. Eur. J. Biochem. 1998; 252: 284-289Crossref PubMed Scopus (104) Google Scholar). Pores of such a diameter cannot be simply formed by a cluster of four helices (Fig. 1D). Either other parts of the molecule contribute to the final oligomeric conductive pore, or the pore is composed partially of lipid molecules from the bilayer. The first possibility requires considerable unfolding of the β-sandwich and its rearrangements in such a way that remaining space between helices is filled with the polypeptide chain. A testable model has been proposed for EqtII involving the formation of a transmembrane β-barrel (12Hinds M.G. Zhang W. Anderluh G. Hansen P.E. Norton R.S. J. Mol. Biol. 2002; 315: 1219-1229Crossref PubMed Scopus (126) Google Scholar). Each molecule would then contribute one β-hairpin, including residues 118–142. This model requires a significant unfolding of the β sandwich, because each hairpin would include β-strands 8 and 9 and C-terminal helix B. The aim of this study is to test such models by clearly defining those regions of the EqtII molecule that undergo conformational rearrangements during the pore formation and which could therefore participate in the building of pore walls. We used disulfide scanning mutagenesis, because this technique was previously successfully used to determine conformationally flexible regions in PFTs (25Duche D. Izard J. González-Mañas J.M. Parker M.W. Crest M. Chartier M. Baty D. J. Biol. Chem. 1996; 271: 15401-15406Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar, 26Schwartz J.L. Juteau M. Grochulski P. Cygler M. Préfontaine G. Brousseau R. Masson L. FEBS Lett. 1997; 410: 397-402Crossref PubMed Scopus (104) Google Scholar, 27Kawate T. Gouaux E. Protein Sci. 2003; 12: 997-1006Crossref PubMed Scopus (61) Google Scholar). The movement of polypeptide chain segments during pore formation can be restricted by a disulfide formation and viewed as a large decrease in permeabilizing activity. We show that apart from the N-terminal segment there are no other parts, specifically β-sandwich and helix B, that undergo gross conformational changes. Interestingly, mutations of broad loops at the bottom of the molecule abolished hemolytic activity completely, showing the need to preserve primary structure for pore formation. The need for an unperturbed β-sandwich was further and independently confirmed by destabilizing mutants in the β-sandwich. We propose that the flexible N-terminal segment and loops at the bottom of the molecule are the solely parts of the molecule directly involved in the steps that follow the initial binding, whereas the β-sandwich is maintained rigid during all steps of pore formation. Materials—1, 2-Dioleoyl-sn-glycero-3-phosphorylcholine (PC) and brain sphingomyelin (SM) were both from Avanti Polar Lipids (Alabaster, AL). All other materials were from Sigma, unless specified differently. Mutant Design—Positions for mutations were selected first by the SSBond program (28Hazes B. Dijkstra B.W. Protein Eng. 1988; 2: 119-125Crossref PubMed Scopus (153) Google Scholar) using EqtII crystal structure as a template (11Athanasiadis A. Anderluh G. Maček P. Turk D. Structure. 2001; 9: 341-346Abstract Full Text Full Text PDF PubMed Scopus (188) Google Scholar) (PDB code 1IAZ). Additional mutants were selected by manually inspecting the structure and searching for amino acid pairs with suitable Cβ–Cβ distances (between 3.8 and 4.8 Å). These additional mutants were employed to cover all regions of the EqtII molecule. All selected pairs were analyzed by the Scwrl3 program (29Canutescu A.A. Shelenkov A.A. Dunbrack R.L. Protein Sci. 2003; 12: 2001-2014Crossref PubMed Scopus (876) Google Scholar) and Swiss-PDB Viewer (www.expasy.org/spdbv/) (30Guex N. Peitsch M.C. Electrophoresis. 1997; 18: 2714-2723Crossref PubMed Scopus (9641) Google Scholar). The most suitable rotamer was selected, either by program, in the case of Scwrl3, or manually by the use of Swiss-PDB Viewer, when Scwrl3 did not report in silico disulfide bridge formation. Disulfide bonds could be formed in silico by all mutants used in this study. Cloning, Expression, and Isolation of Mutants—Double cysteine mutants and Trp117 mutants were prepared by replacing the corresponding EqtII residue with the desired amino acid as described previously (31Anderluh G. Barlič A. Podlesek Z. Maček P. Pungerčar J. Gubenšek F. Zecchini M.L. Dalla S. Menestrina G. Eur. J. Biochem. 1999; 263: 128-136Crossref PubMed Scopus (91) Google Scholar). Wild-type EqtII and selected mutants were expressed in the E. coli BL21(DE3) strain and isolated from the bacterial cytoplasm exactly as described elsewhere (15Malovrh P. Viero G. Dalla Serra M. Podlesek Z. Lakey J.H. Maček P. Menestrina G. Anderluh G. J. Biol. Chem. 2003; 278: 22678-22685Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar, 31Anderluh G. Barlič A. Podlesek Z. Maček P. Pungerčar J. Gubenšek F. Zecchini M.L. Dalla S. Menestrina G. Eur. J. Biochem. 1999; 263: 128-136Crossref PubMed Scopus (91) Google Scholar, 32Anderluh G. Pungerčar J. Štrukelj B. Maček P. Gubenšek F. Biochem. Biophys. Res. Comm. 1996; 220: 437-442Crossref PubMed Scopus (109) Google Scholar). All mutants were purified to homogeneity as observed on the SDS-PAGE gels. Treatment of Mutants with Reductant or Oxidant—Double cysteine mutants were treated either with 20 mm dithiothreitol (DTT) or 0.5 mm 1,10-phenantroline and 0.1 mm CuSO4 for 30 min at 37 °C to obtain reduced or oxidized form of the mutants, respectively. These concentrations were sufficient to obtain fully oxidized or reduced samples. Samples were immediately used for hemolytic activity measurements or erythrocytes binding assay. DTT has no effect on cells in these tests, because it was always diluted below 2 mm. Hemolytic Assay—Bacterial lysates were used to assay hemolytic activity of expressed double cysteine mutants. Mutants were expressed in E. coli BL21(DE3) strain. 20 ml of LB medium supplemented with 100 μg/ml ampicillin was inoculated with 1 ml of overnight culture. The expression of proteins was induced with final concentration of 0.4 mm isopropyl-1-thio-β-d-galactopyranoside, when bacteria reached an absorbance at 600 nm of ∼0.6, and then grown for additional 4 h. To prepare lysates, bacterial cells were harvested by centrifugation at 4 °C and 14,000 × g for 10 min. Pellets were resuspended in 0.5 ml of 50 mm Tris-HCl, pH 8, containing ∼2 mg/ml lysozyme. Incubation of samples at room temperature for 15 min was followed by a 30-s pulse of sonication. Supernatants were collected after centrifugation at 4 °C and 14,000 × g for 15 min. The pellets were washed with 0.25 ml of the same buffer, vortexed, sonicated, and centrifuged as above. Supernatants were merged and stored at –20 °C before use. Hemolytic activity was measured turbidimetrically by the use of a microplate reader (MRX; Dynex Technologies, Deckendorf, Germany) (20Malovrh P. Barlič A. Podlesek Z. Maček P. Menestrina G. Anderluh G. Biochem. J. 2000; 346: 223-232Crossref PubMed Scopus (81) Google Scholar). A suspension of bovine red blood cells (BRBC) with A630 = 0.5 in hemolysis buffer (0.13 m NaCl, 0.02 m Tris-HCl, pH 7.4) was prepared from well washed BRBC. 100 μl of BRBC suspension were added to 100 μl of 2-fold serially diluted bacterial lysates or purified mutants. Hemolysis was monitored by measuring the absorbance at 630 nm for 20 min at room temperature. The amount of bacterial lysate used for the assay, where reduced and oxidized forms were compared, corresponded to the volume of the lysate that exhibits 50% of maximal hemolytic rate at the dilution 32 or 64 (6th or 7th well in the multiwell plate) in the presence of 2 mm DTT. Binding of Double Cysteine Mutants to Erythrocytes—The binding assay was done as described in Hong et al. (13Hong Q. Gutiérrez-Aguirre I. Barlič A. Malovrh P. Kristan K. Podlesek Z. Maček P. Turk D. González-Mañas J.M. Lakey J.H. Anderluh G. J. Biol. Chem. 2002; 277: 41916-41924Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar). Briefly, 0.15 nmol of pretreated protein was incubated with BRBC suspension (A700 = 0.5) in erythrocyte buffer supplemented with 30 mm polyethylene glycol 3350 at 4 °C. After 10 min of incubation, the samples were pelleted and electrophoresis loading buffer was added to the pellets containing erythrocytes with bound toxin. Bound proteins were resolved by SDS-PAGE electrophoresis and Western-blotted as described below. SDS-PAGE Electrophoresis—Mini Protean II system (Bio-Rad) was used. 5 μl of bacterial lysates or 0.15 nmol of purified toxins, treated with reductant or oxidant as described above, were mixed with the same volume of SDS-PAGE sample loading buffer (120.5 mm Tris-HCl, 4% SDS (w/v), 10% glycerol (v/v), and 0.002% (w/v) Bromphenol Blue). Proteins were resolved on 15% SDS-PAGE gels. The gels were stained with Coomassie Blue or Western-blotted. For blotting, proteins were transferred to polyvinylidene difluoride membranes (Millipore), blocked in 4% bovine serum albumin at 4 °C overnight, and incubated with rabbit anti-EqtII serum for 2 h. Bands were stained with goat anti-rabbit antibodies conjugated with horseradish peroxidase using 3-amino-9-ethylcarbazole/H2O2. Permeabilizing Assay Using Liposomes—Large unilamellar vesicles were prepared by extrusion of multilamellar vesicles. Lipids, dissolved in chloroform, were spread on the round-bottom glass flask on a rotary evaporator and dried under vacuum for at least 3 h. The lipid film was resuspended in 1 ml of 60 mm calcein in vesicle buffer (140 mm NaCl, 20 mm Tris-HCl, pH 8.5, 1 mm EDTA) and freeze-thawed six times. Large unilamellar vesicles of 100 nm were extruded through polycarbonate membranes with 100-nm pores by an Avestin lipid extruder (Avestin, Ottawa, Canada) (33MacDonald R.C. MacDonald R.I. Menco B.P. Takeshita K. Subbarao N.K. Hu L. Biochim. Biophys. Acta. 1991; 1061: 297-303Crossref PubMed Scopus (1389) Google Scholar). The excess of calcein was removed from the calcein-loaded liposomes by gel filtration on a small G-50 column. For calcein release experiments, liposomes were stirred at the desired concentration at 25 °C. Toxins were added at final 0.05 μm concentration. Changes in the fluorescence intensity were followed on a Jasco FP-750 spectrofluorimeter. Excitation and emission wavelengths were set to 485 and 520 nm, respectively. Excitation and emission slits were set to 5 nm. The permeabilization induced by the toxins was expressed in percentage as compared with the maximal permeabilization obtained at the end of the assay by the addition of Triton X-100 at a final concentration of 2 mm. Critical Pressure Measurements—A MicroTrough S system (Kibron, Helsinki, Finland) was used to perform surface pressure measurements. The aqueous phase consisted of 0.5 ml of 10 mm HEPES, 200 mm NaCl, pH 7.5, at room temperature. The PC/SM 1/1 mixture of lipids, dissolved in chloroform, was gently spread over the surface, until desired initial pressure was achieved. The protein was injected with a Hamilton microsyringe through a hole connected to the subphase. The final protein concentration was 1 μm. The increment in surface pressure versus time was followed until a stable signal was obtained. ANS Binding Assay—The fluorescence of 1-anilinonaphthalene-8-sulfonic acid (ANS) in the presence of EqtII or Trp117 mutants in 10 mm HEPES, 200 mm NaCl, pH 7.5, was followed at 468 nm with excitation wavelength set to 370 nm. Excitation and emission slits were set to 5 nm. The cuvette was placed in a thermostatted cell holder, and the contents were stirred continuously. Measurements were taken between 15 and 75 °C by using a 1 °C/min heating rate. The concentration of ANS and protein was 15 and 6 μm, respectively. Circular Dichroism Spectroscopy—Circular dichroism (CD) spectra were measured in the far-UV region, 250–185 nm, by Jasco J-810 spectropolarimeter. Bandwidth was set to 2 nm; scanning speed was 50 nm/min. Proteins were placed in a 1-mm path-length stoppered cuvette (Hellma) at 8 μm concentration in 10 mm NaH2PO4, pH 7.0. The reported spectra are averages of 10 scans. The signal at 217 nm was followed for thermal denaturation that was performed between 20–90 °C at a heating rate of 1 °C/min. Crystallization and Data Collection—Crystals of EqtII double cysteine mutant Val8 → Cys, Lys69 → Cys (8–69) were grown by the sitting drop vapor diffusion method. The reservoir contained 1 ml of 15% polyethylene glycol 4000, 0.12 m ammonium sulfate, 0.06 m sodium acetate, pH 6.0. The drop was composed of 3.7 μl of reservoir solution and 2.3 μl of the protein (10 mg/ml). Diffraction data were collected from a single crystal using CuKα radiation from a Rigaku Ru200 rotating anode x-ray generator and recorded on a 345-mm MAR Research image plate detector. Auto indexing and scaling was done using HKL2000 (34Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38617) Google Scholar). The crystal diffracted to 2.3-Å resolution and belonged to the primitive monoclinic space group P21 with cell dimensions a = 36.6 Å, b = 52.7 Å, c = 39.8 Å, β = 92.3°. The asymmetric unit contained one molecule. Structure Determination and Refinement—The structure of 8–69 mutant was solved by the molecular replacement method implemented in the AMoRe (35Navaza J. Saludjian P. Macromol. Cryst. Pt. A. 1997; 276: 581-594Crossref Scopus (368) Google Scholar) program. The crystal structure of the wild-type equinatoxin (11Athanasiadis A. Anderluh G. Maček P. Turk D. Structure. 2001; 9: 341-346Abstract Full Text Full Text PDF PubMed Scopus (188) Google Scholar) was used as the search model. In the subsequent structure determination, the program MAIN (36.Turk, D. (1992) Weiterentwicklung eines Programmes für Molekülgraphik und Electronendichte-manipulation und Seine Anwendungen auf Verschiedene Protein-strukturaufklärungen, Dissertation, Technische Universität München, GermanyGoogle Scholar) was used for model building and refinement. The final refinement included all reflections in the resolution range 10–2.3 Å, with the crystallographic R-value being 0.192. The geometry of the final model was inspected with MAIN and Procheck (37Laskowski R.A. Macarthur M.W. Moss D.S. Thornton J.M. J. App. Crystallogr. 1993; 26: 283-291Crossref Google Scholar). All residues lie in the allowed regions of the Ramachandran plot, 0.913 of residues in most favored regions and 0.087 in additional allowed regions. The coordinates have been deposited in the Protein Data Bank with the accession code 1TZQ. Preparation of Double Cysteine Mutants—The wild-type EqtII is produced in the active form in the cytoplasm of E. coli. The negative control, the soluble fraction of bacterial cells with pT7 expression vector without EqtII gene, is not hemolytically active (see Fig. 2) (32Anderluh G. Pungerčar J. Štrukelj B. Maček P. Gubenšek F. Biochem. Biophys. Res. Comm. 1996; 220: 437-442Crossref PubMed Scopus (109) Google Scholar). Thus any hemolytic activity observed is due to expressed active toxin. The hemolytic assay we use routinely is very sensitive, because nanomolar concentrations of the toxin already produce detectable hemolysis. Conformational changes are expected during pore formation. By introducing two cysteines at appropriate positions in a cysteineless EqtII it is possible to block conformational changes upon disulfide formation. Therefore, twenty-five mutants were designed throughout the molecule covering the N-terminal region (6 mutants), the region around the C-terminal helix B (5 mutants), β-sandwich (11 mutants), and broad loops at the bottom of the molecule (3 mutants) (Fig. 1 and Table I). Positions for mutagenesis were determined by using the program SSBond using EqtII crystal structure as a template. Additional mutants were added to selection, although SSBond did not report them as favorable, on the basis of their Cβ–Cβ distances. The majority of mutants have this distance between 3.8 and 4.8 Å, one of the criteria that needs to be fulfilled for the disulfide formation. This set of mutants covers all regions proposed to be involved in conformational changes during the pore formation, thus enabling us to exclude regions that are not important.Table IMutants used for double cysteine scanningMutantaAmino acids that were mutated to a cysteine are shown in the table by a three-letter code.Structural elements linkedbβ, beta strand; h, helix. Loops between the strands or helices are written without hyphens.Cβ—CβcCβ—Cβ distances were determined by using the SSBond program (28) and from the crystal structure of EqtII (11) (PDB code 1IAZ).Hemolytic activitydHemolytic activity shown is relative to the wild-type in reduced conditions (final 2 mm DTT): ++, more than 25% of wild-type's activity; +, 3-25% of wild-type activity; -, <3% of wild-type activity or not hemolytically active.ÅN-terminal regionAla7-Leu72β1-β54.1++Val8-Lys69β1-β55.0++Gly11-Asp38eCα—Cβ distance is shown.β1hA-β24.1+Leu14-Asp38β1hA-β24.8++Leu14-Ala70β1hA-β53.8-Val22-Ala34hA-β23.8++Helix BVal89-Ala128β6-β8hB4.1++Val89-Tyr133β6-hB5.0++Leu103-Leu136β7-hB3.8++Asn118-Leu136β8-hB4.3-Asn118-Pro142β8-hBβ93.8++LoopsArg31-Lys77β2-β5β64.8-Gly85-Tyr108eCα—Cβ distance is shown.β5β6-β7β84.2-Pro107-Tyr113β7β83.5-β-SandwichGlu40-Ser177β2-β124.8++Lys43-Ser95β2β3-β6β73.9++Thr44-Ser95β2β3-β6β74.5++Ala47-Leu61β3-β3β44.0++Leu48-Ala91β3-β64.2++Tyr51-Ala128β3-β8hB3.9+Ser54-Val87β3-β64.0++Val87-Ser105β6-β74.6-Val89-Leu103β6-β75.5++Val119-Ser160β8-β114.2++Ser160-Val176β11-β125.1++a Amino acids that were mutated to a cysteine are shown in the table by a three-letter code.b β, beta strand; h, helix. Loops between the strands or helices are written without hyphens.c Cβ—Cβ distances were determined by using the SSBond program (28Hazes B. Dijkstra B.W. Protein Eng. 1988; 2: 119-125Crossref PubMed Scopus (153) Google Scholar) and from the crystal structure of EqtII (11Athanasiadis A. And