Portal de Periódicos da CAPES

Evidence for the “Dock, Lock, and Latch” Ligand Binding Mechanism of the Staphylococcal Microbial Surface Component Recognizing Adhesive Matrix Molecules (MSCRAMM) SdrG

2007; Elsevier BV; Volume: 283; Issue: 1 Linguagem: Inglês

10.1074/jbc.m706252200

1083-351X

M. Gabriela Bowden, Alejandro P. Heuck, Karthe Ponnuraj, Elena Kolosova, Damon Choe, Sivashankarappa Gurusiddappa, Sthanam V.L. Narayana, Arthur E. Johnson, Magnus Höök,

Streptococcal Infections and Treatments

Staphylococcus epidermidis is an opportunistic pathogen and a major cause of foreign body infections. The S. epidermidis fibrinogen (Fg)-binding adhesin SdrG is necessary and sufficient for the attachment of this pathogen to Fg-coated materials. Based largely on structural analyses of the ligand binding domain of SdrG as an apo-protein and in complex with a Fg-like peptide, we proposed that SdrG follows a "dock, lock, and latch" mechanism to bind to Fg. This binding mechanism involves the docking of the ligand in a pocket formed between two SdrG subdomains followed by the movement of a C-terminal extension of one subdomain to cover the ligand and to insert and complement a β-sheet in a neighboring subdomain. These proposed events result in a greatly stabilized closed conformation of the MSCRAMM-ligand complex. In this report, we describe a biochemical analysis of the proposed conformational changes that SdrG undergoes upon binding to its ligand. We have introduced disulfide bonds into SdrG to stabilize the open and closed forms of the apo-form of the MSCRAMM. We show that the stabilized closed form does not bind to the ligand and that binding can be restored in the presence of reducing agents such as dithiothreitol. We have also used Förster resonance energy transfer to dynamically show the conformational changes of SdrG upon binding to its ligand. Finally, we have used isothermic calorimetry to determine that hydrophobic interactions between the ligand and the protein are responsible for re-directing the C-terminal extension of the second subdomain required for triggering the β-strand complementation event. Staphylococcus epidermidis is an opportunistic pathogen and a major cause of foreign body infections. The S. epidermidis fibrinogen (Fg)-binding adhesin SdrG is necessary and sufficient for the attachment of this pathogen to Fg-coated materials. Based largely on structural analyses of the ligand binding domain of SdrG as an apo-protein and in complex with a Fg-like peptide, we proposed that SdrG follows a "dock, lock, and latch" mechanism to bind to Fg. This binding mechanism involves the docking of the ligand in a pocket formed between two SdrG subdomains followed by the movement of a C-terminal extension of one subdomain to cover the ligand and to insert and complement a β-sheet in a neighboring subdomain. These proposed events result in a greatly stabilized closed conformation of the MSCRAMM-ligand complex. In this report, we describe a biochemical analysis of the proposed conformational changes that SdrG undergoes upon binding to its ligand. We have introduced disulfide bonds into SdrG to stabilize the open and closed forms of the apo-form of the MSCRAMM. We show that the stabilized closed form does not bind to the ligand and that binding can be restored in the presence of reducing agents such as dithiothreitol. We have also used Förster resonance energy transfer to dynamically show the conformational changes of SdrG upon binding to its ligand. Finally, we have used isothermic calorimetry to determine that hydrophobic interactions between the ligand and the protein are responsible for re-directing the C-terminal extension of the second subdomain required for triggering the β-strand complementation event. The attachment of microbes to host tissues represents the initial step in the pathogenesis of most infections, with specificity and tissue tropism being defined by precise adhesin-ligand interactions. For bacteria that are not obligatory intracellular pathogens, extracellular matrix proteins appear to be preferred targets for bacterial adhesion. The adhesins mediating these interactions have been termed MSCRAMMs (microbial surface components recognizing adhesive matrix molecules). 2The abbreviations used are:MSCRAMMsmicrobial surface components recognizing adhesive matrix moleculesMES4-morpholineethanesulfonic acidPIPES1,4-piperazinediethanesulfonic acidDTTdithiothreitolFRETFörster resonance energy transferFgfibrinogen Multiple MSCRAMMs have been found in many organisms. On Gram-positive bacteria, a family of MSCRAMMs might have arisen from a common ancestor, as indicated by their amino acid sequence relatedness, similar modular design, and common binding domain organization. In general, these structurally related MSCRAMMs (Fig. 1A) contain an N-terminal ∼40 amino acid long signal sequence (S), and C-terminal features that are required for sorting the proteins to the cell wall including a proline-rich wall-spanning region (W), the wall anchoring LPXTG motif, a hydrophobic transmembrane region (M), and a positively charged cytoplasmic tail (C). The ligand binding activity of most of these MSCRAMMs is localized to the N-terminal A regions, which in the staphylococcal proteins are ∼500 amino acids long. microbial surface components recognizing adhesive matrix molecules 4-morpholineethanesulfonic acid 1,4-piperazinediethanesulfonic acid dithiothreitol Förster resonance energy transfer fibrinogen The blood plasma protein fibrinogen is targeted by many MSCRAMMs. In the cases of Staphylococcus epidermidis and Staphylococcus aureus, fibrinogen-binding proteins that share a common structure have been reported to mediate bacterial attachment to fibrinogen-covered surfaces (1McDevitt D. Francois P. Vaudaux P. Foster T.J. Mol. Microbiol. 1995; 16: 895-907Crossref PubMed Scopus (141) Google Scholar, 2Deivanayagam C.C. Wann E.R. Chen W. Carson M. Rajashankar K.R. Höök M. Narayana S.V. EMBO J. 2002; 21: 6660-6672Crossref PubMed Scopus (140) Google Scholar, 3Wann E.R. Gurusiddappa S. Höök M. J. Biol. Chem. 2000; 275: 13863-13871Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). SdrG from S. epidermidis is necessary and sufficient for the attachment of this bacterium to surfaces coated with fibrinogen (4Hartford O.M. O'Brien L. Schofield K. Wells J. Foster T.J. Microbiology. 2001; 147: 2545-2552Crossref PubMed Scopus (108) Google Scholar). Furthermore, SdrG is expressed on the surface of S. epidermidis during infection, as indicated by an increase of anti-SdrG titers in sera from patients infected with S. epidermidis (5McCrea K.W. Hartford O. Davis S. Eidhin D.N. Lina G. Speziale P. Höök M. Microbiology. 2000; 146: 1535-1546Crossref PubMed Scopus (133) Google Scholar). SdrG binds to a site at the N terminus of the fibrinogen β-chain that covers the thrombin cleavage site and the MSCRAMM has anticoagulant properties (6Davis S.L. Gurusiddappa S. McCrea K.W. Perkins S. Hook M. J. Biol. Chem. 2001; 276: 27799-27805Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar). ClfA appears to be the major fibrinogen binding MSCRAMM on S. aureus; this adhesin has been shown to act as a virulence factor in animal models of staphylococcal-induced endocarditis (7Que Y.-A. Haefliger J.-A. Piroth L. Francois P. Widmer E. Entenza J.M. Sinha B. Herrmann M. Francioli P. Vaudaux P. Moreillon P. J. Exp. Med. 2005; 201: 1627-1635Crossref PubMed Scopus (243) Google Scholar), septicemia, and arthritis (8Josefsson E. Hartford O. O'Brien L. Patti J.M. Foster T. J. Infect. Dis. 2001; 184: 1572-1580Crossref PubMed Scopus (243) Google Scholar, 9Palmqvist N. Foster T. Fitzgerald J.R. Josefsson E. Tarkowski A. J. Infect. Dis. 2005; 191: 791-798Crossref PubMed Scopus (49) Google Scholar). In addition to ClfA, the S. aureus MSCRAMMs ClfB and FnbpA also bind to fibrinogen (3Wann E.R. Gurusiddappa S. Höök M. J. Biol. Chem. 2000; 275: 13863-13871Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar, 10Ní Eidhin D. Perkins S. Francois P. Vaudaux P. Höök M. Foster T.J. Mol. Microbiol. 1998; 30: 245-257Crossref PubMed Scopus (321) Google Scholar). ClfA and FnbpA bind to similar sites located at the C terminus of the fibrinogen γ chain (3Wann E.R. Gurusiddappa S. Höök M. J. Biol. Chem. 2000; 275: 13863-13871Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar), whereas ClfB has a more complex binding specificity (11Perkins S. Walsh E.J. Deivanayagam C.C. Narayana S.V. Foster T.J. Höök M. J. Biol. Chem. 2001; 276: 44721-44728Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar). The A-regions of the four fibrinogen-binding staphylococcal MSCRAMMs are very similar. Each A-region is predicted to contain three subdomains with IgG-like folds (12Ponnuraj K.B. Bowden M.G. Davis S. Gurusiddappa S. Moore D. Xu Y. Hook M. Narayana S.V. Cell. 2003; 115: 217-228Abstract Full Text Full Text PDF PubMed Scopus (235) Google Scholar). Fibrinogen binding involves subdomains two and three. For SdrG, ClfA, and FnbpA, synthetic peptides have been described corresponding to the respective binding sites in fibrinogen; these peptides bind to the respective MSCRAMMs with relatively high affinity. We reported the crystal structures of the ligand binding region of SdrG as an apo-protein and in complex with a synthetic peptide ligand (12Ponnuraj K.B. Bowden M.G. Davis S. Gurusiddappa S. Moore D. Xu Y. Hook M. Narayana S.V. Cell. 2003; 115: 217-228Abstract Full Text Full Text PDF PubMed Scopus (235) Google Scholar) (Fig. 1, A and B). The apo-protein folds as an open structure where a ligand binding trench formed between two subdomains (N2 and N3) is accessible for the docking of the peptide. The protein-peptide complex structure has a closed conformation where the ligand binding trench is no longer exposed. Based largely on these structures we proposed a dynamic multistep ligand binding mechanism that we called "dock, lock and latch." We envisioned that the ligand peptide would first dock in the binding trench exposed in the open conformation. The peptide-MSCRAMM interaction would then induce a redirection of the C-terminal extension of the second subdomain (N3). This segment covers the bound peptide, effectively locking the ligand in place. We proposed a final latching event, where a short segment C-terminal of the locking residues would insert through a β-strand complementation in a trench on the surface of the N2 subdomain (Fig. 1B). In the present study we have experimentally examined the dock, lock and latch model. Construction and Purification of SdrG Point Mutants—Recombinant SdrG-(273-597) point mutants were generated with hexahistidine tags at their N termini using the expression vector pQE30 (Qiagen). Each construct was amplified by PCR using SdrG-(273-597) DNA as the template and Pfu (Stratagene) as the polymerase. To obtain the desired point mutations, the primers listed in Table 1 were used. The PCR fragments were purified and cloned into pQE30 at the sites indicated for each primer in Table 1, and the resulting plasmids were transformed into XL1-Blue supercompetent cells (Stratagene). Large-scale expression and preparation of recombinant proteins using HiTrap nickel chelating chromatography were described previously (6Davis S.L. Gurusiddappa S. McCrea K.W. Perkins S. Hook M. J. Biol. Chem. 2001; 276: 27799-27805Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar). Protein concentrations were determined from the absorbance at 280 nm as measured on a Beckman Du-70 UV-visible spectrophotometer. The molar extinction coefficient of the proteins was calculated using the method of Pace et al. (13Pace C.N. Vajdos F. Fee L. Grimsley G. Gray T. Protein Sci. 1995; 4: 2411-2423Crossref PubMed Scopus (3452) Google Scholar). The structural integrity of each SdrG-(273-597) point mutant was verified using circular dichroism (14Sillanpaa J. Xu Y. Nallapareddy S.R. Murray B.E. Hook M. Microbiology. 2004; 150: 2069-2078Crossref PubMed Scopus (73) Google Scholar).TABLE 1Synthetic oligonucleotides used in this studyPrimerSequenceE273F5′-CCCGGATCCGAACAAGGTTCGAATGTTAATE381C forward5′-GTAGATAAATATTGCAATATTAAAGCGE381C reverse5′-CGCTTTAATATTGCAATATTTATCTACS483C forward5′-ATTATCGACGATTGTACAATCATTAAAS483C reverse5′-TTTAATGATTGTACAATCGTCGATAATT580P reverse5′-TTTGGTACCTCATTTTTCAGGAGGCAAGTCACCTT GTCCTTGACCTGAACTTGTAGAGAAAGCAATCGGATTATCATAAGI581P reverse5′-TTTGGTACCTCATTTTTCAGGAGGCAAGTCACCTTGTCCTTGA CCTGAACTTGTAGAGAAAGCCGGTGTATTATCATAAGA582P reverse5′-TTTGGTACCTCATTTTTCAGGAGGCAAGTCACCTTGTCCTTGA CCTGAACTTGTAGAGAACGGAATTGTATTATCATAAGI581P,A582P reverse5′-TTTGGTACCTCATTTTTCAGGAGGCAAGTCACCTTGTCCTTGA CCTGAACTTGTAGAGAACGGCGGGTATTATCATAAGF583A forward5′-ACAATTGCTGCCTCTACAAGTF583A reverse5′-ACTTGTAGAGGCAGCAATTGTF583W forward5′-ACAATTGCTTGGTCTACAAGTF583W reverse5′-ACTTGTAGACCAAGCAATTGTP595W reverse5′-CCCGGTACCTCATTTTTCAGGCCACAAGTCACCTTGP595C reverse5′-CCCGGTACCTCATTTTTCAGGACACAAGTCACCTTG Open table in a new tab Crystallization and Data Collection—SdrG crystals were obtained by the hanging drop method using 50% PEG 2000 mono methyl ether, 0.2 m NaCl, and 50 mm buffer (Tris, HEPES, MES, PIPES, and cacodylate) between pH 5.5 and 7.5. The crystals belong to the orthorhombic space group and diffracted to 2.8 Å. The x-ray diffraction data were collected at 100 K using an in-house x-ray source. The DENZO/SCALEPACK package (15Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38573) Google Scholar) was used for diffraction data processing. The structure of the disulfide bridge closed conformation of SdrG was solved by molecular replacement using AMORE (16Navaza J. Acta Crystallogr. Sect. D Biol. Crystallogr. 2001; 57: 1367-1372Crossref PubMed Scopus (658) Google Scholar), refined with CNS (17Brünger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Grosse-Kunstleve R.W. Jiang J.S. Kuszewski J. Nilges M. Pannu N.S. Read R.J. Rice L.M. Simonson T. Warren G.L. Acta Crystallogr. Sect. D Biol. Sci. 1998; 54: 905-921Crossref PubMed Scopus (16967) Google Scholar), and the model built with QUANTA. The crystal structure of the SdrG-peptide complex (without the bound peptide) (12Ponnuraj K.B. Bowden M.G. Davis S. Gurusiddappa S. Moore D. Xu Y. Hook M. Narayana S.V. Cell. 2003; 115: 217-228Abstract Full Text Full Text PDF PubMed Scopus (235) Google Scholar) was used as the starting model. Synthetic Peptides—The synthesis and purification procedures of the peptides were described previously (6Davis S.L. Gurusiddappa S. McCrea K.W. Perkins S. Hook M. J. Biol. Chem. 2001; 276: 27799-27805Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar). For the following peptides the amino acid residue numbers are given and the sequence follows: peptide β6-20 (NEEGFFSARGHRPLD), peptide β6-20(F3) (NEEGFFFSARGHRPLD), and a scrambled version of peptide β6-20 (GADSHEPFRDNERFL). Isothermal Titration Calorimetry—Isothermal titration calorimetry experiments were carried out with a VP-isothermal titration calorimetry microcalorimeter (MicroCal Inc., Northampton, MA) at 25 °C. In a typical experiment the cell contained 20 μm rSdrG and the syringe contained 200 μm peptide. Both solutions were in HBS buffer (10 mm HEPES, 150 mm NaCl, pH 7.4) and were degassed at 25 °C for 20-30 min. The titration was performed as follows: one preliminary injection of 5 μl and 29 injections of 10 μl with an injection speed of 0.5 μl/s. The stirring speed was 540 rpm and the delay time between the injections was 5 min. To take into account the heats of dilution, two blank titrations were performed: one injecting peptide into buffer and another injecting buffer into the rSdrG solution. The averaged heats of dilution were subtracted from the main experiment. Data were analyzed using MicroCal Origin software (version 5.0) and fitting them to a single binding site model. Fluorescence Polarization—Fluorescence polarization was used to determine the equilibrium dissociation constant (KD) for the interaction of rSdrG (6Davis S.L. Gurusiddappa S. McCrea K.W. Perkins S. Hook M. J. Biol. Chem. 2001; 276: 27799-27805Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar) with peptide β6-20. The peptide was labeled with fluorescein, and increasing concentrations of rSdrG in PBS, pH 7.5, were incubated with 10 nm labeled peptide for 3 h in the dark at room temperature. The reactions were allowed to reach equilibrium. Polarization measurements were taken with a Luminescence Spectrometer LS50B using FL WinLab software (PerkinElmer Life Sciences). The binding data were analyzed by nonlinear regression used to fit a binding function as defined by the following equation: Δp =ΔPmax(protein)/KD + (protein), where ΔP corresponds to the change in fluorescence polarization, ΔPmax is the maximum change in fluorescence polarization, and KD is the equilibrium dissociation constant of the interaction. A single binding site was assumed in this analysis. SdrG Fluorescent Labeling—SdrG(E381C,P595W) and SdrG(S483C,P595W) (120 nm) were incubated with dithiothreitol (DTT) (20 mm, final concentration) and in HBS-EDTA (50 mm Hepes, 100 mm NaCl, 1 mm EDTA) at 26 °C. The proteins were then run through a Sephadex G-25 column to remove the excess DTT. The protein concentration was calculated, and was then incubated with a 10-fold concentration of IAEDANS (Molecular Probes) solubilized in Tris-HCl (50 mm, pH 8.0) for 1 h at 26 °C, shaking. The labeling reaction was terminated adding DTT to a final concentration of 5 mm, and free dye was removed by gel filtration through a Sephadex G-25 column equilibrated with HBS-EDTA buffer. The eluted protein concentration was quantitated reading the SdrG absorbance at 280 nm (extinction coefficient Σ= 37,015 m-1 cm-1), and the concentration of the dye was determined measuring the absorbance of the AEDANS at 336 nm (Σ= 5,700 m-1 cm-1). Labeling stoichiometries were 0.80-0.89 mol of AEDANS/mol of SdrG. The AEDANS label was assumed to bind to the thiol groups in S381C and E483C, this was confirmed by the inability to label the wild-type version of SdrG, which has no C residues in its amino acid sequence. Energy Transfer Measurements—Förster resonance energy transfer (FRET) was measured using an SLM 8100 photon-counting spectrofluorometer (SLM-Aminco, Rochester, NY) and repeated in a PerkinElmer Luminescence Spectrometer LS 50 B (PerkinElmer Life Sciences). All samples were examined at room temperature (20-23 °C) in HBS-EDTA. The excitation and emission wavelengths were 295 and 348 nm for Trp and 295 and 515 nm for IAEDANS, respectively, with 4-nm band pass widths. Shutters were closed except during measurements to minimize photodegradation of the sample. The spectral overlap integral, quantum yield, and R0 were determined as described (18Wu P. Brand L. Anal. Biochem. 1994; 218: 1-13Crossref PubMed Scopus (1148) Google Scholar). Four samples were prepared in parallel for each energy transfer experiment: microcells D (donor containing, 300 nm SdrG unlabeled) and DA (containing donor and acceptor, 300 nm SdrG labeled) received 3 μm (final concentration) of ligand peptide, whereas microcells B (blank) and P (3 μm peptide) contained HBS-EDTA buffer alone (B) or buffer plus peptide (P). The net emission intensity of a sample was obtained by subtracting the signal of the background B from D and peptide P from DA, and then correcting all volumes for the protein dilution after the addition of peptide. The donor-acceptor distance calculation is given by r = (E-1 - 1)1/6R0, where E = 1 - ((FDA - FD(1 - fA))/FDfA) and R0 for Trp as a donor and IAE-DANS as an acceptor is 22 Å. FDA = fluorescence of sample DA, FD = fluorescence of sample D, and fA) = fractional occupancy of the acceptor site (given by the labeling efficiency). The distance between residues was calculated using the κ2 = 2/3 value, assuming a random orientation. For the peptide titration experiments, 500 nm SdrG was mixed with increasing concentrations of peptide (0-10 μm), and the fluorescent emission of Trp (at 348 nm) and IAEDANS (at 515) is given by F=Fcorr-Fpcorr, where Fscorr=(Fs(V/V0)-Fb(V/V0)) and Fpcorr=(Fp(V/V0)-Fb(V/V0)); V = volume at the point of titration, V0 = initial volume, F = fluorescence intensity sample, Fp = fluorescence intensity of the peptide, Fb = fluorescence intensity of the buffer, and F = fluorescence of the sample minus the contribution of the peptide in the buffered solution. SdrG-(273-597) Can Be Constrained into an Open or Closed Conformation—A key feature of the dock, lock and latch model of ligand binding is the transition of the MSCRAMM from an open apo-protein form to the closed form of the MSCRAMM peptide complex (Fig. 1B). To explore if these proposed dynamics are critical for ligand binding, we decided to construct versions of the MSCRAMM that are constrained in one or the other conformation and test their ability to participate in the postulated conformational rearrangements. The strategy we chose for this experiment was to introduce, through site-directed mutagenesis, two Cys residues that could form a disulfide bond and lock the C-terminal extension of N3 on top of the N3 module, preventing the extension from participating in the lock and latch steps. This disulfide bond would constrain the molecule in an open conformation. Alternatively, the Cys residue in the C-terminal strand could be partnered with a Cys residue introduced at the bottom of the N2 domain, hopefully locking the latch segment in the latching trench and stabilizing a closed conformation. Careful examination of the solved open and closed crystal structures allowed us to identify candidate residues to be replaced with Cys and subsequently form disulfide bonds. SdrG(S483C,P595C) and SdrG(E381C,P595C) are predicted to form an open and a closed conformation, respectively, stabilized by their individually introduced disulfide bonds (Fig. 1C, Table 1). The designed mutated forms of SdrG were constructed, expressed, and purified. Analyses of the two Cys mutants and wild-type proteins by SDS-PAGE revealed that when the protein samples were treated with DTT as a reducing agent, all the proteins appeared as a single band with very similar migration behavior (Fig. 2A) demonstrating their purity. In the absence of reducing agent, the wild-type protein still migrated as a single band at the distance seen for the reduced protein form. However, a major portion of both SdrG(E381C,P595C) and SdrG(S483C,P595C) protein migrated significantly faster than the wild-type protein under non-reducing conditions. This behavior could indicate that the formation of disulfide bonds stabilizes a compact form of the protein. Furthermore, under nonreducing conditions, both protein mutant preparations contain several higher molecular weight species; these presumably represent dimers and oligomers. These higher molecular forms were confirmed to be SdrG in Western blots, using anti-SdrG antibodies (not shown). We further purified the compact form of SdrG(E381C,P595C) and were able to crystallize it. The structure of SdrG(E381C,P595C) was solved by molecular replacement methods using the structure of the rSdrG-(276-596) component of the peptide-protein complex as a starting model. The SdrG(E381C,P595C) structure was solved at a 2.8-Å resolution, with 2 molecules in an asymmetric unit of a P212121 cell (Table 2). Structurally, SdrG(E381C,P595C) is very similar to the previously reported structure of rSdrG in complex with the ligand peptide, with the exception of the absence of the peptide and the presence of the Cys residues, which are shown to form a disulfide bond (Fig. 1C). Therefore, we concluded that a re-direction of the C-terminal extension of N3 and insertion of the "latch" segment into the latching trench events, that we proposed occurs after ligand binding, can also occur in the absence of ligand bringing residues 381 and 595 close enough to form a disulfide bond.TABLE 2Crystallographic and refinement data of SdrG(E381C, P595C)Dataa61.507 Åb94.172 Åc129.583 Åα90°β90°γ90°Space groupP212121Resolution2.8 ÅCompleteness97.3Rsym6.2 (11.2)%No. molecules in the asymmetric unit2Resolution range20-2.8 ÅRfactor/Rfree (all data with 0 σ cutoff)0.249/0.301No. of non-hydrogen protein atoms4930No. of water molecules0Overall B-factor27.8Ramachandran plotCore94.5%Allowed regions4.5%Generously allowed2.0%Disallowed regions0.0%Protein data base code2RAL Open table in a new tab The Double Cys Mutant Monomers Do Not Bind to Fg, but the Multimers Do—We observed that the Cys mutant proteins appear as a mix of monomers, dimers, and multimers in solution, resulting from intra- or inter-molecular disulfide bond formation. As indicated above, these multiple forms can be detected by SDS-PAGE under non-reducing conditions. The multimeric forms can be reduced to monomers when the proteins are treated with reducing agents such as DTT or β-mercaptoethanol (Fig. 2A). We tested the ability of these multiple forms to bind to soluble Fg when the proteins are immobilized onto a membrane. As shown in the ligand blot in Fig. 2B, the dimers and multimers present in the double Cys mutants preparation bind Fg avidly, whereas the locked monomers do not react to the soluble Fg. Adding DTT to the proteins presumably reduces the intermolecular and intramolecular disulfide bridges, restoring the gel migration characteristics and ability to bind Fg to wild-type levels. To evaluate the ability of the disulfide-stabilized monomers to bind to Fg, it was necessary to obtain a pure monomer preparation, free of any multimeric forms. To separate the monomers from the multimers, we purified the monomer fraction of each double Cys mutant using gel permeation chromatography. After this treatment, the monomer preparation contains a minor percentage of protein that apparently has not formed the predicted disulfide bond, because it still migrates as the wild-type protein in SDS-PAGE. This undesired fraction was eliminated by running the monomer mixture through a column packed with Activated Thiol Sepharose 4B (GE Healthcare), where the proteins that had reduced SH-groups were retained in the column, and the disulfide-bonded forms were washed through. Once pure disulfide-bonded monomer preparations were obtained, their ability to bind to Fg and the β6-20 peptide was tested. The Constrained Conformations Do Not Bind to Fibrinogen—Binding of the locked SdrG monomers to immobilized Fg was initially measured in enzyme-linked immunosorbent-type assays (not shown). Disulfide-stabilized forms of SdrG do not bind to immobilized Fg, regardless of whether these are locked in an open or closed conformation. However, the ability of the disulfide-stabilized proteins to bind to Fg was restored by preincubating the SdrG constructs with DTT. To rule out the possibility that immobilizing Fg on a solid surface could perturb the ability of the SdrG constructs to access the ligand, we examined the ability of a fluorescein-labeled peptide to bind to both forms of SdrG. Using fluorescence polarization, where both the protein and the ligand peptide are in solution, the stabilized closed and open conformations do not bind to the Fg β6-20 peptide (Fig. 2C). The inability of these constructs to bind to the ligand can be explained using our previous genetic and structural data. The constrained open conformation displays binding kinetics similar to those observed in a C-terminal truncation mutant, where the ligand could access the binding trench, but is not stably bound. On the other hand, the constrained closed conformation does not bind the β6-20 peptide presumably because the ligand binding trench is covered by the amino acids that constitute the locking region (581-586), blocking access of the peptide to the binding trench. When the disulfide-bonded proteins were pre-treated with DTT, their ability to bind to the β6-20 peptide was restored, showing KD values similar to that of the wild-type protein (Fig. 2C). These data demonstrate that the capability to freely transition from open to closed conformation of the MSCRAMM regulates the affinity of SdrG for the ligand, and that locking SdrG in either conformation abolishes its ability to bind to Fg. Conformational Changes upon β6-20 Ligand Binding—To probe the relative motion of the C terminus G″ strand between the apo-protein and the protein-peptide complex in solution, we used FRET spectroscopy (19Fisher C.A. Ryan R.O. Eur. J. Lipid Res. 1999; 40: 93-99Abstract Full Text Full Text PDF PubMed Google Scholar). In this assay, the rSdrG C-terminal residue Pro595 was mutated to tryptophan (Trp), which can act as an energy donor. Recombinant SdrG does not contain any other Trp residues. Another strategically positioned residue (either Ser483 or Glu381) was substituted with a Cys and its sulfhydryl moiety was covalently modified with a fluorophore (IAEDANS). If these two moieties are localized within a short distance, energy transfer from the donor (Trp) to the acceptor (AEDANS) results in measurable emission changes. The structural integrity of the point mutants was confirmed by circular dichroism (14Sillanpaa J. Xu Y. Nallapareddy S.R. Murray B.E. Hook M. Microbiology. 2004; 150: 2069-2078Crossref PubMed Scopus (73) Google Scholar), comparing the spectra of the mutant proteins to those of the unmodified SdrG; whereas the binding capability of the labeled proteins was validated in enzyme-linked immunosorbent-type assays. After ensuring that the binding properties of the protein were not compromised by the mutagenesis and/or labeling, the spectral properties of the Trp residue and the AEDANS group were monitored, both before and after the addition of the β6-20 peptide. Introducing the acceptor at the bottom of the latching trench (Glu381) or within the N3 domain (Ser-483) and the donor Trp at the C terminus (Pro595) allowed us to monitor intramolecular movements in the presence or absence of the ligand. These molecular movements can be detected as changes in the relative distances of the C terminus to the N2 or N3 subdomains. If the fluorophore present at the bottom of the N2 latching trench (position 381) is in proximity of the Trp residue at the C terminus (position 595), we can clearly detect energy transfer (Fig. 3A). In this case, when the β6-2