Nucleotide Binding in an Engineered Recombinant Ca2+-ATPase N

Nov 29, 2016 - A recombinant Ca2+-ATPase nucleotide binding domain (N-domain) harboring the mutations Trp552Leu and Tyr587Trp was expressed and purifi...
1 downloads 7 Views 6MB Size
Download PDF
Article pubs.acs.org/biochemistry

Nucleotide Binding in an Engineered Recombinant Ca2+-ATPase N‑Domain Edgar D. Páez-Pérez, Valentín De La Cruz-Torres, and José G. Sampedro* Instituto de Física, Universidad Autónoma de San Luis Potosí, Manuel Nava 6, Zona Universitaria, CP, 78290 San Luis Potosí, SLP, Mexico S Supporting Information *

ABSTRACT: A recombinant Ca2+-ATPase nucleotide binding domain (N-domain) harboring the mutations Trp552Leu and Tyr587Trp was expressed and purified. Chemical modification by N-bromosuccinimide and fluorescence quenching by acrylamide showed that the displaced Trp residue was located at the N-domain surface and slightly exposed to solvent. Guanidine hydrochloride-mediated N-domain unfolding showed the low structural stability of the α6−loop−α7 motif (the new Trp location) located near the nucleotide binding site. The binding of nucleotides (free and in complex with Mg2+) to the engineered N-domain led to significant intrinsic fluorescence quenching (ΔFmax ∼ 30%) displaying a saturable hyperbolic pattern; the calculated affinities decreased in the following order: ATP > ADP = ADP-Mg2+ > ATP-Mg2+. Interestingly, it was found that Ca2+ binds to the N-domain as monitored by intrinsic fluorescence quenching (ΔFmax ∼ 12%) with a dissociation constant (Kd) of 50 μM. Notably, the presence of Ca2+ (200 μM) increased the ATP and ADP affinity but favored the binding of ATP over that of ADP. In addition, binding of ATP to the N-domain generated slight changes in secondary structure as evidenced by circular dichroism spectral changes. Molecular docking of ATP to the N-domain provided different binding modes that potentially might be the binding stages prior to γphosphate transfer. Finally, the nucleotide binding site was studied by fluorescein isothiocyanate labeling and molecular docking. The N-domain of Ca2+-ATPase performs structural dynamics upon Ca2+ and nucleotide binding. It is proposed that the increased affinity of the N-domain for ATP mediated by Ca2+ binding may be involved in Ca2+-ATPase activation under normal physiological conditions.

T

It is known that phosphorylation and Ca2+ and ATP binding generate long-range effects in the topology of SR Ca2+-ATPase as a whole.6 The mechanism of translocation of Ca2+ into the SR lumen involves large conformational rearrangements among the A-domain, the N-domain, and the P-domain, most of them linked to movements in the TM α-helices.5,7 The determined three-dimensional structures of the Ca2+ATPase have provided valuable information about the Ca2+ATPase in most catalytic steps.8−10 However, as ATP hydrolysis and Ca2+ transport occur in distant sites of the protein, questions remain about the following: (a) the precise mechanism of energy conversion (from chemical to potential), (b) the detailed events taking place through the transitions between the different structural states in the ATPase cycle, (c) the specific sequence of events generated by Ca2+ and nucleotide binding, (d) the role of each of the two Ca2+ binding events in the conformational changes, (e) the sections of the protein displaying significant changes in the transitions,

he sarcoplasmic/endoplasmic reticulum (SR) Ca2+ATPase (EC 3.6.3.8, AT2A1_Rabbit) accounts for 70− 80% of total protein in SR membranes and is one of the main proteins involved in the control of Ca2+ levels in muscle cells, which allows their relaxation.1 The SR Ca2+-ATPase couples ATP hydrolysis to the transport of two Ca2+ ions across the SR membrane. The SR Ca2+-ATPase belongs to the P-type ATPase family, whose members are characterized by the transitory formation of a phosphorylated enzyme intermediate.1 The SR Ca2+-ATPase has a molecular mass of ∼110 kDa (994 amino acid residues) and structurally consists of a large cytoplasmic region composed of three domains (P-domain for phosphorylation, N-domain for nucleotide binding, and A-domain for actuator) and is linked to the SR membrane by 10 transmembrane α-helices (TM domain).1 Since the discovery of the SR Ca2+-ATPase, there has been much interest in its detailed mechanism of function.2 The catalytic cycle involves the formation of a phosphoenzyme intermediate and the interconversion between two main conformations (named E1 and E2) displaying different affinities for Ca2+.3 The TM domain contains the two Ca2+ binding sites located in α-helices M4−M6 and M8.4 There is ample experimental evidence of the conformational changes in the SR Ca2+-ATPase at different steps of the Ca2+ transport cycle.5 © XXXX American Chemical Society

Received: March 1, 2016 Revised: November 11, 2016

A

DOI: 10.1021/acs.biochem.6b00194 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

N-domain fluorescence (∼12%), indicating binding of Ca2+ to the protein, which led to increased ATP and ADP affinities. The physiological significance of Ca2+-mediated ATP increased affinity is discussed. The N-domain α6−loop−α7 motif displayed low stability to Gnd-HCl, which reflected a dynamic structure.22 In addition, circular dichroism (CD) spectroscopic analysis showed that nucleotide binding generates slight modifications of secondary structure, such as an increase in the level of unordered structure, and a decrease in α-helix content. Molecular docking of ATP to the N-domain was performed showing the different binding modes ATP may attain, and the corresponding affinities were calculated. Finally, a fluorescein isothiocyanate (FITC) labeling assay and molecular docking were used to study the binding site formation, stability, interacting amino acids, and affinity.

and (f) the details of intradomain structural changes taking place in the large cytoplasmic portion.4 The ATP binding site is found in the cytoplasm,11,12 and the reported three-dimensional structures of SR Ca2+-ATPase show bound nucleotides (ADP, ATP, and ATP analogues) at different catalytic steps.7 However, information regarding binding site dynamics upon substrate binding and the formation of product complexes is lacking.13 In this regard, tryptophan (Trp) residues have been of assistance in the understanding the SR Ca2+-ATPase function, as both the wavelength of Trp fluorescence emission and the quantum yield are quite sensitive to the local electrostatic environment.14 In fact, it is known that an appropriately placed Trp will undergo fluorescence wavelength and/or intensity changes in any given process a protein performs.15 SR Ca2+-ATPase contains 12 Trp residues in the TM domain and a single Trp residue (Trp552) in the cytoplasmic N-domain.16 The Trp residues in the TM domain have often been used to monitor the binding of the two Ca2+ ions.2,17−19 Trp552 initially was thought to have some potential as a reporter for structural changes in the N-domain. However, it was recently demonstrated that Trp552 does not display any significant change in fluorescence intensity upon nucleotide binding.4 Therefore, the slight fluorescence change (1.5%) observed in the SR Ca2+-ATPase upon nucleotide binding was due to long effects in the environment of Trp(s) in the TM domain.4 SR Ca2+-ATPase intrinsic fluorescence is complex. Nonetheless, it has allowed the study of some structural features and the correlation of conformational states with specific processes.17 Fluorescence decay in the SR Ca2+-ATPase has been described as heterogeneous, showing a continuous distribution of lifetime values18,20 related to the presence of tryptophan residues in very different microenvironments.18 Therefore, the identification of the residues participating in fluorescence changes is difficult to achieve, thus limiting the applications of intrinsic fluorescence. When multiple Trp residues are present, the fluorescence signal reflects the average intensity of all fluorophores and the overall structural fluctuations rather than a specific structural change in a given domain or motif.17 In contrast, information obtained from intrinsic fluorescence measurements may be interpreted easily if a single and properly located Trp is present in the protein. It is not possible to work with native protein samples of the SR Ca2+-ATPase as it lacks a tryptophan sensitive to conformational changes affecting the surrounding microenvironment of the nucleotide binding site.17 However, it was shown recently that in the plasma membrane H+-ATPase (EC 3.6.3.6) (with a properly located Trp residue in the N-domain), nucleotide binding generates large changes in fluorescence intensity, thus allowing the study of nucleotide binding and protein stability.21 In this work, a recombinant SR Ca2+-ATPase N-domain in which Trp552 was mutated to leucine (Leu) and Tyr587 to Trp was constructed. In the wild-type N-domain, Tyr587 is located in an α-helix that moves upon nucleotide binding.5,10 The N-domain mutant (Trp552Leu and Tyr587Trp) was designed to have a Trp residue properly located to study the structural rearrangements accompanying nucleotide binding using intrinsic fluorescence. The purified N-domain displayed large fluorescence changes (∼30%) in the presence of saturating concentrations of nucleotides (ATP and ADP). Notably, nucleotide binding (free and in complex with Mg2+) was hyperbolic, displaying different affinities: ATP > ADP = ADP-Mg2+ > ATP-Mg2+. The presence of Ca2+ quenched the

■

EXPERIMENTAL PROCEDURES Materials. Adenosine triphosphate (ATP), adenosine diphosphate (ADP), guanidine hydrochloride (Gnd-HCl), and N-bromosuccinimide (NBS) were from Sigma-Aldrich Corp. (St. Louis, MO). Acrylamide was from Bio-Rad Laboratories (Hercules, CA). Restriction enzymes (EcoRI HF and BamHI), RNase If, and DNA polymerase (VentR) were from New England Biolabs Ltd. (Hitchin, U.K.). All other reagents were of the best quality available commercially. Synthesis of the Ca2+-ATPase N-Domain Gene. The DNA encoding the Ca2+-ATPase N-domain (residues 360−601 and mutations Trp552Leu and Tyr587Trp) was synthesized de novo by GeneScript (Piscataway, NJ) and cloned in frame into the BamHI and EcoRI restriction sites of vector pET28a(+) DNA-Novagen (Merck-Millipore, Darmstadt, Germany) to generate mutant N-domain expression vector pET28aDMNDCaATPase. N-Domain Expression and Purification. The His-tagged N-domain was expressed in Escherichia coli BL21 Star (DE3) GroEL/GroES as follows. Shaker flasks (2 L) containing 1 L of Luria-Bertani medium (supplemented with 50 μg/mL kanamycin) were inoculated with 1 mL of an overnight preculture. The cells were grown at 37 °C until the absorbance at 600 nm reached a value of 1.0 OD/mL. Protein expression was induced by adding 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and tetracycline (1 μg/mL) following incubation for 16 h at 18 °C. The bacterial cells were harvested by centrifugation at 10000 rpm for 10 min at 4 °C, and the pellets were stored frozen at −72 °C. Protein purification was performed by thawing the pellets and suspending the cells in lysis buffer [0.3 M NaCl, 0.2% IGEPAL CA-630, and 0.05 M sodium phosphate (pH 7.0) supplemented with 0.25 mg/mL lysozyme, 5 mM β-mercaptoethanol, and 1 mM PMSF] and disruption by sonication. Cell debris was removed by centrifugation at 12000 rpm for 40 min at 4 °C. The supernatant was diluted to a 1:10 ratio in binding buffer [50 mM Tris-HCl (pH 8.0) and 300 mM NaCl], and 5 mL aliquots were injected into a 5 mL Ni-Sepharose Fast Flow column (GE Healthcare Life Sciences, Little Chalfont, U.K.). A stepwise increasing elution with imidazole (10, 50, 250, and 500 mM) in buffer [50 mM Tris-HCl (pH 8.0) and 300 mM NaCl] was applied to release the bound protein. The N-domain was desalted and analyzed for protein concentration by the Lowry assay23 (using BSA as the standard) and by absorbance at 280 nm [extinction coefficient (ε) of 11960 M−1 cm−1] with similar results. Protein samples at different purification stages were subjected to sodium dodecyl sulfate−polyacrylamide gel B

DOI: 10.1021/acs.biochem.6b00194 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

fluorescence quenching in the N-domain was evaluated by titration with each nucleotide as described above, including 200 μM CaCl2 in the suspension buffer. The experiments were performed three times. Circular Dichroism (CD) Spectroscopy. The purified Ndomain (5 μM) was suspended in 450 μL of 10 mM phosphate (pH 8.0) at 25 °C. ATP aliquots (1−2 μL) were added stepwise from an ATP stock solution (30 mM) to obtain the indicated concentrations. Far-UV CD spectra (185−260 nm) were recorded at 20 nm/min using a Jasco J1500 spectropolarimeter and a 0.1 cm path length cell at 25 °C. The data interval resolution was 0.1 nm, and the bandwidth was set to 1 nm. The CD spectra of the N-domain were analyzed for fractional secondary structure content by using the CDPro software package (http://sites.bmb.colostate.edu/sreeram/ CDPro/).26 A low root-mean-square deviation (RMSD) was considered in the selection of the analysis method and reference data set. N-Domain Labeling with Fluorescein Isothiocyanate (FITC). The N-domain (5 μM) was incubated for 10 min at room temperature in the presence of 10 μM FITC. When indicated, 1% SDS and 5 mM ATP were included in the reaction mixture. Samples were subjected to SDS−PAGE and photodocumented using a UV transilluminator and a green filter. The image was further processed by converting it to black and white and producing the negative for densitometric analysis using ImageJ. Three-Dimensional Structural Modeling and Molecular Docking. The three-dimensional (3D) model of the mutant N-domain was generated using the online software IntFold (http://www.reading.ac.uk/bioinf/IntFOLD/).27 The closest structural equivalents available in the Protein Data Bank (PDB) were identified by the TM-align server (http:// zhanglab.ccmb.med.umich.edu/TM-align/). Molecular docking of ATP and FITC to the 3D model of the engineered Ndomain Ca2+-ATPase was performed using the software Autodock Vina essentially as described by Forli et al.28,29 Data Analysis. N-Domain Ca2+-ATPase fluorescence quenching by acrylamide was analyzed using the Stern−Volmer equation (eq 1) as described by Eftink and Ghiron.30,31

electrophoresis (SDS−PAGE) and visualized by Coomassie blue staining. The N-domain was >93% pure as analyzed by densitometry using ImageJ (https://imagej.nih.gov/ij/). The purified N-domain Ca2+-ATPase sample was stored at 4 °C until it was used. N-Domain Fluorescence Titration with NBS. N-Domain fluorescence was titrated at 25 °C, as described by Peterman and Laidler,24 using a Shimadzu RF 5301 spectrofluorophotometer equipped with a thermostated cell chamber and constant stirring. Briefly, the N-domain (5 μM) was suspended in 2 mL of 50 mM phosphate buffer (pH 8.0) and allowed to equilibrate 20 min until stable fluorescence intensity was observed. Then, protein fluorescence was titrated by stepwise addition of aliquots (1−10 μL) of a 2 mM NBS aqueous stock solution; the NBS reaction occurs in seconds. The N-domain fluorescence emission spectrum (300−400 nm) was recorded (∼3000 nm/min, 1 nm step size) after 5 min by excitation at 295 nm with a 5 nm slit width. After fluorescence correction for dilution and subtraction of the buffer baseline, the wavelength of maximal fluorescence intensity was determined and used for data analysis. The experiments were performed three times. N-Domain Fluorescence Quenching by Acrylamide. N-Domain [5 μM protein suspended in 2 mL of 20 mM phosphate buffer (pH 8.0)] fluorescence was quenched at 25 °C by the stepwise addition of small aliquots (5 μL) of an 8 M acrylamide solution. After equilibrium had been reached (>5 min), the fluorescence emission spectrum (300−400 nm, 1 nm step size) was obtained by excitation at 295 nm with a 5 nm slit width.21 The wavelength of maximal emission intensity was used to analyze the tryptophan exposure in the N-domain. The experiments were performed three times. N-Domain Unfolding by Guanidine Hydrochloride (Gnd-HCl). N-Domain (5 μM) unfolding by guanidinium chloride was performed at 25 °C in 2 mL of 20 mM phosphate buffer (pH 7.8). After the addition of Gnd-HCl from a 8 M stock solution, the N-domain was allowed to equilibrate for 30 min and the fluorescence spectrum (300−400 nm, 1 nm step size) was recorded by excitation at 295 nm with a 5 nm slit width.21 The wavelength of maximal fluorescence intensity was used to analyze the unfolding of the N-domain. The experiments were performed three times. N-Domain Fluorescence Quenching by Nucleotide Binding. The N-domain (5 μM) was suspended in 50 mM phosphate buffer (pH 8.0).25 The steady state fluorescence spectra (300−400 nm, 1 nm step size) were titrated at 25 °C by stepwise addition of the nucleotide (ATP and ADP) from stock solutions (0.02 or 0.2 M nucleotide, accordingly) and excitation at 295 nm with a 5 nm slit width. At this wavelength, the ATP/ ADP inner filter effect was minimal; nonetheless, fluorescence was corrected for both dilution and inner filter effects. The wavelength of maximal fluorescence intensity was used for data analysis. The experiments described above were performed in the presence of 2 mM MgCl2 to test the effect of Mg2+ in nucleotide-mediated fluorescence quenching. The experiments were performed three times. N-Domain Fluorescence Quenching by Ca2+ and Nucleotide Binding. The N-domain (5 μM) was suspended in 50 mM phosphate buffer (pH 8.0). The intrinsic fluorescence spectrum (300−400 nm, 1 nm step size) was titrated at 25 °C by the stepwise addition of CaCl2 and excitation at 295 nm with a 5 nm slit width. Because Ca2+ does not absorb light, only fluorescence was corrected for dilution. The effect of Ca2+ on nucleotide (ATP and ADP)-mediated

F0/F = (1 + K svQ )eVQ

(1)

where F0 and F are the fluorescence intensities at 338 nm in the absence and presence of the quencher (Q), respectively. The static (V) and collisional (Ksv) quenching constants31 were calculated by nonlinear regression using the iterative software Origin 6.0 from Microcal (Northampton, MA). Gnd-HCl-mediated N-domain unfolding was analyzed using the fluorescence quenching data by fitting a nonlinear regression to eq 2 as described by Pace and co-workers.32,33 yobs =

yN + yU e−[ΔG(H2O)N→ U − mN→ U[Gnd‐HCl]/ RT ] 1 + e−[ΔG(H2O)N→ U − mN→ U[Gnd‐HCl]/ RT ]

(2)

where yobs is the observed fluorescence intensity at a given GndHCl concentration [Gnd-HCl], yN and yU are the calculated signals for the native and unfolded states, respectively, ΔG(H2O)N→U is the free energy change for the unfolding (N → U) process, and mN→U is a measure of the dependence of ΔG on Gnd-HCl concentration.32 Nucleotide binding-mediated N-domain fluorescence quenching was fitted by nonlinear regression to eq 3: C

DOI: 10.1021/acs.biochem.6b00194 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Figure 1. Structure of the mutant Ca2+-ATPase N-domain. (A) Secondary structure. The Ca2+-ATPase crystal structure (PDB entry 1SU4) was used for secondary structure generation (http://webclu.bio.wzw.tum.de/cgi-bin/stride/stridecgi.py). The mutations Trp552Leu and Tyr587Trp are underlined. The α-helix is colored blue (3-10 helix) and red; the β-sheet is colored green, and the turn and coil are colored yellow. (B) Threedimensional (3D) structural model of the engineered N-domain. The 3D model was generated using the online software IntFold (http://www. reading.ac.uk/bioinf/IntFOLD/).27 Mutations Trp552Leu and Tyr587Trp are colored green and red, respectively. The Phe487 residue involved in π−π stacking with adenine in the binding site is colored blue.

F0 − F =

ΔFmax[nucleotide] Kd + [nucleotide]

events involved in catalysis.4 Furthermore, Trp552 does not seem to have a specific role in ATP hydrolysis or Ca2+ transport, as Vmax and Km for Ca2+ were normal in the Trp552Phe mutant.4 It was stated, therefore, that Trp552 is not essential for Ca2+-ATPase function.4 In this regard, fluorescence variations observed upon nucleotide binding were in reality caused by variations in the physicochemical environment of Trp residues located in the transmembrane (TM) domain.4 In this work, a double-mutant N-domain was designed in which the Trp552 residue was changed to Leu (Trp552Leu) and the Tyr587 residue to Trp (Tyr587Trp) to move Trp from position 552 to 587 (Figure 1A,B). Mutations were rationally designed on the basis of the report that the plasma membrane H+-ATPase N-domain from Kluyveromyces lactis displays a high fluorescence sensitivity to nucleotide binding and unfolding.21 This high sensitivity was determined to be mainly due to environmental changes of the single Trp505 residue during both processes.21 The amino acid sequences of both Ca2+- and H+-ATPase N-domains have been aligned, and the position corresponding to Trp505 in H+-ATPase was identified to be that of Tyr587 in the Ca2+-ATPase.38

(3)

where F0 is the fluorescence in the absence of a nucleotide, F is the fluorescence at a given nucleotide concentration, ΔFmax is the maximal change in fluorescence when the binding site is saturated, and Kd is the dissociation constant for dissociation of the nucleotide from the binding site.

■

RESULTS AND DISCUSSION Design of the Double Mutant in the N-Domain of Ca2+-ATPase. In the SR Ca2+-ATPase, intrinsic fluorescence studies show that in the absence of Ca2+, the binding of nucleotides (ATP or ADP) generates a slight increase in fluorescence intensity, which decreases in the presence of Ca2+.2,25,34,35 Fluorescence changes upon nucleotide binding have been largely related to phosphorylation of the Ca2+ATPase at the P-domain.35−37 The single tryptophan residue (Trp552) located in the nucleotide binding domain (Ndomain) was proposed to contribute to such fluorescence changes.37 Recently, Trp552 fluorescence was found to be insensitive to both nucleotide binding and other structural D

DOI: 10.1021/acs.biochem.6b00194 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

is used, guided by local and global quality valuations.27 The generated 3D model showed high structural equivalence to the Ca2+-ATPase in complex with ATP and thapsigargin (PDB entry 3AR4; RMSD = 0.5 Å) and to the Ca2+-ATPase in the absence of Ca2+ (PDB entry 1IWO; RMSD = 0.8 Å). The in silico results showed preliminarily that the Trp552Leu and Tyr587Trp mutations did not disturb the overall folding of the N-domain. Chemical Modification of Tyr587Trp with N-Bromosuccinimide (NBS). In the 3D model of the mutant Ndomain, Tyr587Trp was located at the protein surface as expected (Figure 1B). To determine the surface location of Tyr587Trp, NBS-mediated chemical modification of Trp was performed in the purified N-domain and monitored by following variations in fluorescence intensity. The N-domain was excited at 295 nm, and fluorescence emission spectra were recorded between 300 and 400 nm. The maximal fluorescence intensity was at 338 nm, which was the same as that reported for the wild-type recombinant N-domain.6 In contrast, the wildtype N-domain obtained upon treatment of SR Ca2+-ATPase with proteinase K had a maximal fluorescence intensity at 334 nm.16 The intrinsic fluorescence of the N-domain decreased when NBS was added, and at a NBS/N-domain molar ratio of 3, a dramatic decrease in intrinsic fluorescence was observed (55.0 ± 2.6%) (Figure 3A). Also, addition of NBS led to further

The synthetic gene DMNDCaATP encoding the Ca2+ATPase N-domain and containing the desired nucleotide mutations was generated by GeneScript (Piscataway, NJ), verified for correct nucleotide sequence, and cloned in frame at BamHI and EcoRI restriction sites in expression plasmid pET28a(+). The synthetic gene encoded 242 amino acid residues from Gln360 to Asp601, corresponding to the entire N-domain of the SR Ca2+-ATPase. A similar approach was used by Moutin et al.; however, they expressed in E. coli an SR Ca2+ATPase segment that they named “the large cytoplasmic loop” (LCL), which included the N-domain and a large segment of the P-domain.39 In contrast, Abu-Abed et al. expressed the wildtype Ca2+-ATPase N-domain segment of residues Thr357− Leu600 (without mutations) for spectroscopic and nuclear magnetic resonance (NMR) studies.6,40 N-Domain Expression and Purification. N-Domain expression in E. coli BL21 (DE3) resulted in a high protein yield but in an aggregated state; the presence of inclusion bodies could be easily observed in the cellular homogenate (results not shown). A similar result was reported for the expression of the LCL by Moutin et al.39 However, those authors subjected the inclusion bodies to treatment with 4 M urea and recovered a small fraction of the protein in soluble form, which was used for fluorescence assays.39 In contrast, Abu-Abed did not report any issues surrounding N-domain folding.40 Therefore, we decided to express the N-domain in E. coli strain BL21 Star (DE3) GroEL/ES. The co-expression with chaperones (GroEL/ES) resulted in a completely soluble (folded) protein that was easily purified to homogeneity by affinity chromatography using Ni-sepharose and observed in an SDS−PAGE gel (Figure 2). Three-Dimensional Model of the Engineered NDomain. The three-dimensional (3D) molecular model of the mutant N-domain (Figure 1B) was generated using the online software IntFold (http://www.reading.ac.uk/bioinf/ IntFOLD/)27 in which a multiple-template modeling approach

Figure 3. Chemical modification of the N-domain Trp by Nbromosuccinimide (NBS). (A) Titration of N-domain fluorescence with NBS. Fluorescence emission spectra were recorded via excitation at 295 nm. (B) Plot of fluorescence quenching vs the NBS/N-domain molar ratio. Experiments were performed three times. N-Domain fluorescence intensity at 338 nm was used for data analysis; the standard deviation (SD) of fluorescence intensity was