Interactions between NNM and TMV CP

NNM and TMV CP binding was initiated by adding 5 mM anti-TMV compounds to 0.5 mM (8.7 mg/mL) TMV CP discs and incubating for 1 h. Then, 0–10 mM NNM was added to the reactions for 1 h. The oligomer status of TMV was then analyzed by SEC [6]. SEC was performed at room temperature using a calibrated Superdex 200 10/300 GL column (GE Healthcare) attached to an AKTA Purifier Fast Protein Liquid Chromatography system (GE Healthcare). The column was equilibrated with a buffer containing 10 mM sodium phosphate and 100 mM sodium chloride solution (pH 7.2). The molecular mass standards (Bio-Rad Hercules, CA, USA) used included thyroglobulin (669 kDa), ferritin (440 kDa), Bovine Serum Albumin (67 kDa), β-lactoglobulin (35 kDa), ribonuclease A (13.7 kDa), cytochrome (13.6 kDa), aprotinin (6.51 kDa) and vitamin B12 (1.36 kDa). The protein was monitored by measuring the absorbance at a wavelength of 280 nm. For TEM [6, 22, 23], self-assembled TMV CP discs were incubated as described previously. Briefly, 20 μL of the mixed solution was deposited onto a 300-mesh formvar-carbon-coated copper grid for 2 min, followed by rinsing with ddH₂O. The grid was stained with 20 μL of 2% aqueous solution of tungstophosphoric acid for 90 s as a negative stain. Images were obtained at the Electron Microscope Lab of Zunyi Medical University using a Hitachi H-7650 transmission electron microscope (Tokyo, Japan) with 80 kV accelerating voltage. The 17% native PAGE [29-31] was performed on ice with TMV CP samples equilibrated overnight in a buffer containing 10 mM sodium phosphate and 100 mM sodium chloride (pH 7.2). Then, 20 μL of the samples was mixed with 20 μL of 2× loading buffer (12.5% 0.5 M Tris-HCl (v/v) (pH= 8.7), 0.5% bromophenol blue (w/v) and 30% glycerin (v/v). Subsequently, 8 μL of the samples was loaded onto 17% gels. Electrophoresis was performed using a 1× native PAGE buffer (Tris-Gly, pH 8.8) at 0 °C for 1 h. After electrophoresis, gels were stained with Coomassie Brilliant blue to identify proteins, and then destained with methanol and glacial acetic acid. The fluorescence spectra [32] were recorded with a fluorescence spectrophotometer (Varian Cary Eclipse, Palo Alto, CA, USA) at 20°C. The emission spectra of TMV CP discs were obtained in buffer (1.5 mL) with a quartz cuvette having a 1-cm path length. Fluorescence intensities were measured with an excitation wavelength of 278 nm and an emission wavelength of 325 nm. The concentrations of TMV CP discs were defined at 50 nM. NNM was continuously added into the cuvette until the fluorescence signal no longer changed. All measurements were taken in phosphate buffer (10 mM sodium phosphate, 100 mM sodium chloride [pH 7.2]), and fluorescence titration curves were corrected for the background intensity of the buffer. The apparent dissociation constants were analyzed by the nonlinear least-squares curve-fitting method using Origin 7.0 software. The ITC binding experiments [33] were performed using an ITC 200 Micro Calorimeter (GE Healthcare) at 20°C. All recombinant CP proteins and mutants were formed in 10 mM sodium phosphate buffer and 100 mM sodium chloride solution (pH 7.2) at 20°C for more than 12 h, and then recombinant CP four-layer aggregate discs and mutant discs were collected. The buffer contained 10 mM sodium phosphate and 100 mM sodium chloride (pH 7.2). The compounds (0–10 mM) were titrated into TMV CP discs and mutants (0.5 mM) in a 200 μL sample cell using a 40-μL microsyringe as follows: 0.4 μL for the first injection and 2 μL for the next 19 injections at intervals of 150 s. The integrated heat data were analyzed using the one-set-of-sites model in MicroCal Origin 7.0 according to the manufacturer’s instructions. The first data point was not used in analysis. The binding parameters reaction enthalpy change in cal·mol⁻¹ (ΔH), binding constant in mol⁻¹ (K) and the number of molecules per TMV CP proteins (n) were floating during the fit. The binding free energy, ΔG, and reaction entropy, ΔS, were calculated using the equations, ΔG = −RTlnK (R = 1.9872 cal·mol⁻¹·K⁻¹, T = 298 K) and ΔG = ΔH−TΔS. The Kd was calculated as 1/K. Then, the binding was calculated for MST [34, 35] Monolith NT. 115 (Nano Temper Technologies, Munchen, Germany). A range of ligands from 0 μM to 5 μM were incubated with 0.5 μM of purified recombinant proteins for 5 min with a NT-647 dye (Nano Temper Technologies). These were used in the thermophoresis experiment at a final concentration of ∼20 nM. A 16 point dilution series was made for selected compounds in dimethyl sulfoxide. Each compound’s dilution series was subsequently transferred to protein solutions in 10 mM Tris/HCl and 100 mM sodium chloride, pH 7.4, containing 0.05% Tween-20. After a 15 min incubation of the labeled CP with each dilution point (1:1 mix) at room temperature, samples were placed into standard capillaries (NanoTemper Technologies). Measurements were taken on a Monolith NT.115 MST (NanoTemper Technologies) under the setting of 20% LED and 40% IR laser. The laser’s on time was set at 30 s, and the laser’s off time was set at 5 s. The Kd values were calculated from the duplicate reads of three separate experiments using the mass action equation in the Nano Temper software. For western blot [36], an electrotransfer system (Bio-Rad) was used. Growing leaves of N. tabacum cv. K₃₂₆ were mechanically inoculated with equal volumes of TMV (0.5 mg/mL). After 72 h, 1-cm diameter leaf discs were removed. The leaf discs were floated on solutions of NNM and on buffer (10 mM sodium phosphate and 100 mM sodium chloride solution, pH 7.2) as a negative control. Discs of healthy leaves were floated on buffer as a mock. All of the leaf discs were kept in a culture chamber at 28°C for 48 h, and then, the TMV concentration in the leaf disc was determined. Leaf discs were ground in 5 × protein loading buffer (10% SDS, 5% β-ME, 50% glycerin, 0.5% bromophenol blue, and 250 mM Tris-HCl, pH 6.8), and then, 5 μL of sample were loaded on a polyacrylamide gel (5% stacking gel and 12% separating gel). After SDS-PAGE, TMV protein bands were transferred at 90 mA for 1 h onto a polyvinylidene fluoride membrane (0.46 μm, washed with methanol to activate) using an electrotransfer system (Bio-Rad). The membrane was washed in TBST (20 mM Tris-HCl, pH 8.0; 150 mM NaCl; and 0.05% Tween-20) and blocked with 5% nonfat milk powder in TBST for 1 h at 37°C. The membrane was washed three times, each time for 3 min with TBST, and reacted with a mixture of 1:30,000 alkaline phosphadase-conjugated anti-rabbit IgG (Sigma, Deisenhofen, Germany) and 1:200 polyclonal antibody of TMV for 2 h at 37°C. After it was washed three times, each time for 3 min with TBST, the membrane was incubated in substrate buffer (12.1 g Tris-HCl, pH 9.5; 5.84 g NaCl; 10.2 g MgCl₂; and 800 mL H₂O) with 330 μL/mL nitrotetrazolium blue chloride and 165 μL/mL 5-bromo-4-chloro-3-indolyl phosphate for 3–5 min in the dark until the bands of the CP were clear. For the MD simulation, the X-ray crystal structure of TMV CP was downloaded from (Protein Data Bank) PDB (PDB ID: 4GQH). The initial structure was revised by adding lost residues and hydrogen atoms and checking bonds and bumps. Subsequently, the energy was minimized for 2,000 steps of the steepest descent calculations and 2,000 steps of conjugated gradient calculations by using Sybyl 7.0 (Tripos Inc., St. Louis, MO, USA) and Gaussian03 program at the HF/6-31+G* level [37]. The optimized geometries were used to construct the entire structures. The final structures of different conformations were optimized by fixing the macrocycle with a conjugated gradient in Sybyl 7.0. The different conformations were used as the starting structures for docking studies. Docking calculations were performed on these conformations with AutoDock4.0. The protein and ligand structures were prepared with Autodock Tools [38]. The atomic Gasteiger–Huckel charges were assigned to the ligand and receptor. Most of the parameters for the docking calculation were set to the default values recommended by the software. Each docked structure was scored by the built-in scoring function and was clustered by 0.8 Å of (root-mean-square deviation) RMSD criteria. For each binding model, the molecular mechanics/Poisson–Boltzmann surface area was calculated. Before this calculation, the complex structure was further refined initially with the steepest descent algorithm, followed by the conjugated gradient algorithm using the Amber9 package [39]. During the energy minimization process, the receptor was first fixed, and only the ligand remained free. Then, the ligand and residue side chains remained free. Finally, all atoms of the system were liberated and refined to a convergence of 0.01 kcal/(mol·Å).

Based on the docking results, two binding models were selected for MD simulation. Prior to the simulation, the electrostatic potential and partial atomic charges were determined by performing an electrostatic potential fitting according to the Merz–Singh–Kollman scheme with the Gaussian-optimized geometries [40, 41]. The rescaled electrostatic potential charges of the ligand were produced using the standard protocol implemented in the antechamber module of the Amber9 program [38, 42, 43]. The system was solvated in an octahedral box of TIP3P water, in which crystallographic water molecules were maintained. The edge of the box was at least 10 Å from the solute. Appropriate sodium counterions were added to the system to preserve neutrality. The solvated system belonged to the solute. Before the MD simulation, some energy minimization steps were applied to the system. First, the solute was kept fixed with a constraint of 500 kcal mol⁻¹Å⁻². Water and counterions were minimized. The backbone atoms of the protein were then fixed with the ligand, side chains, and other atoms that are free to move. Finally, the entire system was fully minimized without any constraint. In each step, energy minimization was first performed using the steepest descent algorithm for 2,000 steps, and subsequently, the conjugated gradient algorithm was used for another 3,000 steps. The MD simulation was performed under periodic boundary conditions using the sander module of the Amber9 program. First, the system was fixed to heat only the water and counterions for 10 ps to make sure the solute was fully solvated. Then, the entire system was gradually heated from 10 K to 300 K using the weak-coupling method and equilibrated for 100 ps with the protein backbone fixed [44]. Lastly, the system was switched to a constant pressure equilibration (2 ns). During the MD simulation, the particle mesh Ewald algorithm was used to handle long-range electrostatic interactions with a cutoff distance of 10 Å [45, 46], which was also used for the van der Waals energy terms. All of the angles and bonds involving hydrogen atoms were constrained using the SHAKE algorithm [47]. The time step used for the MD simulations was 2.0 fs, and the coordinates were collected every 1 ps.