Computational protocol: Multivalent Interactions of Human Primary Amine Oxidase with the V and C22 Domains of Sialic Acid-Binding Immunoglobulin-Like Lectin-9 Regulate Its Binding and Amine Oxidase Activity

Similar protocols

Protocol publication

[…] The 3D structural model for Siglec-9-EC sequence (UniProt Knowledgebase (UniProtKB) Q9Y336) was constructed using the X-ray structure of Siglec-5 (Protein Data Bank Identification Code (PDB ID) 2ZG2 []) as a template. Firstly, a 3D model for the V–C21 domains (residues S20-H229) was made based on the multiple sequence alignment of Siglec-3, -5, -6, -7, -9 and -14 using the V–C21 domains of the Siglec-5 crystal structure as a template (). Secondly, the C22 domain (residues S251-L335) was separately modeled using the sequence alignment of the C2 domains of CD-33-related Siglec sequences, including all C21, C22 and C23 domains, and the C21 domain of the Siglec-5 crystal structure as a template (). Thereafter, the linker region (L230-V250) between the C21 and C22 domains was modeled in two different ways. The first complete Siglec-9-EC model of residues S20-L335 (model 1) was done by orienting the domains manually into an extended 3D arrangement and the second model of the same residues (model 2) was based on the 3D arrangement of the immunoglobulin domains in the X-ray structure of neural cell adhesion protein (PDB ID 1QZ1 []), which also has an extended 3D arrangement of the immunoglobulin domains. All sequence alignments were done using Malign [] within Bodil []. For all the 3D models, ten models were generated with Modeller [] and the ones with the lowest objective function were chosen as a representative for each model. The quality of the 3D folds of the V–C21 and C22 homology models were evaluated with PROCHECK [], ProSA web [, ] and QMEAN []. To visualize the location of the sialic acid (SA) binding site in Siglec-9-EC, sialic acid binding mode was modeled based on the Siglec-5 complex (PDB ID 2ZG1 []). Model 1 and model 2 of Siglec-9-EC were refined by molecular dynamics (MD) simulations as described in the following section. [...] Protein Preparation Wizard, as implemented in Maestro v. 9.6 molecular modeling software (Schrödinger, Inc.), was used to prepare model 1 and model 2 of Siglec-9-EC for the MD simulations. All hydrogen atoms were added, bond orders were assigned, and disulphide bridges were created between the following cysteine residues: 36 and 170, 41 and 102, 164 and 213, and 271 and 320. Hydrogen bonds were assigned at pH 7.0 and the protonation states of histidines were selected interactively to optimize the hydrogen bond network. Finally, a restrained energy minimization of hydrogen atoms was run in the OPLS 2005 force field.Energy minimization, thermal equilibration and standard production simulations were performed with the AMBER package (version 12) [] using the AMBER ff03 force field []. All simulations were run in an octahedral box (extending 10.0 Å from the protein), filled with explicit TIP3P water molecules [] and one neutralizing Cl-ion. Periodic boundary conditions, particle-mesh Ewald electrostatics [] and a cut-off of 9 Å for non-bonded interactions were used. A time step of 1 fs (for Langevin dynamics during equilibration) or 2 fs was applied together with the SHAKE algorithm [] to constrain the bonds to hydrogen atoms. The 20-ns production simulations were performed at a constant temperature of 300 K and a pressure of 1 bar. The coupling constants for temperature and pressure [] were 5.0 and 2.0 ps, respectively. Energy minimization was performed with the steepest descent and conjugate gradient methods in six steps, gradually reducing the restraints on the protein atoms to their initial positions. At each step, the restraint force constant was defined as follows: 10, 5, 1, 0.1, 0.01 and 0 kcal/molÅ2. Each minimization step was carried out for a maximum of 200 iterations (of which the 10 first iterations were with the steepest descent method and the rest with the conjugate gradient algorithm). Equilibration simulations were performed in five steps: (i) 10 ps heating of the system from 10 K to 300 K using a Langevin thermostat with a collision frequency (γ) of 1.0 ps-1, constant volume, and restraints on the protein atom positions (restraint force constant of 5 kcal/molÅ2); (ii) same as the previous step but for 20 ps and without restraints on the protein atom positions; (iii) 20 ps MD at 300 K using a Langevin thermostat with γ = 0.5 ps-1 and constant volume, no restraints on the protein; (iv) 50 ps MD at 300 K using a Langevin thermostat with γ = 0.5 ps-1 and constant pressure of 1.0 bar, coupling constant for pressure of 1.0 ps, no restraints on the protein. (v) 400 ps MD at 300 K and at constant pressure of 1 bar, coupling constants for temperature and pressure were 5.0 and 2.0 ps, respectively, no restraints on the protein. The MD simulation trajectories were analyzed with VMD [] and the ptraj module of AMBER. The resulting final frame structures were first minimized with AMBER (similarly to the last step of the initial minimization) and then visually examined with PyMOL (Schrödinger, Inc.). [...] The purified proteins were characterized using fluorescence based thermal stability assay []. In this methodology, a dye intercalates with the exposed hydrophobic regions generated by unfolding of proteins. We used SYPRO® orange dye with maximal absorption of dye-protein complex at 470 nm and maximal emission at 569 nm. Siglec-9-EC was concentrated to 2 mg/ml. A 96-well plate was filled with protein samples, buffers and dye with assay volume of 25 μl per well. The analysis was done on an iCycler machine (Bio Rad Laboratories) and melting curves were generated by increasing the temperature from 20°C to 95°C with a stepwise increment of 1°C. The fluorescent signal is plotted as a function of temperature, and the significant increase in the signal (slope) corresponds to the melting of the protein. The analysis of the results and the melting temperature (Tm) estimation of thermal shift assay were calculated with the Meltdown program []. The program estimates the melting temperature in two ways: by using a quadratic fit to the data around the global minimum of the first derivative curve (this value is used as the Tm in subsequent analyses) and by finding the temperature associated with the midpoint in the fluorescence response between the high point and the low point of the melt curve. The melt curves are considered to be normal by the Meltdown program if the estimated Tm values are within 5°C. [...] We used non-parametric Mann-Whitney U-test for the comparison of means. Analysis of variance was done using Kruskall-Wallis test. The p-values below 0.05 were considered significant. All the analyses were done using IBM SPSS Statistics version 22.0 (SPSS Inc. USA). […]

Pipeline specifications

Software tools Meltdown, SPSS
Applications Miscellaneous, Thermal shift assay
Organisms Homo sapiens
Diseases Neoplasms
Chemicals Amines, N-Acetylneuraminic Acid