• 2019-10
  • 2020-07
  • 2020-08
  • International Journal of Pharmaceutics br For nucleolin immu


    International Journal of Pharmaceutics 568 (2019) 118511
    For nucleolin immunocytochemistry, HeLa STA-21 were seeded at a density of 1.5 × 104 cells/mL in 8 well µ-slides (IBIDI, Germany) and incubated with free AT11 and AT11-B0, or the ligand complexes for 2 h. Nucleolin antibody (1:100, ref. PA3-16875, Invitrogen) was then added during 2 h and secondary antibody (Alexa Fluor 546®, 1:1000) for 1 h, both at room temperature. The excess fluorophores were washed off by rinsing with PBS three times and the cells were imaged using a Zeiss AxioObserver LSM 710 microscope.
    The degree of complexes colocalization is expressed in Pearson correlation coefficients (rp) values (Manders et al., 1993), which were calculated using the colocalization analysis tool in the ImageJ software.
    HeLa and NHDF cells were seeded in 12-well plates at a density of
    50 × 104 cells/well and incubated overnight for cell adhesion. Then, cells were treated for 24 h with the preformed C8-aptamer complexes and free C8 as control. After the incubation period, the wells were washed by rinsing with PBS three times and cells were then trypsinized, resuspended in PBS and analyzed in a BD FACSCanto™ II flow cyto-metry system (BD Life Sciences, US) to evaluate the uptake of the ap-tamer-ligand complexes. C8 fluorescence was monitored using the FITC channel. Non-specific coloring and debris were excluded by analyzing FSC vs SSC density plots.
    Since 3D structure of AT11-B0 G4 is not available, a predicted model was built based on the AT11 G4 structure (PDB: 2N3M) as pre-viously reported (Carvalho et al., 2019b). Briefly, residues T13 and T14 were deleted from 2N3M using Swiss-PDB Viewer mutation tool and converted the nucleotide sequence of AT11 to AT11-B0 (please see Fig. 1 for further details). The built structure was optimized for further experiments by running fully solvated molecular dynamics (MD) si-mulations using GROMACS 2016.3 with the following parameters. The AT11-B0 G4 structure was initially centered in an octahedral solute box and filled with TIP3P water molecules and K+ atoms. The structure was then submitted to 5000-step energy minimization with no restraints. A 100 ps equilibration was then performed under a modified Berendsen thermostat, with a temperature gradient from to 300 K, followed by another 100 ps of equilibration done under aforementioned thermostat and a Parrinello-Raman barometer. Finally, 200 ns of MD were carried out. The final snapshot of the 200 ns simulation was used as optimized model for further experiments.
    2.10. Molecular dynamics
    Both the AT11 (PDB: 2N3M, Fig. S6) and AT11-B0 (built as de-scribed above, Fig. S7) G4 structures, and nucleolin RNA-binding do-mains (RBD1,2 – PDB: 2KRR) were optimized for docking using Dock Prep tool of Chimera 1.11.2. After assigning polar hydrogens and Gasteiger charges, rigid body docking calculations were carried with
    Fig. 1. (A) Sequences of AS1411 and its derivatives AT11 and AT11-B0 (modifications shown in red). (B) Chemical structure of the acridine orange ligands used in this study. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
    Table 1
    Ligand-induced thermal stabilization of the AT11 or AT11-B0 measured by CD melting experiments.
    Fig. 3. CD spectra of (A) AT11 G4 and (B) AT11-B0 G4 in the absence and presence of 4 M eq. of each of the acridine orange ligand. Spectra acquired in 20 mM potassium phosphate buffer containing 65 mM KCl, pH 7. (For interpretation of the refer-ences to colour in this figure legend, the reader is referred to the web version of this article.)
    Fig. 4. CD melting curves of (A) AT11 G4 and (B) AT11-B0 G4 in the absence and presence of 4 M eq. of each of the acridine orange ligands. Data points were recorded at 260 nm. The Boltzmann curve fitting of the data points is shown. (For interpreta-tion of the references to colour in this figure legend, the reader is referred to the web version of this article.)
    AutoDock 4.2 program. The G4 structure was centered in a box sized 125 × 125 × 125 Å along the x, y, and z axes, respectively, with a grid spacing of 0.6 Å. For each G4 structure, 25 runs under Genetic Algo-rithm were performed setting an initial population of 150 random in-dividuals, a maximum number of evaluations of 2.5 × 106, rate of mutation and crossover of 0.02 and 0.8, respectively, and elitism value of 1. The most representative structures were selected based on avail-able literature (Fan et al., 2016) and further processed with MD simu-lations the all-atom force field AMBER99SB of GROMACS 2016.3.
    For MD simulations, the G4-nucleolin complexes were centered in an octahedral box and solvated with TIP3P water molecules and K+ atoms to neutralize the system. After an energy minimization of 1000 steps using the steepest descent algorithm, system was gradually heated to 300 K in 100 ps under the control of STA-21 Berendsen thermostat followed by 100 ps isobaric simulation under the control of the Parrinello-