br J Figueiredo et al br Fig Confocal fluorescence images
J. Figueiredo, et al.
Fig. 9. Confocal fluorescence images of HeLa cells incubated with (A) Cy5-AT11-C8 and (B) Cy5-AT11-B0-C8 after 2 h at 37 °C. Nucleolin is stained with AlexaFluor 546®, C8 emits green fluorescence, Cy5-AT11 and Cy5-AT11-B0 are shown in red. Scale bar: 25 µm. Brightness was adjusted for all images equally. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 10. Confocal microscopy images of HeLa cells incubated with (A) Cy5-AT11 G4-C8 and (B) Cy5-AT11-B0 G4-C8 for 3 days. Cell nuclei are stained with Hoechst 33,342 (blue), C8 emits green fluorescence and Cy5-AT11 G4 or Cy5-AT11-B0 G4 are shown in red. Overlapping of the two stains, observed as yellow/orange regions, can be seen in the merged images of the cells. Scale bar: 25 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 11. Confocal microscopy images of NHDF cells incubated for 3 days with
(A) Cy5-AT11 G4-C8 and (B) Cy5-AT11-B0 G4-C8. Cell nuclei are stained with Hoechst 33,342 (blue), C8 emits green fluorescence and Cy5-AT11 G4 or -AT11-B0 G4 are shown in red. The combination of the two stains, observed as yellow/ orange regions, can be seen in the merged images of the cells. Scale bar: 25 µm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
negatively aﬀect the recognition of nucleolin by AT11 G4 and AT11-B0 G4.
B0 G4 is available, we manually edited the sequence of PDB 2N3M and re-optimized the structure for MD studies to remove unfavorable van der Waals interactions and probe for incorrect KIN59 names and con-nectivity. The final snapshot of the 200 ns MD run for optimization of the AT11-B0 G4 model in shown in Fig. S7. AT11 PDB structure and AT11-B0 constructed model were used as starting scaﬀolds for docking experiments. Previously, it has been shown that AS1411 aptamer binds preferentially the RNA-binding domains 1 and 2 (RBD1,2) of nucleolin (Fan et al., 2016). As AT11 and AT11-B0 are both AS1411 derivatives, we have used RBD1,2 for the calculations. The complexes formed by AT11-RBD1,2 and AT11-B0-RBD1,2 are depicted in Fig. 8. Stable complexes were attained during the 50 ns simulations as seen by the RMSD graphs (Figs. S7 and S8). After an initial increase in the RMSD values, both complexes converged at around 3 ns remaining stable re-latively to this conformer throughout the remaining of the simulation. RBD1,2 was found interacting with the loop region of both G4 struc-tures, establishing multiple H-bonds (green lines) with the G4 residues, particularly the loops thymines (depicted in grey). The 12-residue linker between both RBD seems to create a binding pocket for the G4 structures, with additional interactions with RBD units themselves. In the case of AT11 G4 (Fig. 13A), thymines T4, T14, T16 and T19 interact with RBD1,2 thus demonstrating the preference of protein nucleolin for the loop residues. AT11-B0 on its hand (Fig. 8B), was found similarly bound to RBD1,2 through T4, T17 and T20. These results agree with previous MD studies using a predicted model of AS1411 aptamer (Fan et al., 2016) and experimental reports using loop-modified AS1411 sequences (Kim et al., 2010).
To determine if AT11 G4 and AT11-B0 G4 are able to bind speci-fically to nucleolin, HeLa cells were used and analyzed by confocal microscopy after incubation with Cy5-AT11 or Cy5-AT11-B0. For comparative purposes, complexes AT11-C8 and AT11-B0-C8 were se-lected for confocal microscopy studies to analyze if the conjugates can recognize nucleolin in vitro. The primary anti-nucleolin antibody con-jugated with the secondary antibody AlexaFluor 546® were used to localize cell surface nucleolin. Free Cy5-AT11 and Cy5-AT11-B0, as well as their conjugates with C8, were able to colocalize with nucleolin-positive regions after a short incubation (2 h) (Figs. 9 and S1). This assay demonstrated that AT11 and AT11-B0 are anti-nucleolin apta-mers, because they were able to co-stain this protein thus suggesting cell surface binding. Moreover, the conjugation of C8 with the G4 se-quences does not seem to negatively aﬀect their nucleolin binding as similar behavior was observed for free and conjugated aptamers.
To evaluate the internalization ability of AT11 G4 and AT11-B0 G4 in cervical cancer and non-malignant cell lines, the cellular uptake of Cy5-labelled AT11 and AT11-B0 or ligand-conjugated was monitored by confocal microscopy. Previously, it was demonstrated that cell death starts to occur after prolonged exposure to AT11 (as to AS1411), more specifically between the second and fourth day, depending on the cells type used (Bates et al., 2009; Do et al., 2017). Taking this into account, cells were visualized after three days of incubation with the complexes AT11 G4-C8 and AT11-B0 G4-C8 in order to avoid excessive cell death rate and to enabling the imaging of their internalization in the used cell lines.