*4.2. In Vitro Targeted siRNA Delivery Performance of Double-Tail, Dual Targeting Self-Assembling Amphiphilic Dendron AD*/*E16G6RGDK*

The authors verified that the presence of the E16G6RGDK peptides on the surface of the siRNA/**AD** micelles was indeed effective in specifically targeting cancer cells via dual binding with the designated target receptors ανβ3integrin and Nrp-1. The cellular uptake of the siRNA/**AD**/E16G6RGDK complexes was investigated again with PC-3 cells, in which both receptors were highly expressed [27,28]. As seen in Figure 15a, the amount of siRNA/**AD**/E16G6RGDK complexes internalized by PC-3 cells was three times higher than that measured for the siRNA/**AD** micelle. Further, the cellular uptake of the nanocarrier modified with the targeting moieties was inhibited in the presence of a cyclic RGD peptide. This clearly indicates that this last molecule competed with the peptide-coated nanovector for integrin binding via the RDG motif.

**Figure 15.** (**a**) Flow cytometry analysis of cellular uptake of siRNA/**AD** and siRNA/**AD**/E16G6RGDK nanoparticles in PC-3 cells using a siRNA labeled with the fluorescent Cy5 die. The last column shows the cell uptake results for the siRNA/**AD**/E16G6RGDK obtained in the presence of the cyclic peptide c-RGD, an ανβ3-integrin integrin binder. (**b**) Cellular uptake visualized by confocal microscopy of the siRNA/**AD**/E16G6RGDK complexes in the absence (top panel) and in the presence (bottom panel) of a Nrp-1 antibody. Images from left to right: (**A**) cy5-labelled siRNA; (**B**) nucleus (4 ,6-diamidino-2-phenylindole (DAPI) staining); (**C**) merged (**A**,**B**). The last two images (**D**,**E**) are the bright field version of (**B**,**C**). In all these experiments, the siRNA/**AD**/E16G6RGDK nanoassemblies were prepared using 50 nM siRNA, siRNA/**AD** N/P = 10, and **AD**/E16G6RGDK molar ratio = 5. Adapted from [8] with permission of the American Chemical Society, 2018.

Next, PC-3 cells pretreated with a monoclonal antibody (mAb) directed against Nrp-1 were used to assay the internalization of the siRNA/**AD**/E16G6RGDK nanoassemblies in comparison with untreated cells. As can be easily seen in the corresponding confocal microscopy images (Figure 15b), in the presence of the Nrp-1 mAb (bottom panel), the uptake of the siRNA-loaded peptide-modified **AD** micelles was drastically reduced when the mAb was absent (upper panel). This indicates that

the targeting peptide-decorated siRNA delivery system was effectively cell internalized also via its interaction of its RGDK warhead with the Nrp-1 receptor.

Following verification of the successful attribution of the desired targeting properties to our double-tail self-assembling **AD** nanovector, the authors checked whether its excellent endosomal escape features, required for efficient siRNA delivery and release, was preserved. By confocal microscopy, it was observed that 2 h after transfection, almost all of the siRNA molecules were localized in the cell cytoplasm (Figure A3). This confirms that the peptide-decorated **AD** nanovectors were also able to effectively circumvent endosomal trapping and release their nucleic acid cargo for subsequent RNAi.

The authors further assessed whether the endowment of the double-tail self-assembling dendron **AD** with targeting properties eventually conferred by the RGDK-based peptides rendered this nanovector more effective in siRNA delivery to cancer cells with respect to the undecorated **AD** micelles. Their RNAi performance in silencing Hsp27 in PC-3 cancer cells was compared. The relevant results showed that a remarkable Hsp27 knockdown was achieved both at the mRNA (56%, Figure 16a) and the protein (90%) levels with a very low (20 nM) siRNA concentration (for comparison, a 50 nM siRNA concentration was required to produce similar results either with the covalent G5 PAMAM dendrimer [9] or with the **AD** micelles without targeting peptide decoration (Figure 10)). Moreover, the same experiments performed in the presence of either the c-RGD peptide or the Nrp-1 directed mAb resulted in a decreased gene silencing effect (as shown by the last two columns in Figure 16a). This confirms that the siRNA/AD/E16G6RGDK nanoparticles indeed interact with the two target receptors.

**Figure 16.** (**a**) mRNA levels following knockdown of Hsp27 in PC-3 cells with siRNA/**AD** and siRNA/**AD**/E16G6RGDK nanoparticles. The last two columns show the results of the same experiments performed in the presence of the integrin binder c-RGD and of the mAb directed toward the Nrp-1 receptor, respectively. (**b**) Comparison of the antiproliferative activity exerted on PC-3 cells by siRNA/**AD** and siRNA/**AD**/E16G6RGDK nanoparticles. In both panels, non-treated cells were used for control, siRNA concentration = 20 nM, siRNA/**AD** N/P = 10, and **AD**/E16G6RGDK molar ratio = 5. Adapted from [8] with permission of the American Chemical Society, 2018.

Following siRNA delivery to PC-3 cells with the peptide-decorated **AD** nanomicelles, a drastic anticancer activity was observed again using only a 20 nM siRNA concentration (Figure 16b). This corresponds to an effect of 30% higher than that achieved with the pristine **AD** nanovectors. Concomitantly, the **AD**/E16G6RGDK nanoassemblies exhibited a safe toxicity profile, similar to that presented by the **AD** micelles (Figure A4), and in addition to its notable improvement in hemolytic effects (Figure A4c). All these data led the authors to perform the final testing of the siRNA/AD/E16G6RGDK nanoparticles in vivo.
