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

The in vivo gene silencing and anticancer effects of the peptide-decorated **AD** nanomicelles was investigated again using PC-3 xenograft nude mice. The first analysis was conducted to verify their enhanced tumor targeting ability with respect to the undecorated **AD** nanovectors using a fluorescent-labeled Hsp27 siRNA. Accordingly, ex vivo imaging of isolated tumors and the quantification of the relevant mean fluorescence intensity were performed (Figure 17a,b), from which the substantially higher tumor accumulation of the targeted nanoparticles was substantiated.

**Figure 17.** (**a**) Ex vivo fluorescence images obtained from tumor isolated from PC-3 xenograft mice treated with PBS buffer (delivery medium used as control, left), siRNA/**AD** (center), and siRNA/**AD**/E16G6RGDK nanoparticles (right). (**b**) Quantification of the mean fluorescence intensity for the images in panel **a**. PBS buffer (**c**) Tumor growth in untreated animals (green filled symbols), and in animals treated with siRNA/**AD** (red filled symbols). (**d**) In vivo knockdown of Hsp27 with siRNA/**AD** (-RGKD) and siRNA/**AD**/E16G6RGDK (+ RGDK) nanoparticles. Untreated animals were used as the control. The treatments (0.25 mg/kg siRNA, siRNA/**AD** N/P = 10, and **AD**/E16G6RGDK molar ratio = 5) were administered by intravenous injection twice per week. Adapted from [8] with permission of the American Chemical Society, 2018.

After 4 weeks of mice treatment with a very low siRNA concentration (0.25 mg/kg)—12-fold less than the conventional concentration used for siRNA delivery in mice—both the tumor volume and the corresponding levels of Hsp27 protein were drastically reduced when siRNA was administered via the peptide-decorate **AD** nanovectors, as shown in Figure 17c,d. In particular, at day 31 after treatment with the siRNA/**AD**/E16G6RGDK nanoparticles, tumor growth was reduced by 73% with respect to the control and by 50% relative to siRNA/**AD** treatment (Figure 17c). Consistently, Hsp27 expression in tumors was reduced by 65% and 41% upon siRNA treatment with the targeted nanovectors with respect to the control and **AD**-delivered siRNA, respectively (Figure 17d).

These highly promising results, coupled with the excellent in vivo toxicity profile exhibited by the **AD**/E16G6RGDK nanocarriers confirmed the high potential for these targeting nanosystems as safe and efficient siRNA delivery, gene silencing and consequent anticancer nanotherapeutics.

#### **5. Conclusions**

As defined by Nature.com [29], "self-assembly is the process by which an organized structure spontaneously forms from individual components, as a result of specific, local interactions among the components. When the constitutive components are molecules, the process is termed molecular self-assembly". A key feature of molecular self-assembly is the multivalent, cooperative and synergistic nature of the intermolecular interactions leading to the organization of individual molecular entities into well-defined nanosized structures. This approach presents at least two major conceptual advantages. The first relies on the evidence that the driving forces governing self-assembly lead to the formation of the nano-objects virtually with no flaws, as interactions between the nanomicelle building blocks are mediated by specific molecular recognition ultimately resulting in complex and ordered nanoscale structures. The second, by no means of less important benefit to self-assembly is that very small amounts of material are required to accomplish the process.

In the companion paper [9], the authors presented the design, synthesis and gene silencing activity of modified PAMAM-based covalent nanovectors for siRNA delivery in cancer therapeutics. However, high generation dendrimer synthesis is extremely laborious and time-consuming, since the final product purification is difficult and hampered by the presence of highly similar side products. Thus, notwithstanding the highly promising results achieved with these molecules, the difficulties inherent in large-scale good manufacturing practice (GMP) production of high generation dendrimers led the authors to explore the potential of self-assembly in the design of new efficient siRNA nanocarriers.

Thus, as presented in this second short review, the small PAMAM-based amphiphilic dendron **4**, composed by a PAMAM head and a C18-long hydrocarbon tail, was initially developed. This was safe and effective in delivering siRNA both in vitro and in vivo. The natural evolution of **4** was its double-tail counterpart **AD**. The in vivo gene silencing effect obtained with siRNA delivered by **AD** nanomicelles in cancer treatment was comparable to that obtained with its precursor dendron **4**. However, **AD** nanovectors were also capable of eliciting gene silencing effects in the highly challenging and treatment-refectory human primary cells and stem cells.

The RNAi results of both the amphiphilic dendrons **4** and **AD** could be mainly attributed to passive targeting via the EPR effect in addition to their excellent siRNA delivery and endosomal escape capabilities. The authors focused on equipping **AD** with the dual E16G6RGDK targeting peptide, with the final goal of endowing the resulting nanovectors with cancer cell-directing specificity and, accordingly, higher anticancer activity. Compared to **AD**, the peptide-decorated **AD** nanomicelles exhibited more than 10-fold greaterin vivo RNAi at significantly lower siRNA doses with respect to both covalent PAMAM dendrimers [9] and the non-targeted **AD** nanovectors. Our current activity in the field is progressing along this line, with the computer-assisted design of new targeting moieties for decorating self-assembling nanovectors in the development of gene silencing-based personalized medicine against different highly challenging and deadly (e.g., glioblastoma and pancreatic) human cancers.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1999-4923/11/7/324/s1, Figure S1. Schematic representation of the coarse-grained DPD model of the AD dendron. The different bead types are colored as follows: RC, dark magenta; R, plum; L, dark turquoise; G, chartreuse; C, light gray. Table S1. Example of DPD interaction parameters used to simulate the self-assembling of the amphiphilic dendron AD per se and in the presence of siRNA molecules.

**Funding:** This research was funded by the Italian Association for Cancer Research (AIRC), grant IG17413 to SP. The assistant position (RTDa) of SA is fully supported by the University of Trieste, in agreement with the actuation of the strategic planning financed by the Italian Ministry for University and Research (MIUR, triennial program 2016–2018) and the Regione Friuli Venezia Giulia (REFVG, strategic planning 2016–18), assigned to SP. This award is deeply acknowledged.

**Acknowledgments:** Authors wish to thank Ling Peng and her group for the longstanding, fruitful collaboration, the challenges in siRNA delivery nanovector design and optimization, the inspiring discussions and, above all, the personal friendship.

**Conflicts of Interest:** The authors declare no conflict of interest.
