Polyamidoamine Dendrimer Conjugates with Cyclodextrins as Novel Carriers for DNA, shRNA and siRNA

Gene, short hairpin RNA (shRNA) and small interfering RNA (siRNA) delivery can be particularly used for the treatment of diseases by the entry of genetic materials mammalian cells either to express new proteins or to suppress the expression of proteins, respectively. Polyamidoamine (PAMAM) StarburstTM dendrimers are used as non-viral vectors (carriers) for gene, shRNA and siRNA delivery. Recently, multifunctional PAMAM dendrimers can be used for the wide range of biomedical applications including intracellular delivery of genes and nucleic acid drugs. In this context, this review paper provides the recent findings on PAMAM dendrimer conjugates with cyclodextrins (CyDs) for gene, shRNA and siRNA delivery.


Introduction
Gene therapy has been utilized for vaccination and for the treatment of several diseases, such as genetic diseases, cancers, cardiovascular diseases, infectious diseases and dermatological diseases [1,2]. Approximately 500 clinical trials of gene therapy have been performed in the world from 2006 to 2010 [3]. Meanwhile, RNA interference (RNAi) technology has not only become a powerful tool for functional genomics, but also allows rapid drug target discovery and in vitro validation of these targets OPEN ACCESS in cell culture. Selective gene silencing by RNAi can be achieved essentially by two nucleic acid based methods: (1) nuclear delivery of gene expression cassettes that express short hairpin RNA (shRNA); or (2) cytoplasmic delivery of small double-stranded interfering RNA oligonucleotides (siRNA) [4,5]. However, a standard therapeutic use of plasmid DNA (pDNA), shRNA and siRNA in clinical settings in humans has been hampered by the lack of effective methods to deliver these genes and nucleic acid drugs into the diseased organs and cells [6]. To address these issues, the improvement in transfer activity of a non-viral vector (carrier) is of utmost importance [7], although it is sure that viral vectors have become a major delivery system for shRNA [5].
Polycation-based genes and nucleic acid drug delivery methods have been strongly expected to offer sufficient efficiency in the transportation of therapeutic genes and nucleic acid drugs across various extracellular and intracellular barriers [8]. These barriers include the interactions with blood components, enzymatic degradation, excretion from kidney and sequestration by the reticuloendothelial system (RES) before reaching target cells. Cationic polymers constitute one of the most promising approaches to the use of non-viral vectors for gene and nucleic acid drug therapy. A better understanding of the mechanisms by which genes and nucleic acid drugs can escape from endosomes and, also, how genes enter the nucleus, has triggered new strategies of synthesis and has revitalized research into new polycation-based systems.
Polyamidoamine (PAMAM) Starburst TM dendrimers (dendrimers), which are developed by Tomalia et al., are biocompatible, non-immunogenic and water-soluble [9]. Dendrimers are a unique class of synthetic macromolecules having highly branched, three dimensional, nanoscale architectures with very low polydispersity and high functionality. These features have allowed their application in nanotechnology, pharmaceutical and medicinal chemistry to be attractive [10]. Since dendrimers possess terminal modifiable amine functional group, they form complexes with genes [11,12], shRNA [13][14][15] and siRNA [16][17][18] through the electrostatic interaction as well as the binding to glycosaminoglycans on cell surface [19]. As a result, dendrimers are known to offer efficient transfer activity for genes and nucleic acid drugs. In addition, the high transfection efficiency of dendrimers can be due to their well-defined shape and the proton-sponge effect. Herein, the proton-sponge effect is believed to be caused by cationic polymers that promote endosome osmotic swelling, disruption of the endosomal membrane and intracellular release of DNA and nucleic acid drugs [20]. Generally, it is evident that the nature of dendrimers as non-viral vectors depends significantly on their generations (G). Gene transfer activity of dendrimers with high generations is likely to be superior to that of low generations [21,22], although their cytotoxicity augment as their generations increase. Therefore, there has been a growing interest in developing dendrimers with low generations (<G4) because of their extremely low cytotoxicity [23].
Cyclodextrins (CyDs) were isolated approximately 100 years ago and were characterized as cyclic oligosaccharides [24][25][26]. The α-, β-, and γ-CyDs are the most common natural CyDs, consisting of six, seven, and eight glucose units, respectively. CyDs can improve the solubility, dissolution rate and bioavailability of drugs, and so the widespread use of CyDs is well known in the pharmaceutical field [27,28]. CyDs have been reported to interact with cell membrane constituents such as cholesterol and phospholipids, resulting in the induction of hemolysis of human and rabbit red blood cells (RRBC) [29][30][31], although CyDs cannot enter cells because of their high molecular weight (ca. 1,000) and hydrophilicity [26]. Regarding the delivery of genes and nucleic acid drugs using CyDs, it is acknowledged that CyDs interact with them only very slightly [32]. Thereby, the combination of CyDs with some cell-penetrating carriers was necessary to enter the cells.
Different strategies to promote interactions between CyD conjugates and genetic material have been exploited. Recently, Pack et al. [33] and Ortiz Mellet et al. [34] reported on CyD-based gene delivery systems. Intriguingly, Davis and co-workers have reported the potential uses of β-CyD-containing polycations (CDP) with adamantine-PEG or adamantine-PEG-transferrin for gene, DNAzyme and siRNA transfer [35][36][37][38]. It should be noted that the first targeted delivery of siRNA in humans via self-assembling, CyD polymer-based nanoparticles has been reported [39,40]. In addition, various CyD-appended polymers and polyrotaxanes have been acknowledged [41]; among these CyD-based polymers and supramolecules, such as pDNA, shRNA and siRNA delivery carriers. Arima and colleague originally developed various CyD conjugates with dendrimers ( Figure 1).
Next, we observed intracellular distribution of pDNA complex with α-CDE (G3, DS 2) using a confocal laser scanning microscopy. The fluorescence derived from FITC-pDNA in the α-CDE (G3, DS 2) system was observed in cytoplasm much more than that in the dendrimer system. Additionally, α-CDEs (G3, DS 2, 5) were found to disrupt liposomal membranes, model bilayer membranes, stronger than dendrimer (G3) and α-CDE (G3, DS 1). Collectively, these lines of evidence demonstrate that α-CDE (G3, DS 2) could be ascribed to the improved endosomal-escaping ability via the additive or synergetic effects of the proton-sponge effects of dendrimers and the endosomal membrane-disrupting effects of α-CyD as shown in Figure 2 [46]. However, transfection efficiency of the pDNA complexes with α-CDEs seems to be still low, probably due to the lack of the translocation ability of the carriers into nucleus [47]. Thus, translocation of the pDNA/α-CDE complex into nucleus should be improved in order to more increase gene expression. Finding a safe and effective systemic delivery system is a major obstacle in gene therapy. Although viral vectors showed promise for high transfection rate, the immunogenicity associated with these systems has hindered further development. As an alternative to viral gene delivery, application of novel safe and effective polymeric systems that have shown high transgene expression when administered systemically has been expected [48]. Thereby, Kihara et al. evaluated gene transfer activity of α-CDE (G3, DS 2) after intravenous administration in mice. Twelve hours after intravenous administration of the solution containing pDNA complexes with α-CDE (G3, DS 2) at a dose of 50 μg pDNA/mice and at a charge ratio of 10 (carrier/pDNA), α-CDE (G3, DS 2) delivered pDNA more efficiently in spleen, liver, and kidney with negligible changes in blood chemistry data such as LDH, AST and BUN, compared with dendrimer and other α-CDE (G3, DS 1, 5). In particular, higher gene expression in spleen was observed 12 h after the administration of the pDNA complex with α-CDE (G3, DS 2). These results suggest the potential use of α-CDE (G3, DS 2) as a promising non-viral vector in vitro and in vivo, and these data may be useful for design of α-CyD conjugates with other non-viral vectors, although the further modification of the chemical structure of α-CDE (G3, DS 2) is required to improve the carrier's ability.

GUG-β-CDE (G2) as DNA Carriers
As described above, gene transfer activity of α-CDE (G3, DS 2) should be improved in vitro and in vivo. As described below, Arima and colleagues recently reported that lactosylated α-CDE (Lac-α-CDE (G2)) and pegylated folate-appended-α-CDE (G3) (Fol-PαC (G3)) selectively deliver pDNA to hepatocytes and tumor cells in vitro and in vivo, respectively ( Figure 1) [49,50]. These carriers have glucose and polyethylene glycol (PEG) as a spacer between dendrimer and targeting ligands, respectively, suggesting the importance of a spacer for a cell-specific pDNA delivery. However, it is still unknown whether introduction of a spacer between dendrimer and CyD improves gene transfer activity of α-CDEs. Therefore, Anno et al. [51] as a novel branched CyD because of its high bioadaptability and low hemolytic activity in order to prepare various dendrimer (G2) conjugates with GUG-β-CyD (GUG-β-CDE (G2)) having different DS values of the glucuronyl-glucosyl group (Figure 1) [52].
Of the four GUG-β-CDEs (DS 1. Therefore, it may play an important role in gene transfer activity at the post-cellular uptake process of pDNA complex. Then, we observed the cells after transfection of pDNA complexes with TRITC-carriers for 6 h in A549 cells using a fluorescence microscope. In all systems, the fluorescence was observed over the whole cytoplasm, suggesting effective endosomal escape of the pDNA complexes after cellular uptake. However, the difference in the fluorescence intensity in cytoplasm was not visually obvious among these carriers. Interestingly, the fluorescence of TRITC-GUG-β-CDE (G2, DS 1.8) was strongly observed in the nucleus, compared with those of other carriers. The mechanism for nuclear localization of GUG-β-CDE (G2, DS 1.8) is still not clear. However, some lectins are known to exist in the nuclear membranes, and can recognize sugars such as glucose or galactose [53][54][55][56]. Additionally, the carbohydrate-recognition domain (CRD) of the lectin mainly recognizes 2-, 4-hydroxyl group of monosaccharides [47]. Hence, Anno et al. hypothecated that GUG-β-CDE (G2) having a 2-, 4-hydroxyl group in the spacer domain might be recognized by the nuclear lectins, and studied to address the hypothesis. As a result, Anno et al. revealed that the pDNA complex with GUG-β-CDE (G2, DS 1.8) shows high endosomal escaping ability and nuclear localization in A549 cells. With few exceptions, where local administration is feasible, a progress towards broad clinical application of gene therapies requires the development of effective delivery systems. However, development of a novel non-viral vector suitable for systemic application is strongly expected [57]. Therefore, Anno et al. examined gene transfer activity 12 h after intravenous administration of the solution containing pDNA complex with GUG-β-CDE (G2, DS 1.8) to tail vein of mice [58]. Various carriers showed higher gene transfer activity in kidney than in other tissues. It is noteworthy that gene transfer activity of GUG-β-CDE (G2, DS 1.8) in kidney was much higher than that of α-CDE (G2, DS 1.2) or β-CDE (G2, DS 1.3) [58]. To investigate the safety profile of GUG-β-CDE (G2, DS 1.8), Anno et al. determined some blood chemistry data such as blood urea nitrogen (BUN) and aspartate aminotransferase (AST) 12 h after intravenous administration of its pDNA complexes in mice [58].
The parameters in the GUG-β-CDE (G2, DS 1.8) system were almost equivalent to those in the control system [58]. These results strongly suggest that the pDNA complex with GUG-β-CDE (G2, DS 1.8) has a safety profile even in vivo. Recently, gene therapy directly administered to blood vessels in patients with incurable renal diseases, such as Alport syndrome, polycystic kidney disease, renal cancers, glomerulonephritis and renal fibrosis has been studied. Potentially, the present findings suggest that GUG-β-CDE (G2, DS 1.8) might have the potential as a carrier for the gene therapy of kidney diseases, although it is necessary to reveal the detail gene expression region in kidney such as a glomerulus or a renal tubule.

α-CDE (G3) as shRNA Carriers
Recently, shRNAs expression systems have been developed in order to prolong duration of the RNAi effect [59]. To elicit silencing in these systems, a small DNA insert encoding shRNA against the gene of interest is cloned into the vector downstream of the polymerase III promoter. Once transfected into mammalian cells, the insert-containing vector expresses the shRNA, which is rapidly processed by a Dicer-dependent cleavage in cytoplasm into siRNA, and then each is incorporated into the RNA induced silencing complex (RISC) followed by degradation of target mRNA [60]. As already described above, standard therapeutic use of RNAi in clinical settings in humans has, however, been hampered by the lack of effective methods to deliver the shRNA-expressing plasmid vectors into the diseased organs [61]. However, viral vectors have safety risks such as immunogenicity, oncogenicity and potential viral recombination that need to be solved [62]. For these reasons, the improvement in shRNA transfer activity of a non-viral vector (carrier) is of utmost important.
The potential of α-CDE (G3, DS 2) as a novel carrier of pDNA expressing shRNA against pGL3 luciferase gene (shGL3) was evaluated. That is, the shGL3 transfer activity of α-CDE (G3, DS 2) was compared with that of dendrimer (G3). Regarding the complexation, α-CDE (G3, DS 2) formed a stable and condensed complex with shGL3 and induced a conformational transition of shGL3 from the B-form, a right-handed double helix with 10 bp per turn, to the C-form, a right-handed double helix with a 9.33 bp per turn that is less compact than the B-form of DNA, in solution, even in the low charge ratios. In addition, α-CDE (G3, DS 2) markedly inhibited the enzymatic degradation of shGL3 by DNase I. The shGL3 complex with α-CDE (G3, DS 2) at a charge ratio of 20/1 (carrier/shGL3) elicited the most potent RNAi effects in cells transiently and stably expressing the pGL3 luciferase gene without the off-target effects and cytotoxicity among the complexes with the various charge ratios (Figure 4). Besides, the RNAi effects were markedly enhanced by the further addition of the adequate amounts of siRNA to the shGL3 complex with α-CDE (G3, DS 2). Additionally, the prominent RNAi effects of the shGL3 complex with α-CDE (G3, DS 2) could be attributed to the stabilizing effect of α-CDE (G3, DS 2) on enzymatic degradation of shRNA and negligible cytotoxicity, although the formation of the stable complex may somewhat act as a negative factor. Collectively, these results suggest that α-CDE (G3, DS 2) possesses the potential to be a novel carrier for shRNA. complexes on endogenous gene expression, as well as physicochemical properties, cytotoxicity, local irritation, interferon response, cellular uptake and intracellular distribution of the siRNA complexes. As a result, the siRNA complex with α-CDE (G3, DS 2) showed potent RNAi effects against Lamin A/C and Fas expression with negligible cytotoxicity, compared to those of the transfection reagents in Colon-26-luc cells and NIH3T3-luc cells, which stably express pGL3 luciferase gene [65]. Interestingly, α-CDE (G3, DS 2) delivered fluorescent-labeled siRNA to cytoplasm, not nucleus, after transfection in NIH3T3-luc cells (Figure 6), consistent with the pDNA complex with α-CDE (G3, DS 2). Furthermore, α-CDE (G3, DS 2) suppressed siRNA degradation by serum. Strikingly, the α-CDE (G3, DS 2)/siRNA complex exerted the in vivo RNAi effect on pGL3 luciferase expression in Colon-26-luc cells after not only intratumoral, but also intravenous administrations in tumor-bearing mice. Additionally, the blood chemistry values did not change after intravenous injection of α-CDE (G3, DS 2)/siRNA complex at the same siRNA dose as that showing the RNAi effect. This in vivo safe profile of the α-CDE (G3, DS 2)/siRNA complex is highly likely to be consistent with the in vitro safe profile, e.g. negligible cytotoxicity and hemolytic activity. These results suggest the potential use of α-CDE (G3, DS 2) as a novel carrier for siRNA both in vitro and in vivo.

Potential Use of Polypseudorotaxanes (PPRXs) of Pegylated Dendrimers (PEG-Dendrimers) with CyDs as the Novel Sustained Release Systems for pDNA
CyD-based polyrotaxanes and PPRXs have inspired interesting exploitation as novel biomaterials because of their low cytotoxicity, controllable size, and unique architecture. Actually, the potential applications of CyD-based polyrotaxanes and PPRXs in life science and biotechnology have been reported [66].
Motoyama et al. demonstrated the potential use of PPRXs of PEG (molecular weight 2000)-grafted dendrimer (PEG-dendrimer) with CyDs as novel sustained release systems for pDNA [67]. The PEG-dendrimer/pDNA complex formed PPRXs with α-CyD and γ-CyD, but not with β-CyD, in solutions. In the PEG-dendrimer/CyDs PPRXs systems, 17.9 mol of α-CyD and 8.8 mol of γ-CyD were determined to be involved in the PPRXs formation with one PEG chain by α-CyD and γ-CyD, respectively, in the 1 H-NMR study. In addition, the CyDs PPRX formation resulted in the sustained release of pDNA from PEG-dendrimer complex with pDNA at least 72 h in vitro. In addition, the release of pDNA from CyDs PPRX retarded as the dissolution medium volume decreased. These results suggest that the PEG-dendrimer/CyD PPRX systems can work as a sustained pDNA release system. Potentially, the PPRX formation with CyDs may be useful as a sustained drug delivery technique for other pegylated polymers [67]. Most recently, Motoyama et al. clarified that pegylated α-CDE/CyD PPRXs are promising sustained release systems in vitro and in vivo [68].

Conclusion
Many attempts have been made to design and evaluate CyD conjugates with polymers for DNA, shRNA and siRNA carriers. In this review, the potential of α-CDEs as DNA, shRNA and siRNA carriers were demonstrated for the first time. Secondly, GUG-β-CDE (G2, DS 1.8) was described to be a preferable pDNA carrier to α-CDE (G2, DS 1.2). Thirdly, the PEG-dendrimer or PEG-α-CDE/CyD PPRX systems were shown to work as a sustained pDNA release system. Fourthly, the potential use of lactosylated and pegylated folate-appended α-CDEs as targeting carriers to hepatocytes-and cancer cells, respectively, was introduced. Thus, these α-CDEs and GUG-β-CDEs are likely to be promising carriers for pDNA, siRNA and shRNA, but their potency may be insufficient for clinical use. Further improvement of the potency of α-CDE (G3, DS 2) and GUG-β-CDE (G2, DS 1.8) as carriers for DNA, shRNA and siRNA should be performed. Elaborate studies are further required to develop novel carriers for genes and various nucleic acid drugs such as shRNA, siRNA, decoy DNA, antisense DNA, ribozyme and aptamers. Most recently, we revealed that fucosylated α-CDE shows good potential as a decoy DNA carrier. The future should see various clinical use products using CyD-containing carriers for DNA, shRNA and siRNA.