**3. Delivery systems**

Over the past 20 years, a variety of techniques have been developed for encapsulating both conventional drugs (such as anticancer drugs and antibiotics) and new genetic drugs (plasmid DNA containing therapeutic genes, antisense oligonucleotides and small interfering RNA) within nanoparticles. Nanoparticle delivery systems for gene delivery possess useful inherent attributes, including a diameter of approximately 100 nm or less, a high drug-to-lipid ratio (in the case of lipid based systems), excellent retention of the encapsulated drug, and a long (> 6 h) circulation lifetime. These properties permit nanoparticles to protect their contents during circulation, prevent contact with healthy normal tissues, and accumulate at sites of disease.

### **3.1 Viral and non-viral delivery systems**

A major impediment to the successful application of gene therapy for the treatment of a range of diseases is not a paucity of therapeutic genes, but the lack of an efficient non-toxic gene delivery system. Gene delivery is generally accomplished using one of two categories of delivery vector: 1) Viral delivery vectors derived from engineered retrovirus, adenovirus, herpesvirus, lentivirus or hybrid retro/adeno virus), and 2) Non-viral (synthetic) vectors comprised of polymers and lipids, naked DNA, plasmid-protein conjugates. Other (physical as opposed to chemical) methods include microneedle coating, gene gun, and ultrasound/microbubble-mediated gene delivery.

Viral vectors account for nearly 75% of all clinical trials conducted thus far (http://www.wiley.co.uk/genetherapy/ clinical). These vectors are essentially viruses that have been genetically modified to remove any replication/pathogenic genes, and instead encode a gene of interest (Hesdorffer et al. 1998). This strategy preserves the viruses highly efficient ability to infect cells, while eliminating/attenuating toxicity. The most commonly used viral vectors are retroviral, adenoviral, adeno-associated viral and herpes simplex viral-based (Young et al. 2006). Although Vitravene, an antisense oligonucleotide-based product, is the only gene delivery product approved so far by US-FDA, there are several other products in late stages of clinical trials. Gendicine, an adenovirus encoding the *p53* gene, developed by SiBiono GeneTech Co., Ltd., was recently approved by China's state Food and Drug Administration, for the treatment of head and neck squamous cell carcinoma. Virus-based treatments tend to maximize efficiency of gene transfer, often at the cost of safety, while non-viral options generally capitalize on a higher safety profile, but usually at the expense of Gene of Interest (GOI) expression efficiency. Many viruses including retroviruses, adenoviruses, herpes simplex viruses, adeno-associated viruses (AAV) and pox viruses have been modified to eliminate their toxicity, maintain their high gene transfer capability and long term gene expression.

nanoparticles. Third, to avoid P-glycoprotein-mediated drug efflux, they further designed another delivery vehicle, LPD-II, which showed much higher entrapment efficiency of Dox than LPD. Finally, the authors delivered a therapeutic siRNA to inhibit MDR transporter. Three daily intravenous injections of therapeutic siRNA and Dox (1.2 mg/kg) co-formulated in either LPD or LPD-II nanoparticles showed a significant improvement in tumor growth inhibition (Chen et al. 2010a). Numerous other similar studies highlights a potential clinical use for these multifunctional nanoparticles with an effective delivery property and a

Over the past 20 years, a variety of techniques have been developed for encapsulating both conventional drugs (such as anticancer drugs and antibiotics) and new genetic drugs (plasmid DNA containing therapeutic genes, antisense oligonucleotides and small interfering RNA) within nanoparticles. Nanoparticle delivery systems for gene delivery possess useful inherent attributes, including a diameter of approximately 100 nm or less, a high drug-to-lipid ratio (in the case of lipid based systems), excellent retention of the encapsulated drug, and a long (> 6 h) circulation lifetime. These properties permit nanoparticles to protect their contents during circulation, prevent contact with healthy

A major impediment to the successful application of gene therapy for the treatment of a range of diseases is not a paucity of therapeutic genes, but the lack of an efficient non-toxic gene delivery system. Gene delivery is generally accomplished using one of two categories of delivery vector: 1) Viral delivery vectors derived from engineered retrovirus, adenovirus, herpesvirus, lentivirus or hybrid retro/adeno virus), and 2) Non-viral (synthetic) vectors comprised of polymers and lipids, naked DNA, plasmid-protein conjugates. Other (physical as opposed to chemical) methods include microneedle coating, gene gun, and

Viral vectors account for nearly 75% of all clinical trials conducted thus far (http://www.wiley.co.uk/genetherapy/ clinical). These vectors are essentially viruses that have been genetically modified to remove any replication/pathogenic genes, and instead encode a gene of interest (Hesdorffer et al. 1998). This strategy preserves the viruses highly efficient ability to infect cells, while eliminating/attenuating toxicity. The most commonly used viral vectors are retroviral, adenoviral, adeno-associated viral and herpes simplex viral-based (Young et al. 2006). Although Vitravene, an antisense oligonucleotide-based product, is the only gene delivery product approved so far by US-FDA, there are several other products in late stages of clinical trials. Gendicine, an adenovirus encoding the *p53* gene, developed by SiBiono GeneTech Co., Ltd., was recently approved by China's state Food and Drug Administration, for the treatment of head and neck squamous cell carcinoma. Virus-based treatments tend to maximize efficiency of gene transfer, often at the cost of safety, while non-viral options generally capitalize on a higher safety profile, but usually at the expense of Gene of Interest (GOI) expression efficiency. Many viruses including retroviruses, adenoviruses, herpes simplex viruses, adeno-associated viruses (AAV) and pox viruses have been modified to eliminate their toxicity, maintain their high gene transfer capability and long term gene expression.

function to overcome drug resistance in cancer.

normal tissues, and accumulate at sites of disease.

ultrasound/microbubble-mediated gene delivery.

**3.1 Viral and non-viral delivery systems**

**3. Delivery systems**

However, despite these advantages, there are currently major limitations to the use of viruses as gene delivery vectors. These include the limited carrying capacity of transgenic materials (i.e. size of plasmid) since the packaging capacity of viral vectors (usually up to 5 kb) is constrained; the potential for oncogenesis due to chromosomal integration or activation of oncogenes/inactivation of oncogene regulators; the generation of infectious viruses due to recombination; and concerns regarding instability that challenge production and storage.

The potentially hazardous nature of viral vectors has been observed in a number of gene therapy-related patient mortalities in various clinical trials. Patient complications to date have included the rejection of DNA carriers, resulting in immune response that have led to already one death, Jesse Gelsinger, who died in 1999 from a rare metabolic disorder during a gene-therapy clinical trial at the University of Pennsylvania (Branca 2005;Ledford 2007;Stolberg 1999;Wilson 2010). DNA constructs need to be optimized to carry minimal immunogenic components without compromising both efficient and longterm transgene expression. In this context, technology employing minimal immunological defined gene expression technology (MIDGE) represents a promising future alternative to conventional plasmids in terms of biosafety, improved gene transfer, potential bioavailability, minimal size and low immunogenicity associated with these chemically engineered miniDNA vectors. MIDGEs are non-viral, lcc (linear covalently closed) miniplasmids synthesized *in vitro* via a patented chemical modification of linear open (lo). These plasmids confer the advantage of a minimal coding sequence that relieves complications from expression of additional unwanted genes and CpG motifs that have a 20-fold higher occurrence in bacterial cells. MIDGE lcc plasmid transfection expression was reported to increase luciferase transgene expression from 2.5 to 17 fold *ex vivo* compared to circular covalently closed (ccc), isogenic forms in a tissue-dependent manner (Schakowski et al. 2007) . In addition, the mean numbers of MIDGE vector molecules per cell was also found to be significantly higher, suggesting that linear lcc plasmids transfect cells more efficiently. MIDGE technology has already been applied, with promising results, to the development of a Leishmania DNA vaccine and a colon carcinoma treatment. Here, therapy was based on IL-2 delivery to specific tumor cell lines *ex vivo*, and revealed 2 to 4 fold higher relative transgene expression compared to the ccc control (Schakowski et al. 2001). In combination, the existing studies employing lcc DNA systems suggest that lcc DNA is superior to lo DNA in transfection efficiency, and to ccc DNA in transgene expression. Purified TelN (closely related to Tel) was previously reported to generate lcc plasmids *in vitro* in (*pal*-related) *telRL*-dependent manner and was shown to successfully deliver EGFP to human embryonal kidney cells *in vitro,* as well as IL-12 in an untargeted manner to inhibit metastasis formation in B16F10C57BL/6 melanoma model mice *in vivo* (Heinrich et al. 2002).

At present, our laboratory is involved in design, and construction of bacteriophageencoded recombination systems to generate linear covalently closed (lcc) DNA minivectors. Conditional expression recombinase systems *in vivo* in *E. coli* provide a one step production system for step minivectors from specialized parental plasmids. The exploitation of this recombination system allows us to generate lcc miniplasmids that should have a preferential safety profile, preventing viable vector-chromosome, single recombination products in mammalian cells. In theory, the integration of covalently closed linear exogenous DNA into double-stranded (ds) DNA breaks is unlikely due to the unavailability of terminal ends, and a vector single recombination event into a target

Nanomedicine Based Approaches to Cancer Diagonsis and Therapy 527

Cationic liposomes are one of the most efficient, and among the most widely used non-viral vector systems. They are composed of positively charged lipid bilayers that can be complexed to DNA through electrostatic interactions resulting in complexes, termed lipoplexes. The application of cationic liposomes to gene therapy was first described in 1987 by Felgner (Felgner et al. 1987). The lipoplexes are generally composed of a positively charged lipidic component (see Figure 2), such as dioleylpropyltrimethylammonium chloride, dioleoyl triethylammonium propane (DOTAP), or dimethylaminoethane carbamoyl cholesterol (DC-Chol), that is capable of complexing and condensing the DNA (Giatrellis et al. 2009). Most lipoplex formulations include a ''helper'' lipid, such as dioleoylphosphatidyl-ethanolamine (DOPE, as seen in Figure 2), or cholesterol that provides added stability to the lipoplexes and enhances DNA release from endosomal compartments. While transfection activity of lipoplexes is shown to vary depending on their DNA/cationic lipid ratio, it is necessary for them to be positively charged to interact with the negatively charge cell membranes and exhibit transfection activity. Lipoplex uptake occurs through endosome formation (via a number of mechnisms including caveosomes, clathrin dependent, etc.), followed by disruption of the endosomal membrane by fusion with, or incorporation of the lipoplex lipids resulting in

The advantages of liposomes in delivery system designs include their simplicity in preparation, ability to complex relatively large amounts of DNA, versatility for use with any size or type of DNA/RNA, ability to transfect non-dividing cells and overall stability (Dutta et al. 2010). In addition, lipids are non-immunogenic allowing for repeated administration without adverse immunologic reaction(Barron et al. 1999). The primary disadvantages of lipoplex delivery vectors include low tumor transfection efficiency and lack of tumor

In order to enhance gene transfection, based upon structure-activity relationship, various gemini surfactants (Gemini surfactants consist of two hydrophobic chains and two polar headgroups linked chemically by a spacer group) have been designed. In order to enhance gene transfection based upon structure-activity relationship, various gemini surfactants have been designed, synthesized and tested for gene delivery in our laboratory (Donkuru et al. 2010;Wang and Wettig 2011;Wettig et al. 2007a;Wettig et al. 2007b;Wettig and Verrall 2001). Recent reports have shown that obstacles can be overcome by exploiting receptormediated endocytosis for highly efficient internalization of ligands naturally employed by eukaryotic cells. Advances along this line include the conjugation of mAbs ("immunoliposomes"), ligands such as growth factors, or hormones to liposomes to confer targeting capability. Nanoparticle based anti-cancer gene/drug delivery first reached clinical trial in mid 1980s and the first nanomedicine Doxil® (liposomal encapsulated doxorubicin) was marketed in 1995. Other example of liposome-mediated drug delivery include daunorubicin (Daunoxome), which is also currently being marketed as liposome delivery systems. Numerous new nanoparticle based cancer gene therapy systems are under

Polyplexes differ from lipoplexes in that they are comprised of charged complexes of plasmid DNA and a cationic polymer, such as poly-L-lysine (PLL), polyethylenimine, polyamidoamine (starburst) dendrimers, and chitosan with a net positive charge (See Figure 2). Cationic polymers differ from cationic lipids primarily in that they do not contain

**3.1.2 Cationic liposomes** 

DNA release.

specificity.

development.

**3.1.3 Polyplexes** 

chromosome should theoretically result in the disruption of the chromosome, killing the cell (see Figure 1).

Fig. 1. Plasmid Integration Events: A miniplasmid that undergoes a single recombination event with the host chromosome should be rare due to the removal of all elements except the gene of interest. Any integration events by: A. A lcc miniplasmid should result in chromosomal disruption that is likely lethal and cannot be replicated or segregated; B. A ccc miniplasmid can integrate without disrupting the host chromosome.

Non-viral gene delivery is emerging as a realistic alternative to the use of viral vectors with the potential to have a significant impact on clinical therapies. Synthetic vectors provide flexibility in formulation design and can be tailored to the size and topology of the DNA cargo and the specific route of vector administration, and can be delivered selectively to a specific tissue type through the incorporation of a targeting ligand. Compared to viral vectors, synthetic vectors are potentially less immunogenic, relatively easy to produce in clinically relevant quantities, and are associated with fewer safety concerns (Liu and Huang 2002). As a result of these advantages, their use has rapidly moved from transfection of cell cultures to clinical cancer gene therapy applications.

Synthetic vectors designed for parenteral administration encompass a wide range of formulations. These include unmodified (naked) DNA, which is designed for direct intratissue injection, cationic polymer–DNA complexes, cationic lipid–DNA complexes, and cationic polymer–lipid–DNA ternary complexes (lipopolyplexes), which are mostly aimed at systemic administration.

#### **3.1.1 Naked DNA**

The simplest employed method of transgene delivery is the injection of naked DNA. It shows very little dissemination and transfection at distant sites following delivery and can be re-administered multiple times into mammals (including primates) without inducing an antibody response against itself (i.e., no anti-DNA antibodies generated) (Wolff and Budker, 2005). The simplicity of this approach is offset by serious limitations such as inefficient uptake of the therapeutic gene into the target cells and rapid clearance of the DNA from the circulation. Direct *in vivo* gene transfer with naked DNA was first demonstrated by efficient transfection of myofibers following injection of mRNA or pDNA into skeletal muscle (Wolff JA, 1990). This novel approach was heralded as a superior method of *in vivo* transfection and its application was later advanced to confer high level expression in hepatocytes in mice by the rapid injection of naked DNA in large volumes into the tail vein (Zhang et al., 1999). This hydrodynamic tail vein (HTV) procedure has proven very useful not only to gene expression studies, but also more recently for the delivery of siRNA (Lewis et al., 2002; McCaffrey et al., 2002). Ultrasoundmediated eruption of polyethyleneglycol (PEG)-modified liposomes loaded with naked plasmid-DNA is also a feasible and efficient technique for gene delivery (Negishi et al. 2010).
