**2.1 DC-Chol**

350 Non-Viral Gene Therapy

The third approached is focused upon the use of non-viral vectors. Non-viral vectors are particularly suitable with respect to their simplicity of use, large-scale production and lack of specific immune response. Non-viral vectors can be grouped into three main categories: cationic lipids, cationic polymers, and peptides. In comparison with their viral counterparts, these vectors are currently considerably less efficient, but their well-defined physical and chemical composition coupled with their reduced immunogenicity and toxicity make them promising candidates for gene delivery (El-Aneed, 2004; Dass, 2002;

Another important method of gene delivery is lipofection, a method based on the use of cationic lipids/cationic liposomes for gene transfer (Templeton, Ed. 2010; Huang, et al., Eds.; 2005; Karmali & Chaudhuri, 2007; Tseng et al., 2009). Cationic lipids have many potential advantages and have thus been viewed favorably in comparison with other non-viral vectors, including the significant simplicity and ease of production, good repeatability and biodegradability, potential commercial value, and their wide range of clinical application

The cationic liposomes are formed using cationic lipids, comprising a wide range of chemical compounds with a common structural feature, namely, the presence of both positively charged hydrophilic and hydrophobic domains. Of the most significant interest are biodegradable cationic lipids of natural origin (Lv et al., 2006). Long-chain hydrocarbons, steroids, and diglycerides are used as the hydrophobic domains (Zhi et al., 2010). The cationic hydrophilic domain can be represented by one (monocationic lipids) or more (polycationic lipids) positively charged groups. Monocationic lipids are most often secondary, tertiary or quaternary derivatives of aliphatic or heterocyclic nitrogen bases. In polycationic lipids, natural or synthetic polyamines or amino acids are used as the hydrophilic domains. The stability and toxicity of cationic lipids in biological systems, are

determined by the type of bond connecting hydrophobic and hydrophilic domains.

In addition to cationic lipids, the zwitterionic helper lipid has a major impact upon the structure and activity of lipoplexes. A helper lipid can improve the ability of cationic liposomes to transfect cells. *In vitro* studies show that liposomes composed of an equimolar mixture of dioleoylphosphatidylethanolamine (DOPE) and cationic lipids (DOTMA, DOTAP) can mediate higher levels of transfection than those containing only the cationic lipid (Hui et al., 1996; Mok et al, 1997; Kerner et al., 2001). This fact has been attributed to the ability of DOPE tendency to undergo a transition from a bilayer to a hexagonal configuration under acidic pH, possibly facilitating fusion with, or destabilization of target membranes, in particular endosomal membranes (Zuidam &

Cholesterol initially used as a helper lipid form more stable but less efficient complexes than those containing DOPE *in vitro*. However, cholesterol containing lipoplexes have shown a higher rate of biological activity when compared to lipoplexes with DOPE, when these complexes were utilized *in vivo* (Liu et al., 1997; Sternberg et al., 1998; Smith et al., 1998; Simberg et al., 2003). The significant transfection activity attained was attributed to an improved cell binding and uptake of the lipoplexes promoted by the presence of cholesterol (Crook et al., 1998) and/or better stability of the lipoplex in serum (Simberg et al., 2003). In 1991, Gao et al. reported the synthesis and application of the cholesterol-based cationic

combined with DOPE to transfect mammalian cells (Gao & Huang, 1991). Since then, considerable endeavors have been made in the synthesis of steroidal cationic lipids, due to


Verma & Weitzman, 2005).

Barenholz, 1998; Zuidam et al., 1999).

′*,N*′

lipid 3-β-[*N*-(*N*

and safety.

Cationic lipid DC-Chol (**1a**) was originally synthesized by Gao et al. (Gao & Huang, 1991) and is available commercially. The DC-Chol was formulated as a cationic liposome with the helper lipid DOPE which promotes the fusion of lipoplexes with the cell membrane resulting in an increase of the DNA transfection efficiency (Zuidam & Barenholz, 1998; Lin et al., 2003).

Cationic liposomes DC-Chol/DOPE have been used extensively both *in vitro* and *in vivo,* displaying high transfection efficiencies (TE) (Litzinger et al., 1996; Song et al., 1997; Porter et al., 1998). There are documented reports that cationic liposomes DC-Chol/DOPE work well in various cell lines e.g. A431 human epithermoid carcinoma cells, A549 human lung carcinoma cells, L929 mouse fibroblast cells, YPT minipig primary endothelial cells (Gao & Huang, 1991), COS-7 cells, CFNPE-9o and 16HBE14o epithelial cell lines (Caplen et al., 1995), SKnSH and the primary rat neuronal cells (Ajmani et al.,1999), glioma cells (Esposito et al., 2003).

An experimental study of DNA compaction with the liposomes DC-Chol/DOPE that covered the whole range of mixed lipid composition and several lipid/DNA charge ratios was published (Rodriguez-Pulido et al., 2008; Munoz-Ubeda et al., 2010). A series of experimental techniques (electrophoretic mobility, SAXS, and fluorescence anisotropy), together with a theoretical aggregation-disaggregation model, has attested to the fact that DC-Chol/ DOPE cationic liposomes, with an average hydrodynamic diameter of (120+/-10) nm, properly condense and compact DNA and the liposomes composition is a key factor pertaining to the properties and structure of the resulting lipoplex.

Non-Viral Gene Delivery Systems Based on

**2.2 Monocationic lipids** 

al., 1998).

Cholesterol Cationic Lipids: Structure-Activity Relationships 353

The basic structure of the cholesterol-based cationic lipids used in gene therapy includes four functional domains: 1) Appositively charged headgroup capable to bind with the negatively charged phosphate group of nucleic acid; 2) a hydrophobic cholesterol anchor, which interacts with the cellular membrane; 3) a spacer group; 4) a linker group, which

In order to estimate the contribution of each functional domain into the efficacy of DNA delivery and cytotoxicity, various types of cholesterol cationic lipids were synthesized. In the first investigation of the structure-activity relationship, it was revealed that in order to achieve an efficient transfection, the tertiary ammonium group must be connected to the

In further studies the 3-deoxycholesterol cationic derivatives **2a-с** were synthesized, which are shown to be more efficient than DC-Chol (**1a**) or lipid **1b** (Takeuchi et al. 1996). It was observed that introduction of the ethyl or propyl groups into the cationic head results in the decrease of the transfection efficiency (TE). The value of the surface charge (ζ-potential) for the liposomes based on lipids **1a,b, 2a-с, 3а** and their analogs was in positive correlation with TE of the HeLa, COS-7, and NIH 3T3 cells. Cationic liposomes **2а**/DOPE, having the highest ζ-potential, demonstrated the highest TE in all cell lines tested. Furthermore, the activity of the liposomes derived from the compounds **2с** and **3а** having the lowest ζ-potential, was less than 20% of the activity of lipid **2а** (Takeuchi et al., 1996). Moreover, it was found that to achieve an efficient transfection, the size of the complexes should be neither smaller (<400 nm) nor larger (>1.4 μm) (Kawaura et

Lipid **2d,** containing the 2-hydroxyethyl group was found to be more active than the compound **2a** with dimethylamine group, both in the presence and absence of serum (Okayama et al., 1997). Further modifications of the lipid structure by introduction of additional 2-hydroxyethyl groups (compounds **2e-i**) into the cationic head resulted in the collapse of transfection activity. Similarly, the derivative **2f** containing the primary amino group was almost not active. The TE of the lipid **2d** was 2-fold higher in comparison with the activity of lipids **2a** and **2h** (Hasegawa et al, 2002). Based on fluorescence resonance energy transfer (FRET) it was observed that DNA is released differently from the lipoplexes by means of anionic liposomes. Furthermore, both the release rate and the amount of unbound DNA have a positive correlation with the TE. The transfection efficiency of nanoparticles composed of the lipid **2d** was 11.5-fold higher than the TE of DC-Chol (**1а**) and was comparable with Lipofectamine 2000, DMRIE-C and Tfx-20 (Hattori et al., 2007). The size of the nonoplex was 290 nm, and the highest TE *in vitro* was observed for the nitrogen/phosphate (N/P) ratio of 3, when the nanoplex was positively charged. The *in vivo* delivery directly into the tumor demonstrated that the optimum TE corresponds to the N/P ratio of 1; meanwhile the size of nanocomplexes was around 145 nm, and ξ-potential was negative (-16.9 mV). After the intravenous, intramuscular and

peroral administration, there was no transgene expression detected in any organ.

Lipids containing ester (**3b,c**), ether (**3d-f**) and urethane (**1b**) bonds were synthesized (Ghosh et al., 2000; Ghosh et al., 2002) in order to study the influence of linker type on the TE. It was subsequently demonstrated, that the presence of the urethane-based linkage in the compound **1b** led to the 6-fold decrease of the TE, in comparison to the ether lipid **3d**. However, in our opinion, this kind of comparison is fallacious due to presence of the different type of cationic head in the structure of the lipids **1b** and **3d**. Cholesterol lipids

cholesterol by a short spacer *via* the ester or urethane bond. (Farhood et al., 1992).

connects the positively charged head and the hydrophobic domain.

Commercially available, DC-Chol is widely used for the development of novel approaches for DNA delivery, i.e. as a component of cationic solid lipid nanoparticles (SLNs) (Choi et al., 2008). The SLNs for gene delivery composed of DC-Chol, DOPE, and Tween 80 with TC (tricaprin) as a core in various ratios were produced by the melt homogenization method. The SLNs were bound to the cellular membrane 10 min after transfection, and translocated to the cytosol 60 min later. After 24 h, the SLNs were detectable in the nucleus and cytosol. SLNs showed high transfection efficiency in comparison with commercially available Lipofectin. In fact, PCR analysis indicated that SLNs prolonged the mRNA expression of the plasmid in various organs for up to 5 days. The SLNs-mediated transfection of the p53 gene resulted in the efficient reestablishment of wild-type p53 function in lung cancer cells and restored the apoptotic pathway (Choi et al., 2008).

The use of the most thoroughly investigated cationic lipids DC-Chol, DOTAP, and dimethyldioctadecylammonium (DDA) as vaccine delivery systems to form an antigen depot at the site of injection (SOI) and to induce immunological responses *in vivo* was reported (Henriksen-Lacey et al., 2011). DC-Chol, DOTAP, and DDA liposomes incorporating immunomodulating trehalose dibehenate (TDB) were prepared. DC-Chol/TDB liposomes were stable under storage and were retained at a significantly better degree at the SOI, with nearly 40% of the original dose still detectable 14 days p.i, in comparison with DOTAP/TDB liposomes. With regards to the depot effect at the SOI, the formulations were able to cause antigen retention between the range of 59 and 79% of the antigen dose recovered one day p.i.

The key step in lipid-mediated DNA delivery may be the structural changes of lipid carriers resulting in DNA release (Tarahovsky et al., 2004; Koynova et al., 2006; Hoekstra et al., 2007). In recent times, it was shown that multicomponent lipoplexes are superior in transfection with respect to the binary ones usually employed for gene delivery (Caracciolo et al., 2005a; 2005b; 2006). For instance, the four-component lipid system incorporating cationic lipids DOTAP, DC-Chol and neutral helper lipids dioleoylphosphocholine (DOPC) and DOPE transfer DNA into mouse fibroblast (NIH 3T3) and tumoral myofibroblast-like (A17) cell lines more efficiently than the thoroughly studied DOTAP/DOPC and DC-Chol/DOPE cationic liposomes separately. To answer the question concerning how TE will change with an increasing number of lipid components, the multicomponent lipoplexes were studied incorporating three to six lipid species simultaneously and the TE was then evaluated with respect to mouse fibroblast (NIH 3T3), ovarian (CHO) and tumoral myofibroblast-like (A17) cell lines (Caracciolo et al., 2007). These multicomponent lipoplexes exhibited a much higher TE (about two orders of magnitude) than binary lipoplexes that are more commonly employed for gene delivery. Furthermore, a trend was discovered that the TE increases in correlation with the number of lipid components (with some exceptions as a result of lipid composition). This discovery may be related to the higher fusogenicity and compatibility of vesicles composed of several lipid components with respect to single lipids (Caracciolo et al., 2007). The existence of different regimes of stability was demonstrated for these multicomponent lipoplexes: the most efficient lipoplexes exhibited intermediate 'optimal stability'. To this end, lipoplexes DOTAP/DOPC-DNA were the least resistant mixture to disintegration; DC-Chol/DOPE/DOPC-DNA, the most resistant mixture to disintegration; and DOTAP/DC-Chol/DOPC/DOPE-DNA, the mixture exhibiting an intermediated behavior and characterized by a high TE. The extent of DNA release estimated by electrophoresis was in total concurrence with the structural stability of lipoplexes revealed by SAXS and TE (Caracciolo et al., 2007).
