**5. Chitosan structure and physicochemical behaviour**

The structure of chitin and chitosan correspond to those of poly [β(1→4)-2-acetamido-2 deoxy-d-glucopyranose] and poly[β(1→4)-2-amino-2-deoxy-d-glucopyranose], respectively (Figure 1). Chitosan is mainly manufactured from crustaceans (crab, krill and crayfish) primarily because a large amount of the crustacean exoskeleton is available as a byproduct of food processing. However, depending on the organism considered chitin can adopt polymorphic structures denominated alpha (α), beta (β) and gamma (γ) chitin (Jang et al., 2004).The polymorphism of chitin is due to different arrangements of chitin chains in the lamellas that constitute the crystalline portions. Alpha (α) chitin found in arthropods corresponds to an antiparallel chain packing at which intramolecular and intermolecular hydrogen bonding is favored. β-Chitin, typically extracted from squid pens, is less widely used although it can have higher reactivity than that of α-chitin. The parallel arrangement of the lamellas is responsible for a loose-packing fashion with weak intermolecular interactions. In the gamma (γ) chitin structure arrangements, beta and alpha occur, i.e., two lamellas in a parallel arrangement is intercalated by a lamella arranged in antiparallel packing (Roberts, 1992). The source from which chitosan is prepared is considered very important since chitosan derived from β-chitin exhibits higher reactivity than that derived from α-chitin (Kuritia et al., 1994; Shimojohay 1998).

In general the isolation of chitin from crustacean shell waste consists of three basic steps: demineralization (DM-calcium carbonate and calcium phosphate separation), deproteinization (DP-protein separation), and decolorization (DC-removal of pigments). These three steps are the standard procedure for chitin production (No et al., 1989). Chitosan is obtained after hydrolysis of the acetamide groups of chitin. However in the commercialized samples both units are commonly found, since chitosans having high deacetylation degrees (DA> 99%) are obtained only through of successive hydrolysis with strong bases as KOH and NaOH, and the degree of deacetylation is strongly dependent of the alkali concentration and temperature (Figure.1). The source of chitin and the deacetylation process can change dramatically the properties of the final product and the

The low stability, low buffering capacity and low cell-specificity have also hindered its clinical applications. However, as a nature resource-based polysaccharide, chitosan has more functional groups that can be chemically modified than other cationic polymers, thus has many more potential chemical derivatives to overcome the deficiencies. Chitosan has been experimentally modified using hydrophilic, hydrophobic, pH-sensitive, thermosensitive and cell-specific ligand groups for enhancement of transfection efficiency (Ishii et al., 2001). The degree of deacetylation (DDA) and the molecular weight (MW) of chitosan or its derivatives, can affect the ultimate transfection efficiency. Most chitosan/DNA complexes are highly deacetylated (above 80%), because chitosan with a high degree of deacetylation exhibits an increased DNA binding efficacy (Kiang et al., 2004). Through chain entanglement, chitosan with a higher MW (longer chain length) can become more readily enmeshed with free DNA, once the initial electrostatic interaction has occurred. But it will also delay the disassociation of chitosan and DNA (Huang et al., 2005). Consequently, low MW chitosan requires a higher charge ratio to stably condense DNA for the same DDA, and a lower DDA requires a higher charge ratio to stably condense DNA at equal MW (Lavertu et al., 2006). The charge ratio for minimum complexation can be

The structure of chitin and chitosan correspond to those of poly [β(1→4)-2-acetamido-2 deoxy-d-glucopyranose] and poly[β(1→4)-2-amino-2-deoxy-d-glucopyranose], respectively (Figure 1). Chitosan is mainly manufactured from crustaceans (crab, krill and crayfish) primarily because a large amount of the crustacean exoskeleton is available as a byproduct of food processing. However, depending on the organism considered chitin can adopt polymorphic structures denominated alpha (α), beta (β) and gamma (γ) chitin (Jang et al., 2004).The polymorphism of chitin is due to different arrangements of chitin chains in the lamellas that constitute the crystalline portions. Alpha (α) chitin found in arthropods corresponds to an antiparallel chain packing at which intramolecular and intermolecular hydrogen bonding is favored. β-Chitin, typically extracted from squid pens, is less widely used although it can have higher reactivity than that of α-chitin. The parallel arrangement of the lamellas is responsible for a loose-packing fashion with weak intermolecular interactions. In the gamma (γ) chitin structure arrangements, beta and alpha occur, i.e., two lamellas in a parallel arrangement is intercalated by a lamella arranged in antiparallel packing (Roberts, 1992). The source from which chitosan is prepared is considered very important since chitosan derived from β-chitin exhibits higher reactivity than that derived

In general the isolation of chitin from crustacean shell waste consists of three basic steps: demineralization (DM-calcium carbonate and calcium phosphate separation), deproteinization (DP-protein separation), and decolorization (DC-removal of pigments). These three steps are the standard procedure for chitin production (No et al., 1989). Chitosan is obtained after hydrolysis of the acetamide groups of chitin. However in the commercialized samples both units are commonly found, since chitosans having high deacetylation degrees (DA> 99%) are obtained only through of successive hydrolysis with strong bases as KOH and NaOH, and the degree of deacetylation is strongly dependent of the alkali concentration and temperature (Figure.1). The source of chitin and the deacetylation process can change dramatically the properties of the final product and the

determined by agarose gel electrophoresis.

from α-chitin (Kuritia et al., 1994; Shimojohay 1998).

**5. Chitosan structure and physicochemical behaviour** 

deacetylation in alkaline medium leads to the depolimerization (Domard & Rinaudo, 1983; Tolaimate et al., 2000). However it has been reported that chitin extracted from squid pens can be hydrolyzed under conditions that it allows obtaining chitosans of high molecular weight ( Tolaimate et al., 2003).

Fig. 1. Chemical structure of chitin and chitosan. In the 2-amino-2-deoxy-d-glucopyranose ring is shown the commonly used numbering for the carbon atoms.

The homopolymer is a weak base with a pKa value of the D-glucosamine residue of about 6.3 and is therefore insoluble at neutral and alkaline pH values. In acidic mediums, the amine groups will be positively charged, conferring to the polysaccharide a high charge density. As in all polyelectrolytes, the dissociation constant of chitosan is not constant, but depends on the degree of dissociation at which it is determined. The pka value can be calculated using the Katchalsky's equation (Roberts, 1992).

$$\mathbf{Pk\_a = pH + \log\left[ (1 - \alpha)/\alpha \right] = pK\_0 - \varepsilon \Delta \psi(\alpha)/kT}$$

Where Δψ is the difference in electrostatic potential between the surface of the polyion and the reference, α is the degree of dissociation, k is Boltzman's constant, T is the temperature and ε is the electron charge. Extrapolation of the pKa values to α = 1, where the polymer is uncharged and hence the electrostatic potential becomes zero, makes possible the value of the intrinsic dissociation constant of the ionizable groups, pK0, to be determined. The value obtained does not depend of the degree of *N*-acetylation, whereas the pKa value is dependent on this parameter, since the electrostatic potential will be varied depending of amount of the free amino groups. The pK0 value is called the intrinsic pKa of the chitosan. However chitosans of low molecular weight having degrees of deacetylation higher than 0.4 are also easily soluble in weakly acidic solvents such as acetic acid and formic acid (Lee et al., 1995).

The physicochemical behavior in aqueous solution is highly dependent of pH and degree of acetylation and has received more attention only recently. Bertha et al. working on chitosans from 95 to 175kDa have recently determined the radius of gyration of chitosan (RG) (Bertha et al., 1998a; 2002b). The RG is an alternative measure of the size of the polymer chain and it can be measured by light scattering measurements. RG express the square mean radius of each one of the elements of the chain measured from its center of gravity. The study established the relationship between the molecular weight and radius of gyration (RG*)* of chitosan in aqueous solution, and the author indicated that chitosan behaved more like a Gaussian coil instead of the worm-like chain model found in common polyelectrolytes. At the same time the presence of *N*- acetyl groups on the chitosan backbone imparts hydrophobic properties. Schatz et al. (Schatz et al., 2003) have studied homogeneous series of chitosans with different degrees of acetylation and almost the same degree of

Chitosan-DNA/siRNA Nanoparticles for Gene Therapy 463

Many strategies have been deployed to improve transfection efficiency, taking into account the biological steps involved in gene delivery. Modifications of chitosan structure to impart properties to NPs, such as to increase endosomal escape (Jiang et al., 2010; Yu et al., 2010), attaching of ligands to mediate cell internalization or to promote the nuclear entry of DNA, are among the most common ways. Figure 2 shows representative structures from these chitosan derivatives tested as carriers for gene therapy. A variety of nucleophilic reactions targeting the groups linked to the glucopyranose ring have been employed to improve the

Poly (ethylene glycol) (PEG) has been widely used for attaching to chitosan due to its hydrophilicity and biocompatibility. In general, the terminal hydroxyl group of methoxy poly(ethylene glycol) is modified to generate PEG derivatives able to promote nucleophilic displacements targeting the amino groups of chitosan (Harris et al., 1984; Aiba et al, 1993; Saito et al., 1997; Ouchi et al., 1998). Chitosan nanospheres modified by introducing PEG5000 chains to amine groups were more stable during lyophilization (Leong et al., 1998). These chitosan-DNA nanospheres were effective in tranfecting 293 cells but not HeLa cells, and

Polymers can also be attached to the chitosan main chain using different routes. Poly(vinyl pyrrolidone) (PVP) was also grafted on galactosylated chitosan (GCPVP) and displayed improved physicochemical properties over unmodified chitosan (Park et al., 2003). PVP with a single terminus carboxylic group was coupled to galactosylated chitosan via formation of an amide bond between the amino *complex* group of GC and the terminal carboxyl group of the PVP. The terminal carboxyl group of PVP was activated by the *N*-hydroxysuccinimide (NHS)/EDC. The binding strength of GCPVP 10k/DNA was superior to that of GCPVP 50k/DNA, which was attributed to its higher flexibility because of its smaller size. However, DNase I protection of GCPVP 10k/DNA complex was inferior to that of GCPVP 50k/DNA. The DNA-binding property was shown to be dependent on the MW of chitosan and the composition of PVP (Park et al., 2003). The reaction of chitosan with methoxy poly(ethylene glycol) iodide (mPEG, Mn 2 kDa) in an alkalinized suspension was recently used by Yu *et al.* to attach PEG (Yu et al., 2010). This derivative was subsequently modified by attaching poly(ethylenimine) to the amino groups (Figure 2). Other approaches successfully employed to attach PEI to chitosan were an imine reaction between periodate-oxidized chitosan and amine groups of low MW PEI (Jiang et al., 2007) and the cationic polymerization of aziridine in the presence of water-

A series of new degradable cationic polymers composed of biocompatible chitosan backbones and poly((2-dimethyl amino) ethylmethacrylate) (P(DMAEMA)) side chains were recently synthesized via atom transfer radical polymerization (ATR) (Ping et al., 2010). This synthesis was carried out by introducing alkyl halide initiators onto chitosan, followed by the reaction with DMAEMA. Bromoisobutyryl-terminated chitosan (CS-Br initiators) was prepared via the reaction of primary amines of chitosan with carboxyl group of 2-bromo-2 methylpropionic acid (BMPA), which was previously converted into reactive esters (succinimidyl intermediates) in the presence of EDAC and NHS. The reactive esters underwent nucleophilic substitution reactions with the amine groups of chitosan to form a stable amide linkage and produce the resultant CS-Br initiators for DMAE polymerization

The activation of carboxylic groups is one of the most commonly used procedures to attach different ligands and peptides to chitosan chain. Arginine-modified trimethylated chitosans

tranfection efficiency was not affected by PEG derivatization.

soluble oligo-chitosan ( Wong et al., 2006).

(Ping et al., 2010).

properties of chitosan.

polymerization in ammonium acetate buffer. Their results indicate that the aqueous solution behavior depends only on the degree of acetylation (DA). Three distinct domains of DA were defined and correlated to the different behaviors of chitosans: (i) a polyeletrolyte domain for DA below 20%; (ii) a transition domain between DA = 20% and 50% where chitosan loses its hydrophilicity; (iii) a hydrophobic domain for DAs over 50% where polymer associations can arise. Conformations of chitosan chains varying from 160 to 270kDa were studied by the calculations of the persistence lengths (L(p)). The average value was found to be close to 5 nm, in agreement with the wormlike chain model, but no significant variation of L(p) with the degree of acetylation was noticed. Pa et al. (Pa & Yu, 2001) have also reported that that the particle sizes of chitosan molecules in dilute acetic acid/water solutions increased with decreasing pH value. SLS data also demonstrated that the second virial coefficient (*A2*) increased with decreasing pH value, suggesting that solubility of chitosan in water increased with increasing acetic acid concentration. Signini et al. (Signini et al., 2000) have also shown that acid-free aqueous solutions of chitosan hydrochloride of variable ionic strengths (0:06 M ≤ µ≤ 0.3 M) are free of aggregation as evaluated by the values of the Huggins constants (0.31 ≤ *k* ≤ 0.63).

As other polysaccharides the biodegradation and biocompatibility are important properties of chitosan making it an attractive polymer for a variety of biomedical and pharmaceutical applications. Besides the degradation by chitinases (Hung et al., 2002), chitosanases ( Kuroiwa et al., 2003), papain (Kumar et al., 2004; Lin et al., 2002; Muzzarelli et al., 2002; Terbojevich et al., 1996) and other proteases (Kumar et al., 2004), partially acetylated chitosan may be also degraded by lyzosymes of the human serum (Varurn et al., 1997), by oxidative-reductive depolymerization (Mao et al., 2004) and by acid hydrolysis reactions (Lee et al., 1999). In the acid hydrolysis the protonation of the glycosidic oxygen is recognized as the first step of the mechanism, which leads to formation of a cyclic carbonium –oxonium ion, yielding after the addition of water the reducing sugar end group (Sinnott et al., 1990; Yip & Withers, 2004). Besides enzymatic and acid hydrolysis the alkaline treatment with ultrasonication can be used to obtain either chitosan of decreasing molecular weight (Tang et al., 2003) or oligomers having a few glucosamine units (Tsaih et al., 2003).

#### **6. General strategies for chitosan modification**

In the chitosan structure two groups are particularly susceptible to react through nucleophilics attacks, i.e., the free amine and/or acetamide groups, and the hydroxyl groups linked to the glucopyranose ring. The hydroxyl groups can be modified by substitution of the hydrogen atoms but their reactivities are smaller than that of the amino group. Various procedures targeting the hydroxyl groups employ a sequence of protection/deprotection reactions aiming to obtain derivatives with a well defined structure (Kuitra, 2001). On the other hand under appropriated conditions a variety of other reactions can be easily conducted to selectively modify the free amine groups. The literature presents a wide range of procedures to target the amine group aiming to improve the properties of chitosan for a particular purpose. The modifications include those aiming the separation technologies of chiral molecules (Franco et al., 2001), recovery of metals (Guibal, 2004; Varma et al., 2004), antimicrobial activity (Rabea et al., 2003), anti tumoral carriers (Kato et al., 2004), biomedical applications (Berge et al., 2004a; 2004b) and vectors for gene therapy (Janes et al., 2001; Sinha et al., 2004; Liu et al., 2002; Borchard et al., 2001). Kumar et al (Kumar et al., 2004) and Kurita (Kurta, 2001) reviewed the procedures for the modification of chitosan.

polymerization in ammonium acetate buffer. Their results indicate that the aqueous solution behavior depends only on the degree of acetylation (DA). Three distinct domains of DA were defined and correlated to the different behaviors of chitosans: (i) a polyeletrolyte domain for DA below 20%; (ii) a transition domain between DA = 20% and 50% where chitosan loses its hydrophilicity; (iii) a hydrophobic domain for DAs over 50% where polymer associations can arise. Conformations of chitosan chains varying from 160 to 270kDa were studied by the calculations of the persistence lengths (L(p)). The average value was found to be close to 5 nm, in agreement with the wormlike chain model, but no significant variation of L(p) with the degree of acetylation was noticed. Pa et al. (Pa & Yu, 2001) have also reported that that the particle sizes of chitosan molecules in dilute acetic acid/water solutions increased with decreasing pH value. SLS data also demonstrated that the second virial coefficient (*A2*) increased with decreasing pH value, suggesting that solubility of chitosan in water increased with increasing acetic acid concentration. Signini et al. (Signini et al., 2000) have also shown that acid-free aqueous solutions of chitosan hydrochloride of variable ionic strengths (0:06 M ≤ µ≤ 0.3 M) are free of aggregation as

As other polysaccharides the biodegradation and biocompatibility are important properties of chitosan making it an attractive polymer for a variety of biomedical and pharmaceutical applications. Besides the degradation by chitinases (Hung et al., 2002), chitosanases ( Kuroiwa et al., 2003), papain (Kumar et al., 2004; Lin et al., 2002; Muzzarelli et al., 2002; Terbojevich et al., 1996) and other proteases (Kumar et al., 2004), partially acetylated chitosan may be also degraded by lyzosymes of the human serum (Varurn et al., 1997), by oxidative-reductive depolymerization (Mao et al., 2004) and by acid hydrolysis reactions (Lee et al., 1999). In the acid hydrolysis the protonation of the glycosidic oxygen is recognized as the first step of the mechanism, which leads to formation of a cyclic carbonium –oxonium ion, yielding after the addition of water the reducing sugar end group (Sinnott et al., 1990; Yip & Withers, 2004). Besides enzymatic and acid hydrolysis the alkaline treatment with ultrasonication can be used to obtain either chitosan of decreasing molecular weight (Tang et al., 2003) or oligomers

In the chitosan structure two groups are particularly susceptible to react through nucleophilics attacks, i.e., the free amine and/or acetamide groups, and the hydroxyl groups linked to the glucopyranose ring. The hydroxyl groups can be modified by substitution of the hydrogen atoms but their reactivities are smaller than that of the amino group. Various procedures targeting the hydroxyl groups employ a sequence of protection/deprotection reactions aiming to obtain derivatives with a well defined structure (Kuitra, 2001). On the other hand under appropriated conditions a variety of other reactions can be easily conducted to selectively modify the free amine groups. The literature presents a wide range of procedures to target the amine group aiming to improve the properties of chitosan for a particular purpose. The modifications include those aiming the separation technologies of chiral molecules (Franco et al., 2001), recovery of metals (Guibal, 2004; Varma et al., 2004), antimicrobial activity (Rabea et al., 2003), anti tumoral carriers (Kato et al., 2004), biomedical applications (Berge et al., 2004a; 2004b) and vectors for gene therapy (Janes et al., 2001; Sinha et al., 2004; Liu et al., 2002; Borchard et al., 2001). Kumar et al (Kumar et al., 2004) and Kurita

evaluated by the values of the Huggins constants (0.31 ≤ *k* ≤ 0.63).

having a few glucosamine units (Tsaih et al., 2003).

**6. General strategies for chitosan modification** 

(Kurta, 2001) reviewed the procedures for the modification of chitosan.

Many strategies have been deployed to improve transfection efficiency, taking into account the biological steps involved in gene delivery. Modifications of chitosan structure to impart properties to NPs, such as to increase endosomal escape (Jiang et al., 2010; Yu et al., 2010), attaching of ligands to mediate cell internalization or to promote the nuclear entry of DNA, are among the most common ways. Figure 2 shows representative structures from these chitosan derivatives tested as carriers for gene therapy. A variety of nucleophilic reactions targeting the groups linked to the glucopyranose ring have been employed to improve the properties of chitosan.

Poly (ethylene glycol) (PEG) has been widely used for attaching to chitosan due to its hydrophilicity and biocompatibility. In general, the terminal hydroxyl group of methoxy poly(ethylene glycol) is modified to generate PEG derivatives able to promote nucleophilic displacements targeting the amino groups of chitosan (Harris et al., 1984; Aiba et al, 1993; Saito et al., 1997; Ouchi et al., 1998). Chitosan nanospheres modified by introducing PEG5000 chains to amine groups were more stable during lyophilization (Leong et al., 1998). These chitosan-DNA nanospheres were effective in tranfecting 293 cells but not HeLa cells, and tranfection efficiency was not affected by PEG derivatization.

Polymers can also be attached to the chitosan main chain using different routes. Poly(vinyl pyrrolidone) (PVP) was also grafted on galactosylated chitosan (GCPVP) and displayed improved physicochemical properties over unmodified chitosan (Park et al., 2003). PVP with a single terminus carboxylic group was coupled to galactosylated chitosan via formation of an amide bond between the amino *complex* group of GC and the terminal carboxyl group of the PVP. The terminal carboxyl group of PVP was activated by the *N*-hydroxysuccinimide (NHS)/EDC. The binding strength of GCPVP 10k/DNA was superior to that of GCPVP 50k/DNA, which was attributed to its higher flexibility because of its smaller size. However, DNase I protection of GCPVP 10k/DNA complex was inferior to that of GCPVP 50k/DNA. The DNA-binding property was shown to be dependent on the MW of chitosan and the composition of PVP (Park et al., 2003). The reaction of chitosan with methoxy poly(ethylene glycol) iodide (mPEG, Mn 2 kDa) in an alkalinized suspension was recently used by Yu *et al.* to attach PEG (Yu et al., 2010). This derivative was subsequently modified by attaching poly(ethylenimine) to the amino groups (Figure 2). Other approaches successfully employed to attach PEI to chitosan were an imine reaction between periodate-oxidized chitosan and amine groups of low MW PEI (Jiang et al., 2007) and the cationic polymerization of aziridine in the presence of watersoluble oligo-chitosan ( Wong et al., 2006).

A series of new degradable cationic polymers composed of biocompatible chitosan backbones and poly((2-dimethyl amino) ethylmethacrylate) (P(DMAEMA)) side chains were recently synthesized via atom transfer radical polymerization (ATR) (Ping et al., 2010). This synthesis was carried out by introducing alkyl halide initiators onto chitosan, followed by the reaction with DMAEMA. Bromoisobutyryl-terminated chitosan (CS-Br initiators) was prepared via the reaction of primary amines of chitosan with carboxyl group of 2-bromo-2 methylpropionic acid (BMPA), which was previously converted into reactive esters (succinimidyl intermediates) in the presence of EDAC and NHS. The reactive esters underwent nucleophilic substitution reactions with the amine groups of chitosan to form a stable amide linkage and produce the resultant CS-Br initiators for DMAE polymerization (Ping et al., 2010).

The activation of carboxylic groups is one of the most commonly used procedures to attach different ligands and peptides to chitosan chain. Arginine-modified trimethylated chitosans

Chitosan-DNA/siRNA Nanoparticles for Gene Therapy 465

Chitosan-DNA gene delivery methods must achieve sufficient efficiency in the transportation of therapeutic genes across various extracellular and intracellular barriers. These barriers include interactions with blood components, vascular endothelial cells and uptake by the reticuloendothelial system. Furthermore, the degradation of therapeutic DNA by serum nucleases is a potential obstacle for functional delivery to target cells. DNA should escape from endosomes and traffic to enter the nucleus. Many factors, including the degree of deacetylation (DDA) and the molecular weight (MW) of the chitosan, the pH of the serum, the charge ratio (in some conditions, it equals the ratio of N/P, 'N': the content of Nitrogen atom in cationic polymer; 'P': the content of Phosohorus atom in DNA or RNA) of chitosan to DNA or RNA and the cell type can all affect the transfection efficiency of chitosan during each step of the process. The pKa value of chitosan is around 6.3-6.4, below which the protonated amines in the chitosan structure facilitate their binding to negatively charged DNA. Sato et al. showed the highest transfection efficiency can be obtained at pH 6.8 to 7.0. When pH of the transfection medium increases to 7.4, the transfection efficiency dramatically decreases due to the dissociation of the free plasmid from the complex (Sato et

Even if chitosan/DNA complexes display high transfection efficiency *in vitro*, their transfection efficiency *in vivo* may be low. Chitosan and its derivatives have become of great interest in the field of controlled release due to their favorable biocompatibility and biodegradability. Thiolated chitosan, which can be oxidized to form inter- and intramolecular disulfide bonds, allowing the crosslinking of chitosan, shows a significant enhancement of transfection over that of lipofectin (Lee et al., 2007). Chitosan microspheres for micro-encapsulation of adenoviral vectors has been achieved by ionotropic coacervation of chitosan, using bile salts as counter-anions (Lameiro et al., 2006). A 3-D scaffold composed of chitosan-gelatin complexes with entrapped DNA has been proposed as a

Although hydrophobic modification is not cell-specific, it can also enhance the attachment of complexes on cell surfaces and the subsequent cell uptake. Amphiphobic deoxycholic acidmodified chitosan oligosaccharide (DACO) nanoparticles showed superior gene condensation and high gene transfection efficiency, even in the presence of serum (Chae et al., 2005). After endocytosis, the endosome containing the complexes has to fuse with a lysosome to form an endolysosome. At this point, the complexes will meet a harsh acidic and multienzymatic environment. Nanocomplexes that are successfully protected against dissociation and degradation will finally escape from the endolysosome and enter the cytoplasm. PEI, a classic synthetic polymer with many amino groups to absorb protons (called a proton sponge mechanism), was found to have a better endolysosome buffering ability and caused a quicker release from the endolysosome in its intact form than did chitosan (Kim et al., 2005). The chemical modifications, such as urocanic acid (UA) (Kim et al., 2003), PEI-graft-chitosan (Wong et al., 2006), chitosan-graft-PEI (Jiang et al., 2007), poly(propyl acrylic acid) (PPAA) (Jones et al., 2003), trimethyl chitosan (Germershaus et al., 2008), have similar effects to PEI. Such modifications can be called pH-sensitive modifications that will not only enhance the escape of chitosan/DNA complexes from

endolysosome but also enhance the stability of complexes in different pH situations.

The dissociation of chitosan/DNA complexes and subsequent release of DNA is also a very important step for its rate-limiting effect (Schaffer et al., 2000). Hydrophobic modification,

**7. Chitosan-DNA delivery system** 

promoter of cartilage regeneration (Xia et al., 2004).

al., 2001).

labeled with folic acid have been prepared by activation of the acid group of arginine using EDC/NHS (Morris et al., 2010). The same procedure was utilized by Gao et al and it has proven to increase the transfection efficiency (Gao et al., 2008) and chitosan properties (Liu et al., 2004). A similar procedure was utilized to attach a short peptide (SP) (Sun et al., 2010) to chitosan. The peptide was further combined with GFP/luciferase reporter gene pDNA to form SP-CS/DNA complex. The NPs were able to transfect multiple cell lines, and the results revealed that, compared with CS, SP-CS could intensively augment transfection efficiency nearly to the level of Lipofectamine 2000 (Sun et al., 2010). Reactions targeting the hydroxyl groups are uncommon, however Sato et al. have prepared 6-Amino-6 deoxychitosan from 6-deoxy-6-halo-N-phthaloylchitosan via 6-azidation. The product had high stereoregularity because of the effective and regioselective reactions (Saito et al., 2004; Satoh et al., 2007).

Fig. 2. Chemical structures of chitosan and its derivatives; **1**. PEG (Harris et al., 1984); **2**. trimethylated (Zeng et al., 2007); **3**. folic acid (Mansouri et al., 2006; Fernandes et al., 2008); **4**.galactosylated (Park et al., 2001); **5**. Arginine (Morris et al., 2010); **6**. histidine; **7**. PEI and PEG grafts (Yu et al., 2010); **8.** 6-amino 6-deoxychitosan (Saito et al., 2004; Satoh et al., 2007); **9**. O-hydroethyl (Kwon et al., 2003); **10**. Phosphorylcholine (Case et al., 2009; Tiera et al., 2006); **11**. grafted PDMAEMA (Ping et al., 2010); **12**. PEI (Jiang et al., 2007; Wong et al., 2006).

#### **7. Chitosan-DNA delivery system**

464 Non-Viral Gene Therapy

labeled with folic acid have been prepared by activation of the acid group of arginine using EDC/NHS (Morris et al., 2010). The same procedure was utilized by Gao et al and it has proven to increase the transfection efficiency (Gao et al., 2008) and chitosan properties (Liu et al., 2004). A similar procedure was utilized to attach a short peptide (SP) (Sun et al., 2010) to chitosan. The peptide was further combined with GFP/luciferase reporter gene pDNA to form SP-CS/DNA complex. The NPs were able to transfect multiple cell lines, and the results revealed that, compared with CS, SP-CS could intensively augment transfection efficiency nearly to the level of Lipofectamine 2000 (Sun et al., 2010). Reactions targeting the hydroxyl groups are uncommon, however Sato et al. have prepared 6-Amino-6 deoxychitosan from 6-deoxy-6-halo-N-phthaloylchitosan via 6-azidation. The product had high stereoregularity because of the effective and regioselective reactions (Saito et al., 2004;

Fig. 2. Chemical structures of chitosan and its derivatives; **1**. PEG (Harris et al., 1984); **2**. trimethylated (Zeng et al., 2007); **3**. folic acid (Mansouri et al., 2006; Fernandes et al., 2008); **4**.galactosylated (Park et al., 2001); **5**. Arginine (Morris et al., 2010); **6**. histidine; **7**. PEI and PEG grafts (Yu et al., 2010); **8.** 6-amino 6-deoxychitosan (Saito et al., 2004; Satoh et al., 2007); **9**. O-hydroethyl (Kwon et al., 2003); **10**. Phosphorylcholine (Case et al., 2009; Tiera et al., 2006); **11**. grafted PDMAEMA (Ping et al., 2010); **12**. PEI (Jiang et al., 2007; Wong et al.,

Satoh et al., 2007).

2006).

Chitosan-DNA gene delivery methods must achieve sufficient efficiency in the transportation of therapeutic genes across various extracellular and intracellular barriers. These barriers include interactions with blood components, vascular endothelial cells and uptake by the reticuloendothelial system. Furthermore, the degradation of therapeutic DNA by serum nucleases is a potential obstacle for functional delivery to target cells. DNA should escape from endosomes and traffic to enter the nucleus. Many factors, including the degree of deacetylation (DDA) and the molecular weight (MW) of the chitosan, the pH of the serum, the charge ratio (in some conditions, it equals the ratio of N/P, 'N': the content of Nitrogen atom in cationic polymer; 'P': the content of Phosohorus atom in DNA or RNA) of chitosan to DNA or RNA and the cell type can all affect the transfection efficiency of chitosan during each step of the process. The pKa value of chitosan is around 6.3-6.4, below which the protonated amines in the chitosan structure facilitate their binding to negatively charged DNA. Sato et al. showed the highest transfection efficiency can be obtained at pH 6.8 to 7.0. When pH of the transfection medium increases to 7.4, the transfection efficiency dramatically decreases due to the dissociation of the free plasmid from the complex (Sato et al., 2001).

Even if chitosan/DNA complexes display high transfection efficiency *in vitro*, their transfection efficiency *in vivo* may be low. Chitosan and its derivatives have become of great interest in the field of controlled release due to their favorable biocompatibility and biodegradability. Thiolated chitosan, which can be oxidized to form inter- and intramolecular disulfide bonds, allowing the crosslinking of chitosan, shows a significant enhancement of transfection over that of lipofectin (Lee et al., 2007). Chitosan microspheres for micro-encapsulation of adenoviral vectors has been achieved by ionotropic coacervation of chitosan, using bile salts as counter-anions (Lameiro et al., 2006). A 3-D scaffold composed of chitosan-gelatin complexes with entrapped DNA has been proposed as a promoter of cartilage regeneration (Xia et al., 2004).

Although hydrophobic modification is not cell-specific, it can also enhance the attachment of complexes on cell surfaces and the subsequent cell uptake. Amphiphobic deoxycholic acidmodified chitosan oligosaccharide (DACO) nanoparticles showed superior gene condensation and high gene transfection efficiency, even in the presence of serum (Chae et al., 2005). After endocytosis, the endosome containing the complexes has to fuse with a lysosome to form an endolysosome. At this point, the complexes will meet a harsh acidic and multienzymatic environment. Nanocomplexes that are successfully protected against dissociation and degradation will finally escape from the endolysosome and enter the cytoplasm. PEI, a classic synthetic polymer with many amino groups to absorb protons (called a proton sponge mechanism), was found to have a better endolysosome buffering ability and caused a quicker release from the endolysosome in its intact form than did chitosan (Kim et al., 2005). The chemical modifications, such as urocanic acid (UA) (Kim et al., 2003), PEI-graft-chitosan (Wong et al., 2006), chitosan-graft-PEI (Jiang et al., 2007), poly(propyl acrylic acid) (PPAA) (Jones et al., 2003), trimethyl chitosan (Germershaus et al., 2008), have similar effects to PEI. Such modifications can be called pH-sensitive modifications that will not only enhance the escape of chitosan/DNA complexes from endolysosome but also enhance the stability of complexes in different pH situations.

The dissociation of chitosan/DNA complexes and subsequent release of DNA is also a very important step for its rate-limiting effect (Schaffer et al., 2000). Hydrophobic modification,

Chitosan-DNA/siRNA Nanoparticles for Gene Therapy 467

activity of chitosan/siRNA complexes. Haliza Katas et al. studied the difference between simple complexation, ionic gelation (siRNA entrapment) and adsorption of siRNA onto the surface of preformed chitosan nanoparticles. Ionic gelation gave the strongest stability and the most efficient gene-silencing activity among the three methods tested. For the involvement of tripolyphosphate (TPP) ions during the complexation of ionic gelation, pH became one of the factors that mostly affected the gene-silencing activity. The decrease of pH resulted in a reduction in the charge number of TPP, which subsequently led to the need for more TPP ions for cross-linking of the chitosan by electrostatic forces (Katas & Alpar, 2006). Rojanarata et al. reported that chitosan-thiamine pyrophosphate (TPP)-mediated siRNA enhanced green fluorescent protein (EGFP) gene silencing efficiency depended on the molecular weight and weight ratio of chitosan and siRNA. The chitosan-TPP-siRNA complex with the lowest molecular weight of chitosan (20 kDa) at a weight ratio of 80 showed the strongest inhibition of gene expression (Rojanarata et al., 2008). A novel study of chitosan/siRNA nanoparticles with fluorescent quantum dots was taken to silence HER2/neu and achieved desirable silencing effects (Tahara et al., 2008; Tan et al., 2007). In the field of controlled release, chitosan coating PLGA nanospheres with a high loading efficiency of siRNAs were found to reduce the initial burst of nucleic acid release and to prolong release at later stages, without changing the release pattern (Tahara et al., 2008). Kenneth A. Howard found that the chitosan-based system had the ability for endosome escape through the proton sponge mechanism, because the endosomolytic agent chloroquine did not increase the effect of RNA interference (Howard et al., 2006). In terms of *in vivo* administration of chitosan/siRNAs complexes, only a few studies are available. Nasal administration to silence EGFP expression of the endothelial cells distributed in the bronchioles of transgenic EGFP mouse model has been successfully achieved without showing any adverse effects (Howard et al., 2006). Cross-linking of hyaluronan and chitosan has proven to have a higher efficiency of transfection in ocular tissue over unmodified

chitosan (de la Fuente et al., 2008).

**9. Potential application of chitosan-DNA/siRNA nanoparticles** 

Gene therapy offers new possibilities for the clinical management of different disease conditions that are difficult to treat by traditional surgical or medical means. In the last decade, extensive improvements have been made to optimize gene therapy and have been tested on several disease conditions. The success of chitosan-DNA nanoparticles for delivery plasmid DNA to mucosal surfaces such as the oral and nasal mucosa has already shown (Bivas-Benita et al., 2003; Chen et al., 2004; Khatri et al., 2008). Oral delivery is most attractive due to easy administration. The oral delivery of peptide, protein, vaccine and nucleic acid-based biotechnology products is the greatest challenge facing the drug delivery industry. Mice were fed with plasmid pCMVβ (containing LacZ gene), whether it was wrapped by chitosan or no. The study demonstrated that oral chitosan-DNA nanoparticles can efficiently deliver genes to enterocytes, and may be used as a useful tool for gene transfer (Chen et al., 2004). Hepatitis B virus infection is a major global health concern and is the most common cause of chronic liver disease, new generation of HBV vaccines are urgently needed in order to overcome problems encountered with the immunization of immunocompromised people and more importantly with the potential of using active immunotherapy in treating chronic patients. DNA vaccines have the potential to eliminate many of the limitations of current vaccine technologies. Chitosan nanoparticles loaded with

such as deoxycholic acid modification (Lee et al., 1998), or 5β-cholanic acid modification (Yoo et al., 2005), can attenuate the electrostatic attractions between cationic polymers and anionic DNA. It is actually a contradiction between the stability and dissociation ability of complexes. A temperature-sensitive modification of poly(N-isopropylacrylamide) (PNIPAAm) can control the dissociation of PNVLCS (N-isopropylacrylamide/vinyl laurate copolymer with chitosan) complexes with DNA by a temporary reduction in the culture temperature to 20℃ (Sun et al., 2005)

The cytoplasm, a mesh-like network of microfilaments and microtubules, will limit the diffusion of complexes or DNA about 500-1000-fold. Adenovirus particles naturally bind to dynein and are actively transported towards the nuclear pore complexes once they are inside the cytoplasm. Prior to entry into the nucleus, the viruses dissociate into smaller structures and use their attached transport factors such as importins or karyopherins which have nuclear localization signals (NLS) to recognize the nuclear pore complex (NPC) (Whittaker & Helenius, 1998). Justin Hanes et al. used a new method called multiple particles tracking (MPT) to quantify the intracellular transport of non-viral DNA nanocarriers. They found that PEI/DNA complexes can accumulate in the perinuclear area through a subdiffusive transport, which is a combination of diffusive transport and active transport. This discovery is a dispute to the common belief that non-viral vectors go through the cell cytoplasm in a slow random way. Further investigation showed that actively transported complexes of PEI/DNA are in endosomes undergoing motor protein-driven movement guided by microtubules or physically associated with the motor proteins themselves (Suh et al., 2003). As to chitosan and its derivatives, however, few studies have examined how they pass through the highly structured cytoplasm and eventually enter into the nucleus.
