CHAPTER 1: Functionalization of Cationic Polymers for Drug Delivery Applications

Cationic polymers can be defined as macromolecules that bear positive charges, which can be either intrinsically present in the polymer backbone and/or in the side chains. Most cationic polymers possess primary, secondary or tertiary amine functional groups that can be protonated. They also differ widely in their polymeric structure (linear, branched, hyperbranched and dendrimer-like) and can be further differentiated by the placement of the positive charges (backbone or side chains). The cationic polymers that will be discussed in this chapter are divided into three categories according to their origin: natural, semi-synthetic and synthetic (Figure 1.1). This chapter will focus only on the most prominent examples which have been shown to have applications in drug delivery rather than trying to include all existing cationic polymers.

Natural cationic polymers are derived from renewable sources and possess inherent positive charges. They are biodegradable and often possess low immunogenicity and low toxicity. Numerous natural cationic polymers have functional groups like carboxylic acid groups that can be further modified to carry therapeutic molecules.

Gelatine is a thermally denatured collagen extracted from porcine skin or bovine bone and is commonly used for pharmaceutical and medical applications because of its biodegradability.^1–4  Being categorized as a safe excipient by the US Food and Drug Administration (FDA), gelatine has shown great promise as a component of biomaterials in many medical applications.^5,6  For example, gelatine nanoparticles have been successfully utilized for non-viral plasmid DNA delivery⁷  and cationic gelatine plasmid DNA polyplexes, i.e. complexes formed by the electrostatic interactions of positively charged polymer molecules and negatively charged DNA, were applied for transfection studies on monocyte-derived immature dendritic cells.⁸  The mode of action of non-viral vectors for gene and RNA delivery will be discussed in detail in the next chapters of this book and will therefore not be examined in depth here. In contrast to other cationic polymers, gelatine also possesses carboxyl groups and therefore can have an overall negative charge, depending on the pH of the environment. The isoelectric point of gelatine at physiological pH can be modified during its extraction to yield either negatively charged acidic gelatine using alkaline treatment (classified as B) or positively charged basic gelatine (denoted as A) by acidic treatment. This differentiation is necessary because the extraction process using a base leads to hydrolysis of the amide groups of glutamine and asparagine residues, which increases the content of carboxylic groups in the polymer. As a result of this treatment the isoelectric point of gelatine is lowered (IEP=4.7–5.4) while the acidic extraction does not change the intrinsic properties of the collagen (IEP=6–9). Furthermore, aminated gelatine can be prepared in a one-pot reaction using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) and diamine. It has been demonstrated that this additional modification technique leads to improved release control in the delivery of acidic peptide/protein drugs.⁹ 

This group of cationic polymers includes all those natural polymers that require further modification in order to acquire a cationic character. Therefore, they differ from the biopolymers with inherent cationic properties and those cationic polymers that are produced artificially using polymerization methods. Such polymers often retain their biodegradability, while the introduction of positive charges leads to increased cytotoxicity and therefore decreased biocompatibility.

Chitosan is a copolymer consisting of statistically distributed N-acetylglucosamine and d-glucosamine.¹⁰  Deacetylation of chitin, the second most abundant polysaccharide in nature, with concentrated alkali solution at elevated temperatures leads to the production of chitosan. The carbohydrate backbone is similar to cellulose and consists of β-1,4-linked d-glucosamine, except that the acetylamino group replaces the hydroxyl group on the C2 position. Thus, chitosan is a copolymer consisting of N-acetyl-2-amino-2-deoxy-d-glucopyranose and 2-amino-2-deoxy-d-glucopyranose, where the two types of repeating units are linked by (1→4)-β-glycosidic bonds.¹¹  Chitosan has proved to be a safe excipient in drug formulations over recent decades, being a non-toxic, biodegradable and biocompatible polymer with antioxidant and antibacterial properties. Additionally, it can increase the cellular permeability and improve the bioavailability of orally administered protein drugs, due to its mucoadhesive properties.^12–14  Furthermore, chitosan possesses the versatile trait of pH responsiveness in the pH range 6–6.5.¹¹  By altering the degree of deacetylation and the molecular weight of the polymer, which can be achieved by varying the temperature of the deacetylation process, the charge density of the polymer can be additionally adjusted.^15,16  It is important to mention that chitosan is poorly soluble at physiological pH and it readily swells in aqueous solutions, resulting in rapid drug release in its application as a continuous matrix for controlled drug release.^17,18  Numerous colloidal delivery systems based on chitosan have been reported for the mucosal delivery of hydrophilic drugs, peptides, proteins, vaccines and DNA.^19–22  Chitosan is a polycationic polymer that has two hydroxyl groups in the repeating glucosidic residue and one amino group. The primary amino groups present on the polymer backbone provide reactive sites for a variety of side-group attachments by employing only mild reaction conditions.

Cyclodextrins (CDs) are sugar derivatives that are produced by enzymatic degradation of starch.²³  CDs are cyclic oligosaccharides composed of six to eight α-1,4-linked glucose units. They feature a bulb-shaped topology with a hydrophobic cavity, which is enclosed by a hydrophilic exterior. Based on the amount of glucose units in these cyclic molecules (6, 7 or 8), CDs are divided into three groups (α, β or γ). As biodegradable monodisperse molecules with low toxicity, low immunogenicity and a high number of hydroxyl groups, which can be used for further modification, CDs have attracted increasing attention for medical applications.^24–27  The straightforward chemical modification of CDs has led to the generation of a large number of CD–polymer conjugates such as star-shaped polymers using α-CD as the core equipped with oligomeric ethylenimine arms utilizing 1,1′-carbonyldiimidazole chemistry.²⁸  The incorporation of CDs into cationic polymers has led to improved electrostatic complexation of DNA molecules. Amphiphilic CD-based systems are valuable carriers for gene delivery, as they can be tuned at will to change the density of cationic groups and introducing or decreasing hydrogen bonding functionalities can alter the flexibility of the polymer chains.^29,30  Besides the examples described above, other diverse modifications have been prepared such as “click clusters”,³¹  polyCDplexes for DNA compaction³²  and chiral separation of anionic drug molecules or amino acids³³  by utilizing the host–guest concept.

Dextran is a versatile and widely available natural polymer that exhibits biodegradability and biocompatibility. These properties are due to its water solubility irrespective of the pH of the solution and of its polysaccharide structure consisting of α-1,6-linked glucose units as well. The polysaccharide glycogen, which is commonly found in animals and fungi as an energy storage molecule, possesses the same kind of bond. Utilizing a debranching enzyme, living organisms are able to cleave this chemically stable bond. The three accessible hydroxyl moieties, found on every monomer unit, facilitate modifications such as the incorporation of amino groups. (Diethylaminoethyl)dextran³⁴  and dextran-spermine³⁵  are well-described examples, especially since dextran-spermine exhibits high transfection efficiency for DNA.³⁶ 

Cellulose is a biopolymer with a polysaccharide structure of β-1,4-conjugated glucose units and as the main component in the cell wall of plants it is in fact the most common organic compound in the world. Therefore it is understandable that this biodegradable polymer became the focus of attention in the modern age of renewable resources. The hydroxyl moieties of cellulose are the chemical target for modification and functionalization. In the case of cationic cellulose these functional groups are usually reacted with glycidyl ammonium salts or by utilizing in situ epoxidation.^37,38  While these modifications are convenient for introducing desired properties like hydrophilicity or antibacterial activity, it needs to be mentioned that cellulose is poorly soluble in both polar and non-polar solvents. This negative aspect of derivative synthesis is due to the strong intra- and intermolecular hydrogen bonding of cellulose molecules. In spite of this disadvantage, several cationic derivatives of cellulose have been prepared and analyzed for medical application, such as the self-assembling micelles based on hydrophobically modified quaternized cellulose (HMQC) for the delivery of poorly water-soluble drugs,³⁹  as well as derivatives with short quaternized poly[2-(N,N-dimethylamino)ethyl methacrylate] (PDMAEMA; see below)⁴⁰  or poly(ethylene oxide) (PEO)⁴¹  polymer chains grafted onto the cellulose backbone for enhanced cationic character of the biopolymer.

Synthetic products have acquired a negative association due to synthetic food additives, but they show many valuable properties, especially in medical applications.²³  Synthetic polymers are valuable for therapeutic use since they can be produced in a well-defined and controlled fashion, overcoming the greatest setback of natural polymers: the batch-to-batch variation. Synthetic polymers often exhibit increased cytotoxicity due to the strong positive charge, but since they can be freely modified in order to introduce desired properties, their biocompatibility can be improved, for example by incorporation of biodegradable linkers and bioactive functionalities.⁴² 

Polyethylenimine (PEI) is the most outstanding example for synthetic cationic polymers because of its wide range of applications. It can be synthesized in linear (LPEI)^43,44  as well as in branched (BPEI)⁴⁵  structures. LPEI possesses primary and secondary amino groups, whereas BPEI also features tertiary amino groups. BPEI usually has a ratio of primary to secondary to tertiary amino functionalities of 1:2:1 and up to 25% of these amino groups are protonated under physiological conditions. Such buffer capability can also be utilized for endosomal escape mechanisms. The amino functionalities are, nevertheless, first and foremost the target for further modification in order to introduce therapeutic molecules or to alter the undesired properties of PEI such as the cytotoxicity, the low hemo-compatibility and the lack of biodegradability. In order to overcome the fact that PEI is non-biodegradable, several strategies have been developed. By incorporation of reducible/cleavable disulfide linkages into the polymer, utilizing biodegradable linkers to graft short PEI chains onto other polymer backbones and by introducing acid-labile ester bonds in the polymer chain, the biocompatibility can be increased.^46,47  Other modifications such as the acid-degradable amino ketal branches grafted onto LPEI, which were originally introduced for endosomal escape, also increased the buffering capability and the transfection efficiency.⁴⁸  Modification methods such as drug conjugation⁴⁹  and the introduction of other polymeric chains are also viable pathways to acquire a tailor-made polymer for drug delivery. This aspect is most prominent in the example of PEI-g-PEG-RGD, which has an incorporated integrin-binding peptide (RGD) for more efficient gene delivery through endothelial cell-targeting.^50,51  Besides their susceptibility for modification and functionalization, the amino groups are also an important asset in order to acquire polyplexes, since they can be protonated and therefore equipped with a cationic charge, depending on the pH of the medium.⁵²  It has been demonstrated that BPEI of high molecular weight forms enzymatically stable polyplexes of small size and high transfection efficiency.^53,54 

Poly(l-lysine) (PLL) was one of the first polycations investigated for the formation of polyplexes with nucleic acids.⁵⁵  It possesses a high number of primary amino groups, which enable efficient complexation of polyanions, and it is well suited for gene delivery, in spite of the fact that the ɛ-amino groups of the l-lysine monomers are only partially protonated in physiological environment due to the neighbouring group effect. Although PLL with high molecular weight shows cytotoxic properties, it is a valuable and widely researched polymer, as most of these drawbacks can be overcome using different modification methods.⁵⁶  The precipitation of the PLL polyplexes, for example, has been elucidated using PEG blocks that have increased the water solubility of the complexes.⁵⁷  The introduction of an artery wall-binding peptide to the PLL-b-PEG copolymer led to a specific targeting and drastically increased the transfection efficiency by 18 000% compared to the PLL-b-PEG copolymer without the covalently bound peptide.^58,59  Similar targeting properties were achieved by introducing a leukaemia-specific JL1 antigen to the PLL backbone.⁶⁰  PLL dendrimers and their PEGylated derivatives, including a pH-sensitive linker molecule for the release of the drug, have been prepared and applied for drug delivery purposes.^61–63  Biodegradability was achieved by incorporating succinimidyldipropyldiamine (SPN) into the PLL dendrimers, and by using octa(3-aminopropyl)silsesquioxane as the dendrimer core the transfection efficiency was successfully increased.^64,65  Utilizing a 3-(hydrazinosulfonyl)benzoic acid linker and the terminal amino groups of lysine or SPN, PLL dendrimer derivatives as well as conjugates have been prepared, such as the guanidine end-caped PLL dendrigraft or the PEGylated dendrimers that were further equipped with doxorubicin, a DNA intercalating anthracycline used in cancer chemotherapy.⁶¹ 

The cationic polymer polyamidoamine (PAA) offers a variety of advantageous properties, such as biocompatibility, biodegradability, water solubility and lower inherent cytotoxicity, than other cationic polymers. The large number of tertiary amino and amido groups on the backbone of PAA makes it an excellent scaffold for further chemical functionalization. PAA is usually synthesized by a Michael-type polyaddition or by using “green” catalysts (salts of alkaline earth metals such as CaCl₂). Structural variations (linear vs. branched) can be introduced just by varying the monomers used for the polymerization.^66,67  This straighforward method can also be utilized in order to influence the polymer properties and to introduce the desired functionalities. For example, incorporating ketals and acetals into the polymer backbone leads to pH-sensitive PAA, which hydrolyzes more quickly under the non-physiological conditions found in tumour tissue.^68,69  Another method of increasing the biocompatibility of the polymer is by introducing reducable disulfide bonds into the backbone of linear PAA using oligoamines and disulfide-containing cystamine bisacrylamide as reactants in the polyaddition.^70–72  The disulfide bond is cleavable under the influence of glutathione, a common antioxidant in nature, which can be found at elevated concentrations in tumour cells.⁷³  It has also been shown that, besides its effect on biodegradability, this modification increases the DNA condensation and transfection efficiency.⁷⁴  This approach has been further improved by introducing PEG side-chains to the disulfide bond containing PAA in order to induce a stealth effect.⁷⁵  The cleavage of the disulfide bond, once the PAA drug carrier reaches the tumour enviroment, enhances this effect even further.⁷⁶  PAA can be functionaized via copolymerizing monomers with diamines, which have N-triphenylmethyl-protected primary amino functionalities.⁷⁷  It has been shown that 2,2-bis(acrylamido)acetic acid (BAC) or 1,4-bis(acryloyl)piperazine (BP) can be copolymerized with N-triphenylmethyl-monosubstituted 1,2-diamines to form almost monodisperse polymers (PDI of 1.16) (Scheme 1.1).⁷⁸  These pendant amino groups can then be used for conjugation chemistry. Utilizing this method, doxorubicin was successfully coupled to PAA carrier molecules.⁷⁹ 

Poly(amino-co-ester)s (PAEs) are usually synthesized following the procedure of Lager et al., using primary or bis-secondary amines in a Michael-type polyaddition with diacrylate esters.⁸⁰  Changing the reaction time, the monomer ratios or the reaction temperature can be used to adjust the molecular weight and the end-chain functionality of the polymer. Using this straightforward method, amine-terminated PAEs can be synthesized which show improved transfection efficiency, similar in efficiency to PLL and PEI.^81,82  Copolymerizing 2-(pyridyldithio)ethylamine as the amine monomer leads to the incorporation of pyridyldithio moieties into PAE side chains. This functional group can react with thiol-containing molecules (see Scheme 1.2), e.g. 2-aminoethanethiol or cysteine-bearing targeting peptides like RGDC, which possess high affinity towards integrin receptors and therefore can be used for active targeting.⁸³  This modification diminished the DNA binding strength of the polycation; however, this is not necessarily a disadvantage as it can lead to a more efficient intracellular DNA unpacking.⁸⁴  PAEs possess hydrolytically degradable ester bonds that can be degraded within 5 hours into non-toxic metabolites, which make them attractive polycationic carriers for gene delivery in vivo.⁸⁵  The biodegradability was improved even further by preparing PAE-b-PEG copolymers using a Michael-type polyaddition of monoacrylated PEG, hexane-1,6-diol, diacrylate and 4,4′-trimethylenedipiperidine.⁸⁶  Besides the improvement of the biodegradability, this procedure led to an amphiphilic and pH-responsive block copolymer with enhanced transfection efficiency.⁸⁷  Other preparation methods for PAEs, such as the use of the enzyme Candida antarctica lipase B (CALB) for the coupling of diesters with amino-substituted diols, have been developed.⁸⁸  This type of PAE synthesis facilitates moderate molecular weights (up to 59 000) in narrow distributions (PDI of 1.5) and therefore overcomes the setback of the method proposed by Langer, which usually results in broad molecular distributions (PDI of 6 and higher).⁸⁹ 

Poly[2-(N,N-dimethylamino)ethyl methacrylate] (PDMAEMA) consists of covalently bound 2-(N,N-dimethylaminoethyl) methacrylate monomers and it can be synthesized in a well-defined and controlled fashion using atom-transfer radical polymerization (ATRP).^90,91  PDMAEMA possesses tertiary amino groups on every monomer molecule, which are partially protonated at physiological pH.⁹²  Based on the fact that PDMAEMA has a pK_a value of ∼7.5 it can behave like a proton sponge and destabilize endosomes, resulting in an efficient delivery of DNA.^93,94  A variety of modifications have been investigated to improve PDMAEMA's properties for drug delivery applications.^95,96  Utilizing reversible addition–fragmentation chain transfer (RAFT) polymerization with a bifunctional chain transfer agent, reducible disulfide groups were introduced.⁹⁷  Since the RAFT method results in dithioester terminated polymers, aminolysis and oxidative conditions in successive steps lead to rPDMAEMA of higher molecular weight with reducible disulfide bonds on the backbone (see Scheme 1.3).⁹⁸  This modification method is very potent for increasing the biocompatibility of water-soluble polymers, which are not biodegradable but still show moderate cytotoxicity, such as PDMAEMA.⁹⁹  The described polymer structures are well-suited for medical applications but because of the lack of in vivo degradability their size cannot exceed a limit set by the renal filtration (30–50 kDa, depending on the polymer shape, molecular conformation and flexibility).¹⁰⁰  Non-biodegradable polymers exceeding this limit can be retained and accumulate in the body, leading to complications.¹⁰¹ 

An alternative modification method that has a similar impact on PDMAEMA chains for in vivo applications is the introduction of ester bonds into the polymer backbone by free radical polymerization of a cyclic ketene acetal [5,6-benzo-2-methylene-1,3-dioxepane (BMDO)] and DMAEMA using a PEGylated macro initiator (Scheme 1.4).¹⁰²  This approach, however, results in broad molecular weight distributions (PDI of 2.6–6), since the polymerization method does not belong to the controlled types. Other ways of modification include grafting of PEG chains onto PDMAEMA by means of ATRP, which moderates the formation of aggregates but also activates the innate immune system, since it was shown to induce cytokine production of murine macrophages.^103,104  Utilizing ATRP, chitosan was successfully grafted onto PDMAEMA by applying chitosan as a macroinitiator that was synthesized by phthaloylation of the unalkylated primary amino groups (deminishing the hydrogen bond formation and therefore increasing chitosan solubility in organic solvents) and posterior acylation of chitosan hydroxy groups using α-bromoisobutyryl bromide. The resulting chitosan-g-PDMAEMA polymer showed pH- as well as temperature-responsiveness.¹⁰⁵  Copolymerizing DMAEMA with monomers that possess primary amino, imidazole or carboxylic functionalities has been investigated as well, in order to enhance the endosomal escape, but all of the researched copolymers showed limited transfection.¹⁰⁶ 

The amino group, which is present on numerous cationic polymers, exists in a variety of different forms (primary, secondary, tertiary and quaternary) and is probably the most important functional group for chemical modification. The quaternary amino groups are often useful because they can provide a permanent positive charge and are thereby interesting for the introduction of requested properties, such as water solubility. This is the reason why methylation plays a crucial role in the modification of cationic polymers. Additionally, since modification and functionalization reactions are often done in a basic medium, the quaternary amino group, in contrast to other types, will not lose its cationic character during the reaction. Primary, secondary and tertiary amino groups, however, can be reprotonated depending on whether they have reacted, on their pK_a values as well as on the acidity of the medium, which is used to regenerate the cationic charge. Tertiary amino functionalities are typically sterically hindered as well as thermodynamically stable and therefore not relevant for functionalization. On the other hand, secondary and especially primary amino groups are suitable for purposes such as the amine-coupling reaction, which is a common way to crosslink polymer chains or to conjugate a peptide and a drug molecule. Such reactions usually proceed rapidly in an aqueous environment and in high yield.¹⁰⁷  Utilizing linking reagents, which can be equipped with a variety of spontaneously as well as selectively reactive functional groups, is therefore a standard technique in conjugation chemistry. Additionally, carbodiimide chemistry is a potent and common way to establish a bond between carboxyl and amino groups (Scheme 1.5). Carbodiimides such as 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), 1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide (CMC), diisopropylcarbodiimide (DIC) and dicyclohexylcarbodiimide (DCC) belong to the group of zero-length crosslinking reagents, because they mediate the amide or phosphoramidate bond formation between carboxyl and amino or phosphate and amino groups without introducing additional atoms.¹⁰⁷  The reaction is frequently carried out in neutral or slightly acidic media (pH 4.5–7.5) using 2-(N-morpholino)ethanesulfonic acid (MES) or phosphate buffered saline. During the reaction, carbodiimides form highly reactive and short-lived O-acylisourea compounds with the employed carboxylic acids. This active species reacts in a subsequent step with the nucleophilic primary amine to form an amide bond. Water molecules, which are potent nucleophiles as well, can also react with the active species and regenerate the carboxylate group by cleaving the activated ester intermediate and forming an isourea compound.

The conjugation yield can be improved by using the water-soluble N-hydroxysulfosuccinimide (sulfo-NHS) (Scheme 1.6). This compound attacks the carboxyl group nucleophilically, forming a long-lived NHS ester species, which does not undergo rapid hydrolysis but readily reacts with amino groups to create a stable amide bond. This method is particularly useful if the conjugation is done at low concentrations of the amine, where the hydrolysis would otherwise be the predominant pathway. Using sulfo-NHS esters therefore increases the efficiency and the yield of the conjugation reaction drastically.¹⁰⁸  Sulfo-NHS esters of proteins or peptides can be isolated and reacted with amino group containing polymers in order to achieve conjugation. Nevertheless, the presence of thiol and hydroxyl groups can reduce the yield, since these groups react as well; therefore they should be protected during the reaction.

Considering the possibilities that this functionalization method offers, it is understandable that carbodiimide chemistry is being extensively used in the field of cationic polymers. For instance, carbodiimide-mediated amide bond formation has been used to synthesize chitosan conjugates with covalently attached thiol moieties, which were meant to enhance the permeation properties of the polymer.¹⁰⁹  Moreover, using a similar synthetic route, chitosan-N-inulin graft copolymers with varying degrees of substitution were synthesized by Janciauskaite et al. as surface conditioners.¹¹⁰  Zhang et al. conjugated polyhydroxy fatty acids with cationic polymers such as PEI and chitosan as novel carriers for gene delivery utilizing carbodiimide chemistry.¹¹¹  Carbodiimides represent only one small part of the existing possibilities to functionalize the amino group of cationic polymers. In the next section, some of the most important amine reactive groups in bioconjugation chemistry will be described.

NHS esters constitute the most widespread molecules to create reactive acylating agents. Therefore, it is not surprising that the great majority of the commercially available amino-reactive reagents for crosslinking¹¹²  or conjugation¹¹³  are NHS esters. NHS esters react with nucleophiles to release the NHS leaving group to form an acylated product (Scheme 1.7). Other nucleophilic groups such as thiol or hydroxy groups can also react with NHS esters to yield thioesters and esters, respectively; however, these reactions do not lead to the formation of stable conjugates, as both of these bonds potentially hydrolyze in water or react with neighbouring amines to from amide bonds. Reaction of NHS esters with primary and secondary amines creates stable amide and imine linkages that do not break readily.

Isothiocyanates react with nucleophiles such as amines, thiols and hydroxy groups; however, the only stable product of these reactions is with primary amine groups.¹⁰⁷  Therefore, isothiocyanate compounds can react selectively with primary amines of a cationic polymer (Scheme 1.8). The reaction proceeds with an attack of the nucleophile on the central electrophilic carbon of the isothiocyanate group, where following an electron shift and proton loss, a thiourea linkage between the isothiocyanate-containing compound and the amine-containing polymer is created with no leaving group involved. Isothiocyanates are therefore ideal for selective modification of ɛ-amino groups of PLL or for reactions with terminal amino groups of branched PEI. The reaction often requires extended reaction times (4–24 h) and alkaline buffers (0.1 M NaCO₃ buffer, pH 9) where the target amine will be unprotonated. Although this conjugation technique has a decreasing use for coupling drugs or proteins to the carrier systems, it is still widely applied for the conjugation of chromophores. In a recent work, PEI shells of nanoparticles were fluorescently labelled utilizing the fluorescein isothiocyanate derivative (FITC).¹¹⁴  Other chromophores like bis-styrylbenzyl fluorophores carrying an isothiocyanate group have been conjugated to PEI and applied in two-photon microscopy imaging of gene delivery in live cells.^115,116 

Isocyanates are very similar to isothiocyanates in their structure, where the only difference is that an oxygen atom replaces the sulfur, as well as having a similar reaction procedure (Scheme 1.9). However, this functional group is more reactive and therefore less common in conjugation chemistry, since more efficient ways of coupling have been found.¹⁰⁷  They also differ in their reactivity towards amines. Isocyanates are more reactive, but for the same reason their stability is often a problem, since moisture rapidly decomposes them, releasing carbon dioxide and leaving an aromatic amine. However, the isourea compounds, formed by reactants possessing primary amino and isocyanate functionalities respectively, are more stable than the isothiourea compounds formed using isothiocyanates. Isocyanates can be applied when crosslinking hydroxyl group containing macromolecules (polysaccharides such as dextran or dextrin), where the reaction with alcohols leads to the formation of the stable carbamate bond. This methodology has been successfully applied in order to achieve single-chain polymer nanoparticles¹¹⁷  or antistatic coating compositions.¹¹⁸ 

Sulfonyl chlorides are highly reactive derivatives of sulfonic acid and possess similar properties and reactivity to acid chlorides of carboxylates.¹⁰⁷  The reaction with nucleophiles such as amines requires alkaline conditions (pH 9–10) and proceeds through the formation of an unstable pentavalent transition state (Scheme 1.10). Hydrolysis is a major competing reaction in water; therefore the reaction with amines proceeds with better yield when done in organic solvents. Sulfonyl chlorides have played an important role in bioconjugation chemistry, since sulfonic acids can be easily converted into sulfonyl chlorides using thionyl chloride or phosphorus pentachloride in non-aqueous conditions.¹¹⁹ 

The reactive carbonyl groups of aldehydes and glyoxals readily react with nucleophilic amines to form Schiff base intermediates (Scheme 1.11). Reducing reagents such as sodium cyanoborohydride, which is often used for reductive amination, can be added to the reaction mixture in order to convert the chemically unstable Schiff base into a secondary amine. Stronger reducing agents such as sodium borohydride, amine-boranes or ascorbic acid can also be applied. However, they can reduce the aldehyde group as well and therefore decrease the yield of the reaction.^120,121  Aldehydes that possess vicinal hydroxyl groups, such as reducing sugars, may undergo an Amadori rearrangement to form stable ketoamine structures (see Scheme 1.16 below). This type of functionalization is common in polysaccharide chemistry but it also has applications in other fields. It has been employed for the introduction of self-fluorescence of nanocomposites¹²²  as well as for the transformation of aldehyde end-capped polymers into amine end-capped ones.¹²³ 

Molecules that are equipped with epoxy functionalities react with nucleophiles such as primary amines, thiols or hydroxyl group containing compounds to constitute secondary amines, thioethers or ethers, respectively, by a ring-opening reaction. Although the reaction itself does not proceed selectively, it is possible to control which reactant will react first, based on the pH of the reaction medium. Thiol groups are the first to react, since these functional groups require a neutral pH environment (pH 7.5–8.5). Amines will react at slightly basic conditions (pH 9), while hydroxides on the other hand will only react at strongly alkaline pH (pH 11–12). Therefore, by selecting the proper reaction conditions, it is possible to steer the functionalization in the desired direction. However, a cationic polymer needs to be either free of thiol groups or needs to be masked with protective groups in order to achieve selective conjugation to the amino functionalities. It should be mentioned that hydrolysis is a side reaction that can also occur. This limitation, however, can be exploited intentionally by hydrolysis of the epoxide and oxidizing the corresponding glycol in a subsequent step via periodate (1 mM, 0 °C) to form an aldehyde (Scheme 1.12). The aldehyde can in turn be reacted with the amine by reductive amination, as described above.

Carbonates can be described as diester derivatives of carbonic acid, formed by condensation with hydroxyl compounds. However, carboxylic acids do not possess the necessary reactivity to synthesize such products. Therefore bifunctional reactive molecules such as phosgene or carbonyldiimidazoles like N,N′-carbonyldiimidazole (CDI)¹²⁴  and disuccinimidyl carbonate (DSC) are commonly used (Scheme 1.13). These compounds readily react with nucleophiles to form stable conjugates. They react with hydroxyl-containing molecules to form the amino-reactive carbonate or carbamate intermediates. These intermediates react in a subsequent step with primary amines to form stable carbamate bonds (aliphatic urethanes). The reaction is usually carried out in alkaline media (pH 7–9) and in the absence of competing amino and thiol functionalities. This conjugation method is extensively used for coupling of PEG chains to proteins and other amino group carrying compounds. These established methods in conjugation chemistry can be easily extended to the field of cationic polymers.

Imido esters are among the most specific acylating agents, which are utilized for the modification of primary amines, since they possess minimal cross-reactivity to other nucleophilic groups (Scheme 1.14).^125–128  The reaction products, amidines, are protonated at physiological pH values and can therefore help to increase the charge density of cationic polymers. The formed bonds are stable under acidic conditions, but can be easily hydrolyzed in alkaline media. The reaction itself requires deprotonated amino groups, which is why slightly alkaline media (pH 8.2, 0.2 M triethanolamine, 0.1 M sodium borate) need to be used. This conjugation method can be also used for crosslinking of polymer chains by utilizing bifunctional imido esters. The most prominent example in the field of cationic polymers is the synthesis of thiolated chitosan.¹²⁹ 

The semi-synthetic cationic polymers described in the previous section originate from the naturally abundant carbohydrates. One of the biological applications of such polysaccharides in vivo is glycosylation of proteins. Polymers such as cationic dextran or chitosan possess, aside from the introduced cationic functionalities, additional hydroxyl as well as aldehyde groups, which are suitable for functionalization. The hydroxyl functionalities are easily accessible, although such polymers possess a definite 3D structure, which is mainly based on the character of the glycosidic bond. Utilizing these functionalities, the cationic polymers can be acylated or alkylated in order to introduce additional properties or functional groups (Scheme 1.15). The incorporation of charged quaternary amines in order to synthesize cationic dextran is based on the same chemistry.¹³⁰  Alkylation of the hydroxyl functionalities by using reactive groups such as epoxides or halides leads to stable ether bonds. Acylation, on the other hand, results in unstable and easily hydrolyzable ester bonds. This disadvantage can be overcome by using CDI or DSC to mediate the formation of a stable carbamate bond, as described above. In case the hydroxyl group is not reactive enough to undergo efficient conjugation reactions, activating reagents such as tosyl chloride, chloroformate derivatives or cyanogen bromide can be utilized. These agents are not applicable in aqueous media because the formed active groups would rapidly hydrolyze. This problem, however, can be circumvented by using polar organic solvents such as dimethyl sulfoxide or dimethylformamide. It is also important to mention that in case of chitosan, primary amino groups might be present on the polymer chain, which would readily react with the active species described above. This might be a desired or undesired side-reaction, since it will alter the properties of the polymer but also will lead to decreased yields of the subsequent conjugation reaction. It is therefore necessary to either employ protective groups or to ensure that all amino groups of the cationic polymer are at least tertiary, hence unable to establish stable bonds.

The aldehyde groups are normally located only on the reducible glucose units at the chain end, which can exist in linear form. These functional groups can be exploited for functionalization (Scheme 1.16) using strong nucleophiles such as primary amines or hydrazines. It is also possible to equip the polysaccharide with terminal amino functionalities by reacting it with 2-(4-aminophenyl)ethylamine, which in turn can be used for further conjugation reactions. Similar chemistry is employed for polymers, which possess solely terminal functional groups such as poly(ethylene oxide). However, such a small amount of possible links is unsatisfactory for applications in drug delivery. In such cases it is necessary to be able to easily introduce additional functional groups.

In order to increase the amount of aldehyde functionalities, either oxidases^131,132  or a 10 mM sodium periodate solution is usually utilized (Scheme 1.17). This method cleaves the carbon–carbon bond between adjacent hydroxyl groups and oxidizes the hydroxyl functionalities to aldehydes. It is obvious that crosslinking of the polymer chains might occur in this case as well, which is why the previously described precautions need to be taken.

Reductive amination, which has been described in Section 1.2.1.5, is applicable in this case as well. Using a mild reducing agent, the Schiff bases formed can be reduced and form a stable secondary amine bond (Scheme 1.18).^133,134 

Hydrazines react in the same manner but spontaneously and even more efficiently than primary amines under the same conditions, since the formed hydrazone bond is more stable than the Schiff base (Scheme 1.19).

This type of synthetic approach to create aldehyde groups, which react with nucleophiles in a subsequent step, is common for the modification of carbohydrates.

The era of polymer therapeutics in cancer therapy started in the late 1950s with the synthesis of N-vinylpyrrolidine conjugates of glycyl-l-leucine-mescaline.¹³⁵  Since then, many new concepts of drug delivery such as polymeric micelles,¹³⁶  polymer–drug conjugates and polymer–peptide/protein conjugates have been elaborated (Figure 1.2). Over recent decades, significant progress has been made and documented on the understanding of their mechanism of effect and on improving their design.^{100,137–139}  The concept of polymer–drug conjugates originated from the polymer–anticancer conjugates proposed by Ringsdorf in 1975.¹⁴⁰  In general, polymer–drug conjugates should consist of a minimum of three components: (1) a water-soluble polymeric carrier molecule, (2) a biodegradable polymer–drug linkage and (3) at least one drug molecule. The drug molecules can be of the same type or several drugs can be conjugated to the carrier at the same time to achieve a synergistic effect for a more efficient combination treatment.^141,142  Carrier systems consisting of the three above-mentioned components can be complemented by targeting ligands in order to achieve cell-specific targeting; however, no specific tumour markers have been found so far. This is the reason why modern approaches include targeting of overexpressed proteins on the cell surface or in the cytosol (active targeting) as well as exploiting the enhanced permeability and retention effect (EPR)¹⁴³  for passive targeting. This effect is, on the one hand, the result of the prolonged plasma circulation time made possible by conjugating the drugs/proteins to polymer carriers. On the other hand, it is based on the fact that the angiogenic tumour vessels are hyper-permeable and therefore enable macromolecules and liposomes to pass into the stroma. Based on the lack of lymphatic drainage in tumour tissue, the macromolecules are retained and thus can accumulate, increasing the local anti-tumour–drug concentration 10- to 100-fold compared to the free drug. One major obstacle in this approach is the fact that the vasculature permeability of the tumour tissue varies in the different stages. Tumour growth increases the intratumoural hydrostatic pressure, reducing the effectiveness of the EPR. Moreover, the angiogenetic vessels can only be found in the periphery of the tumour tissue. Small tumours in their early stages are therefore optimal for passive targeting.¹⁴⁴ 

Several key features govern the suitability of polymer therapeutics for development, such as the potential toxicity of the polymer and its primary metabolites (biocompatibility), reproducible manufacture with acceptable degree of heterogeneity, pharmaceutical formulation with satisfactory stability (shelf-life) and suitability for patient administration. Specific cell targeting and intracellular transport also play a major role in the development of new drug carrier systems. Considering those demanding properties, it is remarkable that cationic polymer carriers have been and still remain in the focus of modern drug delivery for the release of therapeutics in both implanted reservoir systems and pulsatile dose delivery products. They have attracted such attention because of their versatile characteristics, including high cellular uptake efficiency and good water solubility, diminishing the need for the incorporation of hydrophilic polymer blocks or cell penetrating peptides. Their capability as drug carriers is greatly influenced by the flexibility of the polymer chain, the formation of hydrogen bonds, electrostatic forces, pK_a and their nucleophilic character. Chitosan, one of the most exploited cationic polymers, has been applied already in drug delivery as a hydrogel and a polyelectrolyte complex, as well as a drug conjugate, and it has been used for biodegradable release systems.¹⁴⁵  Utilizing those delivery system proteins, anti-inflammatory drugs and also antibiotics and growth factors have been successfully administered in vivo. Chitosan polymers have been the focus of many recent studies, such as protein-loaded chitosan microspheres which showed effective and sustained delivery.¹⁴⁶  Other cationic polymers such as PAA and PEI are also intensely investigated as drug carriers and they have already been successfully applied to the interdisciplinary field of nanotherapeutics. In the special case of PAA, an effective intracellular protein delivery system has been developed based on linear PAAs that forms self-assembled cationic nanosized complexes with oppositely charged proteins.¹⁴⁷  Dendrimeric PAA drug conjugates were investigated as drug delivery agents. They showed improved water solubility of the drug, high drug release efficiency and enhanced activity of the drug, while reducing its cytotoxicity.^148–150  PEI, on the other hand, was used for the immobilization of 1,1-diphosphonic acid by ionic cooperative interactions between the cationic polymeric matrix and the deprotonated etidronate. This approach resulted in stable and slow release, which can be accelerated by increasing the pH.¹⁵¹ 

In conclusion, cationic polymers possess a variety of properties preferential for drug delivery purposes and a set of strategies can be utilized in order to improve these features or even introduce additional ones. These functionalizations include the incorporation of cell targeting moieties and further transport domains using conjugation chemistry for increased specificity and efficiency of delivery. Moreover, the introduction of biodegradable bonds, such as ester or disulfide bonds, into the backbone of cationic polymers can diminish one of their biggest setbacks: the intrinsic toxicity. Considering the fact that the progress of cationic polymers in drug delivery has always been hindered because of their non-biodegradability and toxicity, it is expected that functionalization/modification chemistry can lead the way to overcome these hindrances. This is also the reason why in recent years a variety of modifications to commonly applied cationic delivery systems have been made and the continuing research in multidisciplinary areas of cationic polymers will surely lead to further improved designs, which is only achievable by understanding the possibilities and limitations of conjugation and functionalization chemistry.