Chemistry of Bioconjugates: Synthesis, Characterization, and Biomedical Applications

By Ravin Narain

Explores bioconjugate properties and applications of polymers, dendrimers, lipids, nanoparticles, and nanotubes

Bioconjugation has enabled breakthroughs across many areas of industry and biomedicine. With its emphasis on synthesis, properties and applications, this book enables readers to understand the connection between chemistry and the biological application of bioconjugated materials. Its detailed descriptions of methods make it possible for researchers to fabricate and take full advantage of bioconjugates for a broad range of applications. Moreover, the book sets the foundation for the development of new applications, including assays, imaging, biosensors, drug delivery, and diagnostics.

Chemistry of Bioconjugates features contributions from an international team of leading experts and pioneers in the field. These contributions reflect the authors’ firsthand laboratory experience as well as a thorough review of the current literature. The book’s six sections examine:

• General methods of bioconjugation
• Polymer bioconjugates
• Organic nanoparticle-based bioconjugates
• Inorganic nanomaterial bioconjugates, including metals and metal oxides
• Cell-based, hydrogel/microgel, and glyco-bioconjugates
• Characterization, physico-(bio)chemical properties, and applications of bioconjugates

This comprehensive exploration of bioconjugates includes discussions of polymers, dendrimers, lipids, nanoparticles, and nanotubes. References at the end of each chapter serve as a gateway to the most important original research findings and reviews in the field.

By drawing together and analyzing all the latest chemical methods and research findings on the physico-chemical and biochemical properties of bioconjugates, Chemistry of Bioconjugates sheds new light on the significance and potential of bioconjugation. The book is recommended for organic and polymer chemists, biochemists, biomaterial scientists, carbohydrate chemists, biophysicists, bioengineers, and drug and gene delivery scientists.

• Language: English
• Publisher: Wiley
• Release date: Dec 2, 2013
• ISBN: 9781118776377

Book preview

SECTION I

GENERAL METHODS OF BIOCONJUGATION

1

COVALENT AND NONCOVALENT BIOCONJUGATION STRATEGIES

RAJESH SUNASEE¹ AND RAVIN NARAIN²

¹Department of Chemistry, State University of New York, Plattsburgh, NY, USA

²Department of Chemical and Materials Engineering, Alberta Glycomics Centre, University of Alberta, Edmonton, AB, Canada

1.1 INTRODUCTION

Bioconjugation—the process of covalently or noncovalently linking a biomolecule to other biomolecules or small molecules to create new molecules—is a growing field of research that encompasses a wide range of science between chemistry and molecular biology. The tremendous achievement of modern synthetic organic chemistry has led to a variety of bioconjugation techniques [1] available for application in research laboratories, medical clinics, and industrial facilities. While bioconjugation involves the fusion of two biomolecules, for example protein–protein, polymer–protein, carbohydrate–protein conjugates, it also involves the attachment of synthetic labels (isotope labels, fluorescent dyes, affinity tags, biotin) to biological entities such as carbohydrates, proteins, peptides, synthetic polymers, enzymes, glycans, antibodies, nucleic acids, and oligonucleotides (ONTs). The product of a bioconjugation reaction is usually termed as a "bioconjugate" and synthetic macromolecules produced by bioconjugation approaches are commonly referred to as biohybrids, polymer bioconjugates, or molecular chimeras. Modification of biomolecules is an important technique for modulating the function of biomolecules and understanding their roles in complex biological systems [1a]. However, selective biomolecule modification remains challenging and the ease of generating the desired bioconjugate rapidly under physiological conditions is vital for many applications, such as disease diagnosis, biochemical assays, ligand discovery, and molecular sensing. As applications of bioconjugates continue to grow, an expanded toolkit of chemical methods will be required to add new functionality to specific locations with high yield and chemoselectivity.

The aim of this chapter is to provide a comprehensive review of the different types of bioconjugation methods (covalent and noncovalent approaches) available for the modification of biomolecules (proteins, peptides, carbohydrates, polymers, DNA, etc.). Traditional bioconjugation methods will first be elaborated upon, followed by some modern bioconjugation techniques, particularly the emerging role of bioorthogonal chemistry, where the translation of knowledge of chemical reactions to reactions in living systems can be achieved. While the synthetic aspects of the bioconjugates will be the main focus, a brief description of their applications will also be presented.

1.2 COVALENT BIOCONJUGATION STRATEGIES

The covalent bond is the most common form of linkage between atoms in organic chemistry and biochemistry. The reaction of one functional group with another leads to the formation of a covalent bond via the sharing of electrons between atoms (Figure 1.1).

FIGURE 1.1 Schematic representation of covalent bioconjugation strategy.

c01f001

Covalent bioconjugation strategies are generally categorized as random (modification at multiple sites) or site-specific (modification at a single site) bioconjugation. Traditional covalent bioconjugation strategies preclude control over the regiochemistry of reactions, thereby leading to heterogeneous reaction products and eventually, loss of the biological function of the target biomolecule [1(d)]. However, new methods of bioconjugation that are highly site specific and cause minimal change to the active form of the biomolecule have been developed. For instance, bioorthogonal reactions have recently emerged as essential tools for chemical biologists [1(e)]. The following sections survey the covalent modifications of several reactive functional groups (carboxylic acids, aldehydes, ketones, amines, thiols, and alcohols), which are generally present or can be introduced onto macromolecules (proteins, peptides, carbohydrates, nucleic acids, ONTs, etc.).

1.2.1 Carboxyl Modifications

Carboxyl groups are commonly found on the C-terminal ends of proteins and on glutamate (Glu) and aspartate (Asp) amino acid side chains. Carboxylic acids are strong organic acids and the fastest reaction with a nucleophile is removal of the acidic hydrogen to form the carboxylate anion. The resulting anion is resistant to addition reaction with a second nucleophile, and thus makes conjugation through carboxylate group via nucleophilic addition a difficult process. Usually, harsher conditions, acid catalysis, or special reagents are required to promote carboxylic acid-mediated reactions. However, some carboxylate-reactive chemical reactions have been achieved with diazoalkanes and diazoacetyl derivatives (diazoacetate esters and diazoacetamides) and common activating agents such as carbonyldiimidazole (CDI) and carbodiimides to derivatize carboxylic acids. These reactions generate stable covalent linkages namely esters and amides.

1.2.1.1 Diazoalkanes and Derivatives

Diazoalkanes, in particular, diazomethane [2] is a powerful reagent for esterification of carboxylic acids. They react instantaneously with carboxylic acids without the addition of catalysts and may be useful for direct carboxylic acid modification of proteins and synthetic polymers. The reaction mechanism involves nucleophilic attack of the resulting carboxylate anion onto the diazonium ion, followed by an alkylation step to furnish a covalent ester linkage. The driving force of the reaction is the formation of nitrogen, which is a superb leaving group (Scheme 1.1).

SCHEME 1.1 Mechanism of diazomethane esterification reaction.

c01s001

FIGURE 1.2 Fluorescent diazomethane derivatives as labeling reagents.

c01f002

Diazomethane, though easily made, is quite toxic, highly explosive, and requires special glassware for reactions. A less explosive and commercially available reagent, trimethylsilyldiazomethane [3], is commonly employed; however, toxicity is still a major concern. In the past, fluorescent diazomethane derivatives have gained much attention for the derivatization of biologically important molecules, especially the nonchromophoric fatty acids [4], bile acids, and prostaglandins. 9-anthryldiazomethane (ADAM) [5,6] and 1-pyrenyldiazomethane (PDAM) [7,8] are diazomethane derivatives of the fluorescent dyes anthracene and pyrene, respectively, that have commonly been used as fluorescent labeling reagents for liquid chromatographic determination of carboxylic acids. ADAM and PDAM react readily with carboxylic acids at room temperature in both protic and aprotic solvents. ADAM was found to be unstable and decomposed easily upon storage, while PDAM has a much better chemical stability (a 0.1% (w/v) of PDAM in ethyl acetate solution is stable for 1 week at ≤−20°C) [9]. Furthermore, the detection limit for PDAM conjugates (about 20–30 fmol) is reported to be five times better than reported for detection of ADAM conjugates. Fatty acids derivatized with these reagents have been used to measure phospholipase A2 activity [10].

Protocol for reaction of PDAM with fatty acids [9]:

1. Add 100 μL of 1 mg/mL solution of PDAM in ethyl acetate (ethyl acetate stock solution) to 100 μL of 0.01–10 μg/mL fatty acid solution in methanol.

2. React for 90 minutes at room temperature.

3. Inject 5 μL of reaction mixture into an HPLC column.

1.2.1.2 Activating Agents

The direct conversion of a carboxylic acid to an amide with amines is a very difficult process as an acid–base reaction to form a carboxylate ammonium salt occurs first before any nucleophilic substitution reaction happens. As such, amide formation from carboxylic acid is much easier if the acid is first activated (Scheme 1.2) prior to nucleophilic attack by the amine. This strategy converts the poor carboxy −OH leaving group into a better one. Ester linkages can also be formed using this strategy in the presence of alcohols.

SCHEME 1.2 General strategies for the conjugation of carboxylic acid with amines or alcohols via an activating agent.

c01s002

The explosion in the field of peptide chemistry has led to the development of many activating agents that greatly enhance amide formation, but only the most commonly used ones, such as CDI and carbodiimides, will be discussed here (Table 1.1). N, N′-Carbonyldiimidazole (CDI) [11] is a white crystalline solid that is useful for activating carboxylic acids to form amide, ester, and thioester linkages. During the reaction, a reactive intermediate, N-acylimidazole is formed with liberation of carbon dioxide and imidazole as innocuous side products. The N-acylimidazole can then react with amines or alcohols to form stable covalent amide or ester linkages, respectively. CDI is not commonly used in routine peptide synthesis, but nevertheless is quite useful for coupling peptide fragments to form large peptides and small proteins [12]. One unique application of CDI is the synthesis of urea dipeptides [13]. Dicyclohexylcarbodiimide (DCC) and diisopropylcarbodiimide (DIC) are commonly used in organic synthesis for the preparation amides, esters, and acid anhydrides from carboxylic acids. These reagents can also transform primary amides to nitriles, which is a somewhat troublesome side reaction of asparagine and glutamine residues in peptide synthesis. The choice of these carbodiimides depends largely on their solubility properties. DCC was one of the first carbodiimides developed [14] and is widely used in peptide synthesis. It is highly soluble in dichloromethane, acetonitrile, dimethylformamide (DMF), and tetrahydrofuran, but is insoluble in water. The by-product of a DCC-mediated reaction is dicyclohexylurea, which is nearly insoluble in most organic solvents and precipitates from the reaction mixture as the reaction progresses. Thus, DCC is very useful in solution-phase reactions, but is not appropriate for reactions on resin. Another drawback of DCC-mediated coupling is that trace amounts of dicyclohexylurea remains and are often tedious to remove. DIC was developed as an alternative of DCC since being a liquid, it is easier to handle and also forms a soluble urea by-product, which can easily be removed by simple extraction [15].

Table 1.1 Common Activating Agents for Carboxyl-reactive Groups

1-Ethyl-3-[3-dimethylaminopropyl]carbodiimide (EDC or EDAC) is a versatile modern coupling reagent. It is commonly known as a zero-length cross-linking agent used to conjugate carboxyl groups and amines to form stable covalent amide linkages. Amide bonds typically have a half-life of circa 600 years in neutral solution at room temperature [16], and this extraordinary stability renders amide linkages to be very attractive for bioconjugation. This carbodiimide reagent and its urea by-product are both water soluble; hence, the by-product and any excess reagent are removed by aqueous extraction. EDC reacts with a carboxyl to form an amine-reactive O-acylisourea intermediate, which is highly unstable and short-lived in aqueous solution. Thus, hydrolysis is a major competing reaction. It was found that the addition of N-hydroxysulfosuccinimide (Sulfo-NHS) stabilizes the amine-reactive intermediate by converting it to a semistable amine-reactive Sulfo-NHS ester (Scheme 1.3), thereby increasing the efficiency of EDC-mediated coupling reactions [17].

SCHEME 1.3 EDC-mediated protein–carboxylic acid conjugation.

c01s003

Protocol for conjugation of proteins with EDC and Sulfo-NHS [18]:

1. Add EDC (~2 mM) and Sulfo-NHS (~5 mM) to protein #1 solution.

2. React for ~15 minutes at room temperature.

3. Add 2-mercaptoethanol (final concentration of 20 mM) to quench the EDC.

4. Optional step: Separate the protein from excess reducing agent and inactivated cross-linker using a Zeba Desalting Spin Column. Equilibrate the column with activation buffer.

5. Add protein #2 to the reaction mixture or the pooled fractions containing the activated protein at an amount equal to the number of moles of protein #1.

6. React for 2 hours at room temperature.

7. Add hydroxylamine to a final concentration of 10 mM to quench the reaction. (Other means of quenching involve adding 20–50 mM Tris, lysine, glycine, or ethanolamine; however, these primary amine-containing compounds will result in modified carboxyls on protein #1).

8. Remove excess quenching reagent by gel filtration using the same type of column as in Step 4.

A major drawback of carbodiimide activation of amino acid derivatives is that it usually leads to partial racemization of the amino acid. In peptide synthesis, an equivalent of an additive such as triazoles (e.g., 1-hydroxy-7-aza-benzotriazole [19]) is added to minimize this racemization problem. Recently, during the development of prodrugs for the antitumoral agent thiocoraline, a new coupling reagent known as N,N,N′,N′-tetramethylchloroformamidinium hexafluorophosphate (TCFH) [20] was developed for the coupling of the carboxylic group of an amino acid with the quinolic alcohol to generate an ester linkage (Scheme 1.4). In this case, standard coupling reagents and procedures failed to afford the desired target derivatives. A number of conjugates including PEGylated derivatives with higher solubility were synthesized using the TCFH method.

SCHEME 1.4 TCFH-mediated conjugations for ester linkage.

c01s004

1.2.2 Carbonyl Functional Groups

Aldehydes and ketones are organic compounds that incorporate a carbonyl group (C=O) and are good electrophiles. As such, they undergo nucleophilic addition reactions with various nucleophiles such as amines, N-alkoxyamines (or aminooxy groups), hydrazines, or hydrazide to generate products linked by imine-, oxime-, and hydrazone-reactive groups respectively (Scheme 1.5). The facile synthesis of these carbon–nitrogen double bonds in aqueous solutions at neutral pH makes them attractive for bioconjugation and thus, they have found widespread applications in chemical biology, mainly for the synthesis of nucleic acid conjugates [21]. Aldehydes and ketones are also known as chemical reporters that can tag proteins [22], glycans [23], and other secondary metabolites.

1.2.2.1 Conjugation via Reductive Amination

Reductive amination [24] is a process that transforms a carbonyl group (typically aldehydes and ketones) into an amine via an intermediate imine (Schiff base). Under acidic conditions, the carbonyl group first reacts with primary amines to form a hemiaminal species, which subsequently loses a water molecule to generate a reversible unstable imine. The imine is then trapped irreversibly with a reducing agent in a one-pot reaction to afford a stable amine product (Scheme 1.6). The overall two-step sequence is called reductive amination. Borohydrides are common reducing agents with sodium cyanoborohydride (NaBH3CN) being the most widely used due to its high selectivity to imines and relative unreactivity with oxo groups [25]. Sodium triacetoxyborohydride, NaBH(OAc)3, was introduced as an alternative mild and nontoxic reducing agent of NaBH3CN [26].

SCHEME 1.5 Bioconjugation via carbon–nitrogen double bonds.

c01s005

SCHEME 1.6 Reductive amination process.

c01s006

The reaction of a carbonyl group with an amine proceeds with high chemoselectivity and is also compatible with many functional groups present in biomolecules. Carbohydrate–protein conjugates play vital roles in both basic and applied research [27] and thus, significant efforts to develop simple and efficient chemical methods for the covalent attachment of carbohydrate molecules to proteins have been investigated [28]. Reductive amination remains one of the key methods for the direct conjugation of carbohydrates to the amino group of proteins especially from unprotected free mono- and oligosaccharides [29–31] (Scheme 1.7).

SCHEME 1.7 Bioconjugation of carbohydrates with proteins via reductive amination. Reprinted with permission from Reference 31, Copyright 2008, American Chemical Society.

c01s007

Typical reductive amination protocol for conjugation of sugars with proteins [30]:

1. Dissolve sugar (60 mM), protein (200 μM), and NaBH3CN (300 mM) in aqueous sodium borate buffer (200 mM, pH 9.0).

2. React with stirring at 37–50°C in a thermostated incubator for 10–24 hours.

3. Dialysis against water followed by lyophilization.

Recently, an improved procedure for direct coupling of carbohydrates to proteins via reductive amination was reported [31]. It was found that the addition of a salt (sodium sulfate, 500 mM) in the reaction mixture highly improved the conjugation efficiency. The improved conditions are compatible with microgram quantities of sugar and afford carbohydrate–protein active conjugates for use in assays and incorporation in microarrays.

1.2.2.2 Conjugation via Hydrazone and Oxime Formation

Reaction of carbonyl compounds with primary amines results in formation of unstable reversible Schiff bases, with the equilibrium in water favoring the carbonyl. However, reaction of carbonyl compounds with hydrazides or aminooxy groups leads to formation of Schiff bases (hydrazones and oximes), which are favored in aqueous solution (α-effect nitrogens [32]) as well as being quite stable under physiological conditions [33]. Coupling via hydrazone formation is perhaps the oldest method of bioconjugation. Normally hydrazones are pretty stable; however, in some cases, for example when basic treatment is involved, reduction of the resulting hydrazone linkage with sodium cyanohydride is preferable in order to ensure a stronger covalent linkage. It was observed that the stability of hydrazone depends on the nature of the substituent on the nitrogen and decreases in the following order: aromatic hydrazine > aliphatic hydrazine > hydrazine > hydrazides [34]. Zatsepin et al. used the hydrazone bioconjugation method to synthesize ONT–peptide conjugates (Scheme 1.8) [35]. In this case, the hydrazone formed was prone to hydrolysis in both neutral and basic pH range and hence, reduction was required.

SCHEME 1.8 Synthesis of oligonucleotide (ONT)–peptide conjugates by hydrazone formation followed by reduction. Adapted with permission from Reference 35, Copyright 2002, American Chemical Society.

c01s008

Acylhydrazone linkages were used for the immobilization of peptides to generate peptide microarrays that allowed the sensitive detection of antibodies in blood samples [36]. Aldehydes and ketones react chemoselectively with aminooxy groups to form oxime adducts in mild aqueous solutions. The oxime bond is more stable than hydrazone bond and the reaction proceeds with modest rate in acidic conditions, but are less reactive at pH 7. Recent reports by Dirksen et al. [37] have shown that the rates of imine ligations can be greatly enhanced in the presence of aniline, which behaves as a nucleophilic catalyst. Aniline will first react with the aldehyde to generate an imine. The aniline imine is readily protonated and making it more reactive toward aminooxy reagent. Eventually, loss of aniline leads to formation of the stable oxime product (Scheme 1.9) [38].

SCHEME 1.9 Aniline-promoted nucleophilic catalysis of oxime conjugation. Adapted with permission from Reference 38, Copyright 2009, John Wiley & Sons, Inc.

c01s009

Oxime strategies have been used for the glycosylation of peptides, peptoids, proteins, carbohydrates, microarrays, and small molecules of pharmaceutical interest [39]. The aniline strategy has also been applied to accelerate hydrazone conjugation reactions [40]. The method was used to immobilize antibodies onto surfaces for immune-based biosensing platforms [41]. While both hydrazone and oxime are useful conjugates, they do suffer certain limitations, which sometimes restrict their use in biological applications. They are labile to spontaneous hydrolysis of C=N bonds, and alkyl- and acylhydrazones possess short half-lives (about an hour) under physiological conditions [1(d)].

1.2.2.3 Conjugation via Mannich and Morita–Baylis–Hillman Reactions

The Mannich reaction is a multicomponent condensation of a nonenolizable aldehyde, like formaldehyde (or ketone), a primary or secondary amine (or ammonia) and a C–H activated compound (aliphatic or aromatic carbonyl compounds or electron-rich aromatic compounds such as phenols) to furnish aminoalkylated products [42]. This powerful reaction discovered in the early 1900s was only recently applied for the synthesis of bioconjugates since it results in stable covalent bonds [43]. A one-step three-component Mannich-type reaction was used for the conjugation of aniline-containing peptides to native tyrosine residues on proteins [44] (Scheme 1.50).

SCHEME 1.10 Three-component Mannich reaction for conjugation of protein and peptide. Adapted with permission from Reference 44, Copyright 2008, American Chemical Society.

c01s010

This new bioconjugaton method could be useful in cellular uptake and trafficking studies, protein purification, and materials applications. Recently, Xie et al. used the Mannich reaction to conjugate biomolecules to nanoparticles [45]. Iron oxide nanoparticles functionalized with active hydrogen groups were reacted with amine group-containing cyclic Arginine-Glycine-Aspartic acid (RGD) peptides to develop ultrasmall biocompatible nanoparticles for use as in vivo tumor-targeted imaging agents.

Multifunctional bioconjugation by the Morita–Baylis–Hillman (MBH) reaction in aqueous medium was recently developed for the effective modification of oligosaccharides, peptides, and proteins with fluorescent probes/biotin tags [46]. The MBH reaction involves a carbon–carbon bond formation between the α-position of conjugated carbonyl compounds and carbon electrophiles such as aldehydes or activated ketones. The reaction is usually catalyzed by tertiary amine such as 1,4-diazabicyclo[2.2.2]octane (DABCO) [47]. The resulting β-hydroxy-α-methylene-carbonyl moiety could be further modified via conjugate addition by thiol-based biophysical probes and cysteine-containing peptides and proteins (Scheme 1.11). This mild MBH-based bioconjugation strategy will open up a new direction for the rapid assembly of multifunctional bioconjugates with high structural diversity [46].

SCHEME 1.11 MBH-based multifunctional bioconjugation strategy. Adapted from Reference 46 with permission from the Royal Society of Chemistry.

c01s011

1.2.3 Amine Modifications

Amines are organic compounds that possess a basic nitrogen atom with a lone pair. The dominant reactivity of amines is their nucleophilicity and thus most of their reactions involve nucleophilic-to-electrophilic attacks. Virtually all proteins possess lysine moieties, which have a free amine at the N-terminus. Primary amines are especially nucleophilic and this makes them ideal target for conjugation with other reactive groups. Amine bioconjugation strategies have been mainly used to modify proteins, peptides, oligosaccharides, ligands, and other biomolecules. The two common reactions of amines are alkylation (addition of an alkyl group with loss of a H atom to generate amine linkage) and acylation (replacement of a H atom of amino group by an acyl group to form a stable amide linkage), which are typically used for derivatization of the amine-containing side chains of amino acids such as lysine, arginine, and histidine. These side-chain amines behave as good nucleophiles when they are in their unprotonated forms and a moderately basic pH of 8–10 ensures their reactivity. Amine-bearing ligands have been used to conjugate with alkyl halide-bearing surfaces to form stable amine linkages [48]; however, these alkylation reactions proceeded slowly.

Amines also react with other functional groups bearing electrophilic carbon atoms such as isocyanate (−N=C=O) and isothiocyanate (−N=C=S) to form isourea and isothiourea linkages, respectively. Reactions of isocyanate with amine proceed with good efficiency but are prone to hydrolytic cleavage. Isocyanates deteriorate rapidly upon storage. Alternatively, isothiocyanates react well with amines under alkaline conditions (0.1 M sodium carbonate buffer, pH 9) and are quite stable in water and most solvents. Hydrolysis-resistant diisothiocyanates have been employed for many years for labeling ligands with reporter molecules; however, the thiourea linkage is unstable at lower pH and hence, may be unsuitable for investigations of cell–surface interactions [49]. Antibody conjugates synthesized from fluorescent isothiocyanate have found to degrade over time [50]. Despite all these problems, commercially available fluorescein isothiocyanate (FITC) and tetramethylrhodamine isothiocyanate (TRITC) are still widely used reactive fluorescent dyes for preparing fluorescent bioconjugates (Figure 1.3).

FIGURE 1.3 Structures of fluorescent FITC and TRITC.

c01f003

Recently, Roman and Dong [51] prepared fluorescently labeled cellulose nanocrystals by reacting the primary amino group with the isothiocyanate group of FITC to form a thiourea linkage (Scheme 1.12). These fluorescent bioconjugate will be used to study the interaction of cellulose nanocrystals with cells and the biodistribution of cellulose nanocrystals in vivo.

SCHEME 1.12 Synthesis of fluorescently labeled cellulose nanocrystal bioconjugate.

c01s012

Experimental protocol of FITC conjugation to aminated cellulose nanocrystals following the method of Swoboda and Hasselbach [52]:

1. Add FITC (0.32 mmol/g cellulose) to aminated nanocrystals in 50 mM sodium borate buffer solution (50 mL/g cellulose) containing ethylene glycol tetraacetic acid (5 mM), sodium chloride (0.15 M), and sucrose (0.3 M).

2. Stir reaction mixture overnight in the dark.

3. Dialyze for 5 days.

4. Sonicate suspension for 10 minutes, 200 W with ice-bath cooling.

5. Centrifuge for 10 minutes, 4550 G, 25°C and filter through a syringe filter (0.45 m) to remove any aggregates.

Activated esters are also reliable reagents for amine modification since they can form stable amide bond. Typical examples of active esters are succinimidyl (NHS), tetrafluorophenyl (TFP), 4-sulfotetrafluorophenyl (STP), and 4-sulfodichlorophenyl (SDP) esters (Figure 1.4).

FIGURE 1.4 Chemical structures of active esters for amine modification.

c01f004

Succinimidyl esters display good reactivity with aliphatic amines but low reactivity with aromatic amines, alcohols, and phenols. The major competing reaction of succinimidyl ester conjugation is hydrolysis, which can be minimized if the conjugation is performed below pH 9. Some succinimidyl esters are insoluble in aqueous solution, and as such limit their use for certain specific applications. To overcome this limitation, sulfonated esters such as STP were developed since they are more polar and have better water solubility than simple succinimidyl esters, and can avoid the need of organic solvents in conjugation reactions. STP esters can be prepared from the corresponding phenol (4-sulfo-2,3,5,6-tetrafluorophenol) and are easily purified by chromatography [53]. TFP esters also react smoothly with amines and are more resistant to nonspecific hydrolysis than succinimidyl esters. SDP esters are very hydrolytically stable and have better controlled and consistency in reactions as compared to NHS and TFP esters. Amines react with aldehydes and ketones to generate labile imine linkage, which can be stabilized by reduction with NaBH3CN (reductive amination) to a more stable amine linkage (See Section 1.2.2.1 for more details). Formaldehyde and glutaraldehyde are carbonyl reagents that conjugate with amine via Mannich reactions and/or reductive amination. EDC/NHS-mediated amide bond formation by reaction of an amine with an activated carboxylic compound under physiologic to slightly alkaline conditions (pH 7.2–9) is a very popular and practical conjugation method [54], and this has already been discussed in Section 1.2.1.2 (refer to Scheme 1.3). This method has been commonly used to prepare protein conjugates as well as for the labeling and immobilization of antibodies.

There are several other synthetic chemical groups that will form covalent linkages with amines and this has been well exploited, for example, amine conjugation reactions with epoxides, cyclic anhydrides, imidoesters, carbonates, sulfonyl chlorides, and acyl azides (Table 1.2). Ring-opening reactions of amines with cyclic anhydrides and epoxides are somewhat less reactive and also susceptible to hydrolysis. Imidoesters readily react with amines on proteins with little side reaction such as cross-reactivity with other nucleophilic-reactive groups located on the proteins [55]. The conjugation proceeds well at pH 8–9 and the resulting amidine product does not alter the overall charge of the protein, thereby retaining the native conformation and activity of the protein. Sulfonyl chlorides are very reactive with aliphatic amines at high pH, and protein modification with this reagent is best carried at low temperature. However, they are quite unstable in water, but once conjugated, the sulfonamide linkage is extremely stable [56]. Sulfonyl chlorides are unstable in dimethyl sulfoxide (DMSO) solvent [57], and are not recommended for use in conjugation reactions.

Table 1.2 Conventional Amine Covalent Bioconjugation Strategies

1.2.4 Thiol Modifications

Thiols are organosulfur compounds containing −C–SH or R–SH linkage where R = alkane, alkene, or other carbon-containing groups of atoms. They are also typically known as sulfhydryls or mercaptans. The thiol group is widely distributed in biological materials and represents an important functional center in biological systems [70]. Thiols play crucial role in maintaining the appropriate oxidation–reduction state of proteins, cells, and organisms. Thiol groups are mainly present in cysteine residues of proteins or they can be generated by chemical methods, for example, reduction of native disulfide bonds with dithiothreitol (DTT) (Scheme 1.13a), coupling of primary amino group with 2-iminothiolane (Traut's reagent) (Scheme 1.13b), cystaminiumdichloride reactions with carboxylic acids (Scheme 1.13c) and aldehydes (Scheme 1.13d) followed by reduction with DTT [71].

SCHEME 1.13 Incorporation of thiol groups by chemical methods.

c01s013

However, reformation of disulfide bonds via air oxidation is a common problem during the removal of the reducing agents by dialysis or gel filtration. An alternative powerful reducing agent, namely, tris-(2-carboxyethyl)phosphine (TCEP) [76] was found to prevent disulfide formation and also it does not need to be removed prior to thiol bioconjugation reactions. TCEP is odorless, stable at higher pH and temperature [72] than DTT, and is impermeable to cell membranes and protein hydrophobic core. Depending upon reaction conditions, TCEP is known to react with thiol-reactive chemical groups, such as iodoacetamides and male-imides [73].

Succinimidyl thiolating reagents have also been developed for incorporating thiol groups into lipids, proteins, and nucleic acids by reaction with amine functional group (Scheme 1.14) [74,75].

SCHEME 1.14 Thiolation of amine derivatives with succinimidyl 3-(2-pyridyldithio)-propionate (SPDP) and succinimidyl acetylthioacetate (SATA).

c01s014

Thiols generally behave as nucleophiles due to the presence of lone pairs of electrons on the sulfur atom; however, the corresponding thiolate anion (R–S−) is a more powerful nucleophile in aqueous solutions. Thus, most bioconjugation reactions involve the thiolate anion. Typical thiol-reactive chemical groups include iodoacetamides (or α-halocarbonyl), maleimides, arylating agents (fluorobenzene), and aziridine derivatives, which react via an S-alkylation to form stable thioether linkages (Scheme 1.15)

SCHEME 1.15 Formation of thioester bioconjugates via S-alkylation-type reactions.

c01s015

Iodoacetamides react readily with thiols (usually pH > 7.5) located on proteins, peptides, and thiolated polynucleotides to generate stable thioether bonds [77,78]. They are more reactive than the corresponding bromoacetamides. However, special care has to be taken when carrying the bioconjugation reaction, as iodoacetamides are very unstable in the presence of light or reducing agents. Reaction must be carried out in the dark to avoid liberation of free iodine that can react with histidine, tyrosine, and tryptophan residues. Iodoacetamides have been classically used for determining the presence of free cysteine residues in proteins [79] and recently for immobilization and labeling of proteins [80]. Chloroacetamides turn out to have greater specificity than iodoacetamides for cysteine residues [81].

Acryloyl derivatives (R–CH=CH2) possess reactive double bonds that can undergo addition reactions with nucleophilic thiols. A common example of this class of compounds is the maleimide group. Maleimides react irreversibly with thiols in pH range of 6.5–7.5 to afford thioether linkages, which have frequently been used to synthesize neoglycoconjugates [82,83]. The conjugated double bond of maleimides undergoes an alkylation reaction (Michael addition) with thiolates to generate stable succinimidyl thioether bonds. The maleimide/thiol reaction is known to proceed rapidly in neutral aqueous solutions at room temperature, making it ideal for biological applications. Excess maleimides are removed from reaction mixture at the end of the reaction by quenching with free thiols, such as β-mercaptoethanol. Maleimides are more thiol-selective than iodoacetamides and do not react with histidine, tyrosine, or methionine residues. Competitive reactions of maleimides with amines usually require a higher pH than the reaction of maleimides with thiols. Spontaneous hydrolysis of maleimide moiety [84] in basic aqueous media (pH > 8) gives rise to an N-acyl derivative that can no longer react with thiols [85] and thus competes significantly with thiol modification. Hydrolysis also produces isomeric succinamic acid thioethers resulting in undesirable heterogeneity, which eventually can alter the activity of bioconjugates. Recent studies have shown that molybdate and chromate can deliberately catalyze imido hydrolysis near neutral pH, thereby providing a strategy to decrease the heterogeneity of maleimide-derived bioconjugates [86]. While both iodoacetamide- and maleimide-thiol bioconjugation strategies have been commonly employed, it has been reported that iodoacetamide conjugates are more toxic while maleimide adducts are less stable during the intracellular reactivity and toxicity studies of haloacetyl and maleimido thiol-reactive probes in HEK 293 cells [87].

Aziridine [88], the nitrogenous analog of epoxides, represents a valuable synthetic block due to its electrophilic nature and thus, strong reactivity toward nucleophiles. Thiolates react readily with aziridines under slightly alkaline conditions via a nucleophilic ring-opening reaction to afford stable thioether linkage and a free amine group. Reactions have been carried out in the presence of a catalytic amount of boron trifluoride etherate [89] or stoichiometric amounts of thiobenzoic acid [90]. Recently, addition of catalytic amount amine base (DBU) has been found to promote thiol/aziridine ligation reactions [91]. This base-promoted aziridine ring-opening strategy was used to conjugate various complex thiol moieties such as carbohydrates, lipids, and biochemical tags with aziridine-2-carboxylic acid-containing peptides to prepare complex thioglycoconjugates.

Arylating agents react with thiols in a reaction similar to nucleophilic substitution reactions of simple aromatic halides. Reactions proceed rapidly at or below room temperature in the pH range of 6.5–8.0 to yield stable thioether bonds. Some commonly used arylating agents are benzoxadiazole families, such as nitrobenz-2-Oxa-1,3-diazole derivatives (NBD-Cl and the more reactive NBD-F [92], NBD iodoacetate ester (IANBD ester), NBD iodoacetamide (IANBD amide) [93]), 7-fluorobenz-2-oxa-1,3-diazole-4-sulfonamide (ABD-F) [94], Ammonium 7-fluoro-2,1,3-benzoxadiazole-4-sulfonate (SBD-F) and 4-(N,N-dimethylaminosulfonyl)-7-fluoro-2,1,3-benzoxadiazole (DBD-F) (Scheme 1.16) [70]. NBD is a functional analog of dinitrophenyl hapten. Thiol conjugates of ABD-F are much more stable in aqueous solution than the NBD thiol conjugates [95]. DBD-F has similar properties to ABD-F and SBD-F; however, the order of reactivities with thiols is as follows: DBD-F > ABD-F > SBD-F [70].

SCHEME 1.16 Conjugation of fluorinated benzoxadiazole derivatives with thiols.

c01s016

Thiolates also react with disulfide derivatives via a thiol–disulfide interchange reaction. During the process, thiolate attacks one of the sulfur atoms of the disulfide followed by cleavage of the S–S bond toward the formation of a new mixed disulfide derivative (Scheme 1.17). Depending on the amount of thiol used, other mixed disulfide derivatives could be formed.

SCHEME 1.17 Bioconjugation via disulfide interchange reactions.

c01s017

Among the disulfide derivatives, pyridyl disulfide [96] is the most common one since it reacts with thiols over a broad pH range to form a single mixed disulfide product (Scheme 1.18). Pyridine-2-thione is released as a by-product that helps to monitor the progress of the reaction spectrophotometrically (Amax = 343 nm). It can also be removed from the bioconjugates by dialysis or desalting. The disulfide bioconjugation method has been used to covalently conjugate lipids with ONTs [97].

SCHEME 1.18 Chemical conjugation of thiol with pyridyldithiol reagent.

c01s018

While the classic thiol bioconjugation techniques have been widely used, new methods [98] for selective modification of thiol continue to be investigated. For instance, Caddick and coworkers [98(a)] recently reported a new thiol bioconjugation strategy of green fluorescent protein (GFP) via an unexpectedly stable cyclic sulfonium intermediate. The reaction involves the treatment of GFP with bisalkylating agents under Davis' conditions [99] followed by treatment with nucleophiles (Scheme 1.19). This new bioconjugation method will open a new entry to functionalized proteins consisting of useful chemical motifs.

SCHEME 1.19 Bioconjugation via Davis bisalkylation and nucleophilic ring-opening reactions.

c01s019

Davis and coworkers disclosed a two-step protocol for cysteine modification via an oxidative elimination of cysteine into a dehydroalanine followed by a Michael addition with thiol reagents to afford a stable thioether linkage (Scheme 1.60) [98b]. The oxidative elimination step is induced by O-mesitylenesulfonylhydroxylamine (MSH) under alkaline conditions (pH 10–11). The reaction is compatible with methionine residues. However, the Michael addition step is not stereospecific and thus, the stereochemistry of the cysteine is not preserved. This new bioconjugation method allows the preparation of protein–carbohydrate and protein–peptide conjugates.

SCHEME 1.20 Davis' two-step method for cysteine modification. Adapted with permission from Reference 98(b), Copyright 2008, American Chemical Society.

c01s020

1.2.5 Hydroxyl Modifications

The hydroxyl group (−OH) is prevalent in many biologically important molecules, namely carbohydrates, lipids, nucleotides, glycans, peptides/proteins (Thr, Ser, and the phenolic hydroxyl of Tyr), and natural products. In the field of synthetic chemistry, there is an exhaustive list of hydroxyl transformations, and the latter have mainly been applied in the synthesis of natural products. These transformations are generally carried out in an organic solvent and the absence of water. However, hydroxyl bioconjugation reactions turn out to be more challenging as compared to other functional groups such as amines, thiols, or carboxylic acids owing to the hydroxyl's relatively low nucleophilicity. Moreover, hydroxyl bioconjugation reactions are often thwarted by the presence of water. Despite these problems, the covalent bioconjugation of two molecular entities via an alcohol has been developed and is generally placed in the following categories:

1.2.5.1 Hydroxyl Activation

The hydroxyl group is transformed to a good leaving group in the presence of an activating agent (A) followed by nucleophilic attack, usually by an amine, to yield a carbamate linkage (Scheme 1.21).

SCHEME 1.21 Bioconjugation via hydroxyl activation and nucleophilic displacement.

c01s021

Typical activating agents are N,N′-Carbonyldiimidazole (CDI), N,N′-Disuccinimidyl carbonate (DSC) and N-Hydroxysuccinimidyl chloroformate (Figure 1.5). The active intermediate generated by the reaction of CDI with hydroxyl group can react readily with amines to afford a stable carbamate linkage and the concomitant release of imidazole. This bioconjugation strategy has been applied for immobilization of amine-containing affinity tags [100]. Succinimidyl carbonate was used to activate poly(ethylene) glycol for subsequent coupling with proteins [101]. The major disadvantage with these activating agents is that they are not selective for the hydroxyl group, since they preferentially react with amines and carboxylates.

FIGURE 1.5 Chemical structures of hydroxyl activating agents.

c01f005

1.2.5.2 Hydroxyl Oxidation

Hydroxyl groups located on adjacent carbons (commonly known as cis-diols) can be easily oxidized by sodium periodate to the corresponding dialdehydes (Scheme 1.22). Periodate oxidation method has been used for conjugation of amine-containing dyes with sugars and polysaccharides and other molecules possessing cis-diols. The reactive aldehyde formed can then be used to conjugate with other biomolecules via reductive amination, oxime, or hydrazone [102] methods.

SCHEME 1.22 Sodium periodate oxidation of diols followed by conjugation with hydrazide to generate a covalent hydrazone linkage.

c01s022

Sodium periodate oxidation has also been used to oxidize adjacent hydroxyl groups on an aromatic ring (for example, L-3, 4-dihydroxyphenylalanine derivatives) to generate a reactive ortho-quinone intermediate, which can be easily intercepted by a nearby nucleophile through an intramolecular Michael addition (Scheme 1.23) [103].

SCHEME 1.23 Ortho-quinone formation by periodate oxidation and intramolecular nucleophilic attack of pendant amine.

c01s023

This novel covalent bioconjugation method was recently extended for the conjugation of protein with a polysaccharide (Scheme 1.24) [104]. However, in this case, after oxidation, the conjugation occurs via an intermolecular Michael addition of the amine from chitosan.

SCHEME 1.24 Schematic representation for the bioconjugation of protein with a polysaccharide. Reprinted with permission from Reference 104, Copyright 2011, American Chemical Society.

c01s024

Enzymes such as galactose oxidase have also been exploited for the oxidation of hydroxyl groups on polysaccharide chains [1a].

1.2.5.3 Phenolic Hydroxyl Modifications

The phenolic hydroxyl group of tyrosine is only modestly prevalent and is often buried within the protein structure [105]. As such, it represents an attractive target for bioconjugation. Two main sites of tyrosine residues can be targeted, namely the phenolic hydroxyl group or the electron-rich aromatic ring position ortho to the phenolic hydroxyl (Scheme 1.25). In 2004, the Francis group [106] disclosed the first report of a chemoselective method for tyrosine targeting, and further studies established that electron-deficient diazonium salts could be added to tyrosine residues by electrophilic aromatic substitution (Scheme 1.25(A)). Other tyrosine modifications include:

SCHEME 1.25 Covalent strategies for tyrosine modifications. Reprinted from Reference 105 with permission from Royal Society of Chemistry.

c01s025

A three-component Mannich-type coupling involving tyrosine, an aldehyde, and an electron-rich aniline to generate an O-substituted tyrosine moiety (Scheme 1.25(B)). Strategy has been applied to conjugate proteins with synthetic peptides [44].

An ene-type reaction between tyrosine and a cyclic diazocarboxamide to afford a highly stable 1,2,4-triazoldine-3,5-dione derivative (Scheme 1.25(C)) [107]. This mild, aqueous reaction works well over a broad pH range. An integrin-binding cyclic RGD peptide was conjugated to the therapeutic antibody Herceptin using this method.

A water-compatible selective palladium-catalyzed allylic oxidation of tyrosine residues (Scheme 1.25(D)). Method employs electrophilic π-allyl intermediates derived from allylic acetate and carbamate precursors to modify proteins at room temperature. This transition metal-catalyzed technique provides access to the preparation of synthetic lipoproteins [108].

A cerium-based transition metal-catalyzed strategy recently developed by Francis and coworkers [109]. The electron-rich aromatic ring of tyrosine undergoes an oxidative coupling with electron-rich anilines in the presence of cerium(IV) ammonium nitrate (CAN) as a one-electron oxidant (Scheme 1.24(E)). Attributes of this new bioconjugation strategy are excellent chemoselectivity, mild conditions, low concentration of oxidant and coupling partners, short reaction times, and high yields. Proteins were selectively modified with poly(ethylene)glycol (PEG) and smaller peptides.

1.2.6 Native Chemical Ligation and Expressed Protein Ligation

Native chemical ligation (NCL) is a powerful technique for the linking of two or more unprotected peptides to form large peptide–peptide conjugates. The NCL process involves reaction between a peptide with an N-terminal cysteine and a second peptide having a C-terminal thioester to afford an amide linkage (Scheme 1.26). The key step for the formation of the amide bond is via an S–N acyl shift. Historically, in 1953, Wieland et al. disclosed the chemical foundation of this chemical transformation [110], particularly the S–N acyl transfer step during the synthesis of the dipeptide, valine–cysteine. However, it was not until 1994that Kent and coworkers [111] at the Scripps Research Institute reported the ligation of thioesters with N-terminal cysteine residues to yield a native amide or peptide bond. As such, the reaction was termed NCL.

SCHEME 1.26 Schematic representation of NCL.

c01s026

Mechanistically, the reaction proceeds via a reversible, chemoselective, and regioselective transthioesterification, which connects the peptides through an intermediate thioester. The thioester intermediate undergoes a spontaneous irreversible intramolecular S–N acyl shift rearrangement to form the amide bond at the ligation site and regenerate the cysteine side-chain thiol. While the exact nature of the NCL mechanism is still unclear [112,113], a generally accepted proposed mechanism is depicted in Scheme 1.27.

SCHEME 1.27 Proposed mechanism of NCL.

c01s027

Usually, the NCL reaction is enhanced by the presence of thiol catalysts, since the peptide-α-thiolalkyl esters are relatively unreactive under NCL reaction conditions in water at pH = 7. In this case, a thiol–thiolester exchange reaction occurs prior to the transthioesterification step. Common thiol catalysts used for NCL are a benzyl mercaptan (1%)/thiophenol (3%) mixture for chemically synthesized peptide thioesters [114], or 2-mercaptoethanesulfonate sodium salt (MESNa) for recombinant peptide thioesters [115]. However, ligations still usually require long reaction times accompanied with side reactions. A recent study has shown that aryl thiols could be effective catalysts, in particular, (4-carboxymethyl)thiophenol (MPAA) is a highly effective one for NCL [112]. MPAA is water soluble and does not have an offensive odor as one would expect for typical thiol compounds. Chemical ligations are complete within an hour, with high yields.

A modified version of NCL was reported by Tam and coworkers [116], which employed an N-terminal α-bromoAla whereby reaction with a C-terminal thioester gave the covalent thioester linkage (Scheme 1.28).

SCHEME 1.28 Tam's modified version of NCL.

c01s028

NCL reaction is highly chemoselective for the ligation of two diverse functionalized molecules under physiological conditions and does not require any protecting groups. These properties make NCL an attractive and powerful method for modification, synthesis, and semisynthesis of peptides and smaller proteins (chain length < 200 amino acids (aa)). Proteins synthesized by NCL are much larger than the conventional solid-phase peptide synthesis (SPPS) (chain length < 60 aa) [117]. However, larger proteins cannot be easily synthesized by one NCL step. Multi-step NCL of different peptide segments are required [118]. NCL process has been applied to prepare dendrimer–peptide/proteins (GFP) [119] (Scheme 1.29 [120]) and protein–liposome conjugates [121].

SCHEME 1.29 Peptide–dendrimer and fluorescent protein–dendrimer conjugates via NCL.

c01s029

Despite the widespread applications of NCL, the chemical synthesis of thioesters has always been the bottleneck in NCL. Most of the thioesters' preparation relied previously on the Boc strategy [111, 116b] using solid-phase peptide synthesis due to the base lability of the thioester. Recent studies have demonstrated that 9-fluorenylmethoxycarbonyl (Fmoc) strategy [122–124] could be used instead of the Boc method to improve the yield of thioester. Several Fmoc-deprotection methods [125–127] were developed in order to liberate the peptide thioester from the resin in high yields.

NCL applications were further enhanced by its marriage with recombinant protein technology toward new powerful approaches to protein semisynthesis. The resulting combined technology is known as Expressed Protein Ligation (EPL). EPL [115, 128–131] also known as intein-mediated ligation of expressed proteins [132] allows recombinant and synthetic unprotected polypeptides to be chemoselectively and regioselectively combined together via a native peptide bond under mild aqueous conditions. Since its discovery, EPL's applications have grown significantly in order to address complex biological questions [133]. The overall EPL technique for protein semisynthesis involves the following general steps (Scheme 1.70):

a. Recombinant protein to be ligated is first expressed as an N-terminal intein, which is fused with a chitin-based domain (CBD) on the C-terminal side of the intein. The CBD aids in the affinity purification process.

b. NCL is initiated in situ by incubating carboxy-terminal Cys peptides with the protein thioesters in the presence of thiol (e.g., thiophenol) in buffer.

c. Chitin-bound intein is removed by filtration resulting in purified semisynthetic protein.

SCHEME 1.30 Schematic representation of EPL process. Reprinted with permission from Reference 115.

c01s030

EPL exploits inteins as a means to form a C-terminal thioester [1e]. N-terminal Cys polypeptides can be obtained recombinantly, using engineered inteins, or by chemical synthesis (SPPS). While the chemical section can be as small as possible, the expressed part is not limited in size. Thus, EPL allows the synthesis of chemically modified proteins [134] of chain length greater than 500 aa, which overcomes the size limitation of NCL. However, in spite of numerous applications of both NCL and EPL, the requirement of a cysteine residue (mimic) at the ligation site is still a major obstacle. In this respect, recent studies have developed to circumvent this limitation. For example, NCL with cysteine mimetics (N-α-(ethanethiol) or an N-α-(oxyethanethiol)) followed by treatment with Zn/H+ afforded Gly at the ligation site [135], while NCL combined with desulfurization (Ni/H2) led to an Ala residue [136] (Scheme 1.31).

SCHEME 1.31 Alternate ligation methods: (i) Gly and (ii) Ala at ligation sites.

c01s031

A closely related technology to EPL is protein trans-splicing (PTS), which is also based on the use of inteins. In PTS, artificially or naturally split inteins are employed to form a new peptide bond between their flanking exteins [137,138]. PTS has enabled the extension of NCL in living systems for cyclic peptide synthesis [139,140], in vivo semisynthesis of proteins [141], and the study of protein–protein interactions [142].

1.2.7 Cross-linking Reagents for Bioconjugation

One of the most useful and ready tools for bioconjugation is perhaps the cross-linking reagents. Cross-linking generally refers to the process of chemically combining two or more molecules via a covalent linkage. These cross-linking reagents (or cross-linkers) possess reactive end-groups that response to specific functional groups such as amines, thiols, carboxyls, and carbonyls. Proteins have many of these functional groups, and thus proteins [143] and peptides are the most studied biomolecules using cross-linking methods; however, cross-linkers are also often used to modify drugs, nucleic acids, and solid surfaces. Nowadays, there is a long, growing list of cross-linkers that are mostly commercially available from many suppliers (Thermo Scientific Pierce, G-Biosciences, Cyanagen, Sigma-Aldrich, etc.). While this makes it easier from a synthetic point of view, it becomes more overwhelming when choosing the correct cross-linker for a particular application. The choice of cross-linkers usually depends on their chemical reactivity and properties with respect to the application. Cross-linkers are normally selected based on the following important features [144]:

a. Chemical specificity: type of reactive groups; whether the reagent possesses the same or different reactive end-groups. Usually, a cross-linker will have a minimum of two reactive groups at either end.

b. Spacer arm or connectors: the length [145] and nature of the spacer arm; the conformational flexibility, hydrophilicity, or hydrophobicity.

c. Cell-membrane permeability: whether the reagents are permeable or impermeable to cells/membranes.

d. Chemical reactivity: whether the reagent will react spontaneously upon addition or it can be activated at a specific time; whether the reagent is photo reactive.

e. Cleavability: whether the cross-linker could be cleaved or reversed when required.

f. Important moieties: whether the reagent contains moieties that can be radiolabeled or tagged with another label.

Note: Readers are directed to the Pierce website (www.piercenet.com) which contains a user-friendly cross-linker selection guide by which the above-listed features may be chosen and a list of those cross-linkers with those selected features will be quickly generated.

As mentioned earlier, cross-linkers have at least two reactive end-groups and those with two reactive groups are usually termed as bifunctional cross-linkers. Bifunctional cross-linkers are further classified as either homobifunctional or heterobifunctional reagents depending on whether they have the same or different reactive groups (Figure 1.6).

FIGURE 1.6 Schematic representation of homobifunctional (same end-functional groups) and heterobifunctional (different end-functional groups) cross-linkers.

c01f006

1.2.7.1 Homobifunctional Cross-linkers

Homobifunctional cross-linkers possess similar reactive groups at opposite ends of the cross-linker's spacer arm and are usually symmetrical in design. They have the advantage of being reacted in a one-step chemical cross-linking reaction. However, self-conjugation, polymerization, and intracellular cross-linking are common issues with homobifunctional cross-linking agents. Polymerization can be minimized in a two-step protocol by first allowing the cross-linker to react with one biomolecule, followed by removal of excess cross-linker and by-products. The second biomolecule is then added and allowed to react with the previously activated biomolecule. A common problem in the two-step protocol is that the activated biomolecule intermediate can hydrolyze and degrade rapidly prior to the cross-linking step. Despite these issues, homobifunctional cross-linkers continue to be widely used as they do afford effective bioconjugates. Given the vast number of homobifunctional cross-linkers available, the following tables arrange them with respect to their reactivity toward amino groups, thiols, and other important functional groups. A summary of their attributes and some useful literature references are also provided.

Amines, lysine ɛ-amines, and N-terminal α-amines are the most abundant groups on proteins, and thus have been the most common target for cross-linking. The two most common amine-reactive groups that have been targeted for cross-linking are the imidoesters and NHS esters (see also Section 1.2.3). Imidoester homobifunctional cross-linkers are among the oldest cross-linkers [146], developed to react with primary amines and afford amidine linkages (Scheme 1.32, Table 1.3). The cross-linking occurs rapidly at pH 10, although amidine formation is favored at pH 8–10. They are highly water soluble but possess short half-lives [147,148]. The resulting amidine bioconjugate is protonated, and therefore carries a positive charge at physiological pH [149,150]. Imidoesters can penetrate cell membrane and cross-link proteins within the membrane and thus, imidoester homobifunctional cross-linkers have been used for the study of protein structure, molecular associations in membranes and immobilization of proteins onto solid-phase supports. The amidine linkages formed by imidoester cross-linkers are reversible at high pH and, hence, the more stable and efficient NHS–ester cross-linkers were developed (Table 1.4)[151,152]. Due to the latter excellent reactivity at physiological pH, they have slowly replacing imidoester homobifunctional cross-linkers. NHS-ester cross-linkers react with amines in phosphate/carbonate or borate buffers (50–200 mM) to form amide bonds with the release of N-hydroxysuccinimide as a by-product (Scheme 1.32). NHSester cross-linkers can be grouped into two types depending on their water-solubility properties; however, they all have almost same reactivity toward amines. Water-insoluble NHS esters usually need to be first dissolved in an organic solvent (DMSO or DMF) prior to addition to the aqueous reaction mixture. Sulfo-NHS esters were introduced as water-soluble and membrane impermeable cross-linkers and can be used when organic solvents are not tolerable to the reaction conditions [153]. Sulfo-NHS cross-linkers possess better half-lives of hydrolysis than the NHS-ester cross-linkers [154]. They have been mainly employed for cell–surface conjugation since they cannot permeate the membrane.

SCHEME 1.32 Conjugation of amines with imidoesters and NHS esters.

c01s032

Note: The general precaution for amine conjugation is to avoid buffers containing amines such as Tris or glycine.

Coupling through thiol (sulfhydryl) groups is advantageous since it can be site directed, allow for sequential coupling and yield cleavable products. Usually, a protein in a complex mixture can be specifically labeled if it is the only one with a free thiol group on its surface. The common thiol-reactive groups that have been exploited for the design of cross-linkers are maleimides, iodoacetyl, and pyridyl disulfides (Figure 1.7) (see Section 1.2.4 for more details).

FIGURE 1.7 General representation of thiol-reactive homobifunctional cross-linkers: (a) maleimide, (b) iodoacetyl, (c) pyridyldisulfide.

c01f007

Maleimides have been routinely used for the design of thiol-reactive homobifunctional cross-linkers, which typically react at near neutral pH (6.5–7.5) to afford stable irreversible thioether linkages. Maleimides are unreactive toward tyrosine, histidine, or methionine residues. Thiols must be excluded from reaction buffers used with maleimides in order to prevent competition for coupling sites. Excess maleimides can be quenched at the end of a reaction by adding free thiols while ethylenediaminetetraacetic acid (EDTA) is often included in the coupling buffer to minimize oxidation of thiols. Many thiol-reactive homobifunctional cross-linkers with different spacer lengths are reported in the literature [188–198]; however, most of these are not easily available and have to be synthesized. The following table 1.5 lists the ones that are commonly employed and commercially available from many suppliers.

Note: The general precaution for thiol conjugation is to remove reducing agents from the conjugation reaction and add a metal chelating agent (EDTA) as an antioxidant.

Other homobifunctional cross-linkers reactive toward thiols that have been reported include pyridyl disulfide (namely 1,4-di-[3′-(2′-pyridyldithio)propionamido]butane (DPDPB)), bisiodoacetamide, and bisepoxide derivatives (Figure 1.8). DPDPB reacts with thiols to form disulfide linkages [205,206]. DPDPD is water insoluble and has to be dissolved in an organic solvent (25 mM DPDPB in DMSO) prior addition to reaction mixture. The disulfide linkage is cleavable with reducing agents. Haloacetamide cross-linkers react with thiols at physiological pH resulting in stable thioether linkages [196]. Nucleophilic ring-opening reactions of thiols with bisepoxides at pH 7.5–8.5 afford thioether bonds and hydroxyl groups [207]. Bisepoxides can also react with nucleophilic amines.

FIGURE 1.8 Homobifunctional cross-linkers reactive toward thiols: (a) DPDPB, (b) bisiodoacetylamide derivatives, and (c) 1,4-butanediol diglycidyl ether (bisepoxides).

c01f008

Homobifunctional cross-linking reagents reactive toward carbonyl groups (aldehydes or ketones) are mainly derivatives of hydrazides. Hydrazide-activated cross-linkers will react with aldehydes or ketones at pH 5–7 to generate stable hydrazone linkages. While aldehydes do not readily exist in proteins or macromolecules, they can be readily introduced by mild oxidation of vicinal diols in carbohydrates using sodium meta-periodate. The oxidation has to be carried out in the dark at 0–4°C to avoid side reactions.

These bishydrazide cross-linkers could be used in a single step for the bioconjugation process; however, it is more efficient to perform the cross-linking via a two-step protocol. The two-step protocol involves, first, the addition of a huge excess of hydrazide cross-linker to an aldehyde-containing biomolecule 1 resulting in a hydrazide-activated biomolecule. Excess unreacted hydrazide cross-linker is then removed by desalting followed by addition of biomolecule 2 (Scheme 1.33).

SCHEME 1.33 Conjugation of hydrazide homobifunctional cross-linker with aldehyde-containing biomolecules via a two-step method.

c01s033

The two most commonly used and commercially available (from Aldrich) hydrazide homobifunctional cross-linkers are carbohydrazide and adipic acid dihydrazide (ADH) (Figure 1.9). Carbohydrazide [208,209] is a small five-atom spacer homobifunctional cross-linker with excellent water solubility properties. ADH, on the other hand, is a 10-atom spacer homobifunctional cross-linker, which is also water soluble. ADH was introduced in 1980 in glycoconjugate synthesis for the preparation of Haemophilus influenzae type b polysaccharide–protein conjugates and since then has been a popular reagent for constructing lattice-like constructs of polysaccharides with proteins [210–213].

FIGURE 1.9 Chemical structures of carbohydride and ADH.

c01f009

Similarly, the reverse of hydrazide-activation strategy described above, involves the use of dialdehyde derivatives as homobifunctional cross-linking reagents for conjugation of biomolecules. Aldehydes will react with hydrazides and amines at pH 5–7 to form hydrazone and imine linkages respectively. Imine linkage can be subsequently reduced by NaBH3CN to the more stable amine bond. The reaction with hydrazides is typically faster than with amines, rendering them vital for site-specific cross-linking. Among the dialdehyde homobifunctional cross-linkers (glyoxal, malonaldehyde, succinaldehyde), glutaraldehyde (linear five-carbon spacer) is the most popular and has been extensively used by immunologists in the 1970s for the bioconjugation of an enzyme with an antibody [214–216] (Scheme 1.34). Glutaraldehyde's popularity is mainly due to its commercial availability, low cost, high reactivity, and thermally and chemically stable cross-links [217]. In this case also, the two-step experimental protocol is more efficient than the single-step bioconjugation method [216]. In spite of the widespread use of glutaraldehyde as a cross-linker, it is still unclear about the nature of the mechanism of cross-linking with proteins, which has been a subject of huge debate in the past years. This is mainly because glutaraldehyde can exist in different forms even for specific and controlled reaction conditions [218].

SCHEME 1.34 Conjugation of dialdehyde-homobifunctional cross-linker with amine-containing biomolecules via a two-step method.

c01s034

1.2.7.2 Heterobifunctional Cross-linkers

Heterobifunctional cross-linkers have two different reactive groups that allow for sequential (two-stage) conjugations. The major advantage of heterobifunctional cross-linkers is that it helps to minimize undesirable polymerization or intramolecular cross-linking by allowing the least stable group to react first, at such condition where the other group is nonreactive. By far, the most commonly used heterobifunctional cross-linkers are those consisting of an amine-reactive succinimidyl ester (e.g., NHS ester) at one end and a thiol-reactive group on the other end connected by a spacer (Figure 1.10).

FIGURE 1.10 Schematic representation of heterobifunctional cross-linker reactive toward amines and thiols.

c01f010

The thiol-reactive groups are typically maleimides, pyridyl disulfides, and α-haloacetyls. The NHS-ester reactivity is less stable in aqueous solution and is usually reacted first in sequential cross-linking procedures with amines to afford amide linkages. Table 1.6 lists the various commercially available (Thermo Scientific Pierce) noncleavable heterobifunctional cross-linkers (NHS ester-spacer-Maleimide) reactive toward amines and thiols having an aliphatic, aromatic, or a cyclohexane ring as spacer arms. Figure 1.11 depicts heterobifunctional cross-linkers reactive toward amine and thiols possessing different lengths of PEG-based spacers. The general structure is NHS ester-PEGn-Maleimide, which differs in the number of discrete ethylene glycol units (n = 2, 4, 6, 8, 12, or 24). The PEG (or polyethylene oxide (PEO)) spacer arms help in maintaining the solubility of the bioconjugate. Other advantages include increased stability, highly flexible, nontoxic and reduced aggregation, and immunogenicity. Figure 1.12 illustrates NHS ester-spacer-iodoacetyl heterobifunctional cross-linkers whereby NHS ester reacts with primary amines at pH 7–9 to form stable amide bond while haloacetyl group reacts with thiols at pH > 7.5 to afford stable thioether linkage. Figure 1.13 shows cleavable heterobifunctional cross-linkers containing NHS ester-spacer-dipyridylsulfide. Disulfide bond in the spacer arm can be readily cleaved by reducing agents such as 10–50 mM DTT or tris-2-(carboxyethyl)phosphine (TCEP) at pH 8.

FIGURE 1.11 General structures of PEG- or PEO-based heterobifunctional cross-linking reagents commercially available (a) Thermo Scientific Pierce (b), (c), (d) Cyanagen.

c01f011

FIGURE 1.12 NHS ester/iodoacetyl heterobifunctional cross-linking reagents.

c01f012

FIGURE 1.13 Cleavable NHS ester/pyridyldisulfide heterobifunctional cross-linking reagents.

c01f013

Table 1.3 Imidoester Homobifunctional Cross-linkers Reactive toward Amino Groups

Table 1.4 NHS-Esters Homobifunctional Cross-linkers Reactive toward Amino Groups