Neoglycoenzymes - Chemical Reviews (ACS Publications)

Neoglycoenzymes - Chemical Reviews (ACS Publications)https://pubs.acs.org/doi/abs/10.1021/cr400290xCachedSimilarby ML Vi...

0 downloads 259 Views 19MB Size

Download PDF

Review pubs.acs.org/CR

Neoglycoenzymes María L. Villalonga,† Paula Díez,‡ Alfredo Sánchez,‡ María Gamella,‡ José M. Pingarrón,‡,§ and Reynaldo Villalonga*,‡,§ †

Center for Enzyme Technology, University of Matanzas, Matanzas 44740, Cuba Department of Analytical Chemistry, Faculty of Chemistry, Complutense University of Madrid, 28040-Madrid, Spain § IMDEA Nanoscience, Cantoblanco Universitary City, 28049-Madrid, Spain ‡

Notes Biographies Acknowledgments Abbreviations References

1. INTRODUCTION The rational design of proteins with desired functional properties by manipulation of their chemical composition and three-dimensional structure is a challenge that has fascinated researchers in diﬀerent disciplines during the last century. In this context, the ultimate goal of enzyme engineering has been to create novel and highly active enzymes with increased stability, selectivity, and substrate range, able to work in extreme physicochemical and biological conditions.1,2 In nature, the structure−function relationship for enzymes is modulated by genetic, enzymatic, and chemical mechanisms.3 Among these, glycosylation constitutes the most extended natural approach for protein modiﬁcation. In general, carbohydrate units confer important physicochemical and biological properties to glycoproteins such as conformational stability, protease resistance, hydrophilicity, charge, aqueous solubility, cell and biomolecular recognition, and reduced immunogenicity. Thus, an attractive strategy for the tailor-made preparation of enzymes with speciﬁc properties has been the in vitro covalent modiﬁcation with carbohydrates moieties. These artiﬁcially created neoglycoenzymes have attracted noticeable attention from the scientiﬁc community since their ﬁrst preparation in 1960.4 They have been synthesized through diﬀerent chemical, enzymatic, and chemo-enzymatic approaches and by employing a great assortment of low and high molecular weight carbohydrate derivatives. Neoglycoenzymes have been also successfully employed as robust catalysts in homogeneous media, analytical tools in glycohistochemistry, long-circulating and target-speciﬁc protein drugs, and for the preparation of immobilized enzyme biocatalysts and biosensor devices. The coverage of this Review includes aspects concerning neoglycoenzyme synthesis, classiﬁcation, structural and functional properties, and applications. For this purpose, reports of outstanding signiﬁcance in neoglycoenzyme science will be critically discussed. A detailed description of the synthetic methods used for preparing protein glycoconjugates as well as the properties and applications of these derivatives are beyond

CONTENTS 1. Introduction 1.1. Why Neoglycoenzymes? 1.2. Historical Background 2. Strategies for Preparing Artiﬁcial Glycoenzymes 2.1. Structural Considerations 2.2. Site-Speciﬁc versus Group-Speciﬁc Coupling Strategies 2.3. Synthetic Methods for in Vitro Enzyme Glycosylation 2.3.1. Chemical Methods 2.3.2. Enzymatic Methods 2.3.3. Chemoenzymatic Methods 2.3.4. Rational Design and Synthesis of Neoglycoenzymes 3. Classiﬁcation of Neoglycoenzymes 3.1. Monosaccharide-Based Neoglycoenzymes 3.2. Oligosaccharide-Based Neoglycoenzymes 3.3. Polysaccharide-Based Neoglycoenzymes 3.4. Synthetic Glycopolymers-Based Neoglycoenzymes 4. Structural and Functional Properties of Artiﬁcial Glycoenzymes 4.1. Structural and Conformational Characteristics 4.2. Catalytic and Kinetics Properties 4.3. Stability 4.4. Biorecognition Properties 4.5. Other Characteristics 5. Applications 5.1. Neoglycoenzymes for Biomedical Application 5.2. Neoglycoenzymes for Industrial Application 5.3. Neoglycoenzymes in Chemical and Clinical Analysis 5.4. Neoglycoenzymes in Nanotechnology 6. Conclusions and Outlook Author Information Corresponding Author © 2014 American Chemical Society

4909 4910 4911 4911 4911

4868 4869 4869 4870 4871 4872 4872 4873 4889 4891 4891 4894 4895 4896 4896 4897 4898 4898 4898 4899 4900 4901 4901 4901 4906 4907 4908 4909 4909 4909

Received: May 28, 2013 Published: February 4, 2014 4868

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

neous systems. In vitro modiﬁcation also favors the tailor-made design of enzyme derivatives with desirable properties by selecting the appropriate transforming substances. Glycosylation is the most important and complex co- and post-translational protein modiﬁcation process.46,47 This natural transformation serves to expand the structural and functional diversity of the proteome, often resulting in the production of heterogeneous mixtures of glycoforms from a single protein molecule but diﬀering in both the nature and the site of glycosylation. In nature, the majority of proteins are glycosylated, and the carbohydrate chains play a relevant role in the maintenance of the biologically active protein conformation.48,49 In fact, glycosylation is essential for full catalytic activity of several naturally occurring glycoenzymes.50 Glycosylation also confers solubility, structural and thermal stability, protection against proteolytic degradation, and reduces the immunogenicity of these proteins.51−53 With respect to structural diversity, carbohydrates have the ability to far exceed proteins and nucleic acids as agents encoding information for speciﬁc molecular recognition. This fact allows glycoconjugates to act as mediators of an enormous variety of cellular events based on speciﬁc interactions between lectins and complex carbohydrates such as phagocytosis, endocytosis, intracellular traﬃc, signal transduction, cell−cell recognition, inﬂammation processes, cell adhesion, opsonization, cell growth control, cell regulation and diﬀerentiation, acrosome reaction, cellular traﬃcking, cancer cell metastasis, and cell death.13,54 Under these bases, it is reasonable to expect that similar properties can be conferred to nonglycosylated proteins through the manipulation of their surfaces by attaching carbohydrate moieties. This hypothesis opened the new research ﬁeld of neoglycoconjugates in 1929, when Goebel and Avery synthesized for the ﬁrst time an artiﬁcially glycosylated protein by preparing the diazonium salts of 4aminophenol β-D-glucoside and 4-aminophenol β-D-galactoside and further attaching to the aromatic amino acids residues at the surface of horse serum globulin.55 These neoglycoproteins were used as antigens to induce the production of immune sera, demonstrating the presence of anticarbohydrate antibodies in sera prepared with the neoglycoproteins, which were speciﬁc of the sugar borne by the neoglycoprotein used as immunogen but also recognized unrelated proteins containing the same carbohydrate residues. Neoglycoenzymes, which can be deﬁned as enzymes that have been artiﬁcially glycosylated in vitro, belong to the broad concept of neoglycoconjugates. Such deﬁnition excludes those enzyme-based neoglycoconjugates prepared by genetic manipulation of cells, then introducing or changing new sugar pattern in the target protein in vivo.

the scope of this Review and have been excellently covered in previous papers.5−17 Therefore, only the synthetic methods speciﬁcally adapted to enzyme artiﬁcial glycosylation will be discussed. This Review will highlight the new and/or improved properties showed by the artiﬁcial glycoenzymes as well as their most relevant applications. The predicted impact of neoglycoenzymes in new emerging research areas will be also encompassed. 1.1. Why Neoglycoenzymes?

Enzymes are globular proteins with catalytic activity that are responsible for supporting almost all of the chemical reactions occurring in living organisms. They play an essential role in the regulation of the structure and function of organelles, cells, and organisms by catalyzing the synthesis and transformation of biochemical building blocks and macromolecules, the transmission of genetic information, the conversion of chemical energy, the transport of compounds across the membranes, and the motility of organisms.18,19 Enzymes far exceed man-made catalysts in their reaction speciﬁcity and catalytic eﬃciency, operating either in aqueous or in organic solvents under very mild conditions of temperature, pressure, pH, and ionic strength. Enzymes are also capable of increasing the speed of a broad range of chemical transformations with exquisite substrate stereo-, regio-, and chemoselectivity.20,21 These unique characteristics have supported the increased use of enzymes as analytical tools, therapeutic agents, cosmetic and laundry active components, and catalysts for chemical, biotechnological, agricultural, food, and pharmaceutical industries.22−27 Nowadays, over 3000 enzymes have so far been identiﬁed, and this number should be increased by the advances in the “omic” technologies. In addition, the cost for the large-scale production of enzymes has been considerably reduced due to the developments in genetic engineering, cell and tissue culture, and downstream processing, thus opening the possibility to design a great variety of cost-eﬀective and ecological-friendly enzyme-catalyzed applications.23−25,28 However, the number and diversity of enzyme-based applications are still modest, mainly restricted by the intrinsic low structural and functional stability of the enzyme molecules.29,30 In fact, the catalytic capability of enzymes is a consequence of their complex three-dimensional structure, which deﬁnes the composition and geometry of the active site. Unlike other natural polymers, which adopt low ordered conformations in aqueous solutions, the polypeptide chains folded in well-deﬁned structures yielding a characteristic biologically active 3D conformation for each protein. This three-dimensional structure is genetically determined, but environmentally conditioned. Thus, minor changes in the physicochemical characteristics of the surrounding media can produce enzyme denaturation, and, consequently, a signiﬁcant loss in their catalytic activity. Several strategies have been employed to overcome this limitation: screening and isolation of resistant enzymes from extremophiles,31,32 immobilization on macro- and micro-/nanosized supports,33−35 use of protecting additives,36−40 preparation of genetically transformed enzyme variants,41,42 and in vitro modiﬁcation of the enzyme protein surface with low and high molecular weight compounds.43−45 Among these strategies, enzyme modiﬁcation is particularly attractive because it is simple, inexpensive, and allows large-scale preparation of enzyme biocatalysts with noticeable functional stability and able to work in homoge-

1.2. Historical Background

Since the initial stages, the synthesis of artiﬁcial glycoenzymes was enriched with the advances in glycoconjugate chemistry. Several chemical approaches developed to activate polysaccharide-based supports for protein immobilization were also employed for artiﬁcial glycosylation. To our knowledge, neoglycoenzyme science started in 1960 when Maekawa and Liener prepared S-methylglucosylisothiourea, which was further employed to modify bovine pancreatic trypsin (EC 3.4.21.4).4 The glycosylated enzyme, which contained about 6 mol of sugar per mol of protein molecule, was crystallized and characterized according to its catalytic activity, physical, and 4869

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 1. Reactive amino acid residues in the enzyme protein structure.

stability properties. In 1961, Mitz and Summaria modiﬁed αchymotrypsin (EC 3.4.21.1) with a water-soluble carboxymethylcellulose (CMC) azide derivative. In this pioneer work, it was evidenced the direct inﬂuence of the artiﬁcial glycosylation on the improved functional properties of the conjugated enzyme. In the early 1970s, an indirect method for preparing watersoluble enzyme-carbohydrate derivatives was described by Axén et al.,57 who attached α-chymotrypsin to a CNBr-activated Sephadex matrix and then released a soluble neoglycoconjugate by hydrolysis of the cross-linked dextran support with dextranase (EC 3.2.1.11). The resulting soluble conjugate showed catalytic properties similar to those of the native counterpart. Another contemporary attempt used to synthesize artiﬁcial glycoenzymes was based on the preparation of glycopeptides from naturally occurring glycoproteins and their further coupling to the target enzyme.58,59 This method was successful to improve the pharmacokinetics properties of Lasparaginase (EC 3.5.1.1) by modiﬁcation with glutaraldehydeactivated glycopeptides from human ﬁbrin and γ-globulin,59 and to modulate the hepatic uptake of lysozyme (EC 3.2.1.17) by conjugation with the asialoglycopeptide of fetuin using toluene 2,4-diisocyanate as a cross-linking agent.58 In this decade, O’Neill et al. reported the synthesis of a αchymotrypsin-dextran conjugate using 2-amino-4,6-dichloro-striazine as a coupling agent.60 Such neoglycoenzyme retained 81% and 73% of the initial hydrolytic activity toward N-acetyl-Ltyrosine ethyl ester and casein, respectively, and showed enhanced stability against thermal inactivation and autolytic digestion. Wykes et al. evaluated the same coupling approach to modify α-amylase (EC 3.2.1.1) with dextran, diethylaminoethyl (DEAE)-dextran, and CMC.61 The α-amylase derivatives showed up to 67% of the speciﬁc activity of the free enzyme, and their pH optimum was shifted depending on the charge of the polysaccharide. These neoglycoenzymes were resistant to incubation at high temperatures, and the practical application of the CMC-based conjugate was demonstrated in the continuous hydrolysis of starch in an ultraﬁlter reactor. During this time, a capital contribution to neoglycoenzymes development was performed by Marshall et al., which reported the preparation of a variety of enzyme-dextran conjugates using CNBr as the activation method.62−66 They demonstrated the

eﬀect of this chemical glycosylation on the enhanced functional stability in vitro and increased circulatory lifetime in vivo of the neoglycoenzymes. A similar conjugation approach was used by other authors to modify subtilisin (EC 3.4.21.62), lysozyme, αchymotrypsin, β-glucosidase (EC 3.2.1.21), carboxypeptidase G (EC 3.4.17.11), and arginase (EC 3.5.3.1) with dextran derivatives.67−71 Meanwhile, the ﬁrst patent application on neoglycoenzymes was published, claiming the preparation of a water-soluble noncolloidal conjugate of penicillin G acylase (EC 3.5.1.11) and dextran, which eﬃciently hydrolyzed penicillin and allowed the recovery of 6-aminopenicillanic acid with high eﬃciency and purity.72 In 1978, Marshall published the ﬁrst review on neoglycoenzymes.73 Further decades were characterized by an increased number of research papers and patent applications on artiﬁcial glycoenzymes. Nowadays, neoglycoenzyme science remains as an active research area in protein and carbohydrate chemistry, with signiﬁcant impacts in other ﬁelds like biochemistry, biotechnology, analytical and preparative chemistry, food technology, and pharmaceutical sciences.

2. STRATEGIES FOR PREPARING ARTIFICIAL GLYCOENZYMES The glycosylation of proteins into the cells is a complex process involving a wide variety of enzymes. This enzymatic machinery is speciﬁcally controlled by biochemical factors that diﬀer greatly among cell types and species, and ultimately determine the composition, extensiveness, and pattern of the glycosylation chains in the mature glycoenzyme.46 This natural enzymemediated glycosylation manufacture is very diﬃcult to reproduce by in vitro experiments out of the cells. Moreover, the formation of native-like glycosidic bonds between carbohydrates and proteins using chemical methods is by far a more complex task. In general, conventional carbohydrate synthetic approaches involve strictly anhydrous conditions and a series of protection/deprotection steps, which are incompatible with the labile protein structure, dramatically aﬀecting the biological function.16 For these reasons, almost all of the reported neoglycoenzymes have been prepared using alternative coupling strategies, then bearing unnatural linkages between glycan and polypeptide parts. 4870

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

polypeptide structure are most widely employed as target modiﬁcation points for enzyme glycosylation. In particular, Lys is the amino acid most commonly selected for enzyme glycosylation due to the relatively large occurrence in proteins (about 7% of total amino acids), its usual location at the surface of the three-dimensional protein structure as a consequence of its high hydrophilic character, and to that only a few of its residues can be involved in the active site of enzymes. After Lys, carboxylate groups in Asp and Glu residues have been also commonly used for enzyme glycosylation through the linkage of amino-functionalized carbohydrate derivatives via carbodiimide-catalyzed reactions.81−85 Despite these general considerations, it should be mentioned that recombinant techniques have provided the rational use of Cys residues for enzyme glycosylation, allowing its introduction and modiﬁcation at positions located at the protein surface far from the active site of the selected enzyme.86−88 Glutamine residues have been reported as a target point for enzyme glycosylation by using transglutaminase (EC 2.3.2.13) as biocatalyst.77−79 However, this approach is limited by the fact that the glutamine to be modiﬁed should be recognized as substrate by transglutaminase, which is conditioned by the sequence and the three-dimensional arrangement of the target Gln residues in the enzyme protein structure.89 Other residues often located at the enzyme surface, such as His and Arg, as well as other that are generally buried into the threedimensional polypeptide structure, such as Tyr and Trp, have been more rarely selected for enzyme glycosylation. For artiﬁcial glycosylation, special attention should be paid to the structure and composition of the target carbohydrate. Carbohydrates can be deﬁned as organic substances composed of n (n ≥ 1) units of polyhydroxylic aldehyde or polyhydroxylic ketone structures, or a combination of them, linked by glycosidic bonds, which can be functionalized or not with other chemical groups. Carbohydrates are classiﬁed according to the number of polyhydroxylic aldehyde or polyhydroxylic ketone units in their structures in monosaccharides, oligosaccharides (carbohydrates of deﬁned molecular weight with more than two monosaccharide units), and polysaccharides (carbohydrates with average molecular weight and large monosaccharide content).90,91 Every possible combination of positions for the glycosidic linkages, from 1,2- to 1,6-, has been found in nature. Large oligosaccharides and polysaccharides can be linear or branched chain structures. Oligosaccharides can also form cyclic structures such as cyclodextrins (CDs), and polysaccharides can be chemically cross-linked yielding more complex structures. On the other hand, carbohydrates can be neutral, basic, or acidic compounds. In fact, naturally occurring carbohydrates can contain primary and N-acetylated amines, free and esteriﬁed carboxylates, sulfate and phosphate groups, in addition to the hydroxyl and ketone/aldehyde functionalities. These groups can be chemically transformed, yielding derivatives with novel physicochemical properties. A broad spectrum of carbohydrates diﬀering in size, shape, and composition is then available for artiﬁcial conjugation to enzymes. A wide variety of activation and conjugation reactions have been designed to target many of these potential carbohydrate chemical functionalities, and the most relevant examples are detailed in section 2.3.

In general, the successful preparation of an artiﬁcial glycoenzyme with desired biological properties should be carefully planned, and depends on choosing the appropriate conditions based on:74 (i) the type, size, and structure of the enzyme, (ii) the structure, size, and solution properties of the modifying carbohydrate, (iii) the assembly strategy to be employed, (iv) the chemical/enzymatic reaction involved and the synthetic and puriﬁcation conditions needed, and (v) the expected extension/degree of the glycosylation. 2.1. Structural Considerations

The enzyme protein structure is composed of 20 diﬀerent types of L-amino acids linked through peptide bonds, which have a deﬁned content and distribution pattern in each polypeptide chain. Such a chemical arrangement deﬁnes the folded conformation of the protein as well as the composition and 3D structure of the active site, which often makes up only 10− 20% of the total volume of the enzyme.19,75,76 As a general rule, any artiﬁcial glycosylation strategy should be directed to the amino acid residues located on the protein surface, avoiding the modiﬁcation of the amino acids at the active site of the enzyme. Several amino acid residues have reactive functional groups in their side chains, which can be selected as target modiﬁcation points for enzymes. In fact, Cys, Lys, Arg, His, Asp, Glu, Trp, Met, and Tyr can react under mild conditions with speciﬁc reagents to yield chemically modiﬁed enzyme derivatives. The terminal amino and carboxylic groups can be also considered as target sites for conjugation. In addition, Gln can be enzymatically transformed by compounds having primary amino groups through a transglutaminase-catalyzed reaction.77−79 The structure of these reactive amino acid residues is shown in Figure 1. The most signiﬁcant amino acids for modiﬁcation and conjugation are those containing ionizable side chains (Cys, Asp, Glu, Lys, Ag, Tyr, and His), which can act as nucleophiles in addition reactions when they are in the unprotonated state. It is important to notice that the pKa values shown in Figure 1 are for the side chain groups in aqueous solution, but these values can be greatly aﬀected by the local environment of the amino acid residue into the protein structure.75 The order of nucleophilicity for the unprotonated form of the major functional groups in proteins is:80 R−S− > R−NH 2 > R−COO− = R−O−

The strongest nucleophile in protein molecules is the sulfhydryl group of cysteine, particularly in the ionized, thiolated form. Next are the amine groups in their uncharged forms, including the α-amines at the N terminals, the ε-amines of lysine side chains, the secondary amines of histidine imidazolyl groups and tryptophan indole ring, and the guanidino amines of arginine residues. Finally, the less nucleophilic groups are the oxygen-containing ionizable groups, including the α-carboxylates at the C terminus, the β-carboxyl of aspartic acid, the γ-carboxyl of glutamic acid, and the phenolate of tyrosine residues. According to this, it would be expected that Cys was selected as the most appropriate target modiﬁcation point for artiﬁcial enzyme glycosylation. However, Cys residues are scarcely present in proteins (about 2.8% of the total amino acids), and generally play important roles in enzyme catalysis, substrate binding, or the maintenance of the three-dimensional protein structure via disulﬁde bridges formation.76 Therefore, ε-amino groups in Lys and N-terminal amino acid residues in the 4871

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

2.2. Site-Speciﬁc versus Group-Speciﬁc Coupling Strategies

residues in the presence of a plethora of other functional groups within a protein.96 Site-speciﬁc glycosylation is also valuable to understand the precise role of sugar chains in naturally occurring glycoenzymes, as well as to modulate the biological activity of their artiﬁcial counterparts. Side chains of amino acid residues showing relative low abundance in proteins and rarely located on their surface (cysteine, tyrosine, and tryptophan) as well as the N- and Cterminal groups are basically selected for site-selective enzyme glycosylation. The reactivity of these groups can be kinetically controlled by manipulation of the reaction conditions. A common approach is to select a speciﬁc value of pH for the reaction medium to ensure the selective deprotonation of the target residue to increase its nucleophilic reactivity. As the ionization equilibrium of the amino acid side chains is largely modulated by its local environment into the protein structure,75 it is relatively easy to ﬁnd similar amino acid residues with noticeable diﬀerence in their pKa values into the same enzyme, allowing the speciﬁc manipulation of their nucleophilic reactivity by pH control. Transglutaminase-catalyzed enzymatic reactions also oﬀer possibility for the site-selective glycosylation of enzymes. Transglutaminases are a family of transaminases which produce either intra- or intermolecular isopeptide bonds by using the γcarboxamide group of endoprotein or endopeptide glutamine residues as acyl donor substrate and the ε-amino groups of endoprotein lysine residues as acyl acceptor.97 Moreover, reactive lysines may be substituted by compounds containing primary amino groups, giving rise to a variety of protein-(γglutamyl) derivatives.26,77−79,98,99 Although glutamine is relatively abundant in proteins, not all of these residues are recognized as substrates for transglutaminase. This speciﬁcity, which is governed by the primary sequence and/or tertiary structure around the reactive glutamine residue,89 can be explored for the site-selective glycosylation of enzymes. An elegant approach for the site-selective glycosylation of enzymes is based on the introduction of target amino acid residues at selected position into the protein structure, which can be used for subsequent modiﬁcation.86−88 This design has been improved by introducing novel functionalities into proteins through the genetic incorporation of non-natural amino acids. Examples of these methodologies will be provided in section 2.2, and a more detailed description can be found in a recent review.96

The synthetic strategies used for the preparation of artiﬁcial glycoenzymes can be classiﬁed into two categories: convergent and sequential approaches.16 In the convergent methods, the carbohydrate moiety is obtained from natural sources and used in this form or after chemical modiﬁcation. The carbohydrate derivative is further attached to the enzyme surface as a single block through chemical or enzymatic reactions. In the sequential methods, a single mono- or disaccharide unit is ﬁrst linked to the enzyme surface by enzymatic or chemical approaches, and then the oligosaccharide structure is built upon this initiating sugar using a sequential series of glycosyltransferases capable of selectively attaching various monosaccharides to the growing glycan structure. Both convergent and sequential strategies have been employed to synthesize artiﬁcial glycoenzymes with site-speciﬁc or group-speciﬁc (either with single or multipoint linkages) attachment of carbohydrate to the protein surface. However, convergent methods are by far the most used strategy due to relative low cost, simplicity, and the possibility to prepare a high amount of neoglycoconjugates using well-established synthetic methodologies. The rationale to select a site-speciﬁc or a group-speciﬁc strategy for the in vitro glycosylation of an enzyme is mainly dependent on the ﬁnal use of the resulting neoglycoconjugate, which determines the properties to be conferred or improved to the target enzyme. For industrial purposes, the cost-eﬀective preparation of highly stable and soluble enzyme derivatives able to catalyze reactions in homogeneous systems is generally required. Stability is also desired for neoglycoenzymes that will be immobilized on solid supports by polyelectrostatic interactions or lectin-mediated biorecognition mechanisms, as well as for neoglycoconjugates that should work in hard physicochemical conditions such as high temperature, extreme values of pH, and the presence of surfactant or chaotropic substances in the reaction media. Moreover, stable enzyme forms retaining high catalytic activity and substrate aﬃnity are needed for biosensor construction. The most straightforward approaches to synthesize stable neoglycoproteins are based on group-speciﬁc methodologies involving the multipoint attachment of carbohydrates to the enzyme surface.74,92−95 In fact, the conformational motions in polypeptide enzyme chains leading to unfolding can be constrained by “freezing” the protein molecule through multipoint cross-linking with low and high molecular weight compounds, then conferring stability to the biocatalyst.74 For this reason, cross-linking of enzymes with carbohydrates, mainly using lysine and aspartic and glutamic acid residues as target modiﬁcation points, has been used to prepare the vast majority of neoglycoenzymes reported in the literature. Sugar moieties in artiﬁcial and naturally occurring glycoenzymes can also increase the conformational stability and integrity of these proteins by diﬀerent mechanisms. Carbohydrates are highly hydrophilic, and therefore they can stabilize the hydration layer around the enzymes and promote the creation of new hydrogen bonds at their surfaces. Additionally, attached carbohydrates can avoid proteolytic attacks by masking the potential cleavage sites into the enzyme structure. A well-deﬁned structural composition is a mandatory requirement for neoglycoenzymes intended to be used as drugs for enzyme replacement therapies and other biomedical applications. This implies the preferential use of site-speciﬁc methods for in vitro glycosylation, in which carbohydrates should be attached to a single or some speciﬁcally located

2.3. Synthetic Methods for in Vitro Enzyme Glycosylation

As a general rule, any synthetic procedure intended to prepare artiﬁcial glycoenzymes should be carried out under mild conditions to preserve the biologically active conformation of the biocatalyst. Accordingly, the synthetic methods should preferably proceed at low temperature, in buﬀered aqueous solutions with pH values close to or in the range of maximal enzyme stability, under atmospheric pressure, and avoiding the presence of inactivating compounds and irreversible inhibitors. In addition, a proper puriﬁcation scheme should be planned in advance taking into account the characteristics of the initial enzyme and carbohydrate, as well as the expected conjugate. The strategies commonly employed to prepare neoglycoenzymes are based on chemical, enzymatic, and chemoenzymatic methods. Furthermore, a combined use of genetic engineering and chemical/enzymatic transformations has led to the rational design of site-speciﬁcally glycosylated enzymes. 4872

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 2. Carbodiimide-mediated modiﬁcation of enzymes with carbohydrate-containing carboxylic acid groups.

Figure 3. Modiﬁcation of enzymes with N-hydroxysuccinimide-activated carbohydrates.

Figure 4. Modiﬁcation of enzymes with sugar lactones.

2.3.1. Chemical Methods. Artiﬁcial glycosylation based on conventional synthetic methods is by far the strategy most commonly employed. These procedures have been used for the convergent single- and multipoint attachment of carbohydrates to enzyme surfaces, and here they will be classiﬁed according to the enzyme functional group that is modiﬁed. It should be noticed that the glycan structures drawn in the ﬁgures of this paper are only for general guidance, and not always match with the carbohydrates employed in the cited works. 2.3.1.1. Conjugation of Carbohydrates to Enzyme Amino Groups. The relative high abundance in enzyme molecules, the high accessibility at the protein surface, and the nucleophilic character of unprotonated primary amino groups have promoted the development of a multitude of methods for this kind of chemical glycosylation. These approaches are mainly based on acylation, reductive alkylation, imination, amidination, N-alkylation, and N-arylation reactions. Enzyme amino groups can be acylated with carbohydratecontaining carboxylic acid groups giving amide derivatives by using a water-soluble carbodiimide as coupling agent via the

formation of an intermediary O-acylisourea active ester derivative (Figure 2).100−122 Enzyme−carbohydrate conjugates can be prepared by one-pot synthetic procedures in slightly acid aqueous solutions (pH 4.8−6.0), through coupling reactions that usually involve N-ethyl-N′-(3-dimethylaminopropyl) carbodiimide (EDC) as coupling agent and long reaction times.100−103,106−109,111−122 Two-step procedures have been also reported to acylate enzyme amino groups with sugars containing carboxylic acid groups. In such strategies, the carboxylic groups in saccharides are ﬁrst activated with EDC in acidic or neutral solutions, and further reacted with the amino groups on the enzyme surface.104−106,110 Glycosylation of enzymes can be easily performed by acylation with carbohydrates containing spacer arms ended with carboxylic groups previously activated as N-hydroxysuccinimide esters (Figure 3). Coupling of the activated carbohydrate derivative to enzyme amino groups is usually performed at alkaline pH values.123−125 Acylation of enzyme can be also carried out by coupling with carbohydrates containing sugar lactones as it is illustrated in 4873

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 5. Glycosylation of enzymes with sugar acyl azides.

Figure 6. Glycosylation of enzymes through the Ugi reaction.

Figure 7. Modiﬁcation of enzymes with reducing sugars by reductive alkylation.

Figure 4.126 Although the nucleophilic attack of primary amines to the carbonyl group in the cyclic ester is catalyzed by hydroxyl ions, this glycosylation approach should be performed under strict control of pH to avoid the spontaneous hydrolysis of lactone. Successful single point glycosylation was reported by using a high molar ratio of sugar lactone to enzyme amino groups, short reaction times, and controlling the pH in the range of 7.5−8.0. A reliable method to acylate primary amino groups of enzyme with carbohydrates is based on the treatment of the protein with a saccharide acyl azide derivative.56 As it is shown in Figure 5, carbohydrate-containing carboxylic acid groups can be, in a ﬁrst step, transformed to acyl hydrazide derivatives and further oxidized by sodium nitrate to yield activated acyl azides. Primary amino groups in enzymes can be then cross-linked by reaction in alkaline media.

The Ugi multicomponent reaction has been adapted for the acylation of enzyme amino groups with carboxylated saccharides in aqueous solution (Figure 6).127 For this purpose, the enzyme is treated with formaldehyde to form the corresponding methylol derivatives with primary amines, which rapidly undergo condensation to yield aldimine groups. Further treatment with the carbohydrate-containing carboxylate groups and t-butyl isocyanide yields the neoglycoconjugate through the formation of multipoint stable isopeptide structures. However, the presence of carboxylic and thiol groups in the enzyme structure could also lead to the formation of protein−protein cross-linked adducts. Neoglycoenzymes have been extensively prepared through reductive alkylation reactions. The simplest strategy is based on the treatment with naturally occurring mono- or oligosaccharides containing a free reducing end.128 This reducing end exists 4874

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 8. Reductive alkylation of enzymes with carbohydrates functionalized with aldehyde groups via enzymatic (A) or chemical (B) activation.

Figure 9. Reductive alkylation of enzymes with periodate-oxidized carbohydrates.

as an equilibrium mixture composed of the cyclic hemiacetal form (lactol) and the open-chain aldehyde form, which under suitable conditions can condense with amine groups to form an imine derivative. This Schiﬀ base can be further transformed to a secondary amine group by in situ reduction with NaBH3CN (Figure 7).129−133 The process results in ring-opening of the reducing end monosaccharide. The glycosylation degree can be easily controlled by manipulation of the reaction time with the best results achieved by 1−3 days of incubation. It was also demonstrated that such reductive alkylation is strongly temperature- and pH-dependent, yielding successful glycosylation degree at 25−37 °C in solution of pH 7.0−9.0.130

An alternative electrochemical strategy was tentatively applied for the reductive N-alkylation of enzymes with reducing sugars.134 This approach is based on the electroassisted reduction of a mixture of the enzyme and carbohydrate in alkaline solution using high cathodic potentials. Although some level of glycosylation was obtained, this method was much less eﬃcient than those using a hydride donor and showed important limitations. In fact, electroassisted reductive Nalkylation is suitable only for basic proteins, and several functional groups in proteins are potentially reducible at strongly negative electrode potential. Glycans containing galactose moieties can be also enzymatically transformed to C-6 aldehydogalactosyl derivatives by 4875

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 10. Glycosylation of enzymes through the Maillard reaction.

be solved by using sequential reduction steps with NaBH3CN and NaBH4.150−152 Borane−pyridine complex has been also employed as reducing agent for preparing neoglycoenzymes via reductive alkylation with periodate-oxidized carbohydrates.160 It should be mentioned that several neoglycoconjugates have been also prepared by cross-linking periodate-activated polysaccharides with enzymes without using reducing agents.168−176 Multipoint aldimine-based linkages seem to be enough for stabilizing the enzyme−polysaccharide complexes, causing low eﬀect on the catalytic activity of the resulting neoglycoenzyme. Maillard reaction has been employed to modify enzymes with polysaccharides by linking the reducing end group of the carbohydrate to the primary amino groups located at the protein surface. This glycosylation is performed by incubation of a mixture of powdered enzyme and polysaccharide at 60 °C under a relative humidity of 65−80% during various days or weeks.177−183 A wet modiﬁcation of this method has been also reported for the glycosylation of enzymes with maltodextrin at 95 °C in buﬀered solutions of pH 6.5.184 Maillard reaction is a complex process involving many transformations and pathways, leading to advanced glycation end products with undeﬁned structure (Maillard browning reaction), which makes it diﬃcult to predict the structure of the artiﬁcial glycoenzymes prepared in this way. It can be assumed that this transformation started with the formation of a Schiﬀ base between the reducing end sugar in the polysaccharide and the primary amino groups of the enzyme with further rearrangement to the corresponding N-substituted 1-amino-1-deoxy-ketose derivative called the Amadori compound (Figure 10). A solvent-free alternative approach to modify enzymes with reducing sugars in vacuo has been patented.185 Enzymes were easily and eﬃciently glycated by lyophilizing the protein with a

galactose oxidase (EC 1.1.3.9), which can be covalently attached to primary amino groups in enzymes by reductive alkylation (Figure 8A).135 Single aldehyde functionality can be also introduced into the carbohydrate structure by chemical oxidation of the corresponding O-tosyl derivative to allow further glycosylation of enzymes through reductive alkylation, as it is illustrated in Figure 8B.35,113,136 By far, the most employed strategy to prepare neoglycoenzymes is the covalent cross-linking with periodateoxidized oligo- and polysaccharide via reductive alkylation (Figure 9).74,92−95,107,108,116,137−167 Meta periodate ion cleaves C−C bonds that possess adjacent hydroxyl groups by oxidation to highly reactive aldehydes. This approach permits one to control the oxidation level of the carbohydrate by manipulating the m-NaIO4/monosaccharide ratio.146 In general, functionalization of carbohydrates by periodate oxidation is performed in water. It is also recommended to keep the reaction mixture in the dark to avoid photooxidation of the prepared polyaldehydes derivatives.137,139,150,151,153 The aldehydes are able to form azomethine bonds with primary amino groups on the enzyme surface, which can be further transformed to secondary amines by treatment with reducing agents of the hydride donor type at neutral or alkaline pH. Sodium borohydride is commonly employed for these reductive alkylation processes,93,94,107,108,116,139,140,142,148,149 although it reduces both the imine and the aldehyde groups, thereby limiting the number of attachments between the enzyme and the activated carbohydrate. Sodium cyanoborohydride, a mild reducing agent that does not reduce aldehyde groups, has been also used.141,143−145,147,154,160 However, the toxicity associated with the cyanide as well as the inhibitory eﬀect that reactive aldehyde groups cause on the catalytic activity of several enzymes are the main drawbacks of NaBH3CN. The latter can 4876

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 11. Aminidation of enzymes with 2-iminomethoxymethyl thioglycosides.

Figure 12. Glycosylation of enzymes with CNBr-activated carbohydrates.

rely on overcoming problems such as protein inactivation due to solvent-induced denaturation, production of advanced glycation end products by the Maillard reactions, thermal induced unfolding, and enzymatic degradation. Amidinated neoglycoenzymes are easily prepared by attaching 2-iminomethoxymethyl thioglycosides to the enzyme primary amino groups.187−195 2-Iminomethoxymethyl thioglycosides can be synthesized by treating the corresponding

reducing sugar and incubating at elevated temperature (60−85 °C) in vacuo for a few hours. Under these conditions, the sugar is covalently attached to the amino groups on the protein by a stable ketoamine linkage (Amadori compound). This modiﬁcation strategy was tested by using monosaccharides as well as linear bifunctional and multifunctional branched glycolderivatives containing reducing sugars as glycosylation agents.185,186 The advantages of the solvent-free glycosylation 4877

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 13. Modiﬁcation of enzymes with 1,1′-carbonyldiimidazole-activated glycans.

Figure 14. Modiﬁcation of enzymes with carbohydrates activated with p-isothiocyanatophenyl groups via N-arylation (A) or reductive alkylation (B).

cyanomethylthioglycoside with methoxide as it is illustrated in Figure 11. Direct coupling to the amino groups on the enzyme surface is generally accomplished in alkaline media. Amidine linkages have been also employed to modify enzymes with paminophenyl-β-D-galactopyranoside moieties using dimethyl suberimidate as cross-linking agent.196 Enzymes can be modiﬁed with CNBr-activated water-soluble carbohydrates through chemistry similar to that involved on immobilization on CNBr-activated polysaccharide supports.62−71,197−201 It is strongly recommended to perform a previous activation step, followed by puriﬁcation of the resulting reactive carbohydrate before coupling to the enzyme molecule to avoid CNBr-mediated protein cleavage.

Activation of saccharides with CNBr is achieved in alkaline pH and results in the formation of cyclic imidocarbonate and carbamic acid ester groups (Figure 12). Carbamic acid ester groups are stable, inert, and neutral, and then imidocarbonates are responsible for most of the coupling ability of the activated carbohydrate. The reaction of enzyme amino groups with CNBr-activated carbohydrates produces mainly N-substituted imidocarbonates, substituted isoureas, and N-substituted carbamic acid esters.202 This artiﬁcial glycosylation method was popular in the 1970s, but now is rarely employed due to the toxicity of CNBr, the precipitation of highly activated carbohydrates, and the need to use extreme pH values during the coupling reaction, which might aﬀect the enzyme structure and activity.138 4878

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 15. Benzoquinone-mediated cross-linking of enzymes with carbohydrates.

Figure 16. Glycosylation of enzymes with squarate-activated carbohydrates.

carried out by mixing at 4 °C during 2 days in alkaline buﬀered solution. Although enzyme N-nucleophile groups can participate in this substitution reaction, selective glycosylation of primary amino groups can be achieved by controlling the working pH. The eﬀectiveness of this glycosylation approach is similar to those performed with CNBr-activated carbohydrates. However, 1,1′-carbonyldiimidazole is safe to handle and provides reproducible levels of activation, which can be easily controlled.

1,1′-Carbonyldiimidazole, a carbonylating and zero-length cross-linker agent, is also suitable for the activation of carbohydrates to prepare artiﬁcial glycoenzymes (Figure 13).203 Activation of the glycan is performed in DMSO, often in dilute carbohydrate solutions to avoid the formation of insoluble gels by minimizing the unwanted cross-linking phenomena. The amount of 1,1′-carbonyldiimidazole should be also controlled to reduce the hydrophobicity of the resulting imidazolyl carbamate derivative and ensure water solubility of the activated carbohydrate. Coupling to enzymes is further 4879

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 17. Modiﬁcation of enzymes with glycopeptides.

Carbohydrates can be functionalized with p-isothiocyanatophenyl groups by reaction of the corresponding p-aminophenyl derivative with thiophosgene,204 and further coupling to primary amines. Glycans modiﬁed with p-isothiocyanatophenyl groups, either by N-arylation or by reductive alkylation, have been successfully employed to modiﬁed enzymes (Figure 14).205−208 This isothiocyanate-based glycosylation process commonly performed under alkaline conditions speciﬁcally targets the enzyme primary amino groups. Enzymes have been modiﬁed with carbohydrates via an Narylation reaction by using p-benzoquinone as cross-linking agent, yielding 2,5-substituted hydroquinone derivatives through successive addition−oxidation reactions (Figure 15).209−213 Carbohydrate should be ﬁrst activated with pbenzoquinone in slightly acid solutions and darkness, and further properly puriﬁed. During the activation process, nucleophilic attack of a carbohydrate hydroxyl group on a pbenzoquinone molecule will result in a 2-substituted hydroquinone molecule linked to the saccharide backbone. Hydrogen is then eliminated by reaction with a second molecule of pbenzoquinone to give a 2-substituted quinone derivative.214 The activated carbohydrate is then coupled to the enzyme by the nucleophilic attack of a protein primary amino group to the 5-position of the substituted quinone in alkaline buﬀered solution, yielding the 2,5-substituted hydroquinone derivatives. In addition to primary amino groups, other nucleophilic groups can be involved in the coupling process, such as the thiol and the phenolic groups in cysteine and tyrosine, respectively.

Squarate-based chemistry has been employed for the artiﬁcial glycosylation of enzymes using the primary amino groups on the protein surface as target modiﬁcation points (Figure 16).215 Carbohydrates containing primary amino groups are converted to the corresponding squarate derivatives in H2O/DMF media. These derivatives are stable in water and have a unique reactivity for substitution that is driven by the aqueous solvent. This stability and reactivity of amino squarates in water allow their use to modify enzyme proteins through a stable and neutral covalent bond. Coupling of squarate-activated carbohydrates to enzymes often requires long reaction times, up to 3 days, in alkaline media and room temperature. Artiﬁcial glycoenzymes have been also prepared by attaching glycopeptides, which are obtained by enzymatic hydrolysis of naturally occurring glycoproteins, to the amino groups at the surface of the target enzyme (Figure 17). Toluene 2,4diisocyanate and glutaraldehyde were employed as cross-linking agents for these coupling reactions.59,216 2.3.1.2. Conjugation of Carbohydrates to Enzyme Thiol Groups. Several chemical methods have been developed to attach carbohydrate residues to the thiol group at the cysteine residues in enzymes. In fact, the low natural abundance in proteins as well as the strongly nucleophilic character of the sulfhydryl group at the cysteine side chain confer unique relevance to this amino acid for selective and single-point modiﬁcation with carbohydrates.217 Enzymes were artiﬁcially glycosylated by S-alkylation of cysteine residues with carbohydrate-tethered iodoacetamides, 4880

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 18. Modiﬁcation of enzymes with carbohydrate-tethered iodoacetamides.

Figure 19. Glycosylation of enzymes with glycosyl methanethiosulfonates.

Figure 20. Glycosylation of enzymes with glycosyl methanedithiosulfonates.

Attractive approaches based on disulﬁde bond formation were also proposed to synthesize artiﬁcial glycoenzymes. In particular, the glycosylation methods proposed by the Davis group deserve special attention. These strategies are based on the use of glycosyl methanethiosulfonate,86−88,222−227 glycosyl phenylthiosulfonate,228 as well as selenenylsulﬁde derivatives,229 which oﬀer the most selective of the thiol modiﬁcation reactions. Diverse glycosyl methanethiosulfonates can be synthesized with good yield from unprotected or peracetylated glycosyl halides by treatment with sodium methanethiosulfonate at proper temperature (50−90 °C) in polar organic solvent (EtOH, DMF) as it is depicted in Figure 19. Glycosylation of

which were synthesized from protected azido-glycans previously reduced with PtO2/H2,218 and further coupled to αiodoacetic acid using 2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline (EEDQ) as a condensing agent (Figure 18).219,220 Deprotection with sodium methoxide yields the iodoacetamidemodiﬁed glycan, which is ﬁnally coupled to the enzyme thiol groups by incubation during 3 days in the dark in cold buﬀer solution of pH 7.5. Modest yields of glycosylated enzyme are produced under these coupling conditions. Higher degree of glycosylation was achieved by coupling the thiol-activated sugar to the enzyme at pH 9.0, but N-alkylation of lysine residues was observed in addition to the S-alkylation of cysteines. 4881

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 21. Glycosylation of enzymes with glycosyl phenylthiosulfonates.

Figure 22. Synthesis of disulﬁde-linked neoglycoenzymes via selenenylsulﬁde derivatives.

appropriate glycosyl halides using sodium phenylthiosulfonate in acetonitrile at 70 °C in the presence of tetrabutylammonium halide (Figure 21). In general, glycosyl phenylthiosulfonates are prepared with higher yield than the corresponding methylthiosulfonates and showed better β-stereoselectivity. Glycosyl phenylthiosulfonates also exhibited enhanced stability and eﬃciency under coupling reaction conditions, allowing their use at lower glycan/enzyme ratios than methylthiosulfonates. Attachment to enzyme cysteine residue, which is carried out at pH 9.5, produces neoglycoconjugate in quantitative yields with complete in situ deacetylation of the glycan. Two parallel selenenylsulﬁde-mediated conjugation strategies, where the protein cysteine residue plays either the electrophilic or the nucleophilic role, were proposed for the preparation of artiﬁcial glycoenzymes (Figure 22).229 The ﬁrst method involves the transformation of the enzyme thiol group into the corresponding (phenylselenenyl)sulﬁde by treatment with phenylselenenyl bromide at pH 9.5. The resulting S−Se bonds confer electrophilic character to the sulfur atom, which is then capable of nucleophilic substitution by 1-thio glycans. The second method is based on the conversion of the modifying 1-thio glycans into their corresponding selenenyl-

the thiol group in the model enzyme, which only contained one cysteine residue, was almost quantitative after 1 h at pH 9.5 and room temperature. Under these conditions, enzyme conjugation was accompanied by complete in situ deacetylation of the modiﬁed carbohydrate when protected glycosyl methanethiosulfonates derivatives were used. This conjugation approach was successful for the single-point glycosylation of enzymes with monosaccharides, linear and branched oligosaccharides, and glycodendrimers.86−88,222−227 An interesting modiﬁcation of this glycosylation approach implied the preparation of glucosyl methanedithiosulfonate and the further attachment to the enzyme cysteine residue through trisulﬁde linkage (Figure 20). Glucosyl methanedithiosulfonate was synthesized in a workable yield (about 13%) by treatment of acetylated glucosyl bromide with an excess of sodium methanethiosulfonate in dioxane at 70 °C during 70 h.230 Trisulﬁde neoglycoenzymes were easily prepared by conjugation of the glucosyl methanedithiosulfonate derivative to the enzyme cysteine residue at alkaline pH. Glycosyl phenylthiosulfonates have been also proposed as modiﬁcation agents for enzyme thiol groups.228 These derivatives are synthesized by displacement of halide from the 4882

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 23. Synthesis of selenenylsulﬁde-linked neoglycoenzymes.

Figure 24. Synthesis of thioether-linked neoglycoenzymes via conversion of cysteines to dehydroalanines.

Figure 25. Glycosylation of enzymes via cross methathesis reaction.

hydrolysis with sodium metabisulﬁte and Zémplen deacetylation,231 or by direct thionation of protected and unprotected reducing carbohydrates with Lawesson’s reagent.232 Disulﬁde-linked neoglycoenzymes prepared from glycomethanethiosulfonates, glycophenylthiosulfonates, and glycoselenenylsulﬁdes derivatives can be further converted to thioetherlinked neoglycoenzymes.233 This postglycosylation desulfurization process can be achieved by treatment of the modiﬁed enzyme with hexamethylphosphorous triamide in alkaline medium. The resulting S-linked neoglycoconjugates showed enhanced chemical stability with respect to initial disulﬁde-

sulﬁde derivatives also by reaction with phenylselenenyl bromide, which can be subsequently coupled to a nucleophilic thiol group in the enzyme cysteine residues. This coupling reaction is pH-dependent with pH values around 9.5 needed to favor high glycosylation degrees. An important advantage of this selenenylsulﬁde-based glycosylation strategy is the possibility to obviate the need for protecting groups during glycoconjugation due to the high selectivity of S−Se chemistry. In addition, the introduction of S−Se linkages into enzymes allows the coupling of large thionated oligosaccharides, which can be prepared by treatment of the glycosyl halides with thiourea followed by mild 4883

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 26. Glycosylation of enzymes via Suzuki−Miyaura cross-coupling reaction.

Figure 27. Modiﬁcation of enzymes with aminated carbohydrates via carbodiimide-mediated coupling.

protogenic enzyme thiol groups to yield active functionalities able to react with the target glycosylation agents. In this context, alkene groups have been introduced on the enzyme surface by conversion of cysteines to dehydroalanines by oxidative elimination with O-mesitylenesulfonylhydroxylamine at pH 8.0 (Figure 24). The modiﬁed enzyme is further transformed to a thioether derivative by nycleophilic addition of 1-thio glycans in slightly alkaline medium.235 Cross metathesis has been demonstrated to be an alternative and eﬀective way to prepare artiﬁcial glycoenzymes (Figure 25). The reaction of enzymes with O-mesitylenesulfonylhydroxylamine causes the fast conversion of cysteine to dehydroalanine, which can then be coupled with a thiol nucleophile in one-pot reaction. Dehydroalanine is an eﬀective Michael acceptor for thiols, and can be easily converted in S-allylcysteine by reaction with allyl mercaptan. Further glycosylation of the S-allylcysteine containing enzyme can be achieved by cross metathesis with monosaccharide allyl glycosides.236−238 Davis and collaborators evaluated the artiﬁcial glycosylation of enzymes by using the Suzuki−Miyaura cross-coupling

linked neoglycoenzymes, being resistant to reduction with tris(2-carboxyethyl)phosphine. The high reactivity of diselenides in thiol−diselenide exchange has been recently exploited to prepare enzyme− monosaccharide conjugates through selenenylsulﬁde linkages.234 Sugar diselenides are synthesized by reaction of the corresponding protected glycosyl halides with potassium 4methylselenobenzoate and further deprotection/dimerization by treatment with sodium methoxide (Figure 23). The resulting bis(glycosyl)-1,1′-diselenides were almost quantitatively attached to the model enzyme cysteine residue through a thiol−diselenide exchange by simply mixing sugar diselenide derivatives with the protein at pH 8.0 and 4 °C during 0.5−2 h. A relevant advantage of this glycosylation approach is the high stability of the selenenylsulﬁde bond in the SeS-linked neoglycoenzyme, allowing the subsequent modiﬁcation to more complex glycans through endoglycosidase-catalyzed transglycosylation. Several strategies oriented to the speciﬁc glycosylation of cysteine residues are based on the chemical modiﬁcation of the 4884

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 28. Modiﬁcation of enzymes with S-methylglucosylisothiourea.

Figure 29. Glycosylation of enzymes with 2-amino-4,6-dichloro-s-triazine activated carbohydrates.

reaction (Figure 26).239 For this purpose, the model enzyme cysteine residue was functionalized with an aryliodide moiety by treatment with 4-iodobenzyl bromide during 1 h at 37 °C in pH 8.0 buﬀer solution containing DMF. The modiﬁed enzyme was cross-coupled with the corresponding glycosyl vinylboronic acid derivative at 37 °C in slightly alkaline buﬀer solution using a Pd-pyrimidine catalyst. 2.3.1.3. Conjugation of Carbohydrates to Enzyme Carboxylate Groups. Carboxylate groups located at the side chains of aspartic and glutamic acids as well as at the C-terminal residue are protogenic functionalities that can be used to glycosylate enzymes. Conjugation to these acid residues is basically performed by EDC-mediated coupling of aminated carbohydrates (Figure 27).81,240 In general, these glycosylation reactions are carried out by mixing the enzyme with the amino-containing carbohydrate in slightly acid buﬀered solutions (pH 4.0−6.0) in the presence of the water-soluble carbodiimide. Reaction

times ranging from 20 min to 22 h have been reported. Some EDC-based glycosylation approaches include a ﬁrst reaction period of about 1−2 h at room temperature and further longer incubation at 4 °C.82−85,241−252 2.3.1.4. Non Group-Speciﬁc Conjugation of Carbohydrates to Enzymes. Several less group-speciﬁc glycosylation methods leading to the modiﬁcation of two or more chemical functionalities into the enzyme structure have been reported. Although less speciﬁc, these approaches provided artiﬁcial glycosylated enzymes with improved performance. In a pioneer work, Maekawa and Liener synthesized Smethylglucosylisothiourea, which was further employed as reagent to modify protogenic groups on the enzyme surface (Figure 28).4,253 This coupling reaction needs long reaction times, requiring up to 3 days of incubation in alkaline medium. Structural characterization evidenced that the glucosylamidyl moieties in the modiﬁed enzyme were linked either to the εamino groups of lysine residues or to the carbon atoms at C-2 4885

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 30. Modiﬁcation of enzymes with vinyl sulfone carbohydrate derivatives.

modiﬁcation, which seems to occur by reaction of the diazonium salt with the imino group to form a triazen that can undergo rearrangement to form a C-azo compound, can be improved by coupling at pH < 3.0. However, this pHcontrolled selective modiﬁcation is limited by the stability of most of the enzymes at such acidic conditions. Enzymes have been covalently cross-linked with epichlorohydrin-activated polysaccharides through a two-step strategy, as it is illustrated in Figure 32.174 Activation of the polymer is ﬁrst accomplished by stirring with the cross-linking agent during 2 h at 40 °C in NaOH solution. After puriﬁcation, the activated polysaccharide is attached to the nucleophilic amino acid residues located at the enzyme surface in alkaline buﬀered solution. Because of the high reactivity of the epichlorohydrinmodiﬁed carbohydrate, amino, hydroxyl, and thiol groups of enzymes can be involved in this glycosylation process. 2.3.1.5. Conjugation of Carbohydrates to Sugar Chains in Glycoenzymes. Functional properties of glycoenzymes can also beneﬁt from extra-attaching of exogenous carbohydrates to the enzyme, mainly by chemical cross-linking with polysaccharides. However, glycoenzymes are often resistant to covalent modiﬁcation due to masking of the protogenic reactive groups on the enzyme surface by the oligosaccharide chains.257 Therefore, alternative methods for conjugating glycoenzymes to carbohydrates were developed. Sugar moieties have been selected as target modiﬁcation points because carbohydrate chains are often not required for catalytic activity.258 Activation of sugar chains in glycoenzymes by periodate oxidation provided reactive aldehyde groups that can be linked to exogenous carbohydrates bearing primary amino groups through the formation of imine bonds (Figure 33). Reductive alkylation with hydride donors such as NaBH4 and NaBH3CN yields stable neoglycoconjugates.259−261 The main disadvantage of this approach is the possible formation of cross-linked protein−protein adducts due to the presence of reactive amino groups in the enzyme surface. Carbohydrates activated by treatment with hydrazine can be also attached to periodate-oxidized sugar chains in glycoenzymes. Sugars containing reducing end groups can be

and C-4 positions in the imidazole ring of histidine residues, as it is shown in Figure 28. This modiﬁcation method has not been widely employed in further years. Carbohydrates activated by treatment with cyanuric chloride (2,4,6-trichloro-s-triazine) and its derivatives were also employed to modify enzymes. The best results were achieved by using 2-amino-4,6-dichloro-s-triazine as activating agent (Figure 29), although it is not very soluble in water, and mixtures with organic solvents are needed during the activation step.60,61 The activated carbohydrate should be puriﬁed before coupling to the enzyme, which is commonly performed at alkaline pH. In addition to primary amino groups, hydroxy, guanidine, phenol, and thiol enzyme groups can be also modiﬁed with triazine-activated carbohydrates. Vinyl sulfone sugar derivatives have been coupled to the amine and thiol groups present in the proteogenic residues of enzymes through Michael-type addition (Figure 30).254,255 The main advantage of this glycosylation approach relies on the possibility of modifying the enzyme under mild pH conditions compatible with the maintenance of the biological function. However, long reaction times (about three days) at room temperature are usually needed, which can be counterproductive for the glycosylation of proteases and thermolabile enzymes. Although lysine, histidine, and cysteine can act as Michael donors in this coupling reaction, the group speciﬁcity of this glycosylation can be increased by manipulation of the reaction conditions. Another strategy to prepare neoglycoenzymes is the azocoupling of glycosyl diazophenyl derivatives, prepared by diazotization of the corresponding p-aminophenyl glycan, to protogenic nucleophilic groups in enzymes.256 Azo-coupling is a simple and relative inexpensive modiﬁcation approach, which was employed to prepare the ﬁrst neoglycoprotein.55 However, it has not been extensively used due to instability of the diazonium reagents, which forces their in situ preparation just prior to use and makes it diﬃcult to control the stoichiometry of the coupling reaction. Additionally, diazonium derivatives are highly reactive conferring poor selectivity to this glycosylation method as it is illustrated in Figure 31. Selectivity to tryptophan 4886

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 31. Glycosylation of enzymes via azo-coupling reaction.

transformed into their hydrazine derivatives by reaction with N2H4 (Figure 34). The glycosylhydrazine and hydrazone forms are at equilibrium with the glycosylhydrazine being predominant in aqueous solutions.262 Glycosylhydrazines can be further linked to the as-introduced aldehyde groups in glycoenzymes, resulting from the periodate-mediated sugar chains oxidation, forming stable hydrazone bonds. This coupling reaction is generally performed at 37 °C in slightly acid media.263

Carbohydrates functionalized with hydrazide groups were also employed as glycosylation agents for glycoenzymes modiﬁcation (Figure 35A). Zhu et al. reported the preparation of synthetic oligosaccharides anchored with butyl hydrazide groups for coupling to periodate-oxidized enzymes through the formation of hydrazone bonds.264 The main advantages of this glycosylation approach are the high reactivity of hydrazides toward carbonyl groups and their low toxicity in comparison with the hydrazine counterparts. 4887

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 32. Modiﬁcation of enzymes by cross-linking with epichlorohydrin-activated carbohydrates.

Figure 33. Modiﬁcation of sugar chains in glycoenzymes with aminated carbohydrates.

Figure 34. Modiﬁcation of sugar chains in glycoenzymes with hydrazine-activated carbohydrates.

4888

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 35. Modiﬁcation of sugar chains in glycoenzymes with glycosylhydrazides (A) and aminooxy-glycosyl derivatives (B).

glycosylation machine. However, a great variety of glycosidases, glycosynthases, and glycosyltransferases have been employed for the synthesis and remodeling of glycoconjugates.268 Some of these enzymatic approaches have been also evaluated to prepare neoglycoenzymes. These convergent methods take the advantages of the great speciﬁcity of enzyme-catalyzed reactions, the factibility to prepare neoglycoconjugates with deﬁned structure, as well as the possibility to carry out the glycosylation reactions in very mild conditions. A successful and simple strategy to prepare novel artiﬁcial Nglycosylated enzymes is based on the enzymatic attachment of glycans through transglycosylation reaction, as it is illustrated in Figure 37A. The procedure implies ﬁrst a partial deglycosylation of natural occurring glycoenzymes with speciﬁc endoglycosidases to release the undesiderable original sugar chains and provide terminal N-acetylglucosamine moiety as target modiﬁcation points for further expanding the N-glycosylation chains. Speciﬁc endo-β-N-acetylglucosaminidases with transglycosidase activity, either soluble or immobilized on solid supports, were then employed to link the desired oligosaccharide chains to the enzyme.269−272 Endo-β-N-acetylglucosaminidases from Arthrobacter protophormiae (Endo-A, EC 3.2.1.96) is the enzyme most commonly employed for these purposes. Endo-A hydrolyzes the glycosidic bond in the N,N′-diacetylchitobiose moiety of N-linked oligosaccharides of various glycoproteins, but also is capable

An alternative carbonyl-coupling method to generate an oxime bond in the artiﬁcial glycoenzyme is based on the use of aminooxy-derived glycans.265 This approach is achieved by converting glycosylhydrazides to reactive aminooxy-glycosyl derivative by treatment with N-(t-BOC)-aminooxyacetic acid tetraﬂuorophenyl ester, followed by deprotection in triﬂuoroacetic acid. The resulting glycosyl-aminooxy ligand can be coupled with high yield to periodate-oxidized glycoenzymes as it is shown in Figure 35B. Reactive primary amino groups can be placed on the glycosylation chains of periodate-activated glycoenzymes by treatment with diamines and further reductive alkylation of the resulting imine bonds, as shown in Figure 36. These amino groups can be then employed as target modiﬁcation points for the carbodiimide-mediated coupling of anionic carbohydrates containing carboxylate groups.266,267 Stable polysaccharidebased neoglycoenzymes have been prepared through this approach, which are mainly limited by the presence of glutamic and aspartic acid residues on the protein surface, leading to the formation of cross-linked enzyme−enzyme adducts. 2.3.2. Enzymatic Methods. Glycosylation is a co- and post-translational process governed by the combined catalytic action of a plethora of speciﬁc enzymes. Accordingly, use of enzyme-mediated reactions should be the most rational approach to prepare artiﬁcial glycoconjugates. Nevertheless, it is almost impossible to reproduce out of cells the enzymatic 4889

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 36. Modiﬁcation of sugar chains in glycoenzymes with anionic carbohydrates.

Figure 37. Preparation of neoglycoenzymes by enzymatic transglycosylation with endo-β-N-acetylglucosaminidase (A) and polypeptide-R-Nacetylgalactosaminyl transferase (B).

of transferring N-linked oligosaccharides to a suitable acceptor by transglycosylation reactions. It has been reported that the hydrolytic activity of Endo-A can be suppressed and the transglycosylation activity can be enhanced in reaction media containing organic solvents such as acetone, DMSO, and DMF.273 For this reason, reaction mixtures containing 30%

acetone or 20% DMSO (v/v) in buﬀer solutions of pH 6.0 are recommended for the Endo-A-mediated preparation of neoglycoenzymes. Artiﬁcial O-glycosylated enzymes can be prepared in vitro by an enzymatic approach that uses the enzyme polypeptide-R-Nacetylgalactosaminyltransferase (EC 2.4.1.41) (Figure 37B). 4890

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 38. Preparation of neoglycoenzymes via transglutaminase-catalyzed reaction.

This enzyme catalyzes the transfer of the N-acetylgalactosamine sugar from the corresponding uridine diphosphate nucleotide sugar to Thr/Ser residues on an acceptor polypeptide that is at least 11 amino acids long.274 This transferase-mediated preparation of neoglycoenzymes is commonly performed at room temperature in alkaline buﬀer medium. Neoglycoenzymes can be also prepared through transglutaminase-catalyzed enzymatic reactions, as it is shown in Figure 38. This approach allows the covalent linkage of carbohydrates containing primary amino groups to enzyme glutamine residues that can be recognized as glutamyl donor substrates for transglutaminase by the formation of isopeptide bonds.77−79,275 Although this approach is limited by the speciﬁcity of the transaminase enzyme, there is a commercially available variety of transglutaminase, diﬀering in the catalytic and substrate speciﬁcity properties, which can be employed for the in vitro synthesis of neoglycoenzymes. 2.3.3. Chemoenzymatic Methods. Several synthetic methods for the preparation of neoglycoenzymes are based on the chemical coupling of mono- or oligosaccharides to the protein structure and the subsequent enzymatic extension of the glycan chains through transglycosylation reactions.276,277 Such chemoenzymatic strategies beneﬁted from the wide number of chemical approaches that can be employed to attach low molecular weight sugars to enzymes as well as the variety of enzymes capable of catalyzing transglycosylation reactions, which oﬀers a great assortment of possible synthetic combinations. In addition, chemoenzymatic methods take

advantages over the speciﬁcity of the enzymatic sugar chain enlargement and the low steric hindrance of the overall glycosylation process. However, the extension of the enzymatic transglycosylation is diﬃcult to control, which can yield neoglycoconjugates with variable composition and undeﬁned structure. Earlier examples of chemoenzymatic preparation of neoglycoenzymes are the works of Combes et al.,278−281 who described the derivatization of enzymes with CNBr-activated sucrose and the further enzymatic lengthening of the sugar chains with fructose moieties (Figure 39). They evaluated two diﬀerent glycosyltransferases: (1) fructosyltransferase (EC 2.4.1.9) from Aspergillus niger, which transfers fructose residues from sucrose to other sucrose molecules or to its analogues forming new β(1−2) bonds; and (2) levansucrase (EC 2.4.1.10) from Bacillus subtilis, which cleaves the β(1−2) bond in sucrose and synthesizes new β(2−6) bonds between the fructose residues and acceptors containing a terminal fructose forming levan-like polymeric structures. Davis et al. also described the preparation of artiﬁcial glycoenzymes through chemoenzymatic procedures.229,282 They explored the initial chemical synthesis of S-, SeS-, and SS-linked neoglycoproteins and the further enlargement of the glycan chains by Endo-A-catalyzed transglycosylation reactions. 2.3.4. Rational Design and Synthesis of Neoglycoenzymes. In this classiﬁcation, it is possible to sort those methods that allow both the regio- and the glycan-selective glycosylation of enzyme. Those approaches provide highly pure 4891

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 39. Chemoenzymatic preparation of neoglycoenzymes.

residues containing “bio-orthogonal” groups at their side chains.283 This alternative approach oﬀers the advantage that the residues, reagents, and glycosylation reaction will not be compromised by the other protogenic groups at the enzyme, and will not signiﬁcantly aﬀect the enzyme catalytic behavior. An eﬀective method to provide enzymes with alkene bioorthogonal groups is the genetic incorporation of the unnatural amino acid L-homoallylglycine (Hag). Hag can be introduced into the protein structure as a methionine surrogate by growing methionine auxotrophic bacteria in methionine-depleted media supplemented with Hag, forcing the reassignment of the methionine triplet codon to Hag.284 Hag-tagged enzyme can be further glycosylated with 1-glycosyl thiols under mild conditions through a free radical addition hydrothiolation reaction, which is started by the combined use of UV light and the water-soluble initiator 2,2′-azobis[2-(2-imidazolin-2-yl)propane]dihydrochloride (Vazo 44) (Figure 40A). Similarly, the unnatural amino acids homopropargylglycine (Hpg) (Figure 40B) and azidohomoalanine (Aha) (Figure 40C) can be incorporated into enzymes as methionine surrogates, reassigning the methionine triplet codon to an alkyne and an azide tag, respectively.282,285 Azide and alkyne are complementary bio-orthogonal groups that easily form a triazole ring through a copper(I)-catalyzed cycloaddition. The resulting engineered enzymes can be selectively and rapidly glycosylated with glycans containing the complementary bioorthogonal functionality in aqueous buﬀer. NCL has been employed as an alternative approach for the in vitro preparation of homogeneous and structurally deﬁned glycosylated enzymes.286−288 The rationale of this method is based on the design and synthesis of an N-terminal glycopeptide with site-speciﬁc glycosylation and well-deﬁned

and homogeneous neoglycoenzymes with precisely controlled carbohydrate−enzyme structure as it is particularly required for biomedical and pharmaceutical applications and for glycobiological studies. The rational design of these glycoconjugates requires a deep knowledge on the structure/function relationship of the target enzyme to select the proper modiﬁcation site without aﬀecting the catalytic activity and protein stability. The structure of the modifying glycan, the chemical approach to be employed in the glycosylation process, as well as the expected characteristics of the resulting artiﬁcial glycoenzyme are also important factors to be considered. The methods commonly employed for the rational design and synthesis of neoglycoenzymes can be divided as: (i) methods combining genetic manipulation and chemical modiﬁcation, and (ii) methods based on native chemical ligation (NCL). Site-directed mutagenesis provides a powerful tool to introduce selected amino acid residues at speciﬁc places on the enzyme surface, allowing further selective chemical modiﬁcation with activated carbohydrates. This rational strategy, called “tag-and-modify”,283 has been successfully employed to prepare artiﬁcial glycoenzymes. For this purpose, cysteine constitutes the amino acid residue most commonly introduced into protein structure by site-directed mutagenesis, due to its unique reactivity and low abundance in proteins.86,217 A great variety of chemical glycosylation methods were developed by the Davis group to speciﬁcally modify cysteine (see section 2.3.1.2), which were successfully employed for the rational synthesis of neoglycoenzymes.86−88,222−230,232−239 The “tag-and-modify” approach can empower the selective glycosylation of enzymes by incorporating unnatural amino acid 4892

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 40. Bio-orthogonal glycosylation of enzymes containing the unnatural amino acids L-homoallylglycine (A), homopropargylglycine (B), and azidohomoalanine (C).

glycan structure, which should contain a thioester group at its C-termini. This glycopeptide is further ligated to the complementary C-terminal peptide containing an N-terminal cysteine residue through a reversible transthioesteriﬁcation process followed by a spontaneous and irreversible S→N acyl shift to yield a native peptide bond. The glycopeptide prepared through this native chemical ligation approach should match the amino acid sequence of the target enzyme and should yield the catalytic active enzyme form after refolding.

NCL exhibits the advantages of bio-orthogonal chemistry but is limited by the length of the peptides, especially the glycopeptides thioesters, which can be chemically prepared by solid-phase synthesis.287,289 However, large protein scaﬀolds can be easily and economically provided by recombinant techniques. This NCL alternative is commonly referred to as expressed protein ligation.290 Another versatile NCL mode takes advantage of the sequential native chemical ligation strategy based on thiazolidine protection of the N-terminal 4893

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 41. Preparation of neoglycoenzymes by native chemical ligation.

cysteine.291 This approach allows the glycopeptide thioester to be shortened and for preparing the desired glycoenzymes by ligation of three or more constituting peptides, as it is illustrated in Figure 41.288 Until now, only relatively small glycoenzymes have been prepared by native chemical ligation approaches. Despite facilities oﬀered by the sequential strategy, it is expected that the preparation of high molecular weight glycoenzymes with satisfactory catalytic activity should be limited by the occurrence of incorrect protein refolding. However, native chemical ligation has opened new horizons in neoglycoenzyme science by demonstrating the factible rationale and synthesis of

active glycosylated enzymes using convergent chemical methods.292

3. CLASSIFICATION OF NEOGLYCOENZYMES The establishment of rules to classify neoglycoenzymes is diﬃcult to undertake due to the variety of enzymes, carbohydrate derivatives, and coupling procedures that could be involved in the preparation of such neoglycoconjugates. In this sense, the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology (NC-IUBMB) recommended a functional systematic approach to classify enzymes based on the characteristics of the enzyme-catalyzed 4894

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

reaction.293 Accordingly, enzymes are grouped into six major classes: (1) oxidoreductases, (2) transferases, (3) hydrolases, (4) lyases, (5) isomerases, and (6) ligases. On the other hand, carbohydrates can be classiﬁed according to the number of units in their structure as: monosaccharides, which are single units of polyhydroxy aldehydes or polyhydroxy ketones with three or more carbon atoms and their derivatives, but without glycosidic connection to other such sugar unit; oligosaccharides, composed of monosaccharide units joined by glycosidic linkages, and their derivatives; and polysaccharides, which are carbohydrate polymers consisting of a large number of monosaccharide units joined to each other by glycosidic linkages, as well as their derivatives.91 It should be also taken into consideration glycopolymers and glycodendrimers prepared by chemical modiﬁcation of synthetic macromolecules with carbohydrate units because some of these polymers have been successfully employed to synthesize neoglycoenzymes.224,227,294,295 Carbohydrates can be also classiﬁed according to their charge as neutral, basic, and acidic compounds, and that is relevant for neoglycoenzymes due to the direct inﬂuence of the modifying glycan charge on the functional properties of the resulting neoglycoconjugates.93,94 Furthermore, oligosaccharides and polysaccharides can be divided into two groups considering their monosaccharide composition (homo- and heteroglycans). Additionally, polysaccharides and large oligosaccharides can be linear, branched, and cyclic glycans. Important attention should be also paid to the chemical groups and the coupling procedure involved in the synthesis of neoglycoenzymes, which were detailed in section 2, giving a more complex scenario for the systematic classiﬁcation of neoglycoenzymes. To provide researchers with a simple tool for the classiﬁcation of neoglycoenzymes and to avoid misinterpretations and cumbersome descriptions on the nature of an artiﬁcial glycoenzyme, we propose to classify these glycoconjugates into four main groups according to the size of the modifying carbohydrate: (1) monosaccharide-, (2) oligosaccharide-, (3) polysaccharide-, and (4) synthetic glycopolymersbased neoglycoenzymes. In the following sections, the most relevant works dealing with the preparation of each neoglycoenzymes class will be reviewed.

Meyerhof and co-workers also prepared neoglycoenzymes with lectin-binding ability by attaching the p-(isothiocyanato)phenyl-D-glycoside derivatives of α-mannose, α-galactose, and β-N-acetylglucosamine to the enzymes malate dehydrogenase (EC 1.1.1.37) and glucose 6-phosphate dehydrogenase (EC 1.1.1.49). These neoglycoconjugates were successfully employed to prepare a sensitive lectin-based enzyme-linked competitive binding assay for the detection of carbohydrates and glycoproteins in solution.205,206 Kato et al. reported the modiﬁcation of bovine trypsin with 181 D-glucose through the Maillard reaction. Glycosylated trypsin showed high amidolytic and proteolytic activity and retained its original activity after 3 days of incubation at pH 8.0, while native trypsin was largely inactivated through autolytic processes. Glycated trypsin also showed enhanced thermal stability in comparison with the native counterpart. Thermostable glycoconjugates were also prepared by in vacuo glycation of the primary amino groups of trypsin and α-chymotrypsin with D-glucose.186 Thermal stabilization was achieved without altering the activity or speciﬁcity of the enzymes. Similarly, Janecek et al. described the modiﬁcation of Bacillus subtilis αamylase with D-glucono-δ-lactone yielding a highly thermostable enzyme derivative.126 Conjugation with monosaccharides has been also employed to improve the pharmacological and pharmacokinetics properties of several enzymes for therapeutic uses. Mannosylation of Cu,Zn-superoxide dismutase (SOD, EC 1.15.1.1) with 2-imino2-methoxyethyl-l-thiomannoside yielded a neoglycoenzyme that was rapidly cleared from the blood of rats due to uptake by nonparenchymal liver cells. However, administration of mannose-modiﬁed SOD resulted in a lower intrahepatic production of reactive oxygen species (ROS) in rats,192 and was successful to suppress ROS-mediated injury in the alveolar epithelium in rabbits.193 This conjugate, in combination with succinylated catalase (EC 1.11.1.6), also prevented the initial phase of hepatic ischemia/reperfusion injury in mice.195 Modiﬁcation with galactose residues also enhanced the hepatic uptake and biliary excretion of enzymes in rats, as it was demonstrated for SOD and lysozyme chemically modiﬁed with 2-imino-2-methoxyethyl 1-thio-D-galactopyranoside.191 Such hepatic uptake of galactosylated proteins was found to be aﬀected by the extent of galactosylation. Chemical glycosylation was also used to control the tissue distribution of catalase as well as its delivery to hepatocytes and liver nonparenchymal cells by modiﬁcation with galactose and mannose, respectively.296 Accordingly, galactosylated catalase showed high inhibitory eﬀect on hepatic metastasis of colon carcinoma cells in mice.297 Davis and co-workers reported original methods for the controlled site-selective glycosylation of Bacillus lentus subtilisin (EC 3.4.21.14) by using site-directed mutagenesis combined with chemical modiﬁcation with a variety of monosaccharide derivatives. The coupling strategies involved covalent modiﬁcation of the target cysteine residue with glycosyl methanethiosulfonate,86−88,222−227 glycosyl phenylthiosulfonate,228 and glycosyl selenenylsulﬁde derivatives of monosaccharides.229 In general, the monosaccharide-modiﬁed protease forms retained high catalytic activity, and some of them showed a remarkable broadening in stereospeciﬁcity in peptide synthesis. Similarly, Swanwick et al. described a combined site-directed mutagenesis/chemical modiﬁcation approach to prepare homogeneous neoglycoconjugates of dihydrofolate reductase

3.1. Monosaccharide-Based Neoglycoenzymes

Many works deal with the use of monosaccharides to modify enzymes. In general, conjugation can be performed through simple coupling reactions yielding neoglycoenzymes with deﬁned chemical composition and improved functional properties. Krantz et al. used Aspergillus oryzae α-amylase and hen’s eggs lysozyme to synthesize several neutral monosaccharide-based neoglycoenzymes.188 These glycoconjugates were prepared by attaching the 2-imino-2-methoxyethyl 1-thioglycoside and carboxymethyl 1-thioglycoside derivatives of D-galactose, Dglucose, N-acetyl-D-glucosamine, and D-mannose to the primary amino groups of the enzymes via amidination and carbodiimide-mediated amide formation. An azo-coupling strategy was also employed to attach the respective p-aminophenyl thioglycosides to α-amylase and lysozyme. It was demonstrated that the attachment of D-galactose residues by any of the three methods enhanced by several orders of magnitude the ability of these enzymes to bind lectins at rabbit liver membranes. 4895

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

evaluated as lengthening enzymes, the latter being the most eﬀective. This glycosylation strategy was employed to prepare neoglycoconjugates of α-chymotrypsin, lipase (EC 3.1.1.3), and lysozyme. Davis and co-workers developed a convergent and siteselective glycosylation method to attach large oligosaccharides to enzymes and other proteins, based on the preparation of the glycosyl phenylthiosulfonate derivative. This approach was successfully employed to attach the Glcα(1,4)-Glcα(1,4)Glcβtrisaccharide to the cysteine residue in the Bacillus lentus subtilisin S156C mutant with high yield.228

(EC 1.5.1.3) by reaction of the enzyme cysteine thiol group with the glycosyl iodoacetamide derivatives of glucose and Nacetyl glucosamine.219,220 3.2. Oligosaccharide-Based Neoglycoenzymes

Oligosaccharides, obtained from natural sources prepared by either synthetic or enzymatic methods, have been largely employed to prepare neoglycoenzymes by using diﬀerent coupling procedures. Marsh et al. reported the NaBH3CN-mediated reductive alkylation of Escherichia coli L-asparaginase with lactose and Nacetylneuraminyl lactose for pharmacokinetics studies.130 The lactosylated enzyme was cleared more rapidly from the plasma of mice than the native counterpart, but asparaginase modiﬁed with N-acetylneuraminyl lactose was cleared more slowly with a t1/2 that was 2-fold higher than that of the native enzyme. Vaňková and co-workers attached lactose and melibiose to trypsin by reductive coupling with sodium cyanoborohydride, yielding conjugates that contained an average of 5−9 mol of saccharide per mol of enzyme and retained 88−95% of the initial catalytic activity. These neoglycoenzymes showed high thermal stability and resistance to autolytic degradation.133 Similar stabilization eﬀects were found for this enzyme after modiﬁcation with the oligosaccharides maltotriose, raﬃnose, and stachyose, by using a similar coupling approach.132 CDs are cyclic oligosaccharides with relevant applications in neoglycoenzyme science.298 In an early work, Morand and Biellmann described the NaBH3CN mediated reductive alkylation of α-amylase from Bacillus licheniformis with a polyaldehyde derivative of β- and γ-CDs prepared via periodate oxidation.154 The cross-linked enzyme derivative showed about 75% of its initial activity and improved thermal stability. A similar approach was further used by Sebela and co-workers to prepare thermostable neoglycoconjugates of Pisum sativum and Lathyrus sativus amine oxidases (EC 1.4.3.4).158 Trypsin-based neoglycoconjugates with increased esterolytic activity were prepared by EDC-mediated chemical and transglutaminase-catalyzed enzymatic coupling of several monoamino derivatives of α-, β-, and γ-CDs with diﬀerent spacer arms.77,242,243,275,299 Carbodiimide-catalyzed conjugation with β-CDs derivatives containing adipic, pimelic, and dodecanodioic acids also yields trypsin derivatives with enhanced catalytic activity.112 However, both the esterolytic and the proteolytic activities of trypsin were signiﬁcantly aﬀected after modiﬁcation with mono-6-formyl-β-CD.136 All of the resulting neoglycoenzymes exhibited improved stability toward thermal and autolytic inactivation in comparison with the native counterpart. Similar improved stabilization was conferred to α-chymotrypsin, phenylalanine dehydrogenase (EC 1.4.1.20), SOD, lysozyme, and ribonuclease A (RNase A, EC 3.1.27.5) by chemical modiﬁcation with monoactivated CD derivatives.35,113,215,244,252 Several redox enzymes such as glucose oxidase (EC 1.1.3.4), horseradish peroxidase (EC 1.11.1.7), and xanthine oxidase (EC 1.17.3.2) have been chemically modiﬁed with monoactivated β-CD derivatives for biosensing purposes.109,114,246,300−303 An interesting chemoenzymatic glycosylation method was proposed by Longo and Combes to prepare oligosaccharidebased neoglycoenzymes.278−281 The rationale of this approach was based on the modiﬁcation of the primary amino groups of the target enzyme with CNBr-activated sucrose and further lengthening of the glycosidic chains by the action of a glycosyltransferase. Fructosyltransferase and levansucrase were

3.3. Polysaccharide-Based Neoglycoenzymes

In comparison with mono- and oligosaccharides, glycosylation with carbohydrate polymers implies a more drastic structural change in the modiﬁed enzyme. In fact, covalent attachment of polysaccharides to the surface of enzymes produces a noticeable increase in molecular weight as well as high monosaccharide content per mole of transformed enzyme, which signiﬁcantly aﬀect the physicochemical and catalytic properties of the resulting neoglycoenzymes. In general, polysaccharide-based neoglycoenzymes show noticeable increase in surface hydrophilicity, solubility, and hydrodynamic radius. In addition, the high coverage of the protein surface with the polymeric chains, linked either by single or by multiple covalent attachment, protects the three-dimensional conformation of enzymes by formation of intermolecular hydrogen bonds and stabilization of the hydration layer around the protein. Consequently, modiﬁcation of enzymes with polysaccharides usually yields glycoconjugates with improved functional stability, although the catalytic activity is often reduced. Therefore, the preparation of polysaccharide-based neoglycoenzymes has received considerable attention in neoglycoenzyme science. Dextran, a neutral polysaccharide composed of a α(1→6)linked D-glucopyranosyl backbone slightly branched with small chains of D-glucose through α(1→2), α(1→3), and α(1→4)linkages, is the carbohydrate polymer most employed for enzyme modiﬁcation. In early works, neoglycoconjugates with marked functional stability were prepared by cross-linking CNBr-activated dextran to enzymes such as trypsin,64 αamylase,63 cellulase (EC 3.2.1.4),199 α-D-galactosidase (EC 3.2.1.22),198 α-chymotrypsin,70 and cellobiase (EC 3.2.1.21).200 Triazine-activated dextran was also employed to prepare stable neoglycoenzymes from α-amylase61 and α-chymotrypsin.60 However, these glycosylation methods are currently scarcely used due to high toxicity of the activating compounds. Periodate activated dextran was employed to prepare crosslinked conjugates of dextran and penicillin G acylase,171,172 peroxidase,137 and trypsin173 by formation of Schiﬀ bases between the enzyme primary amino groups and the aldehyde groups at the oxidized polysaccharide backbone. Subsequent reductive treatment with NaBH4 or NaBH3CN yielded stable dextran-based glycoconjugates, as it was reported for hyaluronidase (EC 3.2.1.35),304 α-chymotrypsin,95 and αamylase.146,305 Lenders and Crichton employed this later coupling procedure to modify pullulanase (EC 3.2.1.41) and α-amylase with dextrans of diﬀerent molecular weights ranging from 10 to 2000 kDa.138 An apparent correlation between the molecular weight of the neoglycoenzymes and their thermal stability, resistance to urea denaturation, and kinetic parameters was observed. Glycosylation with dextran was demonstrated to be an eﬀective tool for improving pharmacological, pharmacokinetics, 4896

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

polysaccharide.85,245,247 Zhou and co-workers described the carbodiimide-mediated coupling of N-succinyl-chitosan to cellulase and alliinase (EC 4.4.1.4) for preparing glycoconjugates with pH-responsible solubility equilibrium.122,315 Qian et al. reported the modiﬁcation of E. coli L-asparaginase with N,Ocarboxymethyl chitosan, the resulting neoglycoenzyme showing increased plasma half-life in pharmacokinetics experiments.316 SOD was modiﬁed with N,N,N-trimethyl chitosan.317 In vitro experiments revealed that the conjugated enzyme exhibited a noticeable inhibitory eﬀect on superoxide anion release from macrophages and reduced the production of the inﬂammatory cytokines growth factor-β1 and interleukine-1β by irradiated 3T3 ﬁbroblasts. Carboxymethyl chitin was also successfully evaluated by Valdivia et al. as modiﬁcation agent for improving the pharmacokinetics and pharmacological properties of catalase and SOD.115,116 Several polysaccharides were mainly employed to modify enzymes with potential drug applications. Gregoriadis and coworkers proposed the use of colomic acid, an α-(2→8) Nacetylneuraminic acid (sialic acid) polymer, as a substitute of polyethylene glycol to enhance the pharmacological properties of catalase and asparaginase.143−145,318,319 Maksimenko and coworkers employed modiﬁcation with chondroitin sulfate to improve the pharmacokinetics and antithrombotic properties of catalase and SOD.209−212,320−322 Moreover, modiﬁcation with heparin has been proposed as an eﬀective strategy to improve the half-life and tissue targeting of SOD to suppress ROSmediated injury.176,317,323,324 Heparin-modiﬁed enzyme exhibited excellent inhibitory eﬀects on superoxide anion release from macrophages and reduced the radiation-induced inﬂammatory cytokine expression in vitro.317 This conjugate was successful to inhibit bleomycin-induced lung ﬁbrotic lesions, to prevent eﬀects of CCl4-induced acute liver failure and hepatic ﬁbrosis in mice, and to avoid brain reperfusion injury after ischemia in gerbils.325−327 Other natural polysaccharides such as inulin,162 levan,139 sodium alginate,127,153,157,328 sodium hyaluronate,120 xanthan,146 mannan,150,163 and pectin153,266 were also successfully employed for the preparation of enzyme conjugates.

and immunological properties of enzymes with biomedical applications. Cross-linking of L-asparaginase and phenylalanine ammonia lyase (EC 4.3.1.24) with periodate-activated dextrans via reductive alkylation with NaBH4 yielded conjugates with reduced antigenicity and improved resistance to proteolytic degradation.197,306 Reductive alkylation of uricase (EC1.7.3.3) and SOD with polyaldehyde dextran produced neoglycoenzymes exhibiting sustained enzymatic activity in plasma after intravenous injection to rats.307,308 The dextran−SOD conjugate inhibited evolving ﬁbrosis in the rat lungs.308 End-group aminated and carboxylated dextran derivatives were used for the single-point modiﬁcation of catalase yielding conjugates with improved circulatory half-life in rats.79,84,121 Carbodiimide-mediated attachment of monoaminated dextran to SOD not only improved its pharmacokinetics properties, but also increased the antiinﬂamatory activity in experimental rats.248 Polyionic dextran derivatives, such as carboxymethyl dextran and diethylaminoethyl dextran, were also employed to modulate the target delivery and renal disposition of SOD by chemical cross-linking.309−311 Water-soluble cellulose derivatives were also evaluated as modiﬁcation agents for enzymes. Among these, CMC, a negatively charged cellulose derivative, has been widely employed in neoglycoenzyme synthesis. In an early paper, Mitz and Summaria described the conjugation of azideactivated CMC to α-chymotrypsin.56 Enzymes with industrial interest, such as trypsin,94 α-amylase,93 and α-chymotrypsin,95 were alkylated with periodate-oxidized CMC via NaBH4 coupling reduction. A novel trypsin−CMC conjugate has been also prepared through an Ugi four-component reaction.127 In addition, carbodiimide-mediated coupling was employed to attach CMC to the free amino groups, either at the protein surface of enzymes,106,117 or previously introduced into the carbohydrate chains of glycoenzymes,267 to prepare stable neoglycoconjugates. The use of CMC to prepare neoglycoenzymes with potential pharmacological applications was also explored. Covalent attachment of CMC increased the antileukemic activity of Lasparaginase in mice,312 and also improved the antiinﬂamatory activity of SOD in the carrageenan-induced paw edema test rats.107,108 This conjugation also increased the circulatory halflife in rats for SOD and catalase conjugates.116 A β-CD branched CMC derivative was employed to modify SOD. The anti-inﬂammatory activity of this enzyme was 2.2 times increased after conjugation, and its plasma half-life time was 90-fold longer.103 This CMC derivative was also successfully used to prepare stable derivatives of several industrial enzymes, such as α-amylase and trypsin.101,118 The pH-sensitive polymer hydroxypropyl methylcellulose acetate succinate constitutes another example of cellulose derivative that found application in the synthesis of neoglycoenzymes. Cellulase, lysozyme, and chitinase (EC 3.2.1.14) were covalently modiﬁed with this polysaccharide via a carbodiimide-coupling reaction, yielding neoglycoconjugates that showed pH-responsible soluble−insoluble characteristics.104,105,313,314 Several chitin derivatives were successfully employed to prepare neoglycoenzymes with improved characteristics. The attachment of chitosan to the glycosidic chains of invertase (EC 3.2.1.26) and cellulose yielded conjugates with marked thermal stability.259,260 A carbodiimide coupling procedure was employed to modify laccase (EC 1.10.3.2) with this

3.4. Synthetic Glycopolymers-Based Neoglycoenzymes

Several water-soluble synthetic polymers containing carbohydrates moieties, prepared by (i) cross-linking saccharides, (ii) attaching sugar residues to polymeric chains, or (iii) using a monosaccharide core as branching point to growth dendrimer structures, have been used to prepare neoglycoenzymes. Some synthetic polysaccharides prepared by random crosslinking of oligosaccharides with epichlorhydrin were used as glycosylation agents for enzyme modiﬁcation. Sundaram and co-workers described the preparation of highly stable enzyme derivatives by attachment of periodate-oxidized Ficoll, a synthetic branched copolymer of sucrose and epichlorhydrin, to α-chymotrypsin,95 papain (EC 3.4.22.2),156 horseradish peroxidases,92 and trypsin.74 Srimathi and Jayaraman also described the use of Ficoll to modify α-amylase.160 OCarboxymethyl poly-β-CD, prepared by copolymerization of β-CD with epichlorhydrin and further etheriﬁcation with monochloroacetic acid, is another example of synthetic polysaccharide employed to prepare stable neoglycoenzymes.100 Lee and Park reported the synthesis of several temperaturesensitive copolymers containing glucose units in their backbone based on N-isopropylacrylamide and glucosyoxylethyl meth4897

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

polysaccharides as modifying agents contribute to this high molecular weight dispersion for the conjugates. Modiﬁcation of charged amino acids residues at the surface of enzymes leads to changes in their pI due to partial neutralization of the charge on the protein molecule.184,251 The attachment of charged carbohydrates also aﬀects noticeably the pI of the modiﬁed enzyme.131 It has been described that artiﬁcial glycosylation can promote changes in the secondary structure of the enzymes, leading to a helix to turn conversion, as reported by circular dichroism analysis of α-chymotrypsin and α-amylase modiﬁed with dextran.70,146 This transformation was attributed to the insertion of the pedant carbohydrates residues into α helix leading to the conformational conversion of helices to beta turns. Fluorescence spectroscopy studies revealed that glycosylation of enzymes preserved the tertiary structure of the protein.74,95,111,241,252 This conformational stabilization was also observed in the presence of high concentration of guanidinium ion164 or under thermal treatment,74,95,111,167,241,252 which cause partial or complete unfolding of the nonmodiﬁed counterparts. However, Maksimenko et al. noticed that considerable conformation changes could take place at high degree of protein glycosylation, as reported for hyaluronidasemodiﬁed with dialdehyde dextran via reductive alkylation with NaBH4.304 Additionally, Yeboah et al. demonstrated that glycation of lysozyme with D-glucose or D-fructose caused alterations in the conformation of the enzyme provoking an increased exposure of the active site residues and a higher solvent penetration into these active sites.330 They suggested that the enhanced enzymatic activity showed by lysozyme after glycation could be explained by such conformational changes. Covalent cross-linking with polysaccharides was shown to stabilize the quaternary structure of oligomeric enzymes due to cross-linking of several protein subunits with the same polymer chain.259,266,267 However, it was reported that an excess of the modifying polysaccharide in the reaction medium promoted dissociation of oligomeric enzymes into individual subunits.308 This stabilization of the quaternary structure requires a proper length for the modifying carbohydrate, and it has not been observed for oligosaccharide-modiﬁed enzymes.158

acrylate, which resulted in one terminal carboxylic acid group per polymer chain.329 The polymers were used to modify the primary amine groups at the surface of trypsin using a watersoluble carbodiimide as a coupling agent. This glycosylation scheme led to polymer−enzyme adducts with star-shaped conformations, which exhibited reversible precipitation/solubilization behavior over a wide range of temperatures depending on the content of glucosyoxylethyl methacrylate in the copolymer. The same group described the synthesis of a series of temperature-sensitive copolymers composed of N-isopropylacrylamide and acrylamido-2-deoxy-D-glucose, which were covalently attached to α-chymotrypsin through the carbohydrate moieties using periodate activation/reductive alkylation with NaBH3CN.141 The conjugated enzymes were able to be reused several times through temperature-induced reversible precipitation−solubilization cycles. Davis and co-workers reported the synthesis of original monosaccharide-terminated biantennary and second-generation dendrons with a methanethiosulfonate-activated thiol group at the core, which were used for the site-selective glycosylation of the cysteine residue in a Bacillus lentus subtilisin S156C mutant.224−227 Other interesting carbohydrate-containing dendrimers for neoglycoenzyme synthesis were described by Krahmer et al.294 They prepared several branched polymers containing a monosaccharide central unit attached to polyethylene glycol chains at the hydroxyl groups and to an Nhydroxisuccinimide active linkage group at the anomeric position of the carbohydrate central unit. The polymers were further attached to the primary amino groups of L-asparaginase yielding conjugates resistant to proteolytic degradation by trypsin.

4. STRUCTURAL AND FUNCTIONAL PROPERTIES OF ARTIFICIAL GLYCOENZYMES 4.1. Structural and Conformational Characteristics

Glycosylation of enzymes with carbohydrate derivatives leads to a major change in the primary structure of these proteins, caused by modiﬁcation of the target amino acid residues and the consequent introduction of hydrophilic sugar moieties at the enzyme surface. Inevitably, such structural transformation should modulate the conformational, catalytic, stability, solubility, and biorecognition properties of the glycoconjugates. In general, these new characteristics are modulated by several factors including: (i) the nature, location on the threedimensional structure of the protein, and number of target modiﬁed amino acid residues; (ii) the composition, size, and physicochemical properties of the modifying carbohydrate; (iii) the type, size, and structure of the enzyme; and (iv) the coupling approach and conditions employed to prepare the neoglycoenzyme. Within this complex scenario, it is rational to expect speciﬁc structural and conformational properties for each particular neoglycoenzyme. However, some rules can be proposed. In general, as anticipated, carbohydrate−enzyme conjugates show higher molecular weight than their native counterparts.147,158,184 This mass increase depends on the molecular weight of the modifying carbohydrate and the degree of modiﬁcation.153,184,251 A broad molecular weight distribution is often shown by neoglycoenzymes due to formation of enzyme−carbohydrate conjugates of diﬀerent composition.137,162,167 Group-speciﬁc glycosylation and the use of

4.2. Catalytic and Kinetics Properties

There is a fundamental structure−activity relationship in enzymes, in which the 3D conformations have been evolutively optimized to promote the speciﬁc and eﬃcient catalytic transformation of target substrates. Accordingly, transformation of amino acid residues by glycosylation should also aﬀect notably the enzyme-mediated catalysis. In general, alterations in the speciﬁc activity and kinetics parameters can be mediated by the combined action of several factors: (i) modiﬁcation or masking of amino acid residues involved in the catalysis or the enzyme−substrate complex formation; (ii) esteric hindrance for the diﬀusion of substrates to the catalytic site caused by the bulky carbohydrate moieties; and (iii) altered protein ﬂexibility in and around the active site region by intramolecular covalent and noncovalent crosslinking with the glycans. The general trend is the reduction of the speciﬁc activity and turnover number, kcat, of enzymes upon glycosylation. These eﬀects are signiﬁcantly noticed for neoglycoenzymes that catalyze the transformation of high molecular weight substrates, as well as for polysaccharide-based neoglycoenzymes prepared 4898

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

noticed that lower inhibitory eﬀects were shown for higher glycosylated enzyme forms.148 Lower inhibition by substrate was also observed for α-D-galacosidase after modiﬁcation with CNBr-activated dextran.198 Covalent modiﬁcation with carbohydrates often leads to changes in the optimal conditions for the enzyme-catalyzed reaction. Variation of optimum pH can be observed if ionization of the amino acid residues involved in catalysis is aﬀected by glycosylation. Broader ranges of optimum pH have been reported for trypsin modiﬁed with CMC, dextran, and Ficoll, a branched polymer of sucrose cross-linked with epichlorhydrin.78,94 Furthermore, the pH value for maximal catalytic activity was shifted for α-D-galacosidase-dextran,198 subtilisin-DEAE-dextran.67 On the contrary, optimum pH remains the same for trypsin modiﬁed with the mono-6amino-6-deoxy derivatives of α-, β-, and γ-CD,242 and dextran,64 as well as for α-chymotrypsin modiﬁed with dextran, Ficoll, and CMC.95 The optimum temperature is often increased after covalent attachment of carbohydrate moieties to the enzyme surface. Such a change is directly related to an increased conformational rigidity in the enzyme upon glycosylation and have been reported for several polysaccharide-based neoglycoenzymes such as trypsin-modiﬁed with β-CD-branched CMC,118 invertase-pectin,266 cellobiase-dextran,200 and α-chymotrypsinmodiﬁed with O-carboxymethyl-poly-β-CD.119 A similar eﬀect was observed for β-CD-modiﬁed enzymes.113,252 Furthermore, a broad temperature/activity proﬁle was shown by cellulase after modiﬁcation with chitosan, although the optimum temperature remains unchanged.260

by cross-linking reactions. For instance, lower catalytic activity was reported for trypsin-alginate, L-asparaginase-polysialic acid, penicillin G acylase-dextran, and α-amylase-CMC conjugates.93,144,157,172 Moreover, the kcat value of B. amyloliquefaciens α-amylase-maltodextrin and trypsin-carboxymethyl poly-β-CD neoglycoenzymes was lower than that for the native counterparts.100,184 A useful approach to prepare neoglycoenzymes with high catalytic activity is based on the protection of the amino acid residues at the active site by inclusion of substrate in the glycosylation reaction medium. This method was employed to modify bovine trypsin with lactose, maltose, melibiose, maltotriose, raﬃnose, stachyose, α-CD, and β-CD by using benzamidine as protectant. The prepared conjugates retained 70−95% of the initial catalytic activity.132,133 Similarly, a neoglycoenzyme retaining 65% of the original activity was prepared by Tabandeh and Aminlari through the attachment of periodate-oxidized inulin to L-asparaginase via reductive alkylation with NaBH4 using L-asparagine as an active protector of the enzyme.162 Some glycosylations can yield neoglycoenzymes with catalytic activity similar to or even higher than the initial nonmodiﬁed enzymes. Several reports described this eﬀect for enzymes modiﬁed with mono- or oligosaccharides such as the case of trypsin−CD conjugates prepared through chemical or enzymatic approaches,77,112,113,242−244 as well for ribonuclease A (EC 3.1.27.5) and L-asparaginase modiﬁed with lactose.129,130 Higher catalytic activity has been also shown for some polysaccharide-based neoglycoenzymes such as hyaluronidase modiﬁed with dextran,331 and lipase modiﬁed with inulin, dextran, and Ficoll.159 A similar eﬀect was observed for penicillin G acylase conjugated with mannan, dextran, potassium pectate, and sodium alginate.153 The inﬂuence of covalent glycosylation on the apparent Michaelis constant value, KM, varies considerably from case to case. An apparent increase in the aﬃnity for substrate was described for several neoglycoenzymes including phenylalanine dehydrogenase-CD, B. amyloliquefaciens α-amylase-Ficoll, and trypsin-CMC derivatives.94,160,252 On the contrary, similar or higher values of K M were reported for levansucraseglucomannan, Taka-amylase A-carboxymethyl dextran, and B. licheniformis α-amylase-Ficoll glycoconjugates.110,160,169 Consequently, some glycosylation can lead to enhanced catalytic eﬃciency, kcat/KM,112,77,241,252 but in some cases there is a decreased or unchanged eﬀect on these parameters.100,160 Thus, in general, rational manipulation of enzyme catalytic and kinetics properties by glycosylation appears still to be in an empirical stage. However, Davis and co-workers demonstrated that the eﬀect of glycosylation on the catalytic eﬃciency of enzymes is highly inﬂuenced by the location of the target modiﬁed amino acid residues into the protein structure. They reported that the sitespeciﬁc introduction of a representative library of mono- and disaccharides, and their acetylated forms, at key positions within the active site of B. lentus subtilisin mutants caused a diﬀerent eﬀect on the kcat/KM values for esterase activity.87,88 Furthermore, the ratio of amidase to esterase activity was increased relative to wild type for all glycosylated forms of the enzyme.221 Artiﬁcial glycosylation can aﬀect the inhibition of enzymes for selected compounds. Maksimenko et al. described that the inhibition of hyaluronidase by heparin can be controlled through chemical modiﬁcation with aldehyde dextran. They

4.3. Stability

One of the most interesting properties exhibited by neoglycoenzymes is the improved functional stability toward several denaturing/inactivating conditions for proteins. This stabilization includes resistance to thermal treatment, incubation at extreme pH values, denaturation by surfactants or chaotropic substances, proteolytic degradation, and the presence of organic solvents. From a practical point of view, industrial processes catalyzed by thermoresistant enzymes are largely desired. In fact, there are several important technological aspects favored by carrying out enzymatic reactions at high temperatures such as the increase of the reaction rate, operational stability, and solubility of reactants and products. In addition, the use of enzymes at high temperatures can shift the thermodynamic equilibrium to higher product conversion, decrease the viscosity of the reaction medium, and reduce microbial contamination.25 A wide variety of carbohydrate derivatives were successfully employed to stabilize enzymes against thermal treatment. The best results were achieved by using polysaccharides as glycosylation agents. In an early work, Marshall described the stabilization of B. amyloliquefaciens α-amylase against incubation at 65 °C by cross-linking with CNBr-activated dextran.63 Morand and Biellmann reported that cross-linking of B. licheniformis α-amylase with periodate-activated β-CD increased the half-life at 80 °C from 4.7 to 7.0 min.154 Glycation with maltodextrin improved the stability of B. subtilis α-amylase toward incubation at 95 °C, retaining about 13% more activity than the native counterpart after 5 min at this temperature.184 Noticeable thermal stabilization was also reported for αamylase cross-linked with polymerized sucrose,160 and the ionic 4899

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

polysaccharides CMC,93 carboxymethyldextran,110 alginate,157 and β-CD-branched CMC.101 Cellulase, an endoglycosidase with relevant applications in paper and pulp industries, was thermostabilized by modiﬁcation with CNBr-activated dextran,199 chitosan,260 and N-succinylchitosan.122 Increased thermostability was also reported for lysozyme by conjugation with hydroxypropyl methylcellulose acetate succinate, dextran, galactomannan, and mannan,105,178,332 as well as for cellobiase by cross-linking with periodate- and CNBr-activated dextran.147,200 In addition, a noticeable thermal stabilization was reported for invertase after attachment of anionic and cationic polysaccharides to the enzyme sugar chains. High resistance to incubation at 65 °C was described for the chitosan and CMC enzyme conjugates,259,267 and a drastic increase in half-life from 5 min to 2 days at this temperature was reported for the invertase− pectin complex.266 Proteolytic enzymes have been also improved for thermal resistance by modiﬁcation with carbohydrates. Several reports described the thermostabilization of trypsin and α-chymotrypsin by modiﬁcation with monoactivated CD derivatives, polymerized sucrose, and β-CD-based polysaccharides.37,74,95,100,112,113,118,119,136,242−244,275 Increased thermostability was also obtained for trypsin cross-linked with dextran, CMC, and alginate.94,127,173 Glycosylation with dextran was also described as an eﬀective tool to improve thermal stability for pepsin and carboxypeptidase A.249 Themostable derivatives were prepared by conjugation of phenylalanine dehydrogenase with O-carboxymethyl poly-βCD, end-group activated dextran, and β-CD derivatives.111,241,252 Thermal stabilization was also reported for Lasparaginase modiﬁed with levan, inulin, N,O-carboxymethyl chitosan, and lactose.130,139,162,316 Penicillin G acylase conjugates with improved thermal stability were prepared by modiﬁcation with dextran, mannan, potassium pectate, and sodium alginate.150,153,171,172 In general, the improved thermal stabilization exhibited by these neoglycoenzymes can be ascribed to the contribution of several factors for maintaining the active conformation of the enzymes avoiding unfolding of the polypeptide chains and further possible intermolecular aggregation when exposed at elevated temperatures. The most important factor for this conformational stabilization is the covalent and noncovalent intramolecular cross-linking with the carbohydrate moieties.74,95 Higher stabilization is thus expected for neoglycoenzymes prepared by covalent multipoint attachment of polysaccharides or large oligosaccharides to the enzyme structure. Furthermore, the formation of multipoint hydrogen bonds between the attached carbohydrates and the hydrophilic groups at the enzyme surface should also contribute to the stabilization of the three-dimensional structure of the enzymes. Charged carbohydrates also increase conformational stabilization by formation of multipoint electrostatic interactions with opposite charged amino acid residues at the modiﬁed enzyme surface.93,94 Attachment of carbohydrates provides a higher hydrophilicity to the enzyme surface, which also contributes to thermal resistance by stabilization of the hydration layer around the enzymes,333 and also by preventing the intermolecular protein aggregation caused by the exposure of the polar patches at the protein core when unfolding. In the case of neoglycoenzymes prepared with CD derivatives, it has been postulated that the formation of host−guest inclusion complexes between the

hydrophobic amino acids at the enzyme surface and the attached oligosaccharides should also contribute to the noticeable thermal stabilization exhibited by these conjugates.113,136,242,243,275 Several of these factors also contribute to the stabilization often shown by neoglycoenzymes against denaturation/ inactivation when they are exposed to extreme acidic or alkaline condition,93,94,139,157,200 and to chaotropic agents such as urea and guanidinium salts.74,94,95,161,200 Covalent glycosylation can also protect enzymes against exogenous proteolytic degradation or autolysis by modiﬁcation and/or masking of the potential cleavage sites in the protein structure. This stabilization eﬀect was described for artiﬁcial glycoconjugates of α-amylase,110 SOD,176 catalase,79,116,121,143 130,144,162 L-asparaginase, and several pro94,127,173,243,316,329 teases. Improved stability against denaturation by surfactants was reported for glycosidases and proteases covalently cross-linked with polysaccharides.93−95,118,157,200 Some reports also described higher stability for neoglycoenzymes in the presence of organic solvents.94,153 Moreover, improved stabilization against inactivation by oxidizing agents such as H2O2 has been reported for some neoglycoenzymes.248 4.4. Biorecognition Properties

Carbohydrate/lectin interactions are involved in a wide variety of biological events including cellular growth, cell recognition, viral infection, cancer metastasis, opsonization of microorganisms, phagocytosis, cell adhesion and migration, cell activation and diﬀerentiation, and apoptosis.334,335 Such carbohydrate/lectin interactions have been used for many applications including the development of pathogen detection strategies, chemical, clinical and histochemical methods of analysis, drug delivery systems, aﬃnity chromatography methods, antifungal products, antitumor and antiviral drugs, and for the structural characterization of glycoproteins.336 Attachment of carbohydrate moieties to the protein surface confers lectin-biorecognition properties to these conjugates. This fact was early reported by Krantz et al. who modiﬁed αamylase and lysozyme with the 2-imino-2-methoxyethyl 1thioglycosides of D-galactose, D-glucose, N-acetyl-D-glucosamine, and D-mannose, demonstrating the abilities of these neoglycoproteins to bind liver membranes.188 Kim et al. proﬁted the capability of lectins to recognize neoglycoenzymes to determine the glycan structure in naturally occurring glycoproteins.206 This approach was based on the conjugation of malate dehydrogenase with galactose, mannose, and N-acetylglucosamine and their reversible inhibition by speciﬁc interaction with Jacalin, Concanavalin A, and wheat germ agglutinin, respectively. Recovery of enzymatic activity in the presence of speciﬁc glycans allows estimating the type and relative amount of speciﬁc carbohydrate structures in glycoproteins. Moreover, β-galactosidase (EC 3.2.1.23)-based neoglycoenzymes were successfully employed by Gabius and co-workers to detect lectins in solutions and at the surface of tumor cells and tissues.337−342 Modiﬁcation of enzymes with yeast mannan, a highly branched mannose polymer, favored their speciﬁc recognition with Concanavalin A, as it was shown for SOD, α-amylase, and penicillin G acylase derivatives.150−152,155,163 This lectinbiorecognition ability allowed the aﬃnity-based immobilization of mannan-modiﬁed enzymes on Concanavalin A-functionalized solid supports for industrial applications. 4900

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

A glycodendrienzyme, prepared through site-directed glycosylation, was used by Rendle and co-workers to prevent the lectin-mediated oral adhesion and growth of pathogenic microorganisms.227 They described the synthesis of one to four carbohydrate-tipped antennae, activated with sulfhydryl speciﬁc methanethiosulfonate group, and their use for the selective glycosylation of an engineered subtilisin from Bacillus lentus. These derivatives were tested for the inhibition of the coaggregation of Actinomyces naeslundii, a Gram-positive bacterium that aggressively colonizes oral cavities and surgical prostheses with its copathogen Streptococcus oralis.343,344 The glycodendrienzyme with biantennary galactose structure showed a potent inhibitory eﬀect (IC50 = 20 nM) resulting from the combination of the protein degrading activity and multiantennary carbohydrate display in the conjugate, favoring its optimal recognition and further degradation of the galactosebinding adhesin from A. naeslundii. Robinson and co-workers used neoglycoenzyme technology to develop a novel lectin-directed enzyme-activated prodrug therapy (LEAPT).194 This approach is a bipartite delivery system based on the site-selective delivery of an artiﬁcially glycosylated rhamnosidase (EC 3.2.1.40) by sugar-based receptor-mediated endocytosis, and further delivery of a rhamnose-capped prodrug that can be cleaved only by the delivered glycosylated rhamnosidase. For this purpose, naringinase (EC 3.2.1.40) was enzymatically deglycosylated and further reglycosylated with mannose or galactose moieties by treatment with the corresponding 2-imino-2-methoxyethyl1-thioglycoside. Artiﬁcial glycosylation enhanced the rate of enzyme uptake from serum, thereby reducing the time for potential immunogenic exposure, and substantially improved the enzyme stability under degradative conditions that mimic those found in vivo. The conjugates were speciﬁcally delivered into hepatocytes through a receptor-mediated endocytosis mechanism. The therapeutic eﬀectiveness of this lectin-directed enzyme-activated prodrug therapy was tested by using a rhamnosidated derivative of doxorubicin as prodrug, and its application was successful to reduce tumor burden in a hepatocellular carcinoma HepG2 disease model. It should be highlighted that this neoglycoenzyme-based drug delivery system may be applicable to other cell types, with the possibility of speciﬁc carbohydrates allowing speciﬁc cell targeting. Furthermore, the covalent attachment of carbohydrate residues to enzymes can reduce their antigenic properties due to modiﬁcation or masking of the antigenic determinants at the protein surface, thus avoiding biorecognition by T cell receptors or antibodies. Similarly, neoglycoenzymes often show lower immunogenic response than the native counterparts. These immunological properties have contributed to the wide evaluation of neoglycoenzymes as drugs in enzyme replacement therapies. Relevant examples on the inﬂuence of artiﬁcial glycosylation on the antigenic and immunogenic properties of enzymes are discussed in section 5.1.

Such increased hydrophilicity improves the water solubility of enzymes, which is probably associated with the reduced protein aggregation and surface adsorption showed by neoglycoenzymes as compared to that of unmodiﬁed proteins.215 The solubility of covalent glycosylated enzymes can be controlled by manipulation of the solubility characteristics of the modifying carbohydrates. Attachment of polysaccharide derivatives with pH-responsible solubility leads to neoglycoenzymes that show pH-mediated solubilization characteristics as is the case for cellulase, lysozyme, and chitinase104,105,313,314 chemically modiﬁed with hydroxypropyl methylcellulose acetate succinate. A pH-sensitive neoglycoenzyme with reversibly soluble−insoluble characteristics was also prepared by glycosylation of alliinase with N-succinyl chitosan.315 Artiﬁcial glycosylated enzymes can show improved emulsifying properties in comparison with native counterparts.177−180,332 Neoglycoenzymes with signiﬁcantly higher emulsifying characteristics were prepared by controlled Maillard conjugation with neutral polysaccharides such as the lysozymegalactomannan,179,180 lysozyme-dextran,178,332 and lysozymemannan conjugates.332 Improved emulsifying properties were also shown by lysozyme after conjugation with glucose stearic acid monoester through mild Millard-type reaction.345 Several of these neoglycoenzymes retained high emulsifying properties in the presence of relative high NaCl concentration and in acidic pH conversely to that occurring for commercial emulsiﬁers.

5. APPLICATIONS 5.1. Neoglycoenzymes for Biomedical Application

Enzymes are involved in many pathophysiological disorders, and, therefore, their use as therapeutic agents for the prevention or systemic treatments of diseases became an important strategy in medicine and pharmacology.346−349 During the last decades, the impressive development of the pharmaceutical biotechnological industry and molecular biology techniques for gene manipulation favored the production of a number of enzymes in large amounts.350 However, the use of these promissory protein drugs has been limited by several problems such as high immunogenicity and antigenicity as well as rapid body clearance, which is mediated mainly by antibody recognition, proteolytic degradation, clearance through the liver, and glomerular ﬁltration through the kidney.351−353 Undoubtedly, the preparation of novel conjugates with improved pharmacokinetics and pharmacological properties for biomedical use constitutes the major application ﬁeld for neoglycoenzymes. The rationale for the modiﬁcation of therapeutic enzymes with carbohydrate derivatives is the possibility to prolong the plasma half-life by increasing their hydrodynamic volume due the bulky polyhydroxylated chains attached, hence decreasing their excretion rate. Artiﬁcial glycosylation can also prevent the antibodies recognition by masking the epitopes located at the surface of the exogenous enzyme. In addition, neoglycoenzymes can be more resistant to proteolytic and cell attacks due to the steric hindrance caused by the sugar residues. In some cases, glycosylation can favor the biorecognition of the neoglycoenzymes by speciﬁc cell receptors allowing the target delivery of the therapeutic adduct to the aﬀected tissue. The ﬁrst neoglycoenzyme successfully synthesized and commercialized for biomedical use involved the thrombolytic

4.5. Other Characteristics

Artiﬁcial glycosylation increases the hydrophilic characteristics of the enzymes contributing to the formation of a stable hydration layer around the protein molecule. This fact was conﬁrmed by water sorption isotherm experiments on dextranmodiﬁed lipase, demonstrating that the amount of adsorbed water depends on the amount of carbohydrates linked to the enzyme surface.333 4901

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

agent streptokinase (EC 3.4.99.22).354,355 This enzyme was cross-linked with periodate-oxidized dextrans of 35−50 kDa, which signiﬁcantly improved the pharmacokinetics proﬁle of the enzyme, remaining active in human body after 3 days of administration. Glycosylation with dextran also reduced the toxicity of the enzyme, as reﬂected by the reduced allergic reactions, rethromboses, and hemorragic complications in patients. This streptokinase−dextran conjugate is industrially produced in Russia, commercialized under the trademark of Streptodekase, and its use for the treatment of ophthalmological and cardiovascular diseases caused by thrombosis was approved in 1980.356 L-Asparaginase is another example of a therapeutic enzyme that has received signiﬁcant attention for its artiﬁcial glycosylation. Asparaginase is an antitumor enzyme with speciﬁc use in the treatment of acute lymphoblastic leukemia and commonly employed in native form.357,358 However, its rapid clearance and the development of immune-based resistance in several patients are important drawbacks that could be overcome by in vitro glycosylation. Wriston and co-workers early reported the chemical glycosylation of Escherichia coli asparaginase with lactose and N-acetylneuraminyl lactose through reductive alkylation with sodium cyanoborhydride.130 These neoglycoconjugates retained high catalytic activity despite the substantial degree of modiﬁcation achieved. These adducts were also more resistant to thermal inactivation and proteolytic degradation. However, only the enzyme derivative with high content of Nacetylneuraminyl lactose showed a better pharmacokinetics proﬁle than the native counterpart, with a 2-fold increase in half-life. No signiﬁcant change in the pharmacokinetics behavior of asparaginase modiﬁed with glucuronic acid was reported by the same group.131 These results suggest that modiﬁcation of asparaginase with low molecular weight sugars was not an eﬃcient strategy for improving its body circulation. On the contrary, an increased circulatory half-life was described for Erwinia carotovora asparaginase modiﬁed with oxidized dextrans.306,359,360 These conjugates retained about 50% of the initial enzyme activity and showed marked resistance to proteolysis by trypsin and α-chymotrypsin and to inactivation by asparaginase-speciﬁc antibody. The larger molecular weight dextran−asparaginase conjugate showed prolonged circulatory survival in both immune and nonimmune animals and failed to elicit full type III hypersensitivity or anaphylactic reactions when injected into sensitized guineapigs. A signiﬁcant reduction of antigenicity, with maintenance of high catalytic activity, was also reported for an asparaginase− dextran conjugate synthesized with the presence of the enzyme substrate in the reaction medium.361 Sherwood and co-workers reported also that conjugation of asparaginase to dextran signiﬁcantly enhanced its plasma persistence in normal and tumor-bearing mice.71 In addition, a prolonged decrease in arginine concentrations in plasma of tumor-bearing mice was demonstrated by using the dextranlinked arginase. A noticeable improvement in the pharmacokinetics behavior of asparaginase in rabbits was conferred by modiﬁcation with CMC.312 This adduct was also tested for antileukemic activity in mice with inoculated lymphoid leukemia L5178y, showing higher antitumor activity than the native enzyme. A 33-fold increase in the plasma half-life of E. coli L-asparaginase was achieved by conjugation with N,O-carboxymethylchitosan.316 The authors included L-aspartic acid, the normal product of the

asparaginase-catalyzed reaction, in the conjugation medium to preserve the active site during the coupling process. This approach was successful for preparing a neoglycoconjugate with high catalytic activity. The asparaginase−carboxymethylchitosan conjugate showed high resistance toward proteolytic degradation by trypsin and α-chymotrypsin but was rapidly inactivated at high temperatures. A thermostable asparaginase−inulin conjugate was prepared by reductive alkylation of the free amino groups at the surface of the enzyme with the periodate-oxidized polysaccharide in the presence of sodium borohydride.162 This glycosylation also increased 5 times the aﬃnity of the enzyme for the substrate and the resistance to trypsin digestion. The conjugate showed lower immunogenicity than the native enzyme as revealed by the lower production of total IgG after repeated injection in rabbits. Similar polymer activation and enzyme conjugation approaches were employed for coupling levan, a bioactive fructose polymer, to asparaginase.139 The neoglycoconjugate exhibited higher thermal and storage stability, but no possible therapeutic use was investigated. Promissory results were achieved by using colominic acid, a low molecular weight polysialic acid, as glycosylation agent for 318 L-asparaginase. The coupling strategy was based on the activation of colominic acid by periodate oxidation of carbon 7 at the nonreducing end of the polymer, and its further attachment to the free amino groups of the enzyme by reductive alkylation with sodium cyanoborohydride. This approach has the advantage of preparing non-cross-linked enzyme−polymer conjugates with more deﬁned composition and lower polydispersion, properties that are desired for such kind of macromolecular drugs. Asparaginase−polysialic acid conjugates retaining more than 80% of the initial catalytic activity and showing similar aﬃnity for L-asparagine were prepared through this synthetic scheme. These neoglycoenzymes were also resistant to proteolytic degradation after 6 h of exposure to mouse blood plasma, whereas the native enzyme was completely inactivated. In comparison with the native enzyme, a signiﬁcant increase of 250% in the half-life of the polysialylated asparaginase was observed after intravenous administration in mice.144 The total IgG immune response of mice intravenously immunized with the polysialylated asparaginase derivative was lower than that corresponding to the native enzyme. Polysialylation also reduced the antigenicity of asparaginase, and the authors suggested that the improved pharmacokinetics behavior of this neoglycoconjugate is caused by its lower immunoreactivity.145 These results were conﬁrmed by Wang and co-workers, which reported similar improvements in the immunogenic, antigenic, pharmacokinetics, and stability properties for L-asparaginase after conjugation with colominic acid.362 The scavenger enzyme SOD, which catalyzes the disproportion of superoxide radical ion to molecular oxygen and hydrogen peroxide and thus is critical for protecting the cell against the toxic products of aerobic respiration,363 has been postulated as the ideal drug for the prevention and treatment of diseases associated with the oxidative stress.364,365 However, the clinical application of SOD has far been limited by its rapid clearance through the kidney and the occurrence of antigenic reactions when non human enzyme was employed.366 In addition, SOD is rapidly inactivated by its own reaction product H2O2, yielding more toxic oxidant species.367 It should be highlighted that although several formulation strategies were 4902

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

model of experimental silicosis showed that this neoglycoenzyme inhibited evolving ﬁbrosis in the lungs, whereas a not therapeutic eﬀect was observed for the native SOD. The synthesis and pharmacokinetics evaluation of SODcarboxymethyl dextran and SOD-diethylaminoethyl dextran conjugates was described by Fujita et al.309 The authors reported that SOD modiﬁed with diethylaminoethyl dextran rapidly disappeared from plasma and distributed into liver through a nonspeciﬁc electrostatic mechanism. On the contrary, the SOD−carboxymethyl dextran conjugate exhibited a long plasma half-life because of restricted glomerular ﬁltration and tissue interaction.311,374 To avoid the disadvantages associated with the cross-linking of therapeutic enzymes with polysaccharides, a low molecular weight dextran of 5 kDa was end-group functionalized with 1,6hexylenediamine and further attached to SOD via a carbodiimide-catalyzed reaction.248 The conjugate retained about 80% of the initial catalytic activity, and was 12-fold more resistant to inactivation by H2O2. The anti-inﬂammatory activity of this neoglycoenzyme was 2-fold higher, and its plasma half-life was prolonged from 4 min to 3.2 h in comparison with the native SOD. Chondroitin sulfate, a vascular wall glycosaminoglycan with important biomedical application,378 has been extensively evaluated by Maksimenko and co-workers as glycosylation agent for SOD.379 The synthetic strategy employed previous activation of the polysaccharide with benzoquinone and further cross-linking to the surface of the enzyme.209 This conjugate exerted a potent antithrombotic eﬀect in a rat model of arterial thrombosis induced by ferrous chloride, which was associated with the adsorption of the enzyme−polysaccharide adduct on the glycocalyx of the vascular wall cells.211 To prepare a more potent antithrombotic adduct, these authors developed an original approach based on the bienzymatic cross-linking of SOD and catalase with chondroitin sulfate. The bienzymatic neoglycoconjugate showed a noticeable antithrombotic activity, being eﬀective at doses 2 orders of magnitude smaller than those for native SOD and catalase, and an order of magnitude smaller than that for the chondroitin sulfate-modiﬁed enzymes, administered either individually or as a mixture.320,379,380 On the basis of the unique anti-inﬂammatory properties and ability to scavenge hydroxyl radicals,381 sodium hyaluronate was proposed for modifying SOD using EDC as coupling agent.120 The cross-linked conjugate retained 70% of the original superoxide anion-scavenging activity and exhibited a much higher therapeutic activity than unconjugated SOD in experimental models of inﬂammatory diseases. In artiﬁcially induced paw ischemia in mice and carrageenan-induced pleurisy in rats, half dose of modiﬁed SOD showed 2-fold and 6-fold more activity than the native enzyme, respectively. The polymer−SOD conjugate was also eﬀective in suppressing adjuvant arthritis in rats, a model of chronic rheumatoid arthritis in humans. SOD was chemically modiﬁed with CMC through two diﬀerent synthetic procedures: reductive alkylation with the periodate-oxidized polymer and formation of amide linkages via a carbodiimide-catalyzed reaction.108 The cross-linked adducts retained 68−78% of the initial catalytic activity and showed a signiﬁcant improvement in the pharmacokinetics behavior with plasma half-life prolonged to 7−35 h when compared to 5 min for the native SOD. The anti-inﬂammatory activity of the enzyme, tested in the carrageenan-induced paw edema in rats,

developed for improving the therapeutic potential of SOD, including conjugation with polyethylene glycol and other macromolecules,365,367−370 these approaches generally failed to protect SOD against H2O2 inactivation. To overcome these disadvantages, Hashida and co-workers modiﬁed SOD with galactose and mannose residues by reacting the enzyme with the corresponding 2-imino-2-methoxyethyl-1thioglycoside.309 The conjugates retained high catalytic activity and were stable to proteolytic degradation when incubated with mouse serum during 3 h. However, galactosylated and mannosylated enzyme derivatives were very rapidly eliminated from the circulation and taken up by parenchymal and nonparenchymal cells of the liver, respectively, via receptormediated endocytosis.191,192,371,372 This latter characteristic was employed to targeted delivery of the modiﬁed enzyme to the liver. Both neoglycoenzymes were eﬀective in the prevention of hepatic ischemia/reperfusion injury in rats.189 Although mannosylated SOD slightly decreased the intrahepatic production of ROS in rats with ﬁbrotic livers,192 its combination with succinylated catalase was successful to prevent the initial phase of hepatic ischemia/ reperfusion injury,190 as well as the late-phase injury mediated by inﬁltrating neutrophils.195 In addition, both SOD derivatives, particularly the galactosylated enzyme, showed protective eﬀects on cold ischemia/reperfusion injury during orthotropic liver transplantation in rats.373 The accumulation of SOD in kidney was drastically decreased by glycosylation with galactose and mannose, showing reduced tubular reabsorption and increased exposure of the luminal surface to the modiﬁed enzymes.309,374 However, a small but signiﬁcant dose-dependent uptake of mannosylated SOD from the capillary side was observed.311 This kidney uptake, probably associated with a mannose receptor-mediated endocytosis mechanism, promoted renal damage in rats when mannose-modiﬁed SOD was evaluated in an acute renal failure model induced by ischemia/reperfusion.310 Hashida and co-workers described the targeted delivery of mannosylated SOD to inﬂammatory macrophages via mannose receptor-mediated endocytosis, as well as its high inhibitory eﬀect on the superoxide radical ion released by these cells.375 These authors also reported that this neoglycoconjugate did not protect cultured rabbit alveolar cells against ROS-mediated injury due to paraquat poisoning.193 Noticeable improvements in the pharmacological, immunological, and pharmacokinetics properties of SOD were also achieved by glycosylation with polysaccharides. McCord and co-workers early reported the use of Ficoll, a polysucrose-based polymer, for modifying SOD.376 The polysaccharide, previously activated with CNBr, was cross-linked to the enzyme yielding a conjugate that retained 50% of the initial activity. Pharmacokinetics studies revealed that the circulating half-lives of native and Ficoll-modiﬁed SOD preparations were 6 min and 24 h, respectively. Additionally, the intravenous administration of Ficoll-SOD to animals with induced inﬂammation suppressed the inﬂammatory response and inhibited leukocyte inﬁltration into the challenged site. Miyata et al. cross-linked the enzyme with CNBr-activated dextran obtaining a conjugate that retained about 70% of the dismutase activity and was less antigenic but more immunogenic that the native counterpart.377 The authors also observed longer half-life and greater anti-inﬂammatory activities in rats for this conjugate. A similar increase in half-life was reported for SOD cross-linked with periodate-oxidized dextran.308 A rat 4903

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

superoxide anion from macrophages and also reduced the in vitro expression of inﬂammatory cytokines by irradiated 3T3 ﬁbroblasts, lowering the levels of transforming growth factor-β1 and interleukin-β1.317,323 Further studies supported the broad therapeutic potential of this heparin−SOD conjugate demonstrating successful preventive eﬀects on CCl4-induced acute liver failure and hepatic ﬁbrosis in mice.326 Moreover, it was also shown its neuroprotective and anti-inﬂammatory eﬀects by inhibiting upregulation of intercellular adhesion molecule-1 and preventing neuronal cell apoptosis in the brain reperfusion injury after ischemia model in gerbils.327 A recent patent proposes the use of hydroxyethyl starch, previously oxidized with iodide or Cu2+ ions, as a water-soluble cross-linking agent for proteins.382 The authors claim its use for modifying therapeutic enzymes, providing methods for the glycosylation of asparaginase and SOD. Although less extensively used than SOD, catalase is another antioxidant enzyme that has been artiﬁcially glycosylated to improve its therapeutic potential. Modiﬁcation of the enzyme with galactose and mannose was performed by reacting catalase with the corresponding 2-imino-2-methoxyethyl-1-thioglycoside, yielding derivatives with 90% of the initial activity.190 Biodistribution studies showed that galactosylated catalase accumulated selectively in the liver parenchymal cells, whereas the mannosylated derivative was mainly delivered to liver nonparenchymal cells. Pharmacokinetics analysis revealed that the hepatic uptake clearance of both glycosylated derivatives was much greater than that of native catalase. Mannosylated catalase showed a noticeable protective eﬀect in the hepatic ischemia/reperfusion injury model in mice, whereas that of the galactosylated derivative was small. Glycosidation with these monosaccharides did not improve the protective eﬀect of catalase in experimental pulmonary metastasis in mice,383 but the galactosylated enzyme showed a great inhibitory eﬀect on hepatic metastasis of colon carcinoma cells in mice.296,297 Polysialylated catalase, prepared by reductive alkylation of the enzyme with the periodate-oxidized derivative of colominic acid, retained 70% of the initial activity and showed improved resistance to proteolytic degradation.143,318 A similar stability against inactivation by proteases was conferred to catalase by modiﬁcation with mannan, CMC, and carboxymethylchitin.116 These neoglycoenzymes were 1.9−5.7-fold more stable against thermal inactivation at 55 °C in comparison with the native counterpart. Pharmacokinetics studies revealed lower total body clearance for all the enzyme−polymer preparations, but better results were achieved for the carboxymethylchitin− catalase conjugate with 12-fold increase in plasma half-life. Benzoquinone-activated chondroitin sulfate was used as a cross-linking agent for catalase,210 yielding a glycoconjugate that showed a markedly high antithrombotic activity in a rat model of arterial injury induced by ferrous chloride.212 It was further demonstrated that this improved antithrombotic activity was caused by the enhanced accumulation of the conjugate on the surface of the vascular wall cells due to the presence of the attached chondroitin sulfate moieties.320 Higher antithrombotic activity was achieved by the preparation of a bienzymatic neoglycoconjugate of catalase−chondroitin sulfate−SOD, as described above. An original supramolecular-based strategy was used to prepare a bienzymatic conjugate of catalase and SOD, by assembling adamantane-modiﬁed SOD with catalase previously cross-linked with CD-branched CMC.102 The anti-inﬂamma-

was 2−2.4 times increased after conjugation with the polysaccharide. Conjugation of SOD with CMC conferred noticeable resistance for the enzyme against inactivation with H2O2. These conjugates also showed low cytotoxicity in the in vitro experiments according to the results obtained for the cell proliferation inhibition index and cell morphology of human ﬁbroblasts treated with the SOD derivatives.107 A novel CMC derivative, synthesized by attaching mono-6hexylenediamino-6-deoxy-β-CD residues to the periodateoxidized polymer, was described as glycosylation agent for SOD.103 The cross-linked conjugate, prepared via carbodiimide coupling reaction, contained about 1.4 mol of polymer per mol of protein and retained 87% of the initial catalytic activity. The authors reported a signiﬁcant improvement for the pharmacokinetics and anti-inﬂammatory properties of the enzyme by conjugation with this CD-branched polysaccharide. The plasma half-life was prolonged from 5 min to 7.2 h, and the inhibition of paw edema induced in rats by carrageenan was 2.2 times increased. The same group reported the use of another anionic polysaccharide, O-carboxymethylchitin, for modifying SOD.115 The catalytic activity of the enzyme was signiﬁcantly aﬀected by attaching this polysaccharide to the free amino groups of the protein, only retaining 57% of the initial dismutase activity after glycosylation. However, the pharmacokinetics properties were by far the most largely improved for SOD-based neoglycoenzymes with an increase in the plasma half-life from 5 min to 69 h. The enzyme was also remarkably more resistant to inactivation with H2O2, and its anti-inﬂammatory activity was 2.4 times increased after conjugation with the polysaccharide. Another chitin-derived polysaccharide, the polycationic N,N,N-trimethyl chitosan chloride, was recently employed by Liu and co-workers to modify SOD.317 The authors reported an excellent inhibitory eﬀect of this neoglycoenzyme on superoxide anion release from macrophages, which surpassed those of native and heparin-modiﬁed SOD preparations. The eﬀects of this adduct on inﬂammatory cytokine expression in vitro were also evaluated, showing that the conjugate signiﬁcantly lowered the levels of transforming growth factor-β1 and interleukin- β1 expressed by irradiated 3T3 ﬁbroblasts. These results support the potential use of this neoglycoenzyme in the prevention and treatment of ROS-mediated inﬂammatory diseases. Yeast mannan, a highly branched mannose polymer, was activated by periodate oxidation and further cross-linked to SOD via reductive alkylation with sodium borohydride.163 The glycosylated enzyme retained 52% of the initial activity but was 560-fold more resistant to inactivation with H2O2. The antiinﬂammatory activity of SOD was 2-fold increased, and its plasma half-life time was prolonged from about 5 min to 1.7 h after artiﬁcial glycosylation. Modiﬁcation with mannan also conferred lectin-recognition ability with respect to concanavalin A. More recently, promissory results have been achieved by using low molecular weight heparin as modiﬁcation agent for SOD. The conjugate prepared by Zhang and co-workers showed lower immunogenicity, higher anti-inﬂammatory activity, and higher stability toward acid, alkali, heat, and trypsin treatment than native SOD.176 In addition, heparin− SOD conjugate inhibited bleomycin-induced pulmonary ﬁbrotic lesions in vivo, as reﬂected by the decrease of lung hydroxyproline content and lung ﬁbrosis grade.323 This neoglycoenzyme showed inhibitory eﬀect on the release of 4904

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Uricase, a peroxisomal liver enzyme existing in animals but not in humans, has been proposed for the therapy of gout and uric acid nephropathy resulting from hyperuricemia.390,391 Early reports concerning uricase-based glycoconjugates described the use of CNBr-activated dextran, sucrose, and polymerized sucrose as modiﬁcation agents.392 The authors claimed high retention of catalytic activity for these conjugates as well as noticeable increase in solubility and stability, particularly for conjugates prepared with polysaccharides. Although a possible therapeutic application could be expected for these neoglycoenzymes, experimental data supporting this fact were not provided. Yasuda and co-workers performed a complete study for dextran−uricase conjugates using four diﬀerent polymer activation and coupling methods: periodate oxidation, cyanogen bromide, carbodiimide, and cyanuric chloride.307 All conjugates retained high level of catalytic activity, especially those prepared with periodate-oxidized polysaccharide. The immunogenicity and antigenicity of this neoglycoenzyme were 4-fold and 2-fold lower than those corresponding to native uricase, respectively. The pharmacokinetics of the enzyme was also improved by glycosylation, with an about 11-fold increase in the half-life in rats. Furthermore, it was demonstrated that conjugation of uricase with periodate-oxidized diethylaminoethyl dextran resulted in an extremely short plasma half-life, while conjugation with carboxymethyl dextran gave a longer one than that of the native and dextran-modiﬁed enzyme.393 CNBr-activated dextran was also employed by Sherwood and co-workers to modify carboxypeptidase G2 (EC 3.4.17.11),71,394 an enzyme with potential application in cancer therapy.395 Artiﬁcial glycosylation resulted in a 5-fold and 2.5fold increase in plasma persistence of the enzyme in healthy and tumor bearing mice, respectively. The conjugate was also more resistant to proteolysis by trypsin and α-chymotrypsin. Native and dextran-modiﬁed enzyme showed diﬀerent patterns of clearance and excretion rates in mouse.396 Tissue distribution studies demonstrated that there was little or no tissue uptake of native enzyme, whereas dextran-based conjugate was retained in the liver. Within the liver, the modiﬁed protein was rapidly degraded, demonstrating the involvement of the reticuleendothelial system in the clearance of these adduct. These results support the potential use of dextran−carboxypeptidase G2 conjugates in hepatic cancer chemotherapy. Plasmin (EC 3.4.21.7), a protease involved in the coagulation pathway, was modiﬁed with oxidized dextran.397 The conjugate, retaining 85% of the original activity, showed improved pharmacokinetics and immunologic characteristics in comparison with the native enzyme. Another proteolytic enzyme, papain, was also cross-linked with aldehyde dextran in the presence of cysteine.398 This approach allows the preservation of a high catalytic activity in the complex due to the formation of a protective microenvironment from cysteine sulfhydryl groups around the active site of the modiﬁed enzyme. Glycosidation with oxidized dextran-cysteine conferred an extended optimum pH of catalytic activity to the enzyme and increased stability toward the action of extreme pH, proteolysis, and temperature. The therapeutic eﬃciency of the modiﬁed preparation in ophthalmologic lesions models was also demonstrated by the authors. Blomhoﬀ and co-workers reported the use of dextran for the modiﬁcation of β-galactosidase,399−401 a glycosidase with potential application in the treatment of β-galactosidosis and other lysosomal diseases.402 The conjugate retained 90% of the

tory activity of this complex was 4.5-fold higher than that corresponding to native SOD. In addition, supramolecular association with the polysaccharide−catalase conjugate completely preserves SOD against inactivation by H2O2. Low molecular weight dextrans, previously functionalized with a carboxylate or amino group at its reducing end residue, were chemically attached to catalase via a carbodiimide coupling reaction.84,121 Signiﬁcantly, the catalytic activity of catalase increased after modiﬁcation with the monoaminated dextran, while the conjugate with the monocarboxylated polysaccharide retained 77% of the original activity. The plasma half-life of the enzyme was 7.8-fold and 7.3-fold increased after glycosylation with dextrans having the amino and carboxylate group, respectively. An enzymatic approach, based on a transglutaminasecatalyzed coupling reaction, was used by Valdivia and coworkers to modify catalase with the end-group aminated dextran derivative.79 Similarly to the neoglycoenzyme prepared by chemical methods, catalase showed increased catalytic activity after enzymatic glycosylation. The catalase−dextran adduct was 3.8-fold more stable to thermal inactivation at 55 °C and 2-fold more resistant to proteolytic degradation by trypsin. This enzymatic glycosylation also improved the pharmacokinetics behavior of catalase increasing 2.5-fold its plasma half-life in rats. Mammalian hyaluronidase, an endoglycosidase that can increase tissue permeability by depolymerisation of hyaluronic acid, is currently used in clinical applications for the treatment of joint diseases, as well as for dermatological and ophthalmological disorders. Hyaluronidase is a potential drug for decreasing myocardial infarction size, but this therapeutic application is currently limited by its inhibition with heparin.331,384,385 Maksimenko and co-workers intended to overcome this problem by cross-linking the enzyme with polysaccharide derivatives such as oxidized dextran.331 The neoglycoconjugate retained more than 90% of the glycosidase activity, showed noticeable thermostability, and was signiﬁcantly biodistributed to the lung after intravenous administration in mouse. Hyaluronidase−dextran conjugate showed a pronounced inhibitory action on the progress of experimental silicosis in rats, and this eﬀect was signiﬁcantly increased by using a combined application with SOD-modiﬁed with oxidized dextran.386,387 Glycosidation with dextran increased the resistance of hyaluronidase to denaturing agents and extended the optimal pH and ionic strength conditions for maximal endoglycosidase activity. The conjugation also reduced electrostatic eﬀects on the active site of hyaluronidase and eﬃcacy of heparin inhibition, being almost insensitive to heparin at pH 7.5.304 It was further conﬁrmed that this reduction on heparin inhibition directly correlated with the degree of modiﬁcation of the enzyme.149,388 These authors also used artiﬁcial glycosylation with several benzoquinone-activated polysaccharides including dextran sulfate, dextran, and chondroitin sulfate to simulate the glycosaminoglycan microenvironment of hyaluronidase and determine the inﬂuence on the regulation of its endoglycosidase activity and the protective eﬀect against heparin inhibition and protein glycation with mono and disaccharides.213,389 It was demonstrated that chondroitin sulfate increased considerably the resistance of the enzyme to glycation and heparin inhibition due to the higher rigidity conferred to the hyaluronidase structure. 4905

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Gemeiner and co-workers proﬁted the lectin-biorecognition properties of several neoglycoenzymes to perform their aﬃnitybased immobilization on lectin-modiﬁed supports.152 They reported the attachment of yeast mannan to penicillin G acylase and its further biospeciﬁc immobilization on Concanavalin Acellulose beads. The immobilized biocatalyst showed high storage and operational stability, similar to that obtained for the covalently immobilized enzyme. The same approach was also employed by Pérez and Villalonga to immobilize mannanmodiﬁed α-amylase on Concanavalin A-Sepharose support.155 Enzymes modiﬁed with ionic polysaccharides have been immobilized on solid supports, previously coated with opposite-charged polysaccharides, via polyelectrolyte complex formation. A chitosan−invertase conjugate was immobilized on chitin matrix either coated with sodium alginate, hyaluronic acid, pectin, or CMC.404−407 The immobilized enzymes were stable against incubation in high-ionic-strength solutions and were 4−12.6-fold more resistant to thermal treatment at 65 °C than the native counterpart. The immobilized biocatalysts retained about 96−100% of the initial activity after 10 cycles of use and were noticeably stable under storage at 4 °C and continuous operational conditions in packed bead reactors. Lysozyme is a hydrolytic enzyme that causes lysis of bacterial cells, and thus has been proposed as natural antimicrobial or preservative agent in food and pharmaceutical industry. However, the suitability of lysozyme as enzyme-based preservative has been questioned due to its limited lytic spectrum only to Gram positive bacteria and low stability.332 Such limitations can be overcome by covalent modiﬁcation with carbohydrate derivatives,177−180,332 allowing the use of lysozyme-based neoglycoenzymes in food industry. Lysozyme−dextran conjugate, prepared by Maillard reaction, revealed signiﬁcant antimicrobial activity for both Gramnegative and Gram-positive bacteria.178 In addition, higher thermal and pH stability as well as improved emulsifying ability were described for lysozyme-modiﬁed with a variety of carbohydrates, such as mannan, galactomannan, dextran, and glucose stearic acid monoester.177−180,332,345 Enzymatic catalysis is increasingly recognized as an attractive tool for the synthesis of a wide variety of organic compounds with high stereo- and regiospeciﬁcity.408 Modiﬁcation with carbohydrates can provide organic solvent tolerance, catalytic activity, and stability to enzymes, which allows neoglycoenzymes to be proposed as potential biocatalysts for organic synthesis. Several reports have described the use of proteasebased neoglycoenzymes as catalysts for the synthesis of peptides. Fernández et al. reported the use of trypsin−CD conjugates for the preparation of benzoyl-argininephenylalaninamide dipeptides in high ionic strength aqueous media.112 Higher conversion rates up to 70% were observed for the neoglycoenzyme-catalyzed reaction in comparison with the native trypsin. David and co-workers used a combined sitedirected mutagenesis and chemical glycosylation strategy to create chemically modiﬁed mutants of B. lentus subtilisin with broadened substrate speciﬁcity.222,223 These neoglycoenzymes were capable of catalyzing the coupling reactions of not only Lamino acid esters but also D-amino acid esters to give the corresponding dipeptides with good yields. De la Casa et al. described the modiﬁcation of C. rugosa lipase with dextrans of diﬀerent molecular weight, ranging from 6 to 50 kDa, and their use in the enantioselective esteriﬁcation of (R,S)-ibuprofen with n-propanol.333 The glycosylated biocatalysts were more active than the native enzyme in the

initial activity and was 35% and 43% more stable to degradation by isolated hepatocytes and nonparenchymal liver cells, respectively, than the native enzyme. Similar results were achieved by using methylated and acetylated dextrans as glycosylation agents. However, there was not a signiﬁcant diﬀerence between the uptake of native and conjugated enzyme preparations into the liver cells. Upon intravenous infusion into rats, native and conjugated enzymes were cleared from plasma with only a slight diﬀerence in the clearance rate. Modiﬁcation with dextran also conferred noticeable stability to the enzyme toward thermal inactivation and proteolytic degradation with subtilopeptidase A. 5.2. Neoglycoenzymes for Industrial Application

The eﬀective and speciﬁc catalytic properties of enzymes have promoted their wide use in a variety of industrial applications, including detergent, food, textile, fuel alcohol, pulp and paper, animal feed, and personal care manufactorings.25 Such applications are currently dominated by hydrolytic enzymes, mainly proteases, glycosidases, and lipases. In general, highly stable enzymes are required for industrial applications. This requirement opens a promissory place for neoglycoenzymes as potential biocatalysts in industrial processes. However, to the best of our knowledge, the industrial use of carbohydratemodiﬁed enzymes has not been yet accomplished. Excellent candidates for industrial application are neoglycoenzymes prepared by modiﬁcation of hydrolytic enzymes with oligo- and polysaccharides due to the improved thermal stabilization and resistance to extreme values of pH and surfactants. Stable neoglycoenzymes have been prepared from α-amylase,93,101,157,160,184 invertase,259,266,267 lipase,159 cellulase,122,260 chitinase,104 levansucrase,169 and several proteases.37,74,94,95,100,112,113,118,119,127,173,244,249,275 These neoglycoconjugates can be envisaged as robust candidates for such industrial purpose. Glycoconjugates obtained by modiﬁcation of other enzymes of industrial interest such as penicillin G acylase, phenylalanine dehydrogenase, amine oxidase, and laccase could be also useful for industrial applications.111,150,153,158,171,172,241,245,252 Covalent glycosylation often yields enzyme derivatives able to work in homogeneous aqueous solutions under extreme physicochemical conditions, but there are several technological disadvantages when comparing with enzymes immobilized on solid supports. Immobilization of enzymes allows for their multiple reuses, controlled product formation, easy separation from the reaction media, rapid termination of reactions, continuous operation of enzymatic processes, and a wide variety of engineering designs.403 Therefore, a rational approach to prepare highly stable immobilized enzyme forms could be the previous modiﬁcation with carbohydrates and further immobilization on a solid support. Additionally, the attached sugar moieties can be employed as target points for the covalent or oriented immobilization of the neoglycoenzyme. This approach was used by Vaňková et al., who described the initial glycosylation of trypsin with lactose and melibiose, via reductive alkylation with NaBH3CN, and their further oxidation by periodate treatment.133 Trypsin forms with oxidized saccharide moieties were bound to two types of solid support: polyacrylamide gel containing covalently bound amino groups and cellulose beads modiﬁed with adipic acid dihydrazide. The covalently immobilized enzymes retained full catalytic activity after freeze-drying and reswelling in the appropriate buﬀer solutions. 4906

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 42. Lectin-based neoglycoenzyme-linked competitive binding assay for carbohydrate detection.

Figure 43. Electrode with supramolecular architectures based on CD-modiﬁed redox enzymes.

principle was employed by Gabius and co-workers to develop valuable analytical tools for lectin detection in solid-phase assays and glycohistochemistry by using a variety of artiﬁcially glycosylated β-galactosidase.337 These neoglycoenzymes were also evaluated in glycohistochemical and glycocytological studies for the detection of tumor-associated expression of carbohydrate-binding proteins in several malignant cells and tissues.338−342 Meyerhof and co-workers employed also the lectin binding ability of monosaccharide and oligosaccharide modiﬁed enzymes to design a homogeneous lectin-based neoglycoen-

esteriﬁcation reaction, giving a 50% total yield with about 97% of conversion of the (S)-enantiomer. The (S)-enantiopreference of the C. rugosa lipase did not change by the molecular weight of dextran used in the preparation of the glycoconjugates. 5.3. Neoglycoenzymes in Chemical and Clinical Analysis

Neoglycoenzymes have been useful tools for the development of histological and analytical methods based on lectin− carbohydrates interaction. In such methods, the sugar moieties in the glycoconjugate acts as biorecognition element for lectins, and the enzyme part acts as catalytic signaling element. This 4907

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Figure 44. Sensing-actuating nanomachine for neoglycoenzyme-controlled cargo delivery.

with adamantyl-11-pyrrolyl-1-undecyl carboxylic acid amide, previously electropolymerized on platinum and single-walled carbon nanotubes-coated platinum electrodes (Figure 43C,D).246 Highly sensitive electrochemical biosensors for glucose quantiﬁcation were prepared with these neoglycoenzyme-modiﬁed electrodes. Original approaches have been reported by Cosnier and coworkers for the immobilization of β-CD tagged enzymes on carbon nanotubes through double supramolecular junctions for biosensing purposes. These strategies were based on the use of either pyrenebutyric acid adamantyl amide or tris(bispyrenebipyridine)iron(II) as supramolecular bridges for the initial modiﬁcation of carbon nanotubes via π-stacking and further biofunctionalization with the neoglycoenzyme through host− guest interactions (Figure 43E,F).300−303 A sensitive amperometric biosensor for xanthine was recently prepared by the supramolecular immobilization of β-CDmodiﬁed xanthine oxidase on glassy carbon electrodes coated with an hybrid nanomaterial composed of single-walled carbon nanotubes and an electropolymerized network of adamantanefunctionalized gold nanoparticles (Figure 43G).301 This neoglycoenzyme-based biosensor was able to detect xanthine at nanomolar concentration and showed high reproducibility and storage stability. The structural and functional stabilization conferred to enzymes by artiﬁcial glycosylation favored the use of proteasebased neoglycoenzymes in proteomic studies. Sebela et al. demonstrated that modiﬁcation of bovine trypsin with oligosaccharides maltotriose, raﬃnose, and stachyose increased its thermostability and suppressed autolysis.132 This stable enzyme derivative was successfully used for the in-gel and gelfree digestion of cytochrome C, myoglobin, aldolase (EC 4.1.2.13), and bovine serum albumin for proteomic applications.

zyme-linked competitive binding assay to detect monosaccharides, oligosaccharides, and glycoproteins in solution.205−207 The rationale of this method was based on the reversible inhibition of artiﬁcially glycated malate dehydrogenase and/or glucose 6phosphate dehydrogenase by lectins that recognize the attached saccharide moieties on the protein surface. In the presence of selected analytes (saccharides or glycoproteins), the activity of the neoglycoenzymes increased proportionally to the concentration of present carbohydrate. This analytical method was used to quantify human IgA at nanomolar level and was successful to detect glycoprotein possessing sialic acid terminal groups.208 The principle of the analytical approach is illustrated in Figure 42. These authors also reported a variant of this analytical approach involving the use of three diﬀerent lectin/enzyme− saccharide systems in a “ﬁngerprint”-type assay scheme, which was employed to estimate the type and relative amount of certain carbohydrate structures within given glycoproteins.206 Neoglycoenzymes have been also valuable tools for analytical bioelectrochemistry. In particular, CD-modiﬁed redox enzymes have been employed for the construction of electrochemical biosensors due to the enhanced functional stability of the neoglycoconjugates and the ability of CDs to form inclusion complexes with hydrophobic compounds, allowing the design of three-dimensional arrangements with supramolecular architecture at the electrode surfaces.298 Villalonga and co-workers reported that modiﬁcation of xathine oxidase with β-CD-branched CMC allowed its further supramolecular immobilization on a gold electrode coated with a monolayer of 1-adamantanyl residues (Figure 43A). This enzyme electrode was used to construct an amperometric biosensor for xanthine that retained more than 90% of its initial activity after 3 weeks of storage.109 A layer-by-layer supramolecular approach was employed by the same group for the construction of an amperometric enzyme biosensor toward H2O2. The rationale of the arrangement was the host−guest immobilization of alternating layers of horseradish peroxidase, modiﬁed with either 1-adamantane or β-CD-branched CMC residues, on Au electrodes coated with polythiolated β-CD polymer (Figure 43B). It was demonstrated that the analytical response of the bioelectrodes increased when the number of enzyme layers was also increased.114 Glycosylation of glucose oxidase with mono-6-deoxy-6amino-β-CD via a carbodiimide-catalyzed reaction yielded a neoglycoenzyme able to form host−guest inclusion complexes

5.4. Neoglycoenzymes in Nanotechnology

Nanotechnology is an emerging multidisciplinary ﬁeld devoted to the preparation and use of nanosized materials that is advancing many research and technological areas. One of the main goals in nanotechnology is the development of an eﬃcient method for the controlled functionalization of nanomaterials and nanostructured surfaces with active biomacromolecules to prepare biologically inspired nanomaterials and smart nanomachines. In this context, neoglycoenzymes can be envisaged as new tools for the modiﬁcation of nanosized materials oﬀering important advantages as a consequence of the new and/or 4908

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

modiﬁcation agents. Original site-selective glycosylation methods have been also developed for the rational design and synthesis of artiﬁcial glycoenzymes with deﬁned structure and composition. These late approaches have been reinforced by the site-speciﬁc introduction of unnatural amino acids in the enzyme structure through genetic manipulation, and the use of orthogonal coupling chemistry to prepare the artiﬁcial glycoenzymes. Nowadays, new challenges are open for neoglycoenzyme research. Artiﬁcial glycosylation has been postulated as a more biocompatible alternative to PEGylation technology, addressed to prepare enzyme drug formulations with improved pharmacokinetics and pharmacological properties. Future eﬀorts should lead to the establishment of methods for the site-selective glycosylation with large and complex oligosaccharide chains, similar to those found in naturally occurring glycoproteins, to prepare novel and eﬀective neoglycoenzymebased drugs. Controlled site-selective attachment of two or more diﬀerent glycans to the same enzyme should be also an active line for future research,285 allowing the biospeciﬁc recognition of the neoglycoenzyme by diﬀerent lectins and organs. In addition, the tailor-made synthesis of novel carbohydrate derivatives should also contribute to the preparation of neoglycoenzymes with predeﬁned structural and functional properties for therapeutic uses. The fact that artiﬁcial glycosylation can stabilize enzymes speaks about the prospects of neoglycoenzymes in the establishment of new and more eﬃcient industrial and biotechnological processes, based on enzymes able to resist harsh operating conditions. Stable neoglycoenzymes should also contribute to the design of point-of-care biosensor systems for improving patient care through real-time and remote health monitoring. In addition to their catalytic properties, neoglycoenzymes can be designed to interact with other chemical or biological entities through aﬃnity, supramolecular, or electrostatic interactions. This dual capacity can be employed to prepare catalytically active surfaces and nanomaterials with threedimensional architectures for biosensor construction. The assembly of novel nanodevices with lectin-biorecognition properties and neoglycoenzyme-controlled smart delivery of drugs to target aﬀected cells also can be envisioned.

improved structural and functional properties of these enzyme derivatives. As it was commented in section 5.3, several CD-based neoglycoenzymes have been supramolecularly immobilized on diﬀerent nanomaterials, such as carbon nanotubes and electropolymerized gold nanoparticles to construct electrochemical biosensors. Additionally, a bienzymatic supramolecular nanoassembly was prepared on β-CD-capped gold nanoparticles by coimmobilization of adamantane-modiﬁed catalase and a β-CD-SOD neoglycoenzyme.35 This bienzymatic nanocatalyst showed improved thermal stability for catalase and a 90-fold increased resistance for SOD toward inactivation by H2O2. Pečová et al. have described the covalent immobilization of α-, β-, and γ-CD modiﬁed trypsin derivatives on core−shell Fe3O4 magnetic nanoparticles coated with chitosan polymer.409 The immobilized trypsin glycoconjugates showed increased resistance to thermal treatment, high stability toward autolysis, unchanged optimum pH, and a signiﬁcant storage stability and reusability. These neoglycoenzyme-based nanocatalysts were successfully evaluated for the digestion of several model proteins. Recently, Aznar et al. prepared a new gated nanodevice design able to control cargo delivery using glucose as trigger and CD-modiﬁed glucose oxidase as capping agent (Figure 44).410 In this work, mesoporous silica nanoparticles were capped with an active β-CD modiﬁed glucose oxidase derivative through the formation of inclusion complexes between the CDs and the propylbenzymidazole group anchored to the nanosized support. In the presence of glucose, the neoglycoenzyme catalyzes its conversion to gluconic acid, which induces the protonation of benzimidazole, resulting in a dethreating of the inclusion complex and delivery of the cargo compound. This neoglycoenzymes-based approach may allow inspiring the design of novel high-selective on-command delivery smart nanosystems for diﬀerent applications based on enzymecontrolled processes.

6. CONCLUSIONS AND OUTLOOK Covalent attachment of carbohydrates to enzymes constitutes a well-established, cost-eﬀective, and simple method to modulate the structural and catalytic properties of these biocatalysts. This strategy is a valuable tool for increasing stability to enzymes using several denaturing/inactivating conditions for proteins, such as thermal treatment, incubation at extreme values of pH, and the presence of surfactant, chaotropic agents, or organic solvents. Artiﬁcial glycosylation also confers new characteristics to enzymes such as lectin-biorecognition capacity, reduced antigenic and immunogenic properties, enhanced lifetime in biological ﬂuids, and target organ delivery. These novel and/or improved characteristics allow the application of neoglycoenzymes in several ﬁelds, including clinical and chemical analysis, organic synthesis, industrial and biotechnological productions, and the preparation of biofunctionalized nanomaterials. In this context, the evaluation of artiﬁcial glycoenzymes as drugs for enzyme-replacement therapy has received special attention, and some neoglycoenzyme formulations are currently in advanced clinical study or even in the market. For more than 50 years, a great variety of chemical, enzymatic, and chemoenzymatic approaches have been described for the successful synthesis of neoglycoenzymes using mono-, oligo-, and polysaccharide derivatives as

AUTHOR INFORMATION Corresponding Author

*Phone: +34 91 3944315. Fax: +34 91 3944329. E-mail: [email protected]. Notes

The authors declare no competing ﬁnancial interest. 4909

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

Biographies

Alfredo Sánchez was born in 1983 in Madrid, Spain. He received his B.S. degree in Chemical Engineering from Rey Juan Carlos University in 2006, and obtained his Ph.D. degree from the same university in 2011. He is currently working at the Complutense University of Madrid under the supervision of Prof. Reynaldo Villalonga and Prof. José M. Pingarrón in the preparation of enzyme biosensors and enzyme-based smart nanomaterials.

Mariá de Lourdes Villalonga was born in 1971 in Matanzas, Cuba, and received her B.S. degree in Chemistry from the University of Havana in 1994. Currently, she works at the University of Matanzas and joined the new research group of her brother Prof. Reynaldo Villalonga in 1996 for working in neoglycoenzymes. She obtained her M.Sc. degree in Protein Biochemistry in 2001 and her Ph.D. degree in Biology in 2008 from the University of Havana under the direction of Prof. Reynaldo Villalonga for work toward chemical and enzymatic synthesis of neoglycoenzymes. She is currently Full Professor at the University of Matanzas. Her research is focused on neoglycoenzymes, application of CDs and polymers in enzyme technology, enzyme biosensors, and nanotechnology. She has been awarded three prizes from the Cuban Academy of Sciences and the Prize of the Cuban Ministry of Science in 2007. In 2011, she received the TWAS-CONACYT Postdoctoral Fellowship in Biological Science. She is a member of the Latin

Mariá Gamella was born in 1980 in Madrid, Spain, and received her B.S. degree in Chemistry from the Complutense University of Madrid in 2003, where she obtained her Ph.D. degree in 2010. She is currently working under the supervision of Prof. José Manuel Pingarrón in the desing of biosensors. She is also cofounder of the “spin-oﬀ” company Inbea Biosensores S.L.

American Network for Enzyme Technology (RELATENZ) and the Cuban Chemical Society.

́ was born in 1986 in León, Spain, and received her B.S. Paula Diez degree in Chemistry from the Complutense University of Madrid in 2011. She obtained her M.Sc. degree in Analytical Chemistry in 2012

José M. Pingarrón obtained his Ph.D. (1981) from the Complutense University of Madrid. Between 1982 and 1983, he did postdoctoral training at the École Nationale Supérieure de Chimie de Paris. Since 1994, he is a full Professor of Analytical Chemistry at the Complutense University of Madrid. He headed the Department of Analytical Chemistry at the Faculty of Chemistry between 1998 and 2006, and he

from the Complutense University of Madrid, where she is currently pursuing his Ph.D. degree under the guidance of Prof. Reynaldo Villalonga and Prof. José M. Pingarrón. Her research is focused on enzyme biosensors and nanomaterials. 4910

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

from the Spanish Ministry of Science and Innovation (CTQ2011-24355 and CTQ2012-34238), and Comunidad de Madrid S2009/PPQ-1642, Programme AVANSENS, is gratefully acknowledged. R.V. and M.L.V. acknowledge long-term support from the University of Matanzas for their research on neoglycoenzymes.

was the President of the Spanish Society of Analytical Chemistry between 1998 and 2001. He is the recipient of the Faculty of Chemistry Medal, the Complutense University of Madrid Medal, and the research award on Analytical Chemistry of the Spanish Royal Society of Chemistry. His research interests focus on analytical electrochemistry, nanostructured electrochemical interfaces, and electrochemical and piezoelectric sensors and biosensors. He has over 270 publications including peer-reviewed papers, book chapters, and text books. He is currently Vice-President of the Spanish Royal Society of Chemistry and is its representative in the Division of Analytical Chemistry of the European Association for Chemical and Molecular Sciences. He is Associate Editor of the Electroanalysis journal and belongs to the Editorial Advisory Boards of the Journal of Electroanalytical Chemistry, Talanta, Analyst, Chemical Sensors, and ChemElectroChem and is a Member of the Analytical Chemistry Division Committee of IUPAC. Moreover, he is cofounder of the “spin-oﬀ” company Inbea Biosensores S.L.

ABBREVIATIONS CD cyclodextrin CMC carboxymethylcellulose EDC N-ethyl-N′-(3-dimethylaminopropyl) carbodiimide ROS reactive oxygen species SOD Cu,Zn-superoxide dismutase REFERENCES (1) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.; Moore, J. C.; Robins, K. Nature 2012, 485, 185. (2) Otten, L. G.; Hollmann, F.; Arends, I. W. Trends Biotechnol. 2010, 28, 46. (3) Walsh, C. T.; Garneau-Tsodikova, S.; Gatto, G. J. Angew. Chem., Int. Ed. 2005, 44, 7342. (4) Maekawa, K.; Liener, I. E. Arch. Biochem. Biophys. 1960, 91, 108. (5) Davis, B. G. J. Chem. Soc., Perkin Trans. 1999, 1, 3215. (6) Bennett, C. S.; Wong, C. H. Chem. Soc. Rev. 2007, 36, 1227. (7) Nicotra, F.; Cipolla, L.; Peri, F.; La Ferla, B.; Redaelli, C. Adv. Carbohydr. Chem. Biochem. 2007, 61, 353. (8) Lee, Y. C.; Lee, R. T. Neoglycoconjugates: Preparation and Applications; Academic Press: San Diego, CA, 1994. (9) Davis, B. G. Chem. Rev. 2002, 102, 579. (10) Stowell, C. P.; Lee, Y. C. Adv. Carbohydr. Chem. Biochem. 1980, 37, 225. (11) Zhang, J.; Zhang, Q. S.; Tian, G. Y. Chin. J. Chem. 2003, 23, 425. (12) Wong, S. Y. C. Curr. Opin. Struct. Biol. 1995, 5, 599. (13) Monsigny, M.; Roche, A. C.; Duverger, E.; Srinivas, O. Neoglycoproteins. In Comprehensive Glycoscience; Kamerling, J. P., Ed.; Elsevier: Oxford, 2007; pp 477−521. (14) Doores, K. J.; Gamblin, D. P.; Davis, B. G. Chem.-Eur. J. 2006, 12, 656. (15) Gamblin, D. P.; Scanlan, E. M.; Davis, B. G. Chem. Rev. 2009, 109, 131. (16) Khmelnitsky, Y. L. J. Mol. Catal. B: Enzym. 2004, 31, 73. (17) Koeller, K. M.; Wong, C. H. Chem. Rev. 2000, 100, 4465. (18) Mäntsälä, P.; Niemi, J. Enzymes: The Biological Catalysts of Life. In Physiology and Maintenance; Hänninen, O. O. P., Atalay, M., Eds.; UNESCO-EOLSS Publishers: Oxford, 2009; Vol. 2, pp 1−21. (19) Bugg, T. D. H. Introduction to Enzyme and Coenzyme Chemistry, 2nd ed.; Blackwell Publishing Ltd.: Oxford, 2004. (20) Wen, F.; McLachlan, M.; Zhao, H. Directed Evolution: Novel and Improved Enzymes. In Wiley Encyclopedia of Chemical Biology; Begley, T. P., Ed.; John Wiley and Sons, Inc.: New York, 2008; pp 1− 10. (21) Barbayianni, E.; Kokotos, G. ChemCatChem 2012, 4, 592. (22) Faber, K. Biotransformations in Organic Chemistry, 5th ed.; Springer-Verlag: New York, 2011. (23) Illanes, A.; Cauerhff, A.; Wilson, L.; Castro, G. R. Bioresour. Technol. 2012, 115, 48. (24) Sánchez, S.; Demain, A. L. Org. Process Res. Dev. 2011, 15, 224. (25) Kirk, O.; Borchert, T. V.; Fuglsang, C. C. Curr. Opin. Biotechnol. 2002, 13, 345. (26) Di Pierro, P.; Chico, B.; Villalonga, R.; Mariniello, L.; Damiao, A. E.; Porta, R. Biomacromolecules 2006, 7, 744. (27) Díez, P.; Piuleac, C. G.; Martínez-Ruiz, P.; Romano, S.; Gamella, M.; Villalonga, R.; Pingarrón, J. M. Anal. Bioanal. Chem. 2013, 405, 3773. (28) Kaul, P.; Asano, Y. Microb. Biotechnol. 2012, 5, 18. (29) Iyer, P. V.; Ananthanarayan, L. Process Biochem. 2008, 43, 1019. (30) Daniel, R. M.; Danson, M. J. Trends Biochem. Sci. 2010, 35, 584.

Reynaldo Villalonga was born in 1970 in Matanzas, Cuba. He studied Chemistry at the University of Havana, where he graduated with his Gold Diploma in 1993. In 1994 he worked on protein structures at the National Center for Genetic Engineering and Biotechnology and then joined the Laboratory of Bioinorganic Chemistry at the University of Havana. In 1996 he moved to the University of Matanzas, where he founded the Enzyme Technology Group and started works on neoglycoenzymes. He completed his Ph.D. degree in neoglycoenzyme synthesis in 2001 and was appointed as Full Professor and Founding Director of the Center for Enzyme Technology at the University of Matanzas in 2007. In 2010 he moved to the Complutense University of Madrid, Spain, where he held a Ramón y Cajal Research Professorship. He was a visiting Professor at the University of Naples “Federico Segundo”, McGill University, Toyama Prefectural University, Firenze University, Vigo University, and Joseph Fourier University. His research is focused on neoglycoenzymes, carbohydrate chemistry, drug delivery systems, biosensors, and nanotechnology. He has been recognized with several honors and awards including seven prizes from the Cuban Academy of Sciences, the Development Cooperation Prize of Belgium (2001), the Third World Academy of Sciences Award for Young Chemist (2004), the National Award in Chemical Sciences (2006), the Prize of the Cuban Ministry of Science (2007), the Prize of Ministry of Higher Education (2004, 2007), the Japanese Society for the Promotion of Science Fellowship (2004, 2008), and the Carolina Foundation Fellowship (2007). Since 2001 he has promoted and organized the Latin American Network for Enzyme Technology “RELATENZ”.

ACKNOWLEDGMENTS R.V. acknowledges the Ramón & Cajal contract from the Spanish Ministry of Science and Innovation. Financial support 4911

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

(31) Unsworth, L. D.; Van der Oost, J.; Koutsopoulos, S. FEBS J. 2007, 274, 4044. (32) Li, W. F.; Zhou, X. X.; Lu, P. Biotechnol. Adv. 2005, 23, 271. (33) Miletić, N.; Nastasović, A.; Loos, K. Bioresour. Technol. 2012, 115, 126. (34) Fragoso, A.; Caballero, J.; Almirall, E.; Villalonga, R.; Cao, R. Langmuir 2002, 18, 5051. (35) Villalonga, R.; Cao, R.; Fragoso, A.; Damiao, A. E.; Ortiz, P. D.; Caballero, J. J. Mol. Catal. B: Enzym. 2005, 35, 79. (36) Fernández, L.; Gómez, L.; Ramírez, H. L.; Villalonga, M. L.; Villalonga, R. J. Mol. Catal. B: Enzym. 2005, 34, 14. (37) Fernández, M.; Villalonga, M. L.; Caballero, J.; Fragoso, A.; Cao, R.; Villalonga, R. Biotechnol. Bioeng. 2003, 83, 743. (38) Fernández, M.; Villalonga, M. L.; Fragoso, A.; Cao, R.; Villalonga, R. Enzyme Microb. Technol. 2004, 34, 78. (39) Santagapita, P. R.; Gómez, L.; Mazzobre, F.; Ramírez, H. L.; Corti, H. R.; Villalonga, R.; Buera, M. P. Carbohydr. Polym. 2011, 83, 203. (40) Santagapita, P. R.; Gómez, L.; Mazzobre, F.; Ramirez, H. L.; Corti, H. R.; Villalonga, R.; Buera, M. P. Biomacromolecules 2008, 9, 741. (41) Davids, T.; Schmidt, M.; Böttcher, D.; Bornscheuer, W. T. Curr. Opin. Chem. Biol. 2013, 17, 215. (42) Woodley, J. M. Curr. Opin. Struct. Biol. 2013, 17, 310. (43) Villalonga, M. L.; Reyes, G.; Villalonga, R. Biotechnol. Lett. 2004, 26, 209. (44) Ismaya, W. T.; Hasan, K.; Kardi, I.; Zainuri, A.; Rahmawaty, R. I.; Permanahadi, S.; El Viera, B. V.; Harinanto, G.; Gaffar, S.; Natalia, D.; Subroto, T.; Soemitro, S. Appl. Biochem. Biotechnol. 2013, 170, 44. (45) DeSantis, G.; Jones, J. B. Curr. Opin. Biotechnol. 1999, 10, 324. (46) Spiro, R. G. Glycobiology 2002, 12, 43R. (47) Weerapana, E.; Imperiali, B. Glycobiology 2006, 16, 91R. (48) Meynial-Salles, I.; Combes, D. J. Biotechnol. 1996, 46, 1. (49) Shental-Bechor, D.; Levy, Y. Curr. Opin. Struct. Biol. 2009, 19, 524. (50) Eckhardt, M.; Fewou, S. N.; Ackermann, I.; Gieselmann, V. Biochem. J. 2002, 368, 317. (51) Wang, C.; Eufemi, M.; Turano, C.; Giartosio, A. Biochemistry 1996, 35, 7299. (52) Romero, J. M.; Carrizo, M. E.; Montich, G.; Curtino, J. A. Biochem. Biophys. Res. Commun. 2001, 289, 69. (53) Solá, R. J.; Rodríguez-Martínez, J. A.; Griebenow, K. Cell. Mol. Life Sci. 2007, 64, 2133. (54) Bertozzi, C. R.; Kiessling, L. L. Science 2001, 291, 2357. (55) Goebel, W. F.; Avery, O. T. J. Exp. Med. 1929, 50, 521. (56) Mitz, M. A.; Summaria, L. J. Nature 1961, 189, 576. (57) Axén, R.; Myrin, P. A.; Janson, J. C. Biopolymers 1970, 9, 401. (58) Rogers, J. C.; Kornfeld, S. Biochem. Biophys. Res. Commun. 1971, 45, 622. (59) Holcenberg, J. S.; Schmer, G.; Teller, D. C. J. Biol. Chem. 1975, 250, 4165. (60) O’Neill, S. P.; Wykes, J. R.; Dunnill, P.; Lilly, M. D. Biotechnol. Bioeng. 1971, 13, 319. (61) Wykes, J. R.; Dunnill, P.; Lilly, M. D. Biochim. Biophys. Acta 1971, 250, 522. (62) Marshall, J. J.; Rabinowitz, M. L. Arch. Biochem. Biophys. 1975, 167, 777. (63) Marshall, J. J. Carbohydr. Res. 1976, 49, 389. (64) Marshall, J. J.; Rabinowitz, M. L. J. Biol. Chem. 1976, 251, 1081. (65) Marshall, J. J.; Humphreys, J. D. Biotechnol. Bioeng. 1977, 19, 1739. (66) Marshall, J. J.; Humphreys, J. D.; Abramson, S. L. FEBS Lett. 1977, 83, 249. (67) Svensson, B. Biochim. Biophys. Acta 1976, 429, 954. (68) Vegarud, G.; Christensen, T. B. Biotechnol. Bioeng. 1975, 17, 1391. (69) Vegarud, G.; Christensen, T. B. Acta Chem. Scand., Ser. B 1975, 29, 887.

(70) Lasch, J.; Bessmertnaya, L.; Kozlov, L. V.; Antonov, V. K. Eur. J. Biochem. 1976, 63, 591. (71) Sherwood, R. F.; Baird, J. K.; Atkinson, T.; Wiblin, C. N.; Rutter, D. A.; Ellwood, D. C. Biochem. J. 1977, 164, 461. (72) Hupe, F. U.S. Patent 4,167,446, 1979. (73) Marshall, J. J. Trends Biochem. Sci. 1978, 3, 79. (74) Venkatesh, R.; Sundaram, P. V. Protein Eng. 1998, 11, 691. (75) Copeland, R. A. Enzymes: A Practical Introduction to Structure, Mechanism, and Data Analysis, 2nd ed.; Wiley-VCH: New York, 2000. (76) Buxbaum, E. Fundamentals of Protein Structure and Function; Springer: New York, 2007; p 23. (77) Villalonga, R.; Fernández, M.; Fragoso, A.; Cao, R.; Di Pierro, P.; Mariniello, L.; Porta, R. Biotechnol. Bioeng. 2003, 81, 732. (78) Villalonga, M. L.; Villalonga, R.; Mariniello, L.; Gómez, L.; Di Pierro, P.; Porta, R. World J. Microbiol. Biotechnol. 2006, 22, 595. (79) Valdivia, A.; Villalonga, R.; Di Pierro, P.; Pérez, Y.; Mariniello, L.; Gómez, L.; Porta, R. J. Biotechnol. 2006, 122, 326. (80) Hermanson, G. T. Bioconjugate Techniques, 2nd ed.; Elsevier Inc.: Amsterdam, 2008; pp 14−15. (81) Moczar, E. Experientia 1973, 29, 1576. (82) Moczar, E.; Leboul, J. FEBS Lett. 1975, 50, 300. (83) Moczar, E.; Vass, G. Carbohydr. Res. 1976, 50, 133. (84) Pérez, Y.; Valdivia, A.; Ramírez, H. L.; Villalonga, R. Macromol. Rapid Commun. 2005, 26, 1304. (85) Delanoy, G.; Li, Q.; Yu, J. Int. J. Biol. Macromol. 2005, 35, 89. (86) Davis, B. G.; Lloyd, R. C.; Jones, J. B. J. Org. Chem. 1998, 63, 9614. (87) Davis, B. G.; Lloyd, R. C.; Jones, J. B. Bioorg. Med. Chem. 2000, 8, 1527. (88) Davis, B. G.; Maughan, M. A. T.; Green, M. P.; Ullman, A.; Jones, J. B. Tetrahedron: Asymmetry 2000, 11, 245. (89) Tagami, U.; Shimba, N.; Nakamura, M.; Yokoyama, K.; Suzuki, E.; Hirokawa, Y. Protein Eng., Des. Sel. 2009, 22, 747. (90) Cummings, J. H.; Stephen, A. M. Eur. J. Clin. Nutr. 2007, 61, S5. (91) IUPAC-IUBMB.. Pure Appl. Chem. 1996, 68, 1919. (92) Venkatesh, R.; Sundaram, P. V. Ann. N. Y. Acad. Sci. 1998, 864, 517. (93) Villalonga, R.; Gómez, L.; Ramírez, H. L.; Villalonga, M. L. J. Chem. Technol. Biotechnol. 1999, 74, 635. (94) Villalonga, R.; Villalonga, M. L.; Gómez, L. J. Mol. Catal. B: Enzym. 2000, 10, 483. (95) Sundaram, P. V.; Venkatesh, R. Protein Eng. 1998, 11, 699. (96) Díaz-Rodríguez, A.; Davis, B. G. Curr. Opin. Chem. Biol. 2011, 15, 211. (97) Aeschlimann, D.; Paulsson, M. Thromb. Haemostasis 1994, 71, 402. (98) Folk, J. E.; Chung, S. I. Methods Enzymol. 1985, 113, 358. (99) Di Pierro, P.; Chico, B.; Villalonga, R.; Mariniello, L.; Masi, P.; Porta, R. Enzyme Microb. Technol. 2007, 40, 437. (100) Villalonga, M. L.; Reyes, G.; Fragoso, A.; Cao, R.; Fernández, L.; Villalonga, R. Biotechnol. Appl. Biochem. 2005, 41, 217. (101) Darias, R.; Herrera, I.; Fragoso, A.; Cao, R.; Villalonga, R. Biotechnol. Lett. 2002, 24, 1665. (102) Valdivia, A.; Pérez, Y.; Cao, R.; Baños, M.; García, A.; Villalonga, R. Macromol. Biosci. 2007, 7, 70. (103) Valdivia, A.; Pérez, Y.; Ramírez, H.; Cao, R.; Villalonga, R. Biotechnol. Lett. 2006, 28, 1465. (104) Wang, S. L.; Chio, S. H. Enzyme Microb. Technol. 1998, 22, 634. (105) Chen, S. H.; Yen, Y. H.; Wang, C. L.; Wang, S. L. Enzyme Microb. Technol. 2003, 33, 643. (106) Charusheela, A.; Arvind, L. Enzyme Microb. Technol. 2002, 30, 19. (107) Chiumiento, A.; Dominguez, A.; Lamponi, S.; Villalonga, R.; Barbucci, R. J. Mater. Sci.: Mater. Med. 2006, 17, 427. (108) Dominguez, A.; Valdivia, A.; Caballero, J.; Villalonga, R.; Martínez, G.; Schacht, E. H. J. Bioact. Compat. Polym. 2005, 20, 557. (109) Villalonga, R.; Matos, M.; Cao, R. Electrochem. Commun. 2007, 9, 454. 4912

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

(110) Kobayashi, M.; Yanagihara, S.; Ichishima, E. Agric. Biol. Chem. 1989, 53, 3133. (111) Villalonga, R.; Tachibana, S.; Cao, R.; Matos, M.; Asano, Y. Enzyme Microb. Technol. 2007, 40, 471. (112) Fernández, M.; Fragoso, A.; Cao, R.; Baños, M.; AnsorgeSchumacher, M.; Hartmeier, W.; Villalonga, R. J. Mol. Catal. B: Enzym. 2004, 31, 47. (113) Fernández, M.; Villalonga, M. L.; Fragoso, A.; Cao, R.; Villalonga, R. Biotechnol. Appl. Biochem. 2002, 36, 235. (114) Camacho, C.; Matías, J. C.; Cao, R.; Matos, M.; Chico, B.; Hernández, J.; Longo, M. A.; Sanromán, M. A.; Villalonga, R. Langmuir 2008, 24, 7654. (115) Valdivia, A.; Pérez, Y.; Dominguez, A.; Caballero, J.; Gomez, L.; Schacht, E. H.; Villalonga, R. Macromol. Biosci. 2005, 5, 118. (116) Valdivia, A.; Pérez, Y.; Gómez, L.; Ramírez, H. L.; Schacht, E. H.; Villalonga, R. Arch. Pharm. 2006, 339, 372. (117) Nekliudov, A. D.; Ivankin, A. N.; Baburina, M. I. Prikl. Biokhim. Mikrobiol. 1998, 34, 61. (118) Villalonga, M. L.; Fernández, M.; Fragoso, A.; Cao, R.; Villalonga, R. Prep. Biochem. Biotechnol. 2003, 33, 53. (119) Fernández, M.; Villalonga, M. L.; Fragoso, A.; Cao, R.; Baños, M.; Villalonga, R. Process Biochem. 2004, 39, 535. (120) Sakurai, K.; Miyazaki, K.; Kodera, Y.; Nishimura, H.; Shingu, M.; Inada, Y. Glycoconjugate J. 1997, 14, 723. (121) Villalonga, R.; Valdivia, A.; Pérez, Y.; Chongo, B. J. Appl. Polym. Sci. 2006, 102, 4573. (122) Zhou, J. J. Agric. Food Chem. 2010, 58, 6741. (123) Solá, R. J.; Al-Azzam, W.; Griebenow, K. Biotechnol. Bioeng. 2006, 94, 1072. (124) Solá, R. J.; Griebenow, K. FEBS Lett. 2006, 580, 1685. (125) Solá, R. J.; Griebenow, K. FEBS J. 2006, 273, 5303. (126) Janecek, S.; Balaz, S.; Rosenberg, M.; Stredansky, M. Biotechnol. Tech. 1992, 6, 173. (127) García, A.; Hernández, K.; Chico, B.; García, D.; Villalonga, M. L.; Villalonga, R. J. Mol. Catal. B: Enzym. 2009, 59, 126. (128) Gildersleeve, J. C.; Oyelaran, O.; Simpson, J. T.; Allred, B. Bioconjugate Chem. 2008, 19, 1485. (129) Wilson, G. J. Biol. Chem. 1978, 253, 2070. (130) Marsh, J. W.; Denis, J.; Wriston, J. C. J. Biol. Chem. 1977, 252, 7678. (131) Nickle, E. C.; Solomon, R. D.; Torchia, T. E.; Wriston, J. C. Biochim. Biophys. Acta 1982, 704, 345. (132) Sebela, M.; Stosová, T.; Havlis, J.; Wielsch, N.; Thomas, H.; Zdráhal, Z.; Shevchenko, A. Proteomics 2006, 6, 2959. (133) Vaňková, H.; Pospíšilová, M.; Tichá, M.; Turková, J. Biotechnol. Tech. 1994, 8, 375. (134) Cayot, P.; Roullier, L.; Tainturier, G. J. Agric. Food Chem. 1999, 47, 1915. (135) Van Zile, J.; Henderson, L. A.; Baynes, J. W.; Thorpe, S. R. J. Biol. Chem. 1979, 254, 3547. (136) Fernández, M.; Fragoso, A.; Cao, R.; Baños, M.; Villalonga, M. L.; Villalonga, R. Biotechnol. Lett. 2002, 24, 1455. (137) Altikatoglu, M.; Arioz, C.; Basaran, Y.; Kuzu, H. Cent. Eur. J. Chem. 2009, 7, 423. (138) Lenders, J. P.; Crichton, R. R. Biotechnol. Bioeng. 1984, 26, 1343. (139) Vina, I.; Karsakevich, A.; Bekers, M. J. Mol. Catal. B: Enzym. 2001, 11, 551. (140) Benbough, J. E.; Wiblin, C. N.; Rafter, T. N. A.; Lee, J. Biochem. Pharmacol. 1979, 28, 833. (141) Kim, H. K.; Park, T. G. Enzyme Microb. Technol. 1999, 25, 31. (142) Eremin, A. N. Appl. Biochem. Microbiol. 2001, 37, 45. (143) Fernandes, A. I.; Gregoriadis, G. Biochim. Biophys. Acta 1996, 1293, 90. (144) Fernandes, A. I.; Gregoriadis, G. Biochim. Biophys. Acta 1997, 1341, 26. (145) Fernandes, A. I.; Gregoriadis, G. Int. J. Pharm. 2001, 217, 215. (146) Jadhav, S. B.; Singhal, R. S. Carbohydr. Polym. 2012, 90, 1811.

(147) Lenders, J. P.; Germain, P.; Crichton, R. R. Biotechnol. Bioeng. 1985, 27, 572. (148) Maksimenko, A. V.; Petrova, M. L.; Tishchenko, E. G.; Shchechilina, Y. V. Russ. J. Bioorg. Chem. 2000, 26, 322. (149) Maksimenko, A. V.; Petrova, M. L.; Tischenko, E. G.; Schechilina, Y. V. Eur. J. Pharm. Biopharm. 2001, 51, 33. (150) Masárová, J.; Mislovičová, D.; Gemeiner, P.; Michalková, E. Biotechnol. Appl. Biochem. 2001, 34, 127. (151) Mislovičová, D.; Masárová, J.; Svitel, J.; Gemeiner, P. Int. J. Biol. Macromol. 2002, 30, 251. (152) Mislovičová, D.; Masárová, J.; Vikartovská, A.; Gemeiner, P.; Michalková, E. J. Biotechnol. 2004, 110, 11. (153) Mislovičová, D.; Masárová, J.; Bučko, M.; Gemeiner, P. Enzyme Microb. Technol. 2006, 39, 579. (154) Morand, P.; Biellmann, J. F. FEBS Lett. 1991, 289, 148. (155) Pérez, Y.; Villalonga, R. Int. J. Biotechnol. 2004, 6, 385. (156) Rajalakshmi, N.; Sundaram, P. V. Protein Eng. 1995, 8, 1039. (157) Gómez, L.; Ramírez, H. L.; Villalonga, R. Acta Biotechnol. 2001, 21, 265. (158) Sebela, M.; Kopecný, D.; Lamplot, Z.; Havlis, J.; Thomas, H.; Shevchenko, A. Biotechnol. Appl. Biochem. 2005, 41, 77. (159) Siddiqui, K. S.; Cavicchioli, R. Extremophiles 2005, 9, 471. (160) Srimathi, S.; Jayaraman, G. Protein J. 2005, 24, 79. (161) Srimathi, S.; Jayaraman, G.; Narayanan, P. R. Enzyme Microb. Technol. 2006, 39, 301. (162) Tabandeh, M. R.; Aminlari, M. J. Biotechnol. 2009, 141, 189. (163) Valdivia, A.; Pérez, Y.; Dominguez, A.; Caballero, J.; Hernández, Y.; Villalonga, R. Biotechnol. Appl. Biochem. 2006, 44, 159. (164) Venkatesh, R.; Srimathi, S.; Yamuna, A.; Jayaraman, G. Biochim. Biophys. Acta 2005, 1722, 113. (165) Yurkovetskiy, A.; Choi, S.; Hiller, A.; Yin, M.; McCusker, C.; Syed, S.; Fischman, A. J.; Papisov, M. I. Biomacromolecules 2005, 6, 2648. (166) Arslan, A.; Kuzu, H.; Altikatoglu, M. J. Carbohydr. Chem 2010, 29, 222. (167) Altikatoglu, M.; Basaran, Y.; Arioz, C.; Ogan, A.; Kuzu, H. Appl. Biochem. Biotechnol. 2010, 160, 2187. (168) Ö ztürk, D. C.; Kazan, D.; Erarslan, I. World J. Microbiol. Biotechnol. 2002, 18, 881. (169) Ammar, Y. B.; Matsubara, T.; Ito, K.; Iizuka, M.; Minamiura, N. Enzyme Microb. Technol. 2002, 30, 875. (170) Ahmed, S. A.; Saleh, S. A.; Abdel-Fattah, A. F. Aust. J. Basic Appl. Sci. 2007, 1, 313. (171) Ertan, H.; Kazan, D.; Erarslan, A. Biotechnol. Tech. 1997, 11, 225. (172) Kazan, D.; Ertan, H.; Erarslan, A. Appl. Microbiol. Biotechnol. 1997, 48, 191. (173) Kobayashi, M.; Takatsu, K. Biosci. Biotechnol. Biochem. 1994, 58, 275. (174) Maeda, H.; Matsumura, Y.; Asami, O.; Tanaka, H.; Sasaki, I. U.S. Patent 4,844,897, 1989. (175) Siddiqui, K. S.; Parkin, D. M.; Curmi, P. M.; De Francisci, D.; Poljak, A.; Barrow, K.; Noble, M. H.; Trewhella, J.; Cavicchioli, R. Biotechnol. Bioeng. 2009, 103, 676. (176) Zhang, H. W.; Wang, F. S.; Shao, W.; Zheng, X. L.; Qi, J. Z.; Cao, J. C.; Zhang, T. M. Biochemistry (Moscow) 2006, 71, S96. (177) Nakamura, S.; Kato, A.; Kobayashi, K. Agric. Biol. Chem. 1990, 54, 3057. (178) Nakamura, S.; Kato, A.; Kobayashi, K. J. Agric. Food Chem. 1991, 39, 647. (179) Nakamura, S.; Kato, A.; Kobayashi, K. J. Agric. Food Chem. 1992, 40, 735. (180) Nakamura, S.; Kobayashi, K.; Kato, A. J. Agric. Food Chem. 1994, 42, 2688. (181) Kato, Y.; Matsuda, T.; Nakamura, R. Biosci. Biotechnol. Biochem. 1993, 57, 1. (182) Kato, A. Food Sci. Technol. Res. 2002, 8, 193. (183) Arita, K.; Babiker, E. E.; Azakami, H.; Kato, A. J. Agric. Food Chem. 2001, 49, 2030. 4913

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

(184) Sutthirak, P.; Assavanig, A.; Dharmsthiti, S.; Lertsiri, S. J. Food Biochem. 2010, 34, 1157. (185) Kaplan, H.; King, M. C.; Stewart, N. A. S. U.S. Patent 20060122369, 2006. (186) Pham, V. T.; Ewing, E.; Kaplan, H.; Choma, C.; Hefford, M. A. Biotechnol. Bioeng. 2008, 101, 452. (187) Lee, Y. C.; Stowell, C. P.; Krantz, M. J. Biochemistry 1976, 15, 3956. (188) Krantz, M. J.; Holtzman, N. A.; Stowell, C. P.; Lee, Y. C. Biochemistry 1976, 15, 3963. (189) Fujita, T.; Furitsu, H.; Nishikawa, M.; Takakura, Y.; Sezaki, H.; Hashida, M. Biochem. Biophys. Res. Commun. 1992, 189, 191. (190) Yabe, Y.; Nishikawa, M.; Tamada, A.; Takakura, Y.; Hashida, M. J. Pharmacol. Exp. Ther. 1999, 289, 1176. (191) Staud, F.; Nishikawa, M.; Takakura, Y.; Hashida, M. Biochim. Biophys. Acta 1999, 1427, 183. (192) Swart, P. J.; Hirano, T.; Kuipers, M. E.; Ito, Y.; Smit, C.; Hashida, M.; Nishikawa, M.; Beljaars, L.; Meijer, D. K.; Poelstra, K. J. Hepatol. 1999, 31, 1034. (193) Ishimoto, N.; Nemoto, T.; Nagayoshi, K.; Yamashita, F.; Hashida, M. J. Controlled Release 2006, 111, 204. (194) Robinson, M. A.; Charlton, S. T.; Garnier, P.; Wang, X. T.; Davis, S. S.; Perkins, A. C.; Frier, M.; Duncan, R.; Savage, T. J.; Wyatt, D. A.; Watson, S. A.; Davis, B. G. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 14527. (195) Yabe, Y.; Kobayashi, N.; Nishihashi, T.; Takahashi, R.; Nishikawa, M.; Takakura, Y.; Hashida, M. J. Pharmacol. Exp. Ther. 2001, 298, 894. (196) Ghosh, M.; Bachhawat, B. K.; Surolia, A. Biochem. J. 1979, 183, 185. (197) Gilbert, H. J.; Jack, G. W. Biochem. J. 1981, 199, 715. (198) Kuo, J. Y.; Goldstein, J. Enzyme Microb. Technol. 1983, 5, 285. (199) Wongkhalaung, C.; Kashiwagi, Y.; Magae, Y.; Ohta, T.; Sasaki, T. Appl. Microbiol. Biotechnol. 1985, 21, 37. (200) Abdel-Naby, M. A. Microbiol. Res. 1999, 154, 213. (201) Rosemeyer, H.; Körnig, E.; Seela, F. Eur. J. Biochem. 1982, 122, 375. (202) Axén, R.; Ernback, S. Eur. J. Biochem. 1971, 18, 351. (203) Ibrahim, M.; Hubert, P.; Dellacherie, E.; Magdalou, J.; Muller, J.; Siest, G. Enzyme Microb. Technol. 1985, 7, 66. (204) Buss, D. H.; Goldstein, I. J. J. Chem. Soc. C 1968, 0, 1457. (205) Kim, B.; Cha, G. S.; Meyerhoff, M. E. Anal. Chem. 1990, 62, 2663. (206) Kim, B.; Behbahani, I.; Meyerhoff, M. E. Anal. Biochem. 1992, 202, 166. (207) Kim, B.; Buckwalter, J. M.; Meyerhoff, M. E. Anal. Biochem. 1994, 218, 14. (208) Guo, X.; Meyerhoff, M. E. Appl. Biochem. Biotechnol. 1997, 68, 41. (209) Maksimenko, A. V.; Tischenko, E. G. Biochemistry (Moscow) 1997, 62, 1163. (210) Maksimenko, A. V.; Tischenko, E. G. Biochemistry (Moscow) 1997, 62, 1167. (211) Maksimenko, A. V.; Tischenko, E. G.; Golubykh, V. L. Cardiovasc. Drugs Ther. 1999, 13, 479. (212) Maksimenko, A. V.; Golubykh, V. L.; Tischenko, E. G. Metab. Eng. 2003, 5, 177. (213) Maksimenko, A. V.; Schechilina, Y. V.; Tischenko, E. G. Biochemistry (Moscow) 2003, 68, 862. (214) Brandt, J.; Andersson, L. O.; Porath, J. Biochim. Biophys. Acta 1975, 386, 196. (215) Prashar, D.; Cui, D.; Bandyopadhyay, D.; Luk, Y. Y. Langmuir 2011, 27, 13091. (216) Holcenberg, J. S. Annu. Rev. Pharmacol. Toxicol. 1977, 17, 97. (217) Chalker, J. M.; Bernardes, G. J.; Lin, Y. A.; Davis, B. G. Chem. Asian J. 2009, 4, 630. (218) Davis, N. J.; Flitsch, S. L. Tetrahedron Lett. 1991, 32, 6793. (219) Swanwick, R. S.; Daines, A. M.; Tey, L. H.; Flitsch, S. L.; Allemann, R. K. ChemBioChem 2005, 6, 1338.

(220) Swanwick, R. S.; Daines, A. M.; Flitsch, S. L.; Allemann, R. K. Org. Biomol. Chem. 2005, 3, 572. (221) Lloyd, R. C.; Davis, B. G.; Jones, J. B. Bioorg. Med. Chem. 2000, 8, 1537. (222) Matsumoto, K.; Davis, B. G.; Jones, J. B. Chem. Commun. 2001, 10, 903. (223) Matsumoto, K.; Davis, B. G.; Jones, J. B. Chem.Eur. J. 2002, 8, 4129. (224) Davis, B. G. Chem. Commun. 2001, 4, 351. (225) Davis, B. G.; Jones, J. B.; Bott, R. R. U.S. Patent 6,627,744, 2003. (226) Davis, B. G.; Jones, J. B.; Bott, R. R. U.S. Patent 0,272,670, 2005. (227) Rendle, P. M.; Seger, A.; Rodrigues, J.; Oldham, N. J.; Bott, R. R.; Jones, J. B.; Cowan, M. M.; Davis, B. G. J. Am. Chem. Soc. 2004, 126, 4750. (228) Gamblin, D. P.; Garnier, P.; Ward, S. J.; Oldham, N. J.; Fairbanks, A. J.; Davis, B. G. Org. Biomol. Chem. 2003, 1, 3642. (229) Gamblin, D. P.; Garnier, P.; Van Kasteren, S.; Oldham, N. J.; Fairbanks, A. J.; Davis, B. G. Angew Chem., Int. Ed. 2004, 43, 828. (230) Bernardes, G. J. L.; Marston, J. P.; Batsanov, A. S.; Howard, J. A. K.; Davis, B. G. Chem. Commun. 2007, 30, 3145. (231) Johnston, B. D.; Pinto, B. M. J. Org. Chem. 2000, 65, 4607. (232) Bernardes, G. J. L.; Gamblin, D. P.; Davis, B. G. Angew. Chem., Int. Ed. 2006, 45, 4007. (233) Bernardes, G. J. L.; Grayson, E. J.; Thompson, S.; Chalker, J. M.; Errey, J. C.; El Oualid, F.; Claridge, T. D. W.; Davis, B. G. Angew. Chem., Int. Ed. 2008, 47, 2244. (234) Boutureira, O.; Bernardes, G. J. L.; Fernandez-Gonzalez, M.; Anthony, D. C.; Davis, B. G. Angew. Chem., Int. Ed. 2012, 51, 1432. (235) Bernardes, G. J. L.; Chalker, J. M.; Errey, J. C.; Davis, B. G. J. Am. Chem. Soc. 2008, 130, 5052. (236) Lin, Y. A.; Chalker, J. M.; Floyd, N.; Bernardes, G. J.; Davis, B. G. J. Am. Chem. Soc. 2008, 130, 9642. (237) Lin, Y. A.; Chalker, J. M.; Davis, B. G. ChemBioChem 2009, 10, 959. (238) Chalker, J. M.; Lin, Y. A.; Boutureira, O.; Davis, B. G. Chem. Commun. 2009, 25, 3714. (239) Chalker, J. M.; Wood, C. S.; Davis, B. G. J. Am. Chem. Soc. 2009, 131, 16346. (240) Wriston, J. C. FEBS Lett. 1973, 33, 93. (241) Villalonga, R.; Tachibana, S.; Pérez, Y.; Asano, Y. Biotechnol. Lett. 2005, 27, 1311. (242) Fernández, M.; Fragoso, A.; Cao, R.; Baños, M.; Villalonga, R. Enzyme Microb. Technol. 2002, 31, 543. (243) Fernández, M.; Fragoso, A.; Cao, R.; Villalonga, R. J. Mol. Catal. B: Enzym. 2003, 21, 133. (244) Fernández, M.; Fragoso, A.; Cao, R.; Villalonga, R. Process Biochem. 2005, 40, 2091. (245) Vazquez-Duhalt, R.; Tinoco, R.; D’Antonio, P.; Topoleski, L. D.; Payne, G. F. Bioconjugate Chem. 2001, 12, 301. (246) Holzinger, M.; Bouffier, L.; Villalonga, R.; Cosnier, S. Biosens. Bioelectron. 2009, 24, 1128. (247) Cabana, H.; Ahamed, A.; Leduc, R. Bioresour. Technol. 2011, 102, 1656. (248) Pérez, Y.; Valdivia, A.; Gomez, L.; Simpson, B. K.; Villalonga, R. Macromol. Biosci. 2005, 5, 1220. (249) Chaplin, B.; Green, M. L. Biotechnol. Bioeng. 1982, 24, 2627. (250) Hernández, K.; Fernández, L.; Gómez, L.; Villalonga, R. Process Biochem. 2006, 41, 1155. (251) Baek, W. O.; Vijayalakshmi, M. A. Biochim. Biophys. Acta 1997, 1336, 394. (252) Villalonga, R.; Tachibana, S.; Cao, R.; Ramirez, H. L.; Asano, Y. Biochem. Eng. J. 2006, 30, 26. (253) Maekawa, K.; Ishimoto, K. Bull. Chem. Soc. Jpn. 1961, 34, 1221. (254) López-Jaramillo, F. J.; Pérez-Banderas, F.; Hernández-Mateo, F.; Santoyo-González, F. Acta Crystallogr., Sect. F: Struct. Biol. Cryst. Commun. 2005, F61, 435. 4914

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

(290) Payne, R. J.; Wong, C. H. Chem. Commun. 2010, 46, 21. (291) Bang, D.; Kent, S. B. H. Angew. Chem., Int. Ed. 2004, 43, 2534. (292) Davis, B. G. Angew. Chem., Int. Ed. 2009, 48, 4674. (293) NC-IUBMB. Enzyme Nomenclature: Recommendations of the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology on the Nomenclature and Classiﬁcation of Enzymes by the Reactions They Catalyse; Academic Press: San Diego, CA, 1992. (294) Krahmer, R.; Muller, A.; Leenders, F. U.S. Patent 2006/ 0127899/A1, 2006. (295) Bott, R. R.; Cowan, M. M.; Davis, B. G.; Jones, J. B. U.S. Patent 20040097720, 2004. (296) Nishikawa, M.; Hashida, M.; Takakura, Y. Adv. Drug Delivery Rev. 2009, 61, 319. (297) Nishikawa, M.; Tamada, A.; Hyoudou, K.; Umeyama, Y.; Takahashi, Y.; Kobayashi, Y.; Kumai, H.; Ishida, E.; Staud, F.; Yabe, Y.; Takakura, Y.; Yamashita, F.; Hashida, M. Clin. Exp. Metastasis 2004, 21, 213. (298) Villalonga, R.; Cao, R.; Fragoso, A. Chem. Rev. 2007, 107, 3088. (299) Cao, R.; Fragoso, A.; Almiral, E.; Villalonga, R. Supramol. Chem. 2003, 15, 161. (300) Haddad, R.; Holzinger, M.; Villalonga, R.; Neumann, A.; Roots, J.; Maaref, A.; Cosnier, S. Carbon 2011, 49, 2571. (301) Villalonga, R.; Díez, P.; Eguílaz, M.; Martínez, P.; Pingarrón, J. M. ACS Appl. Mater. Interfaces 2012, 4, 4312. (302) Holzinger, M.; Baur, J.; Haddad, R.; Wang, X.; Cosnier, S. Chem. Commun. 2011, 47, 2450. (303) Le Goff, A.; Gorgy, K.; Holzinger, M.; Haddad, R.; Zimmerman, M.; Cosnier, S. Chem.Eur. J. 2011, 17, 10216. (304) Maksimenko, A. V.; Schechilina, Y. V.; Tischenko, E. G. Biochemistry (Moscow) 2001, 66, 456. (305) Jadhav, S. B.; Singhal, R. S. Carbohydr. Polym. 2013, 92, 1724. (306) Wileman, T. E.; Foster, R. L.; Elliott, P. N. J. Pharm. Pharmacol. 1986, 38, 264. (307) Yasuda, Y.; Fujita, T.; Takakura, Y.; Hashida, M.; Sezaki, H. Chem. Pharm. Bull. 1990, 38, 2053. (308) Maksimenko, A. V.; Grigoreva, E. L.; Bezrukavnikova, L. M.; Petrov, A. D.; Tishchenko, E. G.; Arkhipova, O. G.; Torchilin, V. P. Byull. Eksp. Biol. Med. 1991, 112, 265. (309) Fujita, T.; Nishikawa, M.; Tamaki, C.; Takakura, Y.; Hashida, M.; Sezaki, H. J. Pharmacol. Exp. Ther. 1992, 263, 971. (310) Mihara, K.; Oka, Y.; Sawai, K.; Takakura, Y.; Hashida, M. J. Drug Targeting 1994, 2, 317. (311) Mihara, K.; Sawai, K.; Takakura, Y.; Hashida, M. Biol. Pharm. Bull. 1994, 17, 296. (312) Karsakevich, A. S.; Kinstler, O. B.; Dauvarte, A. Zh.; Vina, I. A. Vopr. Med. Khim. 1988, 34, 107. (313) Taniguchi, M.; Hoshino, K.; Watanabe, K.; Sugai, K.; Fujii, M. Biotechnol. Bioeng. 1992, 39, 287. (314) Chen, J. P.; Chang, K. C. J. Chem. Technol. Biotechnol. 1994, 60, 133. (315) Zhou, J. Q.; Wang, J. W. Enzyme Microb. Technol. 2009, 45, 299. (316) Qian, G.; Zhou, J.; Ma, J.; Wang, D.; He, B. Artif. Cells, Blood Substitutes, Immobilization Biotechnol. 1996, 24, 567. (317) Liu, J.; Teng, L.; Liu, C.; Hu, L.; Wang, Y.; Liu, H.; Wang, F. J. Drug Targeting 2009, 17, 216. (318) Gregoriadis, G.; Fernandes, A.; Mital, M.; McCormack, B. Cell. Mol. Life Sci. 2000, 57, 1964. (319) Gregoriadis, G.; Jain, S.; Papaioannou, I.; Laing, P. Int. J. Pharm. 2005, 300, 125. (320) Maksimenko, A. V. Curr. Pharm. Des. 2005, 11, 2007. (321) Maksimenko, A. V.; Golubykh, V. L.; Tischenko, E. G. J. Pharm. Pharmacol. 2004, 56, 1463. (322) Maksimenko, A. V.; Vavaev, A. V.; Bouryachkovskaya, L. I.; Mokh, V. P.; Uchitel, I. A.; Lakomkin, V. L.; Kapelko, V. I.; Tischenko, E. G. Acta Nat. 2010, 2, 82. (323) Liu, J.; Zhao, T.; Tan, H.; Cheng, Y.; Cao, J.; Wang, F. Biomed. Pharmacother. 2010, 64, 686.

(255) Morales-Sanfrutos, J.; Lopez-Jaramillo, J.; Ortega-Muñoz, M.; Megia-Fernandez, A.; Perez-Balderas, F.; Hernandez-Mateo, F.; Santoyo-Gonzalez, F. Org. Biomol. Chem. 2010, 8, 667. (256) Kieda, C.; Delmotte, F.; Monsigny, M. FEBS Lett. 1977, 76, 257. (257) Rajput, Y.; Gupta, M. Biotechnol. Appl. Biochem. 1988, 10, 242. (258) Knezević-Jugović, Z. D.; Bezbradica, D. I.; Mijin, D. Z.; Antov, M. G. Methods Mol. Biol. 2011, 679, 99. (259) Gómez, L.; Ramírez, H. L.; Villalonga, R. Biotechnol. Lett. 2000, 22, 347. (260) Darias, R.; Villalonga, R. J. Chem. Technol. Biotechnol. 2001, 76, 489. (261) Gonera, A.; Mischnick, P.; Ukeda, H. Enzyme Microb. Technol. 2004, 34, 248. (262) Tolvanen, M.; Gahmberg, C. G. J. Biol. Chem. 1986, 261, 9546. (263) Zhu, Y.; Li, X.; Kyazike, J.; Zhou, Q.; Thurberg, B. L.; Raben, N.; Mattaliano, R. J.; Cheng, S. H. J. Biol. Chem. 2004, 279, 50336. (264) Zhu, Y.; Li, X.; McVie-Wylie, A.; Jiang, C.; Thurberg, B. L.; Raben, N.; Mattaliano, R. J.; Cheng, S. H. Biochem. J. 2005, 389, 619. (265) Zhu, Y.; Jiang, J. L.; Gumlaw, N. K.; Zhang, J.; Bercury, S. D.; Ziegler, R. J.; Lee, K.; Kudo, M.; Canfield, W. M.; Edmunds, T.; Jiang, C.; Mattaliano, R. J.; Cheng, S. H. Mol. Ther. 2009, 17, 954. (266) Gómez, L.; Villalonga, R. Biotechnol. Lett. 2000, 22, 1191. (267) Ramírez, H. L.; Chico, B.; Villalonga, R.; Hoste, K.; Schacht, E. H. J. Bioact. Compat. Polym. 2002, 17, 161. (268) Schmaltz, R. M.; Hanson, S. R.; Wong, C. H. Chem. Rev. 2011, 111, 4259. (269) Takegawa, K.; Tabuchi, M.; Yamaguchi, S.; Kondo, A.; Kato, I.; Iwahara, S. J. Biol. Chem. 1995, 270, 3094. (270) Takegawa, K.; Fujita, K.; Fan, J. Q.; Tabuchi, M.; Tanaka, N.; Kondo, A.; Iwamoto, H.; Kato, I.; Lee, Y. C.; Iwahara, S. Anal. Biochem. 1998, 257, 218. (271) Takegawa, K.; Fan, J. Q. Methods Enzymol. 2003, 362, 64. (272) Fujita, K.; Tanaka, N.; Sano, M.; Kato, I.; Asada, Y.; Takegawa, K. Biochem. Biophys. Res. Commun. 2000, 267, 134. (273) Fan, J. Q.; Takegawa, K.; Iwahara, S.; Kondo, A.; Kato, I.; Abeygunawardana, C.; Lee, Y. C. J. Biol. Chem. 1995, 270, 17723. (274) Ramakrishnan, B.; Boeggeman, E.; Qasba, P. K. Bioconjugate Chem. 2007, 18, 1912. (275) Villalonga, R.; Fernández, M.; Fragoso, A.; Cao, R.; Mariniello, L.; Porta, R. Biotechnol. Appl. Biochem. 2003, 38, 53. (276) Wang, L. X. Carbohydr. Res. 2008, 343, 1509. (277) Rabuka, D. Curr. Opin. Chem. Biol. 2010, 14, 790. (278) Longo, M. A.; Combes, D. FEBS Lett. 1995, 375, 63. (279) Longo, M. A.; Combes, D. J. Mol. Catal. B: Enzym. 1997, 2, 281. (280) Longo, M. A.; Combes, D. J. Biotechnol. 1997, 58, 21. (281) Longo, M. A.; Combes, D. J. Chem. Technol. Biotechnol. 1999, 74, 25. (282) Fernández-González, M.; Boutureira, O.; Bernardes, G. J. L.; Chalker, J. M.; Young, M. A.; Errey, J. C.; Davis, B. G. Chem. Sci. 2010, 1, 709. (283) Chalker, J. M.; Bernardes, G. J.; Davis, B. G. Acc. Chem. Res. 2011, 44, 730. (284) Floyd, N.; Vijayakrishnan, B.; Koeppe, J. R.; Davis, B. G. Angew. Chem., Int. Ed. 2009, 48, 7798. (285) Van Kasteren, S. I.; Kramer, H. B.; Jensen, H. H.; Campbell, S. J.; Kirkpatrick, J.; Oldham, N. J.; Anthony, D. C.; Davis, B. G. Nature 2007, 446, 1105. (286) Mezzato, S.; Schaffrath, M.; Unverzagt, C. Angew. Chem., Int. Ed. 2005, 44, 1650. (287) Piontek, C.; Ring, P.; Harjes, O.; Heinlein, C.; Mezzato, S.; Lombana, N.; Pöhner, C.; Püttner, M.; Varón Silva, D.; Martin, A.; Schmid, F. X.; Unverzagt, C. Angew. Chem., Int. Ed. 2009, 48, 1936. (288) Piontek, C.; Varón Silva, D.; Heinlein, C.; Pöhner, C.; Mezzato, S.; Ring, P.; Martin, A.; Schmid, F. X.; Unverzagt, C. Angew. Chem., Int. Ed. 2009, 48, 1941. (289) Kajihara, Y.; Yamamoto, N.; Okamoto, R.; Hirano, K.; Murase, T. Chem. Rec. 2010, 10, 80. 4915

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

(324) Liu, J. F.; Wang, X.; Tan, H. N.; Liu, H.; Wang, Y. G.; Chen, R. Q.; Cao, J. C.; Wang, F. S. Drug Discoveries Ther. 2010, 4, 355. (325) Liu, J.; Wang, X.; Wang, F.; Teng, L.; Cao, J. Biomed. Pharmacother. 2009, 63, 484. (326) Liu, J.; Tan, H.; Sun, Y.; Zhou, S.; Cao, J.; Wang, F. Mol. Cell. Biochem. 2009, 327, 219. (327) Qi, J.; Li, Y.; Zhang, H.; Cheng, Y.; Sun, Y.; Cao, J.; Zhao, Y.; Wang, F. Brain Res. 2009, 1260, 76. (328) Le-Tien, C.; Millette, M.; Lacroix, M.; Mateescu, M. A. Biotechnol. Appl. Biochem. 2004, 39, 189. (329) Lee, H.; Park, T. G. Biotechnol. Prog. 1998, 14, 508. (330) Yeboah, F. K.; Alli, I.; Yaylayan, V. A.; Yasuo, K.; Chowdhury, S. F.; Purisima, E. O. Bioconjugate Chem. 2004, 15, 27. (331) Maksimenko, A. V.; Konovalova, O. I.; Berdichevskii, V. R.; Arkhipova, O. G.; Torchilin, V. P. Byull. Eksp. Biol. Med. 1986, 102, 567. (332) Scaman, C.; Nakai, S.; Aminlari, M. Food Chem. 2006, 99, 368. (333) De la Casa, R. M.; Guisán, J. M.; Sánchez-Montero, J. M.; Sinisterra, J. V. Enzyme Microb. Technol. 2002, 30, 30. (334) Ghazarian, H.; Idoni, B.; Oppenheimer, S. B. Acta Histochem. 2011, 113, 236. (335) Rabinovich, G. A.; van Kooyk, Y.; Cobb, B. A. Ann. N. Y. Acad. Sci. 2012, 1253, 1. (336) Lam, S. K.; Ng, T. B. Appl. Microbiol. Biotechnol. 2011, 89, 45. (337) Gabius, S.; Hellmann, K. P.; Hellmann, T.; Brinck, U.; Gabius, H. J. Anal. Biochem. 1989, 182, 447. (338) Gabius, H. J.; Gabius, S.; Brinck, U.; Schauer, A. Pathol., Res. Pract. 1990, 186, 597. (339) Gabius, S.; Schirrmacher, V.; Franz, H.; Joshi, S. S.; Gabius, H. J. Int. J. Cancer 1990, 46, 500. (340) Gabius, S.; Joshi, S. S.; Gabius, H. J.; Sharp, J. G. Anticancer Res. 1991, 11, 793. (341) Gabius, H. J.; Gabius, S. Methods Mol. Biol. 1993, 14, 263. (342) Gabius, S.; Wawotzny, R.; Martin, U.; Wilholm, S.; Gabius, H. J. Ann. Hematol. 1994, 68, 125. (343) Ruhl, S.; Cisar, J. O.; Sandberg, A. L. Infect. Immun. 2000, 68, 6346. (344) Wust, J.; Steiger, U.; Vuong, H.; Zbinden, R. J. Clin. Microbiol. 2000, 38, 929. (345) Takahashi, K.; Lou, X. F.; Ishii, Y.; Hattori, M. J. Agric. Food Chem. 2000, 48, 2044. (346) Veronese, F. M.; Morpurgo, M. Farmaco 1999, 54, 497. (347) Whyte, M. P.; Greenberg, C. R.; Salman, N. J.; Bober, M. B.; McAlister, W. H.; Wenkert, D.; Van Sickle, B. J.; Simmons, J. H.; Edgar, T. S.; Bauer, M. L.; Hamdan, M. A.; Bishop, N.; Lutz, R. E.; McGinn, M.; Craig, S.; Moore, J. N.; Taylor, J. W.; Cleveland, R. H.; Cranley, W. R.; Lim, R.; Thacher, T. D.; Mayhew, J. E.; Downs, M.; Millán, J. L.; Skrinar, A. M.; Crine, P.; Landy, H. N. Engl. J. Med. 2012, 366, 904. (348) Lawlor, M. W.; Armstrong, D.; Viola, M. G.; Widrick, J. J.; Meng, H.; Grange, R. W.; Childers, M. K.; Hsu, C. P.; O’Callaghan, M.; Pierson, C. R.; Buj-Bello, A.; Beggs, A. H. Hum. Mol. Genet. 2013, 22, 1525. (349) Tajima, G.; Sakura, N.; Kosuga, M.; Okuyama, T.; Kobayashi, M. Mol. Genet. Metab. 2013, 108, 172. (350) Walsh, G. Eur. J. Pharm. Biopharm. 2003, 55, 3. (351) Katre, N. V. Adv. Drug Delivery Rev. 1993, 10, 91. (352) Veronese, F. M. Biomaterials 2001, 22, 405. (353) Pasut, G.; Veronese, F. M. Prog. Polym. Sci. 2007, 32, 933. (354) Banerjee, A.; Chisti, Y.; Banerjee, U. C. Biotechnol. Adv. 2004, 22, 287. (355) Karthikeyan, G.; Math, R. S.; Mathew, N.; Shankar, B.; Kalaivani, M.; Singh, S.; Bahl, V. K.; Hirsh, J.; Eikelboom, J. W. Circulation 2009, 120, 1108. (356) Torchilin, V. P.; Voronkov, J. I.; Mazoev, A. V. Ter. Arkh. 1982, 54, 21. (357) Douer, D. Best. Pract. Res., Clin. Haematol. 2008, 21, 647.

(358) Schiﬀer, C. A. Acute lymphoblastic leukemia in adults. In Encyclopedia of Cancer, 2nd ed.; Bertino, J. R., Ed.; Elsevier Science Ltd.: Amsterdam, 2002; Vol. 1, pp 1−10. (359) Wileman, T.; Bennett, M.; Lilleymann, J. J. Pharm. Pharmacol. 1983, 35, 762. (360) Wileman, T. E. Adv. Drug Delivery Rev. 1991, 6, 167. (361) Cao, S. G.; Tian, J.; Zhao, Q. Y.; Ding, Z. T.; Wang, S. Q.; Cheng, Y. H. Acta Pharm. Sin. 1990, 25, 732. (362) Wang, Y. D.; Guo, L.; Qian, S. J.; Meng, G. Z.; Zhang, S. Z. Chin. J. Biotechnol. 2000, 16, 517. (363) Perry, J. J. P.; Shin, D. S.; Getzoff, E. D.; Tainer, J. A. Biochim. Biophys. Acta 2010, 1804, 245. (364) Veronese, F. M.; Caliceti, P.; Schiavon, O.; Sergi, M. Adv. Drug Delivery Rev. 2002, 54, 587. (365) Yasui, K.; Baba, A. Inflammation Res. 2006, 55, 359. (366) Nakaoka, R.; Tabata, Y.; Yamaoka, T.; Ikada, Y. J. Controlled Release 1997, 46, 253. (367) Kojima, Y.; Akaike, T.; Sato, K.; Maeda, H.; Hirano, T. J. Bioact. Compat. Polym. 1996, 11, 169. (368) Takakura, Y.; Mahato, R. I.; Nishikawa, M.; Hashida, M. Adv. Drug Delivery Rev. 1996, 19, 377. (369) Mehvar, R. J. Pharm. Pharm. Sci. 2000, 3, 125. (370) Tsubokawa, T.; Jadhav, V.; Solaroglu, I.; Shiokawa, Y.; Konishi, Y.; Zhang, J. H. Stroke 2007, 38, 1057. (371) Opanasopit, P.; Shirashi, K.; Nishikawa, M.; Yamashita, F.; Takakura, Y.; Hashida, M. Am. J. Physiol.: Gastrointest. Liver Physiol. 2001, 280, G879. (372) Nishikawa, M.; Hirabayashi, H.; Takakura, Y.; Hashida, M. Pharm. Res. 1995, 12, 209. (373) Mizoe, A.; Kondo, S.; Azuma, T.; Fujioka, H.; Tanaka, K.; Hashida, M.; Kanematsu, T. J. Surg. Res. 1997, 73, 160. (374) Takakura, Y.; Mihara, K.; Hashida, M. J. Controlled Release 1994, 28, 111. (375) Takakura, Y.; Masuda, S.; Tokuda, H.; Nishikawa, M.; Hashida, M. Biochem. Pharmacol. 1994, 47, 853. (376) Petrone, W. F.; English, D. K.; Wong, K.; McCord, J. M. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 1159. (377) Miyata, K.; Nakagawa, Y.; Nakamura, M.; Ito, T.; Sugo, K.; Fujita, T.; Tomoda, K. Agric. Biol. Chem. 1988, 52, 1575. (378) Malavaki, C.; Mizumoto, S.; Karamanos, N.; Sugahara, K. Connect. Tissue Res. 2008, 49, 133. (379) Maksimenko, A. V. Pharm. Chem. J. 2007, 41, 235. (380) Maksimenko, A. V.; Vavaev, A. V.; Tischenko, E. G. J. Biotechnol. 2008, 136, S87. (381) Miki, Y.; Teramura, T.; Tomiyama, T.; Onodera, Y.; Matsuoka, T.; Fukuda, K.; Hamanishi, C. Inflammation Res. 2010, 59, 471. (382) Hemberger, J.; Orlando, M.; Sommermeyer, K.; Eichner, W.; Frie, S.; Lutterbeck, K.; Jungheinrich, C.; Scharpf, R. U.S. Patent 0233847A1, 2009. (383) Nishikawa, M.; Tamada, A.; Kumai, H.; Yamashita, F.; Hashida, M. Int. J. Cancer 2002, 99, 474. (384) Astériou, T.; Vincent, J. C.; Tranchepain, F.; Deschrevel, B. Matrix Biol. 2006, 25, 166. (385) Girish, K. S.; Kemparaju, K. Life Sci. 2007, 80, 1921. (386) Maksimenko, A. V.; Arkhipova, O. G.; Yaglov, V. V.; Torchilin, V. P. Aldehyde dextran modiﬁed enzymes for medical application. In ̀ Enzyme Technologies, Progress in Biotechnology; Blazej, Interbiotech 87, A., Zemek, J. Eds.; Elsevier Science Ltd.: Amsterdam, 1988; pp 509− 522. (387) Maksimenko, A. V.; Bezrukavnikova, L. M.; Grigorieva, E. L.; Petrov, A. D.; Tischenko, E. G.; Yaglov, V. V.; Torchilin, V. P. Antiﬁbrosis eﬀect of enzyme compositions (superoxide dismutase, catalase, hyaluronidase) in experimental silicosis. In Biochemical Engineering for 2001, Proceeding of Asia-Paciﬁc Biochemical Engineering Conference; Furusaki, S., Endo, I., Matsuno, R., Eds.; Springer Verlag: Berlin, 2001; pp 646−648. (388) Maksimenko, A. V.; Petrova, M. L.; Tishchenko, E. G.; Shchechilina, Y. V. Russ. J. Bioorg. Chem. 2000, 26, 322. 4916

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Chemical Reviews

Review

(389) Maksimenko, A. V.; Turashev, A. D.; But, A. S.; Tischenko, E. G.; Trifonov, E. V.; Nurminski, E. A.; Likhatskaya, G. N. J. Biotechnol. 2008, 136, S202. (390) Garay, R. P.; El-Gewely, M. R.; Labaune, J. P.; Richette, P. Jt., Bone, Spine 2012, 79, 237. (391) Stamp, L. K.; O’Donnell, J. L.; Chapman, P. T. Intern. Med. J. 2007, 37, 258. (392) Gloger, M.; Heinle, J.; Schlumberger, H. U.S. Patent 4,460,683, 1984. (393) Fujita, T.; Yasuda, Y.; Takakura, Y.; Hashida, M.; Sezaki, H. J. Controlled Release 1990, 11, 149. (394) Melton, R. G.; Wiblin, C. N.; Foster, R. L.; Sherwood, R. F. Biochem. Pharmacol. 1987, 36, 105. (395) Krackhardt, A.; Schwartz, S.; Thiel, E. Onkologie 2000, 23, 538. (396) Melton, R. G.; Wiblin, C. N.; Baskerville, A.; Foster, R. L.; Sherwood, R. F. Biochem. Pharmacol. 1987, 36, 113. (397) Chazov, E. I.; Smirnov, V. N.; Torchilin, I. M.; Moskivichev, V. P.; Grimberg, G. M.; Skuja, A. Z.; Kleiner, G. I. U.S. Patent 4,446,316, 1984. (398) Maksimenko, A. V.; Nadirashvili, L. A.; Romaschenko, A. D.; Erkomaishvili, G. S.; Abramova, V. V.; Torchilin, V. P. J. Controlled Release 1989, 10, 131. (399) Blomhoff, H. K.; Blomhoff, R.; Christensen, T. B. Biochim. Biophys. Acta 1983, 757, 202. (400) Blomhoff, H. K.; Blomhoff, R.; Christensen, T. B. Int. J. Biochem. 1984, 16, 695. (401) Blomhoff, R.; Blomhoff, H. K.; Tolleshaug, H.; Christensen, T. B.; Berg, T. Int. J. Biochem. 1985, 17, 1321. (402) Beck, M. IUBMB Life 2010, 62, 33. (403) Dicosimo, R.; McAuliffe, J.; Poulose, A. J.; Bohlmann, G. Chem. Soc. Rev. 2013, 42, 6437. (404) Gómez, L.; Ramírez, H. L.; Villalonga, M. L.; Hernández, J.; Villalonga, R. Enzyme Microb. Technol. 2006, 38, 22. (405) Gómez, L.; Ramírez, H. L.; Neira-Carrillo, A.; Villalonga, R. Bioprocess Biosyst. Eng. 2006, 28, 387. (406) Gómez, L.; Ramírez, H. L.; Villalonga, R. Prep. Biochem. Biotechnol. 2006, 36, 259. (407) Gómez, L.; Ramírez, H. L.; Cabrera, G.; Simpson, B. K.; Villalonga, R. J. Food Biochem. 2008, 32, 264. (408) Drauz, K.; Gröger, H.; May, O. Enzyme Catalysis in Organic Synthesis, 3rd ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Darmstadt, 2012. (409) Pečová, M.; Šebela, M.; Marková, Z.; Poláková, K.; Cuda, J.; Šafárová, K.; Zboril, R. Nanotechnology 2013, 24, 125102. (410) Aznar, E.; Villalonga, R.; Giménez, C.; Sancenón, F.; Marcos, M. D.; Martínez-Máñez, R.; Díez, P.; Pingarrón, J. M.; Amorós, P. Chem. Commun. 2013, 49, 6391.

4917

dx.doi.org/10.1021/cr400290x | Chem. Rev. 2014, 114, 4868−4917

Neoglycoenzymes - Chemical Reviews (ACS Publications)

Recommend Documents