Principles, Applications, and Recent Advances - Semantic Scholar

1 downloads 0 Views 9MB Size Report
Nov 17, 2015 - enzymatic conversion of UDP-N-acetylglucosamine (UDP-. GlcNAc) into N-acetylmannosamine (ManNAc), which serves as a committed ...
DOI: 10.1002/cbic.201500344

Reviews

Metabolic Remodeling of Cell-Surface Sialic Acids: Principles, Applications, and Recent Advances Bo Cheng, Ran Xie, Lu Dong, and Xing Chen*[a]

ChemBioChem 2016, 17, 11 – 27

11

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews Cell-surface sialic acids are essential in mediating a variety of physiological and pathological processes. Sialic acid chemistry and biology remain challenging to investigate, demanding new tools for probing sialylation in living systems. The metabolic glycan labeling (MGL) strategy has emerged as an invaluable chemical biology tool that enables metabolic installation

of useful functionalities into cell-surface sialoglycans by “hijacking” the sialic acid biosynthetic pathway. Here we review the principles of MGL and its applications in study and manipulation of sialic acid function, with an emphasis on recent advances.

Introduction In vertebrates, all cell surfaces are decorated with a dense layer of glycans, which are complex biopolymers made of nine kinds of monosaccharide building blocks.[1] Among these building blocks are sialic acids, a family of negatively charged nine-carbon monosaccharides that often reside at the outermost ends of glycan chains.[2, 3] N-Acetylneuraminic acid (Neu5Ac) and N-glycolylneuraminic acid (Neu5Gc) are two major sialic acids in mammals. A third sialic acid form, 2-keto-3-deoxy-d-glycerod-galacto-nonulosonic acid (KDN), exists as a minor component. In addition, various modifications such as O-acetylation, Omethylation, and sulfation may occur on the hydroxy groups, producing more than 50 chemically diverse sialic acids on cell surfaces. In the literature, sialic acid is often used either as a generic term or to refer to Neu5Ac Scheme 1. The de novo biosynthetic pathway of sialic acids. A) N-Acetylmannosamine (ManNAc) is synthesized when other sialic acid forms are from UDP-N-acetylglucosamine (UDP-GlcNAc) through the action of UDP-GlcNAc 2-epimerase. ManNAc is phosnot specifically investigated. We phorylated to yield ManNAc-6-phosphate (ManNAc-6-P), which is subsequently condensed with phosphoenolpyrfollow the same convention in uvate (PEP) to yield Nue5Ac-9-phosphate (Neu5Ac-9-P). Dephosphorylation of Neu5Ac-9-P results in Neu5Ac. Neu5Ac is transported into the nucleus, where it is activated to CMP-Neu5Ac by the CMP-Neu5Ac synthase. CMPthis review. Neu5Ac is transported into the Golgi apparatus and utilized by the sialyltransferases. The resulting sialylated glyThe generation of cell-surface coconjugates are delivered to cell surfaces. B) CMP-Neu5Gc is synthesized from CMP-Neu5Ac by CMAH. CMP-KDN sialoglycans is governed by the is synthesized from mannose. CMP-Sia refers to the nucleotide sugar donor of three sialic acid forms and is often sialic acid biosynthetic machi- used for Neu5Ac when other sialic acid forms are not specifically investigated. nery (Scheme 1 A). The de novo biosynthesis of Neu5Ac involves to ManNAc-6-phosphate, followed by condensation with phosenzymatic conversion of UDP-N-acetylglucosamine (UDPGlcNAc) into N-acetylmannosamine (ManNAc), which serves as phoenolpyruvate to yield Neu5Ac-9-phosphate. Dephosphorya committed precursor for Neu5Ac. ManNAc is phosphorylated lation results in Neu5Ac, which is converted into the nucleotide sugar donor CMP-Neu5Ac in the nucleus. In most mammals, CMP-Neu5Ac can also be converted into CMP-Neu5Gc by [a] B. Cheng, Dr. R. Xie, L. Dong, Prof. Dr. X. Chen CMP-N-acetylneuraminic acid hydroxylase (CMAH, Scheme 1 B). Beijing National Laboratory for Molecular Sciences College of Chemistry and Molecular Engineering Humans lack the capability for de novo synthesis of Neu5Gc Synthetic and Functional Biomolecules Center due to the loss of CMAH. CMP-KDN is derived from KDN, and Peking-Tsinghua Center for Life Sciences, Peking University which is synthesized from mannose (Scheme 1 B). CMP-sialic Beijing 100871 (China) E-mail: [email protected] acid is transported into the Golgi compartment, in which it ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

12

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews serves as the substrate for sialylation catalyzed by sialyltransferases. The resulting sialoglycoconjugates, such as sialylated proteins and lipids, are then delivered to the cell surfaces. Cell-surface sialic acids play an important role in diverse biological and pathological processes. For example, cell-surface

sialoglycans serve as ligands for sialic-acid-recognizing proteins,[4] mediating leukocyte homing,[5] immune responses,[6] cell–cell recognition,[7] and pathogen invasion.[8] In contrast to displaying ligands, sialic acids can also mask the underlying glycan epitopes, therefore preventing protein binding.[9] Sialylation has been implicated in fertilization,[10] embryonic development,[11] brain development and cognition,[12] and cardiac function.[13] Hypersialylation has commonly been found in tumor cells and is believed to correlate with metastatic potential in a number of cancer types.[14] The cellular level of sialoglycans is dynamically regulated by the expression and activity of enzymes involved in sialoglycan biosynthesis, such as sialyltransferases.[15] Furthermore, cell-surface sialylation can be locally regulated by post-Golgi glycosidases such as sialidases.[16–18] In addition, dynamic spatial organization of sialoglycans forms microdomains and clusters on cell surfaces.[19] Despite the involvement of sialoglycans in so many important biological processes, elucidating the dynamics and function of sialylation remains a challenge. Historically, genetic mutation or overexpression of sialyltransferases has been used to alter the structure and expression levels of sialoglycans and to study the functional consequences.[20–22] However, glycans are not genetically encoded, so genetic manipulations of glycosyltransferases only provide indirect perturbation of the sialoglycan structure and expression level. Furthermore, the redundancy of sialyltransferases often complicates the genetic studies.[15] On the other hand, glycan labeling or tagging is an essential tool for detection and visualization of sialoglycans in living systems and for enriching and profiling sialylated glycoproteins. Sialoglycan-recognizing lectins or antibodies have long been employed for sialoglycan imaging and glycoproteomic analysis.[23–25] The lectin- and antibody-based approaches, however, suffer from low affinity, toxicity, and low tissue penetration. Alternatively, cell-surface sialylation can be perturbed or labeled by “hijacking” the sialic acid biosynthetic pathway to incorporate unnatural sialic acid analogues into cell-surface sialoglycans (Figure 1). The enzymes involved in the sialoglycan

Bo Cheng received his bachelor’s degree in chemistry from Huazhong Agricultural University in 2011. He is currently a fifth-year graduate student in Prof. Xing Chen’s group at Peking University, and is working on developing new sialic acid chemical reporters.

Ran Xie received his bachelor’s degree in polymer chemistry from Shanghai Jiaotong University in 2010 and his Ph.D. from Peking University in 2015. During his graduate study with Prof. Xing Chen, Dr. Xie developed a liposome-assisted strategy for cell-selective and tissue-specific metabolic glycan labeling. He is currently a postdoctoral researcher at Harvard University. Lu Dong received his bachelor’s degree in chemistry from Peking University in 2012, and is currently a fourth-year graduate student in Prof. Xing Chen’s lab. He is studying the biological function of glycosylation in stem cells.

Dr. Xing Chen completed his undergraduate degree in chemistry at Tsinghua University in 2002. Dr. Chen then obtained his Ph.D. in chemistry from the University of California, Berkeley in 2007, where his research focused on chemical biology and bionanotechnology. After completing postdoctoral work at Harvard Medical School in the field of structural biology and immunology, he joined the Peking University faculty in September 2010. Dr. Chen’s current research interest is focused on chemical glycobiology.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

Figure 1. MGL (or MOE) of cell-surface sialic acids. Unnatural sugars containing a chemical functional group X, either ManNAc analogues or sialic acid analogues, are taken up by cells through diffusion, enter the sialic acid biosynthetic pathway, and are incorporated into cell-surface sialoglycans. Unnatural sugars can also be delivered into cells with the aid of liposomes, which undergo receptor-mediated endocytosis.

13

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews biosynthesis are promiscuous for unnatural sialic acid analogues with certain subtle chemical modifications (i.e., unnatural functionalities), and this has been exploited to introduce new functional moieties metabolically into sialic acids within cell-surface sialoglycans. The technique is termed metabolic oligosaccharide engineering (MOE) or metabolic glycan labeling (MGL), particularly in conjunction with bioorthogonal labeling methods. MOE and MGL have been used interchangeably in the literature. Since its first introduction by the Reutter group in 1992,[26] MGL has emerged as a powerful tool for probing sialylation.[27, 28] Furthermore, the MGL strategy has been expanded to study other types of glycosylation including mucin-type O-linked glycosylation, fucosylation, and O-GlcNAcylation.[29, 30] In this review we focus on the principles and applications of MGL of cell-surface sialoglycans, with an emphasis on recent advances.

mediates is much more challenging. ManNAc is the first committed intermediate for the de novo biosynthesis of sialic acids. Unnatural functionalities are commonly installed at the N-acyl position of ManNAc. The corresponding sialic acid analogues containing functionalities at the N-acyl position can also be used. The sialic acid analogues bypass several enzymatic reactions including the rate-limiting step, the phosphorylation of ManNAc,[32] and therefore enable more versatile choices of the unnatural functionalities. In addition, modifications at positions other than N-acyl, such as C-9 of sialic acids, can be explored. 1.3. Cellular uptake of unnatural sugars Monosaccharides or their unnatural analogues are generally hydrophilic and usually diffuse through the plasma membrane with modest efficiency (Figure 1). When free unnatural ManNAc analogues are used, concentrations in the millimolar range are usually needed to ensure efficient metabolic incorporation. Ester derivatization has been employed to increase lipophilicity and thus facilitate the cellular uptake of unnatural sugars.[33–36] As an example, peracetylation of ManNAc analogues greatly enhances the uptake efficiency. Once the esters are inside the cells they are hydrolyzed by nonspecific esterases and liberate free sugars. Interestingly, ester derivatization of sialic acid analogues improves the uptake efficiency, but to a lesser extent than in that of ManNAc analogues.[35] One of the possible reasons for this is that free sialic acid analogues may be taken up by the cells through an active uptake mechanism, it having been shown that Neu5Gc uptake is through pinocytic/endocytic pathways.[37] Cellular uptake of unnatural sugars can also be facilitated and modulated by liposomes (Figure 1). Through the use of ligand-targeted liposomes encapsulating unnatural sugars, the sialic acid analogues are delivered into the cells with high efficiency by receptor-mediated endocytosis of liposomes. This strategy not only provides an alternative cellular uptake mechanism with enhanced efficiency, but also enables cell-selective and tissue-targeted metabolic labeling of sialoglycans in vivo.[38, 39]

1. Principles of Metabolic Glycan Labeling of Sialic Acids 1.1. Overview The concept of MOE (or MGL) hinges on the substrate promiscuity of the sialic acid biosynthetic pathway (Figure 1). An unnatural analogue of the precursor or intermediate of sialic acid biosynthesis is taken up by the cell and intercepts the biosynthetic pathway. The biosynthetic enzymes tolerate the subtle chemical modification on the analogue and install it into cellular sialoglycans in the same way as its natural counterpart. The outcome is the display of an unnatural functionality (¢X) on cell-surface sialic acids. The incorporated unnatural functionality can be exploited to modulate the properties of cell-surface sialoglycans and the underlying cellular processes such as receptor binding, cell growth, and signal transduction. The scope of unnatural functionalities that can be introduced into cell-surface sialoglycans is greatly expanded when the chosen unnatural functionality (¢X) is a bioorthogonal functional group, also termed a bioorthogonal chemical reporter. A bioorthogonal chemical reporter is a small functional group that is not found in nature, is essentially inert in biological systems, and is selectively reactive with a complementary functional group in living systems with high efficiency and low cytotoxicity.[31] Bioorthogonal chemistry enables the installation of various functionalities, such as imaging probes and affinity tags, onto sialoglycans. In principle, any functionality of interest can be introduced onto sialoglycans in the second step by using the two-step chemical reporter strategy.

1.4. Scope of the metabolizable unnatural functional groups The unnatural sugar analogues must traverse the sialic acid biosynthetic pathway from their entry point. The chemical modifications need to be subtle so that all the enzymes involved may tolerate the unnatural functionalities. In general, increasing the size or steric bulk of the unnatural functional groups decreases the metabolism and incorporation efficiency. Because converting ManNAc into ManNAc-6-phosphate is a rate-limiting step, sialic acid analogues are capable of incorporating functional groups with a bigger size by bypassing the bottleneck enzyme.[32] As a rule, functional groups with no branching are well accommodated by ManNAc analogues, whereas sialic acid analogues can deliver bulkier functional groups. Notably, exceptions to this rule, such as ManNAc ana-

1.2. Entry point to the sialic acid biosynthetic pathway Two kinds of analogues—unnatural ManNAc and unnatural Neu5Ac—have been developed to allow the introduction of unnatural functionalities into the sialic acid biosynthetic pathway. Other intermediates are either phosphorylated compounds or the nucleotide sugar. Practically, these compounds cannot enter mammalian cells due to their negative charge. Furthermore, chemical synthesis of derivatives of these interChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

14

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews logues with branching or bulky groups, have been seen, but often show low incorporation efficiency.[32, 40]

cuity of the sialic acid biosynthetic pathway for remodeling cell-surface glycans was then demonstrated by the Reutter group with N-propanoylmannosamine (ManNProp), a ManNAc analogue with an elongated N-acyl group.[26] Thereafter, a series of ManNAc and Neu5Ac analogues containing alkyl chains,[26, 43–45] fluorinated alkyl groups,[46, 47] the iodo group,[47] sulfo-containing groups,[47] and aryl groups[44, 48] were synthesized and evaluated for metabolic incorporation (Table 1). In this section we focus on these ManNAc and Neu5Ac analogues containing unreactive functionalities, which are categorized as the first-generation unnatural sugars so as to differentiate them from those containing reactive functionalities described in the next section. Metabolic engineering of the chemical structures of cell-surface sialic acids with these unreactive functional groups has been exploited to modulate the biological function of sialoglycans and the underlying biological processes. Cell-surface sialoglycans serve as host receptors to initiate infection by certain viruses.[8] Therefore, it has been of great interest to modulate virus binding and infection by metabolically remodeling the chemical structures of sialoglycans of host cells. Pawlita and co-workers, for example, displayed unnatural sialic acids with elongated alkyl chains of three to five carbon atoms at the N-acyl position on the surfaces of BJA-B and Vero cells, using ManNProp, ManNBut, and ManNPent to modulate the infection by two polyoma viruses: B-lymphotropic papovavirus (LPV) and human polyoma virus BK (BKV).[43] All three analogues decreased the binding of LPV to BJA-B cells. For binding of BKV to Vero cells, elongation by one methylene group, from N-acetyl to N-propanoyl, resulted in sevenfold enhancement of binding, whereas further elongation (ManNPent) reduced infection and produced an inhibitory effect, thus indicating that the N-acyl side chain acts as an important determinant and that substitution at this position can modulate binding affinities of virus particles. A follow-up study by the same group employed a nonradioactive virus binding assay to quantify the receptors on 3T6 mouse fibroblast for murine polyomavirus.[49] Treating 3T6 cells with ManNProp, ManNBut, and ManNPent resulted in the incorporation of about one-third of the unnatural sialic acids, and the number of the natural receptors was decreased. Molecular modeling revealed that elongation of the N-acyl side chain leads to steric hindrance, thus abolishing virus binding. Interestingly, metabolic incorporation of 9iodo-Neu5Ac, 9-deoxy-Neu5Ac, and Neu5FAc in BJA-B cells was shown to enhance susceptibility to infection by LPV.[47] For influenza A virus, binding and infection in MDCK-II cells was inhibited by ManNProp, ManNBut, and ManNPent, with efficiencies of up to 80 %.[50] Similarly, the etiologic agent of Chagas disease—Trypanosoma cruzi, an infectious typomastigote—adheres to host cells in a sialic-acid-dependent manner. Lieke et al. showed that metabolic incorporation of ManNAc analogues bearing elongated N-acyl side chains reduced infection by T. cruzi.[51] Cell-surface sialic acids are implicated in cell adhesion. Sialyl Lewis X (sLeX), for example, is a sialic-acid-containing tetrasaccharide epitope and is expressed on leukocytes. Binding of sLeX to selectins plays an important role in mediating leuko-

2. MGL with Unreactive Functionalities From a historical perspective, the introduction of unnatural functionality into cell-surface sialoglycans can probably be traced back to “failed” efforts to inhibit sialylation with the fluorinated ManNAc analogue N-trifluoroacetylmannosamine (ManNF3Ac, Table 1). This instead resulted in the corresponding unnatural sialic acids metabolically installed into cell-surface glycans.[41, 42] The feasibility of exploiting the substrate promis-

Table 1. Compilation of representative ManNAc and Neu5Ac analogues containing unreactive functionalities.

ManNAc analogues

Name

Substituents

ManNAc[a] ManNGc[a,b] ManNProp[a]

R1 = ¢CH3 ¢CH2OH ¢CH2CH3

Ref(s).[c]

natural sugar [57]

[26, 32, 43, 44, 48–51], [53, 55, 56, 59–65], [67–71, 72, 75]

precursor for sialic acid biosynthesis

ManNBut[a]

¢(CH2)2CH3

ManNPent[a]

¢(CH2)3CH3

ManNHex[a] ManNiBu

¢(CH2)4CH3 ¢CH(CH3)2

Neu5Ac analogues

metabolic replacement of sialic acid

[59, 67–70] [43, 32, 49–51, 59], [68, 69, 72] [32] [44] [51]

cyclic ManNProp [a,b]

[43, 32, 44, 48–51],

ManNF3Ac ManNF3Prop[a,b] ManNF3But[a,b] ManNBz ManNPhAc

¢CF3 ¢CH2CF3 ¢(CH2)2CF3 ¢Ph ¢CH2Ph

Neu5Ac Neu5Gc

R2/R3 = ¢COCH3/¢OH ¢COCH2OH/¢OH

Analogues in which R3 =¢OH, R2 = Neu5FAc ¢COCH2F Neu5F3Pent ¢CO(CH2)3CF3 Neu5F7Pent ¢CO(CH2)2CF2CF3 5-N-thioac-Neu ¢CSCH3 Neu5iPent ¢COCH2CH(CH3)2 Neu5iHex ¢CO(CH2)2CH(CH3)2 Analogues in which R2 =¢COCH3, R3 = 9-deoxy¢H Neu5Ac 9-iodo-Neu5Ac ¢I 9-SCH3-Neu5Ac ¢SCH3

[42, 41, 47] [46] [46] [48] [44, 78–79]

natural sugar natural sugar

[47] [46] [46] [47] [54] [54]

[47]

[47] [47]

[a] Peracetylated ManNAc analogues (where all hydroxy groups are ¢COCH3) with these R1 groups also have been reported. [b] Neu5Ac analogues [although not shown in the second part of this table (Neu5Ac analogues)] with these R1 groups have been reported. [c] The first reported synthesis of the corresponding compound is listed in the first reference.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

15

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews cyte trafficking.[52] Horstkorte et al. reported that incubation of the myeloid leukemia cell line HL60 with ManNProp increased cell binding to selectins, presumably due to the incorporation of SiaNProp into sLeX.[53] Interestingly, Kumar and co-workers found that displaying fluorinated sialic acids decreased cell adhesion to fibronectin- and selectin-coated surfaces.[46, 54] In addition, cell-surface adhesion receptors such as integrins and CEACAM1 are sialylated glycoproteins. Reutter and co-workers increased the half-life of CEACAM1, which mediates hemophilic cell–cell adhesion, by metabolically incorporating SiaNProp into sialoglycans through the use of ManNProp.[55] In contrast, the stability of integrin a1, which mediates cell–matrix interactions, is not affected. Interestingly, Villavicencio-Lorini et al. showed that metabolic engineering of sialoglycans in HL60 cells activated b1-integrin and resulted in an increased adhesion to fibronectin.[56] In addition, peracetylated N-glycolylmannosamine (Ac5ManNGc) can serve as a ManNAc analogue, being metabolically converted into Neu5Gc and incorporated into cell-surface sialoglycans. The resulting Neu5Gc-containing sialoglycans on NG108-15 cells abrogate the binding of myelin-associated glycoprotein (MAG), a sialic-acid-binding lectin.[57] Metabolic remodeling of cell-surface sialoglycans has been exploited to regulate cell growth and proliferation, because sialylation is essential for development and regeneration.[11, 12, 58] ManNProp, ManNBut, or ManNPent were used by Reutter and co-workers to remodel the surfaces of human diploid fibroblasts; this attenuated density-dependent inhibition of cell growth, thus suggesting that sialylation is important in regulating cell–cell contact-dependent growth.[59] Furthermore, Schmidt et al. applied ManNProp to cultures of various cells from rat brains, and showed that proliferation of astrocytes and microglia was stimulated owing to the metabolic incorporation of SiaNProp, but the proliferation of oligodendrocyte progenitor cells was not affected.[60] In addition, the number of oligodendrocyte progenitor cells expressing the early oligodendroglial surface marker A2B5 epitope was increased by ManNProp treatment. The incorporation of SiaNProp into cellsurface sialoglycans was also found to stimulate neurite outgrowth in PC12 cells and cerebellar neurons.[61] Other cell types can also be probed by MGL. For example, the proliferation rate of human peripheral blood mononuclear cells was significantly increased by treatment with ManNProp.[62] Many cell-surface proteins are sialylated glycoproteins. Modulating sialoglycans with unnatural functionalities can alter the function of sialylated receptors and the underlying signal transduction. Schmidt et al. demonstrated that metabolic remodeling of sialoglycans in oligodendrocytes by use of ManNProp-induced GABA-dependent (GABA: g-aminobutyric acid) calcium oscillation.[63] Because ionotropic GABA receptors are sialylated glycoproteins, it is possible that the calcium signaling modulation was induced by the metabolic incorporation of SiaNProp into GABA receptor glycans. However, ManNProp metabolically alters the sialoglycans of various proteins, so direct characterization of the GABA receptor glycans and the functional consequences will be interesting to pursue. Similarly, treatment of HL60 cells with ManNProp resulted in an inChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

creased release of intracellular calcium after application of thapsigargin (an inhibitor of SERCA Ca2 + -ATPases), but the molecular mechanism of the observed response is not completely clear.[64] Furthermore, the incorporation profile of SiaNProp into glycans and whether ManNProp metabolism alters the glycome should be considered. An interesting work by the Valmu group provides structural information on the N-glycome of ManNProp-treated mesenchymal stromal cells based on mass spectrometric analysis, showing that the N-glycome profile was altered and that number of glycans carrying multiple fucose residues was increased.[65] As well as terminating cell-surface glycan chains through a(2,3) and a(2,6) linkages, sialic acids also form polysialic acid (PSA), a homopolymer of a(2,8)-linked sialic acid. The neural cell adhesion molecule (NCAM) is the major polysialylated protein in the brain, and PSA-NCAM plays an important role in nervous system development.[66] Various ManNAc analogues have been examined from the point of view of metabolically modulating PSA biosynthesis. The Bertozzi group reported that treating NT2 neurons with ManNBut resulted in abolished staining with the monoclonal antibody (mAb) to PSA (12F8), whereas ManNProp did not inhibit 12F8 binding.[67] Furthermore, when HeLa cells stably expressing NCAM and the polysialyltransferase STX were treated with ManNBut and used as substratum cells, sensory neurons from embryonic click dorsal root ganglia exhibited reduced neurite outgrowth.[68] In vitro PSA biosynthesis assay showed that polysialyltransferases STX and PST use CMP-SiaNBut less efficiently that CMP-SiaNProp, and both are polymerized more slowly than natural CMP-Sia. These results seemed to suggest that PSA biosynthesis is altered in ManNBut-treated cells by affecting the polysialyltransferase activity. Gerardy-Schahn and co-workers later investigated the effects of ManNProp, ManNBut, and ManNPent on the PSA biosynthesis in several cell lines, including PC12, HL60, and NT2 cells. They found that ST8SiaII was significantly inhibited by unnatural substitutions on the N-acyl position of Sia, whereas the activity of ST8SiaIV was not significantly inhibited.[69] Using a mAb specifically recognizing PSA with SiaNBut or SiaNProp, Jennings and co-workers clearly showed that both ManNBut and ManNProp can be metabolized, resulting in the unnatural PSA in NT2 cells, and that the endogenous PSA expression is lowered but not completely inhibited.[70] In addition, Gagiannis et al. injected ManNProp into living mice, which resulted in 1 % incorporation of SiaNProp in the brain and a decrease in PSA on NCAM.[71] Gnanapragassam et al. recently showed that human neuroblastoma SH-SY5Y cells treated with ManNProp or ManNPent completely lost cell-surface PSA and that cell migration and invasion ability was significantly reduced.[72] These studies highlight the potential utility of MGL of sialic acids as a powerful means to regulate polysialylation. Many of the cancer-associated glycans are sialylated and have emerged as attractive targets for cancer immunotherapy.[14, 73] One promising strategy is metabolic remodeling of the sialylated tumor antigens with unnatural functional groups, thus generating unique antigens in vivo. For immunotherapy, the unnatural antigens are used as vaccines to generate anti16

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews bodies specific for tumors bearing the same unnatural antigens.[74] Jennings and co-workers demonstrated that incubation of tumor cells with ManNProp resulted in the modified PSA, which can be targeted with a mAb specific for the unnatural PSA to induce cell death.[75] The same strategy was then extended to engineer the GD3 ganglioside, and the modified GD3-specific mAb was found to protect ManNAc-analogue-treated mice effectively from melanoma tumor grafting.[48] The Bertozzi group exploited a ManNAc analogue containing a ketone group (see also the next section for its use as a bioorthogonal chemical reporter) to generate the unnatural glycan antigen.[76] Guo and co-workers sought to generate unnatural glycan antigens based on GM3 and sTn antigens and evaluated four ManNAc analogues (ManNProp, ManNBut, ManNiBu, ManNPhAc).[44, 77] ManNPhAc was found to generate unnatural antigens effectively in cancer cells. The unnatural-antigen-specific antibody was generated by using a synthetic vaccine made of GM3NPhAc and KLH in mice and exhibited strong cytotoxicity to ManNPhAc-treated cancer cells. It was then demonstrated that vaccination with a conjugate vaccine made of GM3NPhAc, followed by ManNPhAc treatment, resulted in tumor growth inhibition and prolonged survival of tumor-bearing mice.[78, 79] Recently, Wong and co-workers developed a Globo H analogue bearing an azide for vaccination in mice. Interestingly, the azido-Globo H elicited a strong IgG immune response, and the resulting antibodies were shown to recognize both azido-Globo H and the unmodified Globo H antigen.[80]

physiological pH. Recently, it has been demonstrated that oxime ligation can be dramatically accelerated by addition of aniline as a nucleophilic catalyst.[87, 88] By exploiting the improved aldehyde condensation reaction, Paulson and co-workers developed an efficient method for labeling sialoglycans on live cells, which were metabolically incorporated with 9-benzaldehyde sialic acid (9BA-Neu5Ac, Scheme 2), an aldehyde-modified sialic acid analogue.[89] Notably, aldehydes can also be conjugated by use of a recently developed Pictet–Spengler ligation, which offers improved kinetics but has not been tested on glycans.[90] 3.2. Azides Another drawback of aldehydes and ketones is their natural existence inside cells, so they are not truly bioorthogonal. To overcome these limitations, Bertozzi and co-workers developed Staudinger ligation between azides and triarylphosphines, which was the first bioorthogonal reaction per se.[91] The azide is small in size and chemically stable and does not naturally exist in cells. N-Azidoacetylmannosamine (ManNAz, Scheme 2, tetra-O-acetyl derivative shown) is metabolically converted into the corresponding N-acetylneuraminic acid (5Az-Neu5Ac or SiaNAz), which is incorporated into cell-surface sialoglycans. Cell-surface azides can then be treated with phosphine probes for detection by flow cytometry and for imaging by fluorescence microscopy in live cells.[91, 92] MGL of cell-surface sialoglycans with Ac4ManNAz and subsequent conjugation with phosphine probes through Staudinger ligation was later successfully demonstrated in living mice.[93] In that study, although Staudinger ligation was performed in living animals for the first time, fluorescent detection of the cell-surface azides was done ex vivo. The main obstacle for in vivo imaging of sialoglycans is the somewhat slow kinetics of Staudinger ligation, which has a typical second-order rate constant of 0.002 m¢1 s¢1. The requirement for high concentrations of phosphine probes results in high background signal, which frustrates efforts to achieve in vivo labeling and imaging. In addition, phosphine reagents are slowly oxidized in biological systems. These limitations promoted the development of other bioorthogonal reactions with better kinetics.[31, 94] The most popular bioorthogonal reactions to date have been a class of reactions based on 1,3-dipolar cycloaddition between azides and alkynes. The Sharpless group and the Meldal group independently reported CuI-catalyzed azide– alkyne cycloaddition (CuAAC), based on the classic Huisgen azide–alkyne cycloaddition.[95, 96] Exhibiting fast reaction kinetics and high compatibility with a variety of functional groups, CuAAC is also commonly known as click chemistry. It is almost bioorthogonal, except that the CuI catalyst is toxic to cells. Two strategies have been used to circumvent the CuI toxicity and improve the biocompatibility of click chemistry. A panel of CuI ligands that increase the reaction rates and decrease cytotoxicity have been developed by the Finn group and the Wu group.[97–99] Alternatively, Bertozzi and co-workers developed strain-promoted azide–alkyne cycloaddition (SPAAC) as a copper-free version of click chemistry, in which strained cyclo-

3. MGL with Bioorthogonal Chemical Reporters 3.1. Ketones and aldehydes Building on Reutter’s finding that the cellular machinery of sialoglycan biosynthesis can tolerate alkyl substitutions on sialic acids,[26, 43] the Bertozzi group expanded the scope of MOE (or MGL) by introducing the ketone group into cell-surface sialoglycans in 1997.[81] N-Levulinoylmannosamine (ManLev, Scheme 2) was synthesized and metabolically incorporated. The resulting cell-surface SiaLev was treated with hydrazidebiotin and detected with avidin-fluorophore conjugates. This work set the foundation for chemical remodeling of cell-surface sialoglycans with exogenous reagents by coupling metabolic labeling with bioorthogonal chemistry. Chemoselective ligation between ketones (or aldehydes) and hydrazides (or aminoxy groups) represents an early developed bioconjugation reaction that has been broadly classified as bioorthogonal chemistry.[82] Various functional agents can be installed onto cell surfaces by this two-step strategy. ManLev-treated cells, for example, have been treated with aminoxy- and hydrazine-functionalized glycans to remodel cell-surface glycan epitopes for lectin binding,[83] with aminoxy-bearing magnetic resonance contrast reagents,[84] with hydrazine-bearing polymeric nanoparticles,[85] and with hydrazine-bearing receptors for conjugating gene transfer vectors.[86] One of the limitations of ketone or aldehyde condensation is the relatively slow reaction rates at ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

17

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews

Scheme 2. Chemical structures of representative ManNAc and Neu5Ac analogues with reactive functionalities.

octynes react with azides without the need for a CuI catalyst.[100] Subsequently, a variety of cyclooctynes offering finely tunable reactivity, solubility, and stability have been synthesized.[101, 102] The advances in bioorthogonal chemistry based on azide–alkyne cycloaddition have enabled labeling of sialoglycans in living systems for various applications. The bioorthogonal chemistry has also been used to label other glycans as well as other biomolecules,[103] but this is not covered in this review. Cell-surface sialoglycans are attractive targets for molecular imaging because of their dynamic changes in response to the ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

cell’s physiological state. At the cellular level, the azidosugarincorporated cells are treated with an alkyne-bearing probe, followed by imaging by fluorescence microscopy (Figure 2 A). By use of difluorinated cyclooctyne conjugated to Alexa Fluor 488 (DIFO-AF488), sialoglycans on cells treated with Ac4ManNAz were fluorescently labeled within minutes on live cells, and time-lapse fluorescence microscopy revealed the internalization kinetics of sialoglycoconjugates.[104] The internalization half-life of a subset of cell-surface sialoglycoconjugates was determined to be as short as … 15 min in CHO cells. 18

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews

Figure 2. Metabolic labeling of sialoglycans with azides, coupled with bioorthogonal chemistry, enables visualization and glycoproteomic identification of cellsurface sialoglycans. A) Schematic representation of the experimental processes. Cells are incubated with azide-bearing ManNAc analogues or Neu5Ac analogues, such as Ac4ManNAz or 9Az-Neu5Ac, resulting in metabolic incorporation of azides into cell-surface sialoglycans. The cell-surface azides are subsequently treated with an alkyne-bearing fluorophore for fluorescence imaging or with an alkyne-bearing affinity tag for enrichment and proteomic identification. B) Confocal fluorescence microscopy imaging of cell-surface sialylated glycans metabolically labeled with azides (green) in HaCaT cells during and after EMT. F-Actin (red) was visualized by use of rhodamine-phalloidin, and the nuclei were visualized by staining with Hoechst 33 342 (blue). Scale bars: 20 mm. C) Relative changes in identified sialylated proteins between the 72 and 0 h groups. HaCaT cells fed with 50 mm Ac4ManNAz for 48 h were treated with 100 pm TGF-b for 0, 24, and 72 h. The cell-surface azides were treated with alkyne-biotin. The cells were lysed, and biotin-labeled sialylated proteins were pulled down and subjected to MS analysis. Spectral counting was used to access the relative abundances of individual glycoproteins. Hits from two sets of conditions were combined for this analysis. The proteins were sorted by the maximum spectral counts of the two groups (x-axis). The insert shows overlap of identified sialylated proteins from three groups. Reprinted and modified, with permission, from ref. [107]. Copyright: The American Society for Biochemistry and Molecular Biology, 2015.

Super-resolution fluorescence microscopy has also been applied to image cell-surface sialoglycans by the Sauer group.[105] Ac4ManNAz and CuAAC were used to label cell-surface sialoglycans, and direct stochastic optical reconstruction microscopy (dSTORM) revealed a homogeneous distribution of cell-surface sialic acids. MGL coupled with click chemistry has also been explored for imaging modalities other than fluorescence imaging. Witte et al., for example, recently demonstrated the application of Ac4ManNAz in conjugation with copper-free click chemistry for cellular imaging sialoglycans by xenon MRI.[106] Dynamic changes in sialylation during important cellular processes can also be imaged and monitored by using this strategy. For example, our group recently applied MGL with Ac4ManNAz and CuAAC assisted by the BTTAA ligand to visualize dynamic sialylation in epithelial–mesenchymal transition (EMT), a fundamental process in embryonic development and organ formation.[107] It was discovered that sialylation underChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

goes a downregulation during EMT and is then reverted and upregulated in the mesenchymal state after EMT (Figure 2 B). Granell and colleagues imaged azide-modified sialoglycans in cultured Drosophila third instar larvae CNS neurons and found that the endogenously expressed sialic acid synthase DmSAS is required for sialic acid biosynthesis.[108] Kang et al. recently reported a tissue-based metabolic labeling method based on the use of Ac4ManNAz for imaging sialoglycans on cultured primary hippocampal neurons.[109] Furthermore, MGL enables enrichment and proteomic profiling of the sialylated glycoproteins (Figure 2 A). For glycoproteomics purposes, the Ac4ManNAz-treated cells were treated with an alkyne-bearing affinity probe (e.g., alkyne-biotin). The treated cells were lysed and treated with streptavidin beads, followed by gel-based proteomic identification by tandem mass spectrometry. By this chemical glycoproteomics approach, we identified a list of cellsurface sialylated proteins, the biosynthesis of which was dy19

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews namically regulated during EMT, and the relative abundances of individual sialylated glycoproteins between different time points can be quantitatively compared (Figure 2 C).[107] Similar approaches have been used by several other groups for identifying sialylated proteins in Ac4ManNAz-treated cells.[110–112] The conjugation of affinity probes can be performed either on cell surfaces or in cell lysates. This technical variation may be used to exclude intracellular membrane-associated sialoglycoproteins from or include them in the identified proteome. SiaNAz, the corresponding azido sialic acid of ManNAz, can also be employed for metabolically installing azides into sialoglycans.[35] By bypassing several enzymes, sialic acid derivatives can accommodate larger functional groups, such as aryl-azide (AAz) at the C-5 position (i.e., 5AAz-Neu5Ac or SiaNAAz, Scheme 2), which cannot be incorporated using ManNAc analogues. The azide can also be attached at other positions of ManNAc and Neu5Ac besides the N-acetyl position. Mçller et al. synthesized a ManNAc analogue bearing an azide at position C-4 [N-acetyl-(1,3,6-O-acetyl)-4-azido-4-deoxymannosamine (Ac34AzManNAc); Scheme 2] and demonstrated its metabolic incorporation into cell-surface sialoglycans.[113] Interestingly, 4AzManNAc appeared preferentially to label O-glycosylated proteins, but not N-linked glycosylated proteins. Position C-9 of sialic acids can be substituted with azides. Both 9Az-Neu5Ac and 9AAz-Neu5Ac (Scheme 2) were shown to be incorporated into cell-surface sialoglycans.[114, 115] Imaging sialoglycans in living organisms has attracted great interest. Although Staudinger ligation appeared not to be wellsuited for labeling azide-incorporating sialoglycans in mice for in vivo imaging,[93] Brindle and co-workers reported the metabolic labeling of tumor-associated sialoglycans by intraperito-

neal injection of Ac4ManNAz in a murine tumor model, followed by in vivo treatment with intraperitoneally injected phosphine-biotin.[116] The biotin was conjugated subsequently with intravenously injected NeutrAvidin-Dylight649. Wholeanimal fluorescence imaging exhibited a tumor-to-background ratio (TBR) of … 3 for Ac4ManNAz-administrated mice, in comparison with a TBR of … 2 for vehicle-treated mice. These results are consistent with the modest performance of Staudinger ligation in living animals. Copper-free click chemistry for in vivo imaging was first demonstrated by visualizing membrane-associated O-linked glycan in developing zebrafish treated with peracetylated Nacetylgalactosamine (Ac4GalNAz).[117] Subsequently, Dehnert et al. reported in vivo imaging of sialoglycan in zebrafish by means of a similar methodology.[118] Zebrafish embryos were incubated with Ac4ManNAz, followed by treatment with DIFOAF488. Fluorescent microscopy revealed the spatiotemporal dynamics of sialylation in live zebrafish embryos. In the hope of visualizing sialylation changes during cardiac hypertrophy in rodents, an adaptive response of the heart to hemodynamic overload that often progresses to cardiac dysfunction and heart failure, we developed a heart imaging strategy for fluorescence imaging of sialylation at the intact organ level, based on in vivo metabolic labeling and copper-free click chemistry (Figure 3).[13] Living rats were intraperitoneally administrated with Ac4ManNAz, followed by isolation and perfusion of the hearts on the Langendorff system in the presence of azo-dibenzocyclooctyne-Fluor 488 (DBCO-F488, Figure 3 A). The Langendorff perfusion system, an experimental setup commonly used for studying heart function in basic and preclinical research, not only simplifies the click-labeling but also permits optical imaging of the heart tissue by confocal fluorescence

Figure 3. In vivo metabolic glycan labeling enables visualization of cardiac sialoglycans in intact rat hearts. A) Living rats are injected intraperitoneally with Ac4ManNAz, which is metabolically incorporated into cell-surface sialylated glycans. The azides then serve as a chemical reporter for glycan imaging. B) Photos of Langendorff-perfused hearts from rats treated with vehicle and Ac4ManNAz followed by bioorthogonal labeling with DBCO-F488 under UV illumination. Note the green fluorescence of the Ac4ManNAz-treated heart. C) Confocal fluorescence images of azide-incorporating sialylated glycans in intact hearts that were isolated, perfused, and subsequently treated with DBCO-F488 by copper-free click chemistry. D) Zoom-in view of the SiaNAz-labeled glycans in intact hearts, showing the structure of the T-tubule system. Scale bars: 20 mm. Reprinted and modified, with permission, from ref. [13]. Copyright: The American Chemical Society, 2014.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

20

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews microscopy with high spatial resolution without the complications involved in whole-animal click reactions and imaging (Figure 3 B). Confocal fluorescence imaging on the labeled hearts revealed the distribution of sialylated glycans on cardiomyocyte cell surfaces (Figure 3 C). The imaging results also showed that the transverse tubule (T-tubule) network, the orderly spaced invaginations of cell-surface membrane that play an essential role in synchronizing cell-wide excitation and excitation-contraction coupling, contains sialoglycans and that the glycans are actively synthesized (Figure 3 D). Upon induction of cardiac hypertrophy, the imaging results revealed that the upregulation of cell-surface sialylation is associated with hypertrophy. Notably, the click-labeling step could also be performed in living rats. Furthermore, we applied the chemical Figure 4. A liposome-based strategy for cell-selective metabolic labeling of sialoglycans. A) Ligand-targeted liposomes encapsulating azidosugars are selectively recognized by cell-surface receptors, and induce the internalizaglycoproteomics method to tion of the liposomal azidosugars into a specific cell type of interest. The azidosugars are utilized in biosynthetic enrich and identify more than pathways and are incorporated into sialoglycans in live cells or in living mice. B) Folate-targeted liposomes encap200 cardiac sialoglycoproteins, sulating 9AzSia (f-LP-9AzSia) were used to achieve selective labeling of the cell-surface sialoglycans on HeLa cells and quantitative proteomic anal- with upregulated expression of folate receptors. The nuclei were visualized by staining with Hoechst 33 342 (blue). Scale bar: 20 mm. C) Cell-selective metabolic glycan labeling in vivo. Liposomal azidosugars are intravenously inysis revealed a list of sialylated jected into tumor-bearing mice. The selectively incorporated azidosugars may be directly visualized by copperproteins that were upregulated free click chemistry, and the tumor sialylated proteins can be selectively enriched and identified by MS analysis. Reprinted and modified, with permission, from refs. [39] and [114]. Copyright: Wiley-VCH, 2014, and The American during hypertrophy. Our group has recently devel- Chemical Society, 2012. oped a liposome-based strategy for incorporation of azidosugars into sialoglycans in vivo (Figure 4).[38, 39, 114] One of the motivatissue selectivity, but it dramatically improves the metabolic labeling efficiency in vivo. Similarly, chemical glycoproteomic tions for developing this strategy is to make MGL cell-selective. Relying on passive diffusion into the interiors of cells, MGLanalysis can be performed to identify and quantify the newly based azidosugars cannot selectively label glycans in a specific synthesized sialylated glycoproteins during tumor growth (Figtissue of interest in vivo, but are instead incorporated into siaure 4 C). loglycans in various tissues.[93] This limits MGL for in vivo glycan imaging in which probing of sialylation in a specific 3.3. Terminal alkynes tissue or cell type is desired. To overcome this limitation, we encapsulated 9Az-Neu5Ac into ligand-targeted liposomes and As the reacting partner of the azide for CuAAC, the terminal demonstrated selective labeling of sialoglycans of specific cell alkyne system also possesses good characteristics for serving types in cell culture (Figure 4 B).[114] Furthermore, ligand-targetas a bioorthogonal chemical reporter. It is stable, is not naturaled liposomes encapsulating 9Az-Neu5Ac were intravenously ly present in cells, and has a small size compatible with that of injected into a melanoma xenograft mouse model (Figthe azido group. Wong and co-workers first reported an ure 4 C).[39] The liposomal nanoparticles selectively bound to alkyne-containing ManNAc analogue, peracetylated N-(4-pentythe tumor-specific receptors and metabolically installed 9Aznoyl)mannosamine (Ac4ManNAl, Scheme 2), which can be metNeu5Ac into the melanoma sialoglycans in a tissue-specific abolically converted into SiaNAl and incorporated into cellular manner. After treating the azides with a far-red fluorescent sialoglycans.[119] Chang et al. then demonstrated that the metaprobe, DBCO-Cy5, by means of copper-free click chemistry, bolic efficiency of Ac4ManNAl is higher than that of Ac4ManNAz tumor-associated sialoglycans were imaged in living mice with in several cell lines and in living mice.[120] CuAAC has been the a TBR of … 177.0. Not only does the liposomal strategy provide reaction of choice for conjugating probes onto terminal alChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

21

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews kynes. Initially, the cytotoxicity of the copper catalyst prevented the labeling of SiaNAl in live cells. Nevertheless, alkyne-incorporated sialoglycans can be labeled in fixed cells, cell lysates, and tissue lysates by CuAAC. Liu et al. employed Ac4ManNAl and CuAAC to enrich and identify sialoglycoproteins in two lung cancer cell lines. From the glycoproteomics results, they further determined that sialylation of epidermal growth factor receptor (EGFR) inhibits its dimerization and activation.[121] Another terminal-alkyne-containing ManNAc analogue—peracetylated N-(propargyloxycarbonyl)mannosamine (Ac4ManNProc, Scheme 2)—was reported to be metabolized into sialoglycans and conjugatable by CuAAC.[122] Notably, the Proc group can also undergo a bioorthogonal elimination reaction (see below). Presumably, SiaNAl and SiaNProc can also be used for MGL of cell-surface sialoglycans. N-(1-Oxohex-5-ynyl)neuraminic acid (Scheme 2), a similar Neu5Ac analogue that varies only by having one more methylene group at C-5 than SiaNAl, was reported and shown to be metabolized into sialoglycans in human larynx carcinoma HEp-2 cells.[123] The development of ligand-assisted CuAAC has enabled MGL with Ac4ManNAl, followed by conjugation of fluorescent probes in living cells.[98, 124] Jiang et al. applied MGL with Ac4ManNAl and STORM imaging to resolve the distribution of sialoglycans on the dynamic membrane nanotubes in live cancer cells.[125] Furthermore, alkynyl ManNAc analogues provide an alternative choice when azides need to be avoided. For example, our group recently developed a FRET-based strategy for proteinspecific imaging of cell-surface sialylation, by exploiting Ac4ManNAl for labeling cell-surface sialoglycans in conjunction with site-specific labeling of a specific protein of interest with an azide.[124] Two fluorophores forming a FRET donor–receptor pair were allowed to react with the azide and alkyne, respectively. The intramolecular donor–acceptor distance for the protein of interest meets the nanometer proximity requirement of FRET, whereas intermolecular FRET is disfavored, thus enabling visualization of the sialylation of a specific protein on the cell surface. The azide and alkyne systems have so far been the most popular bioorthogonal chemical reporters for labeling cell-surface sialoglycans. This has enabled imaging, enrichment and proteomic identification, and therapeutic targeting of cell-surface sialoglycans.

et al. demonstrated that a methylcyclopropene group can also be installed at the N-acyl position of ManNAc and that the resulting Ac4ManNCyc can be used for metabolic labeling of cellsurface sialoglycans (Scheme 2).[129] Patterson et al. followed with Ac4ManCCp, a variant of a methylcyclopropene-containing ManNAc analogue that contains a carbamate linkage between cyclopropene and amide.[130] It is argued that Ac4ManCCp might have improved metabolism efficiency because the branched b-carbon of the N-acyl position of Ac4ManNCyc might not be well tolerated by the synthetic pathway. Furthermore, the carbamate linkage avoids the electron-withdrawing amide to locate at the C-3 position of cyclopropene; this hampers the IED-DA reaction. Recently, Ye and co-workers reported that a cyclopropene system without methyl substitution can serve as a minimal and yet stable chemical reporter for sialoglycan labeling.[131] It was demonstrated that Ac4ManNCp is stable and metabolically labels sialoglycans both in live cells and in living mice. Terminal alkenes were also explored as chemical reporters for MGL of sialoglycans.[132] Three terminal-alkene-containing ManNAc analogues—peracetylated N-pentenoylmannosamine (Ac4ManPtl, Scheme 2), peracetylated N-hexenoylmannosamine (Ac4ManNHxl), and peracetylated N-pentenyloxycarbonylmannosamine (Ac4ManNPeoc)—were synthesized and successfully used to label cell-surface sialoglycans. The reaction rate between terminal alkene and tetrazine is slower than that of a strained cyclic alkene and is comparable with that of Staudinger ligation, which makes terminal alkenes useful reporters for sialoglycans. 3.5. Others With the expanding toolkit of bioorthogonal labeling, it may be expected that more and more chemical reporters should be evaluated for MGL of sialic acids. For example, an isonitrile group has been installed at the N-acyl position for serving as a chemical reporter.[133, 134] Ac4ManN-n-Iso-treated cells (Scheme 2) were conjugated with tetrazine probes for visualization of cell-surface sialoglycans. Similarly, Raines and coworkers recently showed that the diazo-functionalized ManNAc—ManDiaz (Scheme 2)—undergoes sialic acid metabolism and labels cell-surface sialoglycans by reacting with DIBAC probes through cycloaddition.[135] These new chemical reporters provide alternatives in addition to the more broadly used azides and alkynes. More importantly, choosing the right combinations of chemical reporters should enable the use of multiple reactions that are orthogonal to each other, therefore enabling multi-color labeling and multi-functional studies.

3.4. Alkenes Motivated by the broad applications of bioorthogonal chemistry in biological discovery,[103] extensive efforts have been devoted to the development of new bioorthogonal reactions and chemical reporters.[31, 126] Some of the newly developed chemical reporters have been exploited for MGL of sialoglycans. Cyclopropenes have emerged as a useful chemical reporter class, owing to their small size and reactivity with tetrazine probes through inverse-electron-demand Diels–Alder (IED-DA) reactions.[127] Prescher and co-workers synthesized 9Cp-Neu5Ac (Scheme 2), a Neu5Ac analogue containing a methylcyclopropene group at position C-9; it was incorporated into cell-surface sialoglycans and reacted with a tetrazine probe.[128] Cole ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

4. MGL with Raman Reporters MGL coupled with bioorthogonal chemistry has become a useful strategy for molecular imaging of sialoglycans in living systems. In particular, this two-step strategy has enabled conjugation of fluorophores onto sialoglycans for fluorescence imaging. One interesting question we asked was whether an imaging probe can be derivatized onto ManNAc or Neu5Ac for 22

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews ure 5 B).[137] A sialic acid analogue containing a bioorthogonal Raman reporter will therefore produce a specific Raman signal that is not interfered with by the cellular background signals. Coincidentally, azides and alkynes display Raman vibration in the silent region, so they might also serve as bioorthogonal Raman reporters. To detect and image SiaNAz and SiaNAl directly in cell-surface glycans, we employed the surface-enhanced Raman scattering (SERS) technique, which exploits the electromagnetic enhancement of Raman intensity near the surfaces of metallic nanostructures such as gold and silver nanoparticles, and has a detection sensitivity on a par with that of fluorescence spectroscopy. In addition to azides and alkynes, a panel of other bioorthogonal reporters, including the nitrile group and the C¢D bond, which cannot undergo bioorthogonal reactions (Scheme 2), have been developed. To simplify the SERS imaging process, silicon wafers and glass slides were coated with arrays of gold or silver nanoparticles, and the cells metabolically labeled with Raman reporters can be cultured on the resulting substrates for direct imaging by SERS microscopy (Figure 5 C).[138] Alternatively, a recently developed enhanced Raman microscopy technique based on coherent Raman scattering, stimulated Raman scattering (SRS) microscopy, was also exploited for direct Raman imaging of sialoglycans metabolically incorporated with alkynes (Figure 5 D).[139] SRS does not

direct metabolic incorporation, so that the burden of bioorthogonal chemistry would be removed. Obviously, typical organic fluorophores are too big, and ManNAc or Neu5Ac analogues functionalized with fluorophores are not well tolerated by the sialic acid biosynthetic pathway. We therefore sought to explore an alternative imaging modality and turned to Raman microscopy, because a Raman reporter can in principle be as small as a chemical bond. Chemical bonds of glycans including the C¢H bond, the O¢H bond, and the C¢O bond produce characteristic Raman vibrational signals. However, those signals are not specific for glycans because the same chemical bonds are shared by other biomolecules. Apart from the signal specificity issue, the fact that intrinsic Raman scattering has a sensitivity order of magnitude lower than fluorescence is another challenge for Raman imaging of sialoglycans. We developed a bioorthogonal Raman imaging strategy in which a bioorthogonal Raman reporter is functionalized onto ManNAc or Neu5Ac analogues for direct metabolic incorporation and enhanced Raman imaging techniques are used for improving the imaging sensitivity (Figure 5 A).[136] A bioorthogonal Raman reporter is defined as a Raman tag that possesses a vibration within the Raman-silent region of a cell, approximately 1800 to 2800 cm¢1, in which all naturally existing molecules produce negligible Raman signals (Fig-

Figure 5. Bioorthogonal Raman imaging of cell-surface sialoglycans. A) ManNAc or Neu5Ac analogues, each containing a bioorthogonal Raman reporter, are metabolically incorporated into cell-surface sialoglycans. This enables direct visualization by Raman microcopy without the need for a second-step bioorthogonal chemical reaction. B) The Raman spectrum on a live HeLa cell shows the Raman-silent region between 1800 and 2800 cm¢1. A bioorthogonal Raman reporter is defined as a Raman tag that produces a vibration in the silent region of a cell. Typical bioorthogonal Raman reporters include the azide, alkyne, and nitrile groups, as well as the C¢D bond. C) SERS imaging of HeLa cells metabolically incorporated with 9Az-Neu5Ac, [D3]Neu5Ac, and Neu5Ac. D) Schematic illustration of the SRS process and the instrumental setup of SRS microscopy. SRS images of K20 cells treated with 400 mm Ac4ManNAl or with vehicle for 72 h. Images shown are the alkyne on-resonance (2120 cm¢1) of each sample. All images (256 Õ 256 pixels) were obtained by using a 40 ms pixel dwell time. Scale bars: 20 mm. Reprinted and modified, with permission, from refs. [136], [138] and [139]. Copyright: Elsevier, 2015, and Wiley-VCH, 2014.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

23

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews require an enhancing substrate, and the signal is quantitatively dependent on the concentration of the Raman reporters. SERS is much more sensitive, but quantification is challenging. The bioorthogonal Raman imaging of sialoglycans should provide a complementary means to fluorescence imaging. Notably, this new imaging strategy has also been extended to other biomolecules, including nucleic acids, proteins, and lipids.[136]

9, and demonstrated the first use of the metabolically incorporated photo-crosslinking sialic acids for investigating sialic-acidmediated interactions.[115] 9AAz-Neu5Ac was incorporated into the sialoglycans in the B-cell line BJA-B K20 and used to define the cis ligands of the B-cell receptor CD22. Upon UV irradiation and subsequent immunoblot analysis, it was determined that CD22 recognizes sialoglycans on neighboring CD22 proteins as cis ligands and forms homomultimeric complexes. Subsequently, metabolic incorporation of 9AAz-Neu5Ac was used in conjugation with a quantitative proteomics approach to identify the trans ligand of CD22 in homotypic B-cell interactions.[143] The homomultimeric complexes can also be captured by SiaDAz, another photo-crosslinking Neu5Ac analogue containing a diazirine group at C-5, developed by Kohler and co-workers.[144] Probably because a diazirine group is a little smaller than the aryl azide, ManNDAz can also be metabolically incorporated with decent efficiency (Scheme 2). In addition to CD22 interactions, the Kohler group applied the photocrosslinking sugars to study ganglioside–protein interactions.[145, 146]

5. MGL with Thiols The Yarema group exploited MGL to enable metabolic installation of thiol groups into cell-surface sialoglycans, by using peracetylated ManNTGc (Ac5ManNTGc, Scheme 2), a ManNAc analogue containing a thiol group at the N-acyl position.[140] Although thiol groups exist naturally on cell surfaces, thiols of cysteine residues in membrane proteins are usually less accessible. The incorporated Neu5TGc can be detected by treatment with a maleimide-biotin conjugate. Notably, the incorporated thiols tend to form disulfide bonds owing to the oxidizing extracellular environment, and treatment of the cells with a reducing reagent greatly increases the amounts of free thiols on cell-surface sialoglycans. Furthermore, remodeling of cell surfaces with thiols offers a means to control cell adhesion to thiolreactive surfaces (for example, maleimide-coated surfaces or gold surfaces) and to modulate the fate of stem cells.[140, 141] The thiol group is unusual because it is a reactive functional group that has been well used for bioconjugation, but not as a bioorthogonal chemical reporter. The work by the Yarema group demonstrates that the biosynthetic pathway of sialoglycans can be explored for installing a wide range of functional groups. In fact, in an earlier work by Reutter and co-workers, it was shown that an Neu5Ac analogue with a thiol group at position C-9, 9thiol-Neu5Ac, can be metabolically incorporated into cell-surface sialoglycans and affects lectin binding.[47] Therefore, both the ManNAc and Neu5Ac analogues containing thiols are useful tools with which to modulate cell adhesion and cell fate.

7. MGL with Caging Groups It is suspected that Neu5Ac undergoes de-N-acetylation on the cell surface to generate neuraminic acid (Neu) in situ. There is evidence in support of this hypothesis, particularly in nerve cells and cancer cells.[147–149] Nevertheless, the natural existence of Neu on the cell surface remains debatable. We were interested in developing a means to generate Neu on a cell surface to elucidate its biological function. Because Neu is chemically unstable, it is unlikely to be metabolically incorporated on its own. The Chen group recently reported that the Proc group can undergo a Pd-catalyzed bioorthogonal elimination reaction to generate free amine.[150] In collaboration with them, we developed a chemical decaging strategy for in situ generation of Neu on cell surfaces.[151] The cells were metabolically labeled with ManNProc or Neu5Proc, and treatment with Pd nanoparticles resulted in efficient depropargylation, which converts cellsurface Neu5Proc to Neu. The resulting cell-surface Neu partially neutralized the negative charge and induced cell aggregation.

6. MGL with Photo-crosslinkers Sialoglycan–protein interactions govern many important biological processes. However, glycan–protein interactions are difficult to capture because they are typically transient and lowaffinity. Photo-crosslinking has emerged as a valuable tool for probing glycan–protein interactions through the introduction of a covalent bond between the binding partners in the cellular environment, enabling subsequent isolation and characterization of the stabilized interacting partners.[142] The aryl azide was the first photo-crosslinker that was metabolically introduced into cell-surface sialoglycans.[35, 115, 143] 5AAzNeu5Ac (Scheme 2), a Neu5Ac analogue containing an aryl azide system at position C-5, was synthesized, and its metabolic incorporation was confirmed by Staudinger ligation and subsequent flow cytometry analysis, but no photo-crosslinking experiment was performed in this study.[35] The Paulson group then developed a similar Neu5Ac analogue, 9AAz-Neu5Ac (Scheme 2), which contains an aryl azide system at position CChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

8. MGL with Two Functionalities Unnatural functional groups can be installed either at position C-5 or C-9 of Neu5Ac, and this prompted us to develop bifunctional sialic acid analogues with two distinct functionalities at both C-5 and C-9.[152] 9AzSiaNAl (Scheme 2), a bifunctional sialic acid analogue containing an azide at C-9 and an alkyne system at C-5, was synthesized and used for two-color imaging of cell-surface sialoglycans. Another bifunctional sialic acid analogue, 9AzSiaDAz (Scheme 2), possesses a photo-crosslinker and an azide, which can be potentially used for identification of sialic-acid-binding proteins in a high-throughput manner. This work demonstrates that the biosynthetic pathway of sialoglycans can tolerate sialic acid substrates with simultaneous modifications at both C-5 and C-9, thus expanding the capability of MGL of sialic acids. 24

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews Outlook

[9] D. O. Croci, J. P. Cerliani, T. Dalotto-Moreno, S. P. M¦ndez-Huergo, I. D. Mascanfroni, S. Dergan-Dylon, M. A. Toscano, J. J. Caramelo, J. J. Garc†a-Vallejo, J. Ouyang, E. A. Mesri, M. R. Junttila, C. Bais, M. A. Shipp, M. Salatino, G. A. Rabinovich, Cell 2014, 156, 744 – 758. [10] P.-C. Pang, P. C. N. Chiu, C.-L. Lee, L.-Y. Chang, M. Panico, H. R. Morris, S. M. Haslam, K.-H. Khoo, G. F. Clark, W. S. B. Yeung, A. Dell, Science 2011, 333, 1761 – 1764. [11] M. Schwarzkopf, K.-P. Knobeloch, E. Rohde, S. Hinderlich, N. Wiechens, L. Lucka, I. Horak, W. Reutter, R. Horstkorte, Proc. Natl. Acad. Sci. USA 2002, 99, 5267 – 5270. [12] B. Wang, Annu. Rev. Nutr. 2009, 29, 177 – 222. [13] J. Rong, J. Han, L. Dong, Y. Tan, H. Yang, L. Feng, Q.-W. Wang, R. Meng, J. Zhao, S.-Q. Wang, X. Chen, J. Am. Chem. Soc. 2014, 136, 17468 – 17476. [14] M. M. Fuster, J. D. Esko, Nat. Rev. Cancer 2005, 5, 526 – 542. [15] A. Harduin-Lepers, V. Vallejo-Ruiz, M.-A. Krzewinski-Recchi, B. SamynPetit, S. Julien, P. Delannoy, Biochimie 2001, 83, 727 – 737. [16] S.-K. Cha, B. Ortega, H. Kurosu, K. P. Rosenblatt, M. Kuro-O, C.-L. Huang, Proc. Natl. Acad. Sci. USA 2008, 105, 9805 – 9810. [17] S. Z. Gadhoum, R. Sackstein, Nat. Chem. Biol. 2008, 4, 751 – 757. [18] T. Miyagi, K. Yamaguchi, Glycobiology 2012, 22, 880 – 896. [19] M. Cohen, A. Varki, OMICS 2010, 14, 455 – 464. [20] L. T. Martin, J. D. Marth, A. Varki, N. M. Varki, J. Biol. Chem. 2002, 277, 32930 – 32938. [21] H. Hu, K. Eggers, W. Chen, M. Garshasbi, M. M. Motazacker, K. Wrogemann, K. Kahrizi, A. Tzschach, M. Hosseini, I. Bahman, T. Hucho, M. Mìhlenhoff, R. Gerardy-Schahn, H. Najmabadi, H. H. Ropers, A. W. Kuss, Am. J. Hum. Genet. 2011, 89, 407 – 414. [22] J. B. Lowe, J. D. Marth, Annu. Rev. Biochem. 2003, 72, 643 – 691. [23] F. Lehmann, E. Tiralongo, J. Tiralongo, Cell. Mol. Life Sci. 2006, 63, 1331 – 1354. [24] E. L. Bird-Lieberman, A. A. Neves, P. Lao-Sirieix, M. O’Donovan, M. Novelli, L. B. Lovat, W. S. Eng, L. K. Mahal, K. M. Brindle, R. C. Fitzgerald, Nat. Med. 2012, 18, 315 – 321. [25] D. F. Zielinska, F. Gnad, K. Schropp, J. R. Wis´niewski, M. Mann, Mol. Cell 2012, 46, 542 – 548. [26] H. Kayser, R. Zeitler, C. Kannicht, D. Grunow, R. Nuck, W. Reutter, J. Biol. Chem. 1992, 267, 16934 – 16938. [27] O. T. Keppler, R. Horstkorte, M. Pawlita, C. Schmidt, W. Reutter, Glycobiology 2001, 11, 11R – 18R. [28] J. Du, M. A. Meledeo, Z. Wang, H. S. Khanna, V. D. P. Paruchuri, K. J. Yarema, Glycobiology 2009, 19, 1382 – 1401. [29] S. T. Laughlin, C. R. Bertozzi, Proc. Natl. Acad. Sci. USA 2009, 106, 12 – 17. [30] S. H. Rouhanifard, L. U. Nordstrøm, T. Zheng, P. Wu, Chem. Soc. Rev. 2013, 42, 4284 – 4296. [31] C. R. Bertozzi, Acc. Chem. Res. 2011, 44, 651 – 653. [32] C. L. Jacobs, S. Goon, K. J. Yarema, S. Hinderlich, H. C. Hang, D. H. Chai, C. R. Bertozzi, Biochemistry 2001, 40, 12864 – 12874. [33] C. L. Jacobs, K. J. Yarema, L. K. Mahal, D. A. Nauman, N. W. Charters, C. R. Bertozzi, Methods Enzymol. 2000, 327, 260 – 275. [34] S. J. Luchansky, H. C. Hang, E. Saxon, J. R. Grunwell, C. Yu, D. H. Dube, C. R. Bertozzi, Methods. Enzymol. 2003, 362, 249 – 272. [35] S. J. Luchansky, S. Goon, C. R. Bertozzi, ChemBioChem 2004, 5, 371 – 374. [36] E. J. Kim, S.-G. Sampathkumar, M. B. Jones, J. K. Rhee, G. Baskaran, S. Goon, K. J. Yarema, J. Biol. Chem. 2004, 279, 18342 – 18352. [37] M. Bardor, D. H. Nguyen, S. Diaz, A. Varki, J. Biol. Chem. 2005, 280, 4228 – 4237. [38] R. Xie, S. Hong, X. Chen, Curr. Opin. Chem. Biol. 2013, 17, 747 – 752. [39] R. Xie, L. Dong, R. Huang, S. Hong, R. Lei, X. Chen, Angew. Chem. Int. Ed. 2014, 53, 14082 – 14086; Angew. Chem. 2014, 126, 14306 – 14310. [40] Z. Zhou, G. Liao, S. Stepanovs, Z. Guo, J. Carbohydr. Chem. 2014, 33, 395 – 407. [41] E. L. Schwartz, A. F. Hadfield, A. E. Brown, A. C. Sartorelli, Biochim. Biophys. Acta Mol. Cell Res. 1983, 762, 489 – 497. [42] A. F. Hadfield, S. L. Mella, A. C. Sartorelli, J. Pharm. Sci. 1983, 72, 748 – 751. [43] O. T. Keppler, P. Stehling, M. Herrmann, H. Kayser, D. Grunow, W. Reutter, M. Pawlita, J. Biol. Chem. 1995, 270, 1308 – 1314.

The diverse functions of sialic acids in living systems, one of the most important topics in glycobiology, have yet to be fully elucidated. The MGL strategy builds on the understanding of the biosynthetic machinery and advances in the development of chemical tools. MGL of sialoglycans in live cells and living animals has emerged as a powerful tool for studying sialic acid chemistry and biology. Examples outlined in this review demonstrate its applications in gaining insights into the underlying biological processes. Moreover, MGL enables the remodeling of cell-surface sialic acids with new and useful functionalities, providing a chemical methodology for modulating cell-surface biology. On the other hand, several important issues need to be taken into consideration when applying MGL in vivo. Firstly, because exogenous sialic acid analogues are provided, the question of whether these unnatural sialic acids significantly alter the levels and functions of cellular sialoglycans should be carefully examined in specific biological systems. Secondly, the metabolic fates of unnatural sialic acids, such as their distribution between sialylated glycolipids and glycoproteins, and in various sialoglycan structures, remain to be investigated in detail. Thirdly, it is critical to choose the right unnatural functionality and right bioorthogonal chemistry for in vivo labeling. We envision that MGL will continue to facilitate the resolution of important issues that remain to be studied in sialobiology. For example, it will be of interest to understand the distribution of sialoglycans in the brain and the dynamic regulation of brain sialylation.[58] Furthermore, MGL provides a means to explore the dynamic changes in sialylation in response to physiopathological environments for diagnostic and therapeutic applications. Finally, whether MGL can be used to engineer cell functions for use in cell-based therapies is an attractive direction to pursue.[153, 154]

Acknowledgements We acknowledge financial support from the National Natural Science Foundation of China (No. 21172013), No. 91127034, and No. 91313301), the National Basic Research Program of China (No. 2012CB917303), and the National Instrumentation Program (No. 2011YQ030124). Keywords: bioorthogonal chemistry · chemical reporters · click chemistry · metabolic glycan labeling · sialic acids [1] A. Varki, R. D. Cummings, J. D. Esko, H. H. Freeze, G. W. Hart, M. E. Etzler, Essentials of Glycobiology, 2nd ed., Cold Spring Harbor Laboratory Press, New York, 2008. [2] T. Angata, A. Varki, Chem. Rev. 2002, 102, 439 – 469. [3] X. Chen, A. Varki, ACS Chem. Biol. 2010, 5, 163 – 176. [4] A. Varki, Nature 2007, 446, 1023 – 1029. [5] S. D. Rosen, Annu. Rev. Immunol. 2004, 22, 129 – 156. [6] P. R. Crocker, J. C. Paulson, A. Varki, Nat. Rev. Immunol. 2007, 7, 255 – 266. [7] P. H. H. Lopez, R. L. Schnaar, Curr. Opin. Struct. Biol. 2009, 19, 549 – 557. [8] J. E. Stencel-Baerenwald, K. Reiss, D. M. Reiter, T. Stehle, T. S. Dermody, Nat. Rev. Microbiol. 2014, 12, 739 – 749.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

25

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews [44] P. Chefalo, Y. Pan, N. Nagy, Z. Guo, C. V. Harding, Biochemistry 2006, 45, 3733 – 3739. [45] N. W. Charter, L. K. Mahal, D. E. Koshland, C. R. Bertozzi, Glycobiology 2000, 10, 1049 – 1056. [46] L. Dafik, M. d’Alarcao, K. Kumar, Bioorg. Med. Chem. Lett. 2008, 18, 5945 – 5947. [47] C. Oetke, R. Brossmer, L. R. Mantey, S. Hinderlich, R. Isecke, W. Reutter, O. T. Keppler, M. Pawlita, J. Biol. Chem. 2002, 277, 6688 – 6695. [48] W. Zou, S. Borrelli, M. Gilbert, T. Liu, R. A. Pon, H. J. Jennings, J. Biol. Chem. 2004, 279, 25390 – 25399. [49] M. Herrmann, C. W. von der Lieth, P. Stehling, W. Reutter, M. Pawlita, J. Virol. 1997, 71, 5922 – 5931. [50] O. T. Keppler, M. Herrmann, C. W. von der Lieth, P. Stehling, W. Reutter, M. Pawlita, Biochem. Biophys. Res. Commun. 1998, 253, 437 – 442. [51] T. Lieke, D. Grçbe, V. Blanchard, D. Grunow, R. Tauber, M. Zimmermann-Kordmann, T. Jacobs, W. Reutter, Glycoconjugate J. 2011, 28, 31 – 37. [52] R. D. Wright, D. Cooper, Glycobiology 2014, 24, 1242 – 1251. [53] R. Horstkorte, K. Rau, W. Reutter, S. Nçhring, L. Lucka, Exp. Cell Res. 2004, 295, 549 – 554. [54] L. Dafik, M. d’Alarcao, K. Kumar, J. Med. Chem. 2010, 53, 4277 – 4284. [55] R. Horstkorte, H. Y. Lee, L. Lucka, K. Danker, L. Mantey, W. Reutter, Biochem. Biophys. Res. Commun. 2001, 283, 31 – 35. [56] P. Villavicencio-Lorini, S. Laabs, K. Danker, W. Reutter, R. Horstkorte, J. Mol. Med. 2002, 80, 671 – 677. [57] B. E. Collins, T. J. Fralich, S. Itonori, Y. Ichikawa, R. L. Schnaar, Glycobiology 2000, 10, 11 – 20. [58] R. L. Schnaar, R. Gerardy-Schahn, H. Hildebrandt, Physiol. Rev. 2014, 94, 461 – 518. [59] J. R. Wieser, A. Heisner, P. Stehling, F. Oesch, W. Reutter, FEBS Lett. 1996, 395, 170 – 173. [60] C. Schmidt, P. Stehling, J. Schnitzer, W. Reutter, R. Horstkorte, J. Biol. Chem. 1998, 273, 19146 – 19152. [61] B. Bìttner, C. Kannicht, C. Schmidt, K. Lçster, W. Reutter, H.-Y. Lee, S. Nçhring, R. Horstkorte, J. Neurosci. 2002, 22, 8869 – 8875. [62] C. Roehlecke, R. Horstkorte, W. Reutter, Glycoconjugate J. 2013, 30, 813 – 818. [63] C. Schmidt, C. Ohlemeyer, H. Kettenmann, W. Reutter, R. Horstkorte, FEBS Lett. 2000, 478, 276 – 280. [64] R. Horstkorte, K. Rau, S. Laabs, K. Danker, W. Reutter, FEBS Lett. 2004, 571, 99 – 102. [65] S. Natunen, M. Lampinen, H. Suila, I. Ritamo, V. Pitk•nen, A. V. Nairn, J. R•bin•, S. Laitinen, K. W. Moremen, W. Reutter, L. Valmu, Glycobiology 2013, 23, 1004 – 1012. [66] U. Rutishauser, Nat. Rev. Neurosci. 2008, 9, 26 – 35. [67] L. K. Mahal, N. W. Charter, K. Angata, M. Fukuda, D. E. Koshland, C. R. Bertozzi, Science 2001, 294, 380 – 381. [68] N. W. Charter, L. K. Mahal, D. E. Koshland, C. R. Bertozzi, J. Biol. Chem. 2002, 277, 9255 – 9261. [69] R. Horstkorte, M. Mìhlenhoff, W. Reutter, S. Nçhring, M. ZimmermannKordmann, R. Gerardy-Schahn, Exp. Cell Res. 2004, 298, 268 – 274. [70] R. A. Pon, N. J. Biggs, H. J. Jennings, Glycobiology 2007, 17, 249 – 260. [71] D. Gagiannis, R. Gossrau, W. Reutter, M. Zimmermann-Kordmann, R. Horstkorte, Biochim. Biophys. Acta Gen. Subj. 2007, 1770, 297 – 306. [72] V. S. Gnanapragassam, K. Bork, C. E. Galuska, S. P. Galuska, D. Glanz, M. Nagasundaram, M. Bache, D. Vordermark, G. Kohla, C. Kannicht, R. Schauer, R. Horstkorte, PLoS ONE 2014, 9, e105403. [73] D. H. Dube, C. R. Bertozzi, Nat. Rev. Drug Discovery 2005, 4, 477 – 488. [74] Z. Guo, Q. Wang, Curr. Opin. Chem. Biol. 2009, 13, 608 – 617. [75] T. Liu, Z. Guo, Q. Yang, S. Sad, H. J. Jennings, J. Biol. Chem. 2000, 275, 32832 – 32836. [76] G. A. Lemieux, C. R. Bertozzi, Chem. Biol. 2001, 8, 265 – 275. [77] Q. Wang, J. Zhang, Z. Guo, Bioorg. Med. Chem. 2007, 15, 7561 – 7567. [78] L. Qiu, X. Gong, Q. Wang, J. Li, H. Hu, Q. Wu, J. Zhang, Z. Guo, Cancer Immunol. Immunother. 2012, 61, 2045 – 2054. [79] L. Qiu, J. Li, S. Yu, Q. Wang, Y. Li, Z. Hu, Q. Wu, Z. Guo, J. Zhang, Oncotarget 2015, 6, 5195 – 5203. [80] H.-Y. Lee, C.-Y. Chen, T.-I. Tsai, S.-T. Li, K.-H. Lin, Y.-Y. Cheng, C.-T. Ren, T.J. R. Cheng, C.-Y. Wu, C.-H. Wong, J. Am. Chem. Soc. 2014, 136, 16844 – 16853. [81] L. K. Mahal, K. J. Yarema, C. R. Bertozzi, Science 1997, 276, 1125 – 1128.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

[82] E. M. Sletten, C. R. Bertozzi, Acc. Chem. Res. 2011, 44, 666 – 676. [83] K. J. Yarema, L. K. Mahal, R. E. Bruehl, E. C. Rodriguez, C. R. Bertozzi, J. Biol. Chem. 1998, 273, 31168 – 31179. [84] G. Lemieux, K. Yarema, C. Jacobs, C. Bertozzi, J. Am. Chem. Soc. 1999, 121, 4278 – 4279. [85] Y. Iwasaki, H. Maie, K. Akiyoshi, Biomacromolecules 2007, 8, 3162 – 3168. [86] J. H. Lee, T. J. Baker, L. K. Mahal, J. Zabner, C. R. Bertozzi, D. F. Wiemer, M. J. Welsh, J. Biol. Chem. 1999, 274, 21878 – 21884. [87] A. Dirksen, T. M. Hackeng, P. E. Dawson, Angew. Chem. Int. Ed. 2006, 45, 7581 – 7584; Angew. Chem. 2006, 118, 7743 – 7746. [88] A. Dirksen, S. Dirksen, T. M. Hackeng, P. E. Dawson, J. Am. Chem. Soc. 2006, 128, 15602 – 15603. [89] Y. Zeng, T. N. C. Ramya, A. Dirksen, P. E. Dawson, J. C. Paulson, Nat. Methods 2009, 6, 207 – 209. [90] P. Agarwal, J. van der Weijden, E. M. Sletten, D. Rabuka, C. R. Bertozzi, Proc. Natl. Acad. Sci. USA 2013, 110, 46 – 51. [91] E. Saxon, C. R. Bertozzi, Science 2000, 287, 2007 – 2010. [92] P. V. Chang, J. A. Prescher, M. J. Hangauer, C. R. Bertozzi, J. Am. Chem. Soc. 2007, 129, 8400 – 8401. [93] J. A. Prescher, D. H. Dube, C. R. Bertozzi, Nature 2004, 430, 873 – 877. [94] H.-W. Shih, D. N. Kamber, J. A. Prescher, Curr. Opin. Chem. Biol. 2014, 21, 103 – 111. [95] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew. Chem. Int. Ed. 2002, 41, 2596 – 2599; Angew. Chem. 2002, 114, 2708 – 2711. [96] C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67, 3057 – 3064. [97] V. Hong, N. F. Steinmetz, M. Manchester, M. G. Finn, Bioconjugate Chem. 2010, 21, 1912 – 1916. [98] D. Soriano Del Amo, W. Wang, H. Jiang, C. Besanceney, A. C. Yan, M. Levy, Y. Liu, F. L. Marlow, P. Wu, J. Am. Chem. Soc. 2010, 132, 16893 – 16899. [99] C. Besanceney-Webler, H. Jiang, T. Zheng, L. Feng, D. Soriano Del Amo, W. Wang, L. M. Klivansky, F. L. Marlow, Y. Liu, P. Wu, Angew. Chem. Int. Ed. 2011, 50, 8051 – 8056; Angew. Chem. 2011, 123, 8201 – 8206. [100] N. J. Agard, J. A. Prescher, C. R. Bertozzi, J. Am. Chem. Soc. 2004, 126, 15046 – 15047. [101] J. C. Jewett, C. R. Bertozzi, Chem. Soc. Rev. 2010, 39, 1272 – 1279. [102] M. F. Debets, S. S. van Berkel, J. Dommerholt, A. J. Dirks, F. P. J. T. Rutjes, F. L. van Delft, Acc. Chem. Res. 2011, 44, 805 – 815. [103] M. Grammel, H. C. Hang, Nat. Chem. Biol. 2013, 9, 475 – 484. [104] J. M. Baskin, J. A. Prescher, S. T. Laughlin, N. J. Agard, P. V. Chang, I. A. Miller, A. Lo, J. A. Codelli, C. R. Bertozzi, Proc. Natl. Acad. Sci. USA 2007, 104, 16793 – 16797. [105] S. Letschert, A. Gçhler, C. Franke, N. Bertleff-Zieschang, E. Memmel, S. Doose, J. Seibel, M. Sauer, Angew. Chem. Int. Ed. 2014, 53, 10921 – 10924; Angew. Chem. 2014, 126, 11101 – 11104. [106] C. Witte, V. Martos, H. M. Rose, S. Reinke, S. Klippel, L. Schrçder, C. P. R. Hackenberger, Angew. Chem. Int. Ed. 2015, 54, 2806 – 2810; Angew. Chem. 2015, 127, 2848 – 2852. [107] J. Du, S. Hong, L. Dong, B. Cheng, L. Lin, B. Zhao, Y.-G. Chen, X. Chen, J. Biol. Chem. 2015, 290, 12000 – 12013. [108] A. E. V. B. Granell, K. B. Palter, I. Akan, U. Aich, K. J. Yarema, M. J. Betenbaugh, W. B. Thornhill, E. Recio-Pinto, ACS Chem. Biol. 2011, 6, 1287 – 1295. [109] K. Kang, S. Joo, J. Y. Choi, S. Geum, S.-P. Hong, S.-Y. Lee, Y. H. Kim, S.-M. Kim, M.-H. Yoon, Y. Nam, K.-B. Lee, H.-Y. Lee, I. S. Choi, Proc. Natl. Acad. Sci. USA 2015, 112, E241 – E248. [110] L. Yang, J. O. Nyalwidhe, S. Guo, R. R. Drake, O. J. Semmes, Mol. Cell. Proteomics 2011, 10, M110.007294. [111] F. Autelitano, D. Loyaux, S. RoudiÀres, C. D¦on, F. Guette, P. Fabre, Q. Ping, S. Wang, R. Auvergne, V. Badarinarayana, M. Smith, J.-C. Guillemot, S. A. Goldman, S. Natesan, P. Ferrara, P. August, PLoS ONE 2014, 9, e110316. [112] C. M. Woo, A. T. Iavarone, D. R. Spiciarich, K. K. Palaniappan, C. R. Bertozzi, Nat. Methods 2015, 12, 561 – 567. [113] H. Mçller, V. Bçhrsch, J. Bentrop, J. Bender, S. Hinderlich, C. P. R. Hackenberger, Angew. Chem. Int. Ed. 2012, 51, 5986 – 5990; Angew. Chem. 2012, 124, 6088 – 6092. [114] R. Xie, S. Hong, L. Feng, J. Rong, X. Chen, J. Am. Chem. Soc. 2012, 134, 9914 – 9917.

26

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reviews [115] S. Han, B. E. Collins, P. Bengtson, J. C. Paulson, Nat. Chem. Biol. 2005, 1, 93 – 97. [116] A. A. Neves, H. Stockmann, R. R. Harmston, H. J. Pryor, I. S. Alam, H. Ireland-Zecchini, D. Y. Lewis, S. K. Lyons, F. J. Leeper, K. M. Brindle, FASEB J. 2011, 25, 2528 – 2537. [117] S. T. Laughlin, J. M. Baskin, S. L. Amacher, C. R. Bertozzi, Science 2008, 320, 664 – 667. [118] K. W. Dehnert, J. M. Baskin, S. T. Laughlin, B. J. Beahm, N. N. Naidu, S. L. Amacher, C. R. Bertozzi, ChemBioChem 2012, 13, 353 – 357. [119] T.-L. Hsu, S. R. Hanson, K. Kishikawa, S.-K. Wang, M. Sawa, C.-H. Wong, Proc. Natl. Acad. Sci. USA 2007, 104, 2614 – 2619. [120] P. V. Chang, X. Chen, C. Smyrniotis, A. Xenakis, T. Hu, C. R. Bertozzi, P. Wu, Angew. Chem. Int. Ed. 2009, 48, 4030 – 4033; Angew. Chem. 2009, 121, 4090 – 4093. [121] Y.-C. Liu, H.-Y. Yen, C.-Y. Chen, C.-H. Chen, P.-F. Cheng, Y.-H. Juan, C.-H. Chen, K.-H. Khoo, C.-J. Yu, P.-C. Yang, T.-L. Hsu, C.-H. Wong, Proc. Natl. Acad. Sci. USA 2011, 108, 11332 – 11337. [122] L. A. Bateman, B. W. Zaro, K. N. Chuh, M. R. Pratt, Chem. Commun. 2013, 49, 4328 – 4330. [123] A. Homann, R.-U. Qamar, S. Serim, P. Dersch, J. Seibel, Beilstein J. Org. Chem. 2010, 6, 24. [124] W. Lin, Y. Du, Y. Zhu, X. Chen, J. Am. Chem. Soc. 2014, 136, 679 – 687. [125] H. Jiang, B. P. English, R. B. Hazan, P. Wu, B. Ovryn, Angew. Chem. Int. Ed. 2015, 54, 1765 – 1769; Angew. Chem. 2015, 127, 1785 – 1789. [126] D. M. Patterson, L. A. Nazarova, J. A. Prescher, ACS Chem. Biol. 2014, 9, 592 – 605. [127] J. Sˇecˇkute˙, N. K. Devaraj, Curr. Opin. Chem. Biol. 2013, 17, 761 – 767. [128] D. M. Patterson, L. A. Nazarova, B. Xie, D. N. Kamber, J. A. Prescher, J. Am. Chem. Soc. 2012, 134, 18638 – 18643. [129] C. M. Cole, J. Yang, J. Sˇecˇkute˙, N. K. Devaraj, ChemBioChem 2013, 14, 205 – 208. [130] D. M. Patterson, K. A. Jones, J. A. Prescher, Mol. Biosyst. 2014, 10, 1693 – 1697. [131] D.-C. Xiong, J. Zhu, M.-J. Han, H.-X. Luo, C. Wang, Y. Yu, Y. Ye, G. Tai, X.-S. Ye, Org. Biomol. Chem. 2015, 13, 3911 – 3917. [132] A. Niederwieser, A.-K. Sp•te, L. D. Nguyen, C. Jìngst, W. Reutter, V. Wittmann, Angew. Chem. Int. Ed. 2013, 52, 4265 – 4268; Angew. Chem. 2013, 125, 4359 – 4363. [133] S. Stairs, A. A. Neves, H. Stçckmann, Y. A. Wainman, H. Ireland-Zecchini, K. M. Brindle, F. J. Leeper, ChemBioChem 2013, 14, 1063 – 1067. [134] Y. A. Wainman, A. A. Neves, S. Stairs, H. Stçckmann, H. Ireland-Zecchini, K. M. Brindle, F. J. Leeper, Org. Biomol. Chem. 2013, 11, 7297 – 7300. [135] K. A. Andersen, M. R. Aronoff, N. A. McGrath, R. T. Raines, J. Am. Chem. Soc. 2015, 137, 2412 – 2415.

ChemBioChem 2016, 17, 11 – 27

www.chembiochem.org

[136] S. Hong, L. Lin, M. Xiao, X. Chen, Curr. Opin. Chem. Biol. 2015, 24, 91 – 96. [137] L. Lin, X. Tian, S. Hong, P. Dai, Q. You, R. Wang, L. Feng, C. Xie, Z. Q. Tian, X. Chen, Angew. Chem. Int. Ed. 2013, 52, 7266 – 7271; Angew. Chem. 2013, 125, 7407 – 7412. [138] M. Xiao, L. Lin, Z. Li, J. Liu, S. Hong, Y. Li, M. Zheng, X. Duan, X. Chen, Chem. Asian J. 2014, 9, 2040 – 2044. [139] S. Hong, T. Chen, Y. Zhu, A. Li, Y. Huang, X. Chen, Angew. Chem. Int. Ed. 2014, 53, 5827 – 5831; Angew. Chem. 2014, 126, 5937 – 5941. [140] S.-G. Sampathkumar, A. V. Li, M. B. Jones, Z. Sun, K. J. Yarema, Nat. Chem. Biol. 2006, 2, 149 – 152. [141] J. Du, P.-L. Che, Z.-Y. Wang, U. Aich, K. J. Yarema, Biomaterials 2011, 32, 5427 – 5437. [142] S.-H. Yu, A. M. Wands, J. J. Kohler, J. Carbohydr. Chem. 2012, 31, 325 – 352. [143] T. N. C. Ramya, E. Weerapana, L. Liao, Y. Zeng, H. Tateno, L. Liao, J. R. Yates, B. F. Cravatt, J. C. Paulson, Mol. Cell. Proteomics 2010, 9, 1339 – 1351. [144] Y. Tanaka, J. J. Kohler, J. Am. Chem. Soc. 2008, 130, 3278 – 3279. [145] M. R. Bond, H. Zhang, J. Kim, S.-H. Yu, F. Yang, S. M. Patrie, J. J. Kohler, Bioconjugate Chem. 2011, 22, 1811 – 1823. [146] M. R. Bond, C. M. Whitman, J. J. Kohler, Mol. Biosyst. 2010, 6, 1796 – 1799. [147] K. I. Hidari, F. Irie, M. Suzuki, K. Kon, S. Ando, Y. Hirabayashi, Biochem. J. 1993, 296, 259 – 263. [148] T. Bulai, D. Bratosin, A. Pons, J. Montreuil, J.-P. Zanetta, FEBS Lett. 2003, 534, 185 – 189. [149] I. Popa, A. Pons, C. Mariller, T. Tai, J.-P. Zanetta, L. Thomas, J. Portoukalian, Glycobiology 2007, 17, 367 – 373. [150] J. Li, J. Yu, J. Zhao, J. Wang, S. Zheng, S. Lin, L. Chen, M. Yang, S. Jia, X. Zhang, P. R. Chen, Nat. Chem. 2014, 6, 352 – 361. [151] J. Wang, B. Cheng, J. Li, Z. Zhang, W. Hong, X. Chen, P. R. Chen, Angew. Chem. Int. Ed. 2015, 54, 5364 – 5368; Angew. Chem. 2015, 127, 5425 – 5428. [152] L. Feng, S. Hong, J. Rong, Q. You, P. Dai, R. Huang, Y. Tan, W. Hong, C. Xie, J. Zhao, X. Chen, J. Am. Chem. Soc. 2013, 135, 9244 – 9247. [153] M. A. Fischbach, J. A. Bluestone, W. A. Lim, Sci. Transl. Med. 2013, 5, 179ps7. [154] D. S. Chen, I. Mellman, Immunity 2013, 39, 1 – 10.

Manuscript received: July 8, 2015 Final article published: November 17, 2015

27

Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim