Stabilizing Impact of N-Glycosylation on the WW Domain Depends

Letters pubs.acs.org/acschemicalbiology

Stabilizing Impact of N‑Glycosylation on the WW Domain Depends Strongly on the Asn-GlcNAc Linkage Brijesh K. Pandey,† Sebastian Enck,‡,§ and Joshua L. Price*,† †

Department of Chemistry and Biochemistry, Brigham Young University, Provo, Utah 84602, United States Department of Chemistry and Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 N. Torrey Pines Road, La Jolla, California 92037, United States

‡

S Supporting Information *

ABSTRACT: N-glycans play important roles in many cellular processes and can increase protein conformational stability in specific structural contexts. Glycosylation (with a single GlcNAc) of the reverse turn sequence Phe-Yyy-Asn-Xxx-Thr at Asn stabilizes the Pin 1 WW domain by −0.85 ± 0.12 kcal mol−1. Alternative methods exist for attaching carbohydrates to proteins; some occur naturally (e.g., the O-linkage), whereas others use chemoselective ligation reactions to mimic the natural N- or O-linkages. Here, we assess the energetic consequences of replacing the Asn linkage in the glycosylated WW domain with a Gln linkage, with two natural O-linkages, with two unnatural triazole linkages, and with an unnatural αmercaptoacetamide linkage. Of these alternatives, only glycosylation of the triazole linkages stabilizes WW, and by a smaller amount than N-glycosylation, highlighting the need for caution when using triazole- or α-mercaptoacetamide-linked carbohydrates to mimic native N-glycans, especially where the impact of glycosylation on protein conformational stability is important.

A

WW domain of the human protein Pin 1 (hereafter called WW),15,16 which harbors a type I β-bulge turn with the following sequence: Phe16-Ala17-Asn19-Gly20-Thr21 (the numbering scheme refers to the native Pin WW domain,17 which has a six-residue loop in place of the five-residue type I βbulge turn; position 18 is present in the six-residue loop but not in the five-residue type I β-bulge turn).18 Phe16 occupies the i position of this type I β-bulge turn, whereas Asn19 and Thr21 occupy the i + 2 and i + 4 positions, respectively. Published structural data indicate that the conformation of this reverse turn places the Phe16, Asn19, Thr21 side chains close together on the same side of the turn (Figure 1A), so that when Asn19 is glycosylated with a single N-acetyl glucosamine (GlcNAc), GlcNAc engages in stabilizing native-state interactions with Phe16 and Thr21 (Figure 1B).18 These interactions stabilize WW by −0.85 ± 0.12 kcal mol−1 (Table 1, compare protein 1 and glycoprotein 1g).14 An Nglycan within a type I β-bulge in the adhesion domain of the human protein CD2 engages in similar stabilizing interactions.8,10,13 Consistent with published structural data,18 mutagenesis experiments reveal that the stabilizing interaction between GlcNAc and Phe16 derives in part from burial of the nonpolar

pproximately one-third of the proteome traverses the secretory pathway of eukaryotic cells,1 and many of these proteins are glycosylated when the oligosaccharyl transferase complex attaches a large oligosaccharide to the side-chain amide nitrogen of Asn residues within a consensus sequence called the N-glycosylation sequon: Asn-Xxx-Thr/Ser (where Xxx is any amino acid but Pro).2 N-glycans play important roles in many cellular processes, modulating protein function, extending protein half-life, shielding immunogenic epitopes, and mediating cell−cell recognition, among many other functions.3 One of the most important is to shuttle Nglycoproteins into the calnexin/calreticulin-assisted folding vs degradation pathway, which provides time for unfolded or misfolded glycoproteins to acquire the proper fold or else to be targeted for ER-associated degradation.4 The conformational stability of many glycoproteins depends strongly on the presence of disulfide bonds.5 However, the Nglycan itself can also increase protein conformational stability,6−12 especially when placed in appropriate sequence/ structural contexts. Kelly and co-workers recently discovered that placement of the consensus sequence Phe-Yyy-Asn-XxxThr (where Xxx is any amino acid but proline and Yyy is any amino acid) within a five-residue reverse turn comprising a type I β-turn followed by a G1 bulge (hereafter called a type I βbulge turn) creates a structural context in which Asnglycosylation can increase protein conformational stability.13,14 This structural context is present in an engineered variant of the © 2013 American Chemical Society

Received: June 21, 2013 Accepted: August 12, 2013 Published: August 12, 2013 2140

dx.doi.org/10.1021/cb4004496 | ACS Chem. Biol. 2013, 8, 2140−2144

ACS Chemical Biology

Letters

α face of GlcNAc with the π system of the aromatic ring of Phe, driven primarily by dispersion forces.18 Other contributors to the enhanced conformational stability of 1g relative to 1 could include close packing of the methyl group of Thr21 against the N-acetyl group of GlcNAc and/or hydrogen bonding between the side-chain carbonyl oxygen of Asn19 and the side-chain hydroxyl of Thr21.18,19 The Phe-Yyy-Asn-Xxx-Thr sequence motif from 1g can be installed in type I β-bulge turns in other proteins, where glycosylation of the i + 2 Asn results in similar increases to protein conformational stability.13 We wondered whether the stabilizing impact of glycosylation in this structural context depends strongly on the identity of the Asn linker between GlcNAc and the protein backbone. There are many alternatives for attaching carbohydrates to proteins: some occur naturally (e.g., O-glycosylation of the side-chain hydroxyl group of Ser and Thr side3) whereas others derive from chemoselective ligation reactions designed to mimic the natural N− or O− linkages.20,21 Can these alternatives effectively recapitulate the stabilizing impact of attaching GlcNAc of Asn19 in 1g? To address this question, we generated WW variants 2−7, in which the i + 2 Asn19 of 1 has been replaced by Gln, Ser, Thr, propargylglycine (PrG), propargyloxyphenylalanine (PrF), or Cys, respectively. We then prepared glycosylated variants 2g− 7g (Figure 2). In glycoprotein 2g, GlcNAc is attached via a βglycosidic linkage to the side-chain amide nitrogen of the i + 2 Gln in protein 2. In glycoproteins 3g and 4g, GlcNAc has been attached via a β-glycosidic linkage to the side-chain hydroxyl oxygens of the i + 2 Ser and the i + 2 Thr of proteins 3 and 4, respectively. In glycoproteins 5g and 6g, a GlcNAc azide has been attached via a triazole linkage to the side-chain alkyne groups of the i + 2 PrG and the i + 2 PrF of proteins 5 and 6, respectively, using the well-known chemoselective coppercatalyzed azide−alkyne cycloaddition.22−24 In glycoprotein 7g, a GlcNAc bromoacetamide has been attached via a chemoselective nucleophilic substitution reaction25−28 to the sidechain thiol of the i + 2 Cys in protein 7. We used variable temperature circular dichroism to assess the conformational stability of each glycosylated WW variant relative to its corresponding nonglycosylated counterpart (Figure 3). Protein 2 (with an i + 2 Gln) has an apparent melting temperature (Tm) of 52.8 ± 0.6 °C (Table 1; the thermal unfolding behavior of 2 was not consistent with the two-state folding model we typically use to extract thermodynamic information from variable temperature CD experiments; however, we estimated an apparent melting temperature from these experiments by calculating the slope at every data point along the melting curve; the point with highest slope is the apparent melting temperature). In contrast, glycoprotein 2g (Gln-GlcNAc at i + 2) has a Tm of 49.6 ± 0.9 °C, indicating that glycosylation of the i + 2 Gln destabilizes WW (recall that glycosylation of the i + 2 Asn stabilizes 1g by −0.85 ± 0.12 kcal mol−1 relative to 1). Gln differs from Asn by the addition of a single methylene (CH2) group in the side chain. It is possible that the longer Gln linker changes the position of GlcNAc, thereby preventing it from interacting optimally with Phe16 and Thr21. It is also possible that the carbonyl oxygen of the longer Gln side-chain is not able to hydrogen bond as effectively with the side-chain OH of Thr21 as is the Asn side chain in 1g. Attaching GlcNAc via a β-glycosidic linkage to the i + 2 Ser or the i + 2 Thr in 3 or 4, respectively, is similarly destabilizing. Glycoprotein 3g is 0.42 ± 0.05 kcal mol−1 less stable at 55 °C

Figure 1. Stick representation of (a) the N-terminal type I β-bulge turn of protein 1 (ref 18) and (b) the N-terminal type I β-bulge turn of protein 1g (ref 18). Nonpolar hydrogen atoms are hidden for clarity, except for residues at the i, i + 2, and i + 4 positions, which are highlighted in yellow, and for the Asn-linked GlcNAc, which is highlighted in green. Thick black dashes represent hydrogen bonds. All structures were rendered in Pymol.

Table 1. Melting Tempertaures and Folding Free Energies (at 328.15 K) for Proteins 1−7, 6p, 7a, and Glycoproteins 1g−7ga protein 1 1g 2 2g 3 3g 4 4g 5 5g 6 6g 6p 7 7g 7a

Tm (°C) 64.5 73.5 52.8 49.6 57.3 52.8 55.7 48.4 54.9 57.5 49.4 53.3 49.5 63.2 55.3 54.7

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.3 0.6 0.6b 0.9 0.3 0.4 0.2 0.4 0.6 0.6 0.6 0.8 0.9b 0.7 0.6 0.4

ΔTm (°C) 9.0 ± 0.7 −3.2 ± 1.1 −4.5 ± 0.5 −7.3 ± 0.5 2.7 ± 0.8 3.9 ± 1.0 0.1 ± 1.1 −7.8 ± 0.9 −8.5 ± 0.8

ΔGf (kcal/mol) −0.97 ± −1.82 ± --0.47 ± −0.22 ± 0.20 ± −0.07 ± 0.55 ± 0.02 ± −0.26 ± 0.62 ± 0.19 ± --−0.77 ± −0.03 ± 0.03 ±

0.04 0.11 0.08 0.03 0.04 0.02 0.04 0.06 0.06 0.08 0.09 0.07 0.05 0.04

ΔΔGf (kcal/mol) −0.85 ± 0.12 --0.42 ± 0.05 0.61 ± 0.05 −0.27 ± 0.08 −0.43 ± 0.11 --0.73 ± 0.09 0.80 ± 0.08

Tabulated data are given as mean ± standard error at 55 °C for 5 μM solutions of WW variants in 20 mM sodium phosphate buffer (pH 7), except 6 and 6g, which were characterized at 4.3 μM, and 6p, which was characterized at 8.7 μM. Variable temperature CD data from which these values are derived appear in SI Figures S74−S80. The ΔTm and ΔΔGf values for proteins 6p, 7a, and glycoproteins 1g−7g are given relative to their unmodified counterparts, proteins 1−7. b The thermal unfolding behavior of 2 and 6p were not consistent with the two-state folding model; therefore, we only report an apparent melting temperature for these compounds (see Supporting Information for details). cThe ΔTm and ΔΔGf values for 7a and 7g are relative to protein 7. a

α face of GlcNAc against the Phe side-chain, though the effect is not as large as what would be expected for burial of Phe in the hydrophobic interior of a protein.18 Another stabilizing component comes from the interaction of C−H bonds on the 2141

dx.doi.org/10.1021/cb4004496 | ACS Chem. Biol. 2013, 8, 2140−2144

ACS Chemical Biology

Letters

Figure 2. Amino acid sequences of proteins 1−7, 6p, 7a, and glycoproteins 1g−7g, along with structures of the amino acids at position 19 (the i + 2 position of the type I β-turn).

relative to 1 do not apply in a straightforward way to β-linked O-glycans. Of the alternative linkers investigated here, the triazole linkage most successfully mimics the stabilization associated with glycosylating Asn. Glycosylation of the i + 2 PrG in 5 with a GlcNAc azide via the copper-catalyzed azide−alkyne cycloaddition gives glycoprotein 5g (Figure 2), which is −0.27 ± 0.08 kcal mol−1 more stable than 5 at 55 °C (Table 1). Glycosylation of the i + 2 PrF side-chain in 6 with a GlcNAc-azide under similar conditions gives glycoprotein 6g (Figure 2), which is −0.43 ± 0.11 kcal mol−1 more stable than 6. Glycosylation of PrG or PrF is not as favorable as glycosylation of Asn, indicating that these unnatural triazole linkers are imperfect mimics of the native Asn linkage, possibly due to their increased length (Figure 2). The increased stability of 5g relative to 5 and of 6g relative to 6 may come from interactions of GlcNAc with Phe16 and/or Thr21 as observed in 1g, or from hydrogen bonding between one of the triazole nitrogen atoms and the side-chain OH of the i + 4 Thr (as observed for the Asn linkage in 1g). Glycoprotein 6g may also benefit from additional interactions with other aromatic residues on the WW surface (e.g., Tyr23 or Phe34), made possible by the increased length of the triazole linker in this variant relative to the Asn linker in 1g. Alternatively, it is possible that 5g and 6g benefit from aromatic stacking interactions between Phe16 and their respective triazole groups, which are not present in 5 and 6. If so, variants of 5g or 6g that lack the carbohydrate but retain the triazole should be stabilized relative to 5 or 6. To test this hypothesis, we prepared protein 6p, in which a short, noncarbohydrate azide has been attached to the alkyne of 6 via the copper-catalyzed azide−alkyne cycloaddition (Figure 2). Variable temperature CD data for 6p showed a hightemperature transition in addition to the usual unfolding transition (see Supporting Information), possibly consistent with high-temperature aggregation. Consequently, we only report an apparent Tm for 6p here (estimated by fitting the variable temperature data prior to the high-temperature transition to a two-state unfolding model). The apparent Tm of 6p is 49.5 ± 0.9 °C (Table 1), which is substantially lower than the Tm of 6g (53.5 ± 0.8 °C) but indistinguishable from that of 6 (49.4 ± 0.6 °C), suggesting that the increased stability

Figure 3. Variable temperature CD data for (a) protein 1 vs glycoprotein 1g at 5 μM protein concentration in 20 mM sodium phosphate, pH 7; and (b) protein 6 vs glycoprotein 6g at 4.3 μM protein concentration in 20 mM sodium phosphate, pH 7.

than its nonglycosylated counterpart 3 (Table 1). At the same temperature, glycoprotein 4g is 0.61 ± 0.05 kcal mol−1 less stable than nonglycosylated 4 (Table 1). The Ser and Thr side chains in glycoproteins 3g and 4g link GlcNAc to the protein backbone via a three-atom linker; this linker is one atom shorter than the linker provided by the Asn side-chain in 1g. The shorter linker length may prevent GlcNAc from interacting optimally with Phe and Thr and may partially explain why glycosylation of the i + 2 residue in 3 and 4 decreases WW conformational stability. In any case, these results suggest that the interactions that allow Asn glycosylation to stabilize 1g 2142

dx.doi.org/10.1021/cb4004496 | ACS Chem. Biol. 2013, 8, 2140−2144

ACS Chemical Biology

Letters

conformational stability relative to the wild-type unmodified sequence (e.g., protein 1, Tm = 64.5 ± 0.3 °C), the αmercaptoacetamide linker in 7g (Tm = 55.3 ± 0.6 °C) might be a better candidate than the triazole linker in 6g (Tm = 53.3 ± 0.8 °C).

of 6g relative to 6 depends on the presence of GlcNAc and that the triazole group does not substantially stabilize 6g. Finally, linking GlcNAc to Cys via an α-mercaptoacetamide linkage was not substantially stabilizing. Glycoprotein 7g is 0.73 ± 0.09 kcal mol−1 less stable than nonglycosylated protein 7. Protein 7a is a derivative of 7, in which the Cys side-chain has been functionalized with nonglycosylated 2-bromoacetamide (Figure 2). The folding free energies of proteins 7a and 7g are indistinguishable, suggesting that the decreased stability of 7g relative to 7 is predominantly due to the α-mercaptoacetamide linker, not GlcNAc. This result is surprising in light of previous work by Flitsch and co-workers,26,28 in which glycosylation of a Cys-containing variant of dihydrofolate reductase with glycosyl iodoacetamides within an unstructured loop led to increases in protein conformational stability. It is possible that the αmercaptoacetamide-modified Cys19 in 7a and 7g has conformational preferences that are substantially different from those of Cys19 in 7 or Asn19 in 1, thereby preventing GlcNAc in 7g from engaging in stabilizing interactions with Phe16 and/or Thr21. N-glycosylation of the Phe-Yyy-Asn-Xxx-Thr sequence in the context of a five-residue type I β-bulge turn within an engineered variant of the WW domain of Pin 1 increases protein stability via a specific tripartite interaction among the i position Phe16, the glycosylated i + 2 position Asn19, and the i + 4 position Thr21.13,14 Here, we assessed the energetic consequences of changing the linker between GlcNAc and the i + 2 position to other alternative linkers (Figure 2). None of these other linkers provides as much GlcNAc-based stabilization as does the native Asn linkage. Glycosylation of some linkers (Gln, Ser, Thr, Cys) is destabilizing; glycosylation of others is moderately stabilizing at best (PrG, PrF). This diminished GlcNAc-based stabilization is not a simple function of linker length but also appears to depend on the identity and position of the functional groups within the linker. The flexibility, hydrogen bonding capability, and conformational preferences of the linker may also play an important role. Understanding the impact of N-glycosylation on protein function and conformational stability remains an important goal, made difficult by the challenge of generating large quantities of pure homogeneously glycosylated proteins.29 Convergent protein synthesis (e.g., native chemical ligation), enzymatic carbohydrate remodeling, and re-engineering of cellular glycosylation machinery are all promising strategies for addressing this problem.30,31 An alternative approach is to use chemoselective reactions to attach glycans selectively to specific bioorthogonal functional groups (natural or unnatural) on protein surfaces.20,21 The triazole and α-mercaptoacetamide linkages we investigated here can be generated via chemoselective bioorthogonal reactions between glycosyl azides and unnatural alkyne-containing side chains22−24 or between glycosyl haloacetamides and Cys side-chains,25−28 respectively. These alternative linkers are interesting as potential mimics of the native Asn linkage present in natural N-glycans. Similar unnatural linkers have been used previously in contexts that did not disturb protein folding or function.26−28 However, our results indicate that these triazole and α-mercaptoacetamide linkers do not always completely recapitulate the conformational impact of N-glycosylation and should be used with caution, especially in cases where the impact of glycosylation on protein conformation and thermodynamic stability is important. On the other hand, if the objective is simply to introduce an artificial glycosylation site with a minimum reduction in

■

METHODS

■

ASSOCIATED CONTENT

Protein Synthesis. WW domain variants were synthesized via solid phase peptide synthesis, using the standard Fmoc protecting group strategy, as described in the Supporting Information. The syntheses of Fmoc-protected glycosylated derivatives of L-asparagine, L-glutamine, L-serine, L-threonine, and p-propargyloxy-L-phenylalanine are described in the Supporting Information. Following cleavage from resin, WW variants were purified by reverse-phase high pressure liquid chromatography (HPLC) on a C18 column using a linear gradient of water in acetonitrile with 0.1% v/v TFA. The identity of each WW domain was confirmed by matrix-assisted laser desorption/ionization time-of-flight spectrometry and purity was evaluated by analytical HPLC. CD Measurements. CD spectra and variable temperature CD data were collected using an Aviv 420 spectropolarimeter. We fit the variable temperature CD data to obtain Tm and ΔGf values for each protein (Table 1), as described in the Supporting Information.

S Supporting Information *

Complete experimental methods, compound characterization, and circular dichroism data. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address §

BASF SE, Ludwigshafen D-67056, Germany

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank J. Kelly, E. Powers, C. Wong, and M. Andrus for helpful discussion. J.L.P. acknowledges the support of start-up funds from the Dept. of Chemistry and Biochemistry at Brigham Young University.

■

REFERENCES

(1) Huh, W.-K., Falvo, J. V., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S., and O’Shea, E. K. (2003) Global analysis of protein localization in budding yeast. Nature 425, 686−691. (2) Kornfeld, R., and Kornfeld, S. (1985) Assembly of asparaginelinked oligosaccharides. Annu. Rev. Biochem. 54, 631−664. (3) Varki, A. (1993) Biological roles of oligosaccharides: All of the theories are correct. Glycobiology 3, 97−130. (4) Moremen, K. W., and Molinari, M. (2006) N-linked glycan recognition and processing: The molecular basis of endoplasmic reticulum quality control. Curr. Opin. Chem. Biol. 16, 592−599. (5) Kajihara, Y., Tanabe, Y., Sasaoka, S., and Okamoto, R. (2012) Homogeneous human complex-type oligosaccharides in correctly folded intact glycoproteins: Evaluation of oligosaccharide influence on protein folding, stability, and conformational properties. Chem.Eur. J. 18, 5944−5953. (6) Wormald, M. R., Wooten, E. W., Bazzo, R., Edge, C. J., Feinstein, A., Rademacher, T. W., and Dwek, R. A. (1991) The conformational effects of N-glycosylation on the tailpiece from serum IgM. Eur. J. Biochem. 198, 131−139.

2143

dx.doi.org/10.1021/cb4004496 | ACS Chem. Biol. 2013, 8, 2140−2144

ACS Chemical Biology

Letters

(7) Imperiali, B., and Rickert, K. W. (1995) Conformational implications of asparagine-linked glycosylation. Proc. Natl. Acad. Sci. U.S.A. 92, 97−101. (8) Wyss, D. F., Choi, J. S., Li, J., Knoppers, M. H., Willis, K. J., Arulanandam, A. R. N., Smolyar, A., Reinherz, E. L., and Wagner, G. (1995) Conformation and function of the N-linked glycan in the adhesion domain of human CD2. Science 269, 1273−1278. (9) Live, D. H., Kumar, R. A., Beebe, X., and Danishefsky, S. J. (1996) Conformational influences of glycosylation of a peptide: A possible model for the effect of glycosylation on the rate of protein folding. Proc. Natl. Acad. Sci. U.S.A. 93, 12759−12761. (10) Hanson, S. R., Culyba, E. K., Hsu, T.-L., Wong, C.-H., Kelly, J. W., and Powers, E. T. (2009) The core trisaccharide of an N-linked glycoprotein intrinsically accelerates folding and enhances stability. Proc. Natl. Acad. Sci. U.S.A. 106, 3131−3136. (11) Shental-Bechor, D., and Levy, Y. (2008) Effect of glycosylation on protein folding: A close look at thermodynamic stabilization. Proc. Natl. Acad. Sci. U.S.A. 105, 8256−8261. (12) Price, J. L., Shental-Bechor, D., Dhar, A., Turner, M. J., Powers, E. T., Gruebele, M., Levy, Y., and Kelly, J. W. (2010) Contextdependent effects of asparagine glycosylation on pin WW folding kinetics and thermodynamics. J. Am. Chem. Soc. 132, 15359−15367. (13) Culyba, E. K., Price, J. L., Hanson, S. R., Dhar, A., Wong, C. H., Gruebele, M., Powers, E. T., and Kelly, J. W. (2011) Protein nativestate stabilization by placing aromatic side chains in N-glycosylated reverse turns. Science 331, 571−575. (14) Price, J. L., Powers, D. L., Powers, E. T., and Kelly, J. W. (2011) Glycosylation of the enhanced aromatic sequon is similarly stabilizing in three distinct reverse turn contexts. Proc. Natl. Acad. Sci. U.S.A. 108, 14127−14132. (15) Jäger, M., Zhang, Y., Bieschke, J., Nguyen, H., Dendle, M., Bowman, M. E., Noel, J. P., Gruebele, M., and Kelly, J. W. (2006) Structure−function−-folding relationship in a WW domain. Proc. Natl. Acad. Sci. U.S.A. 103, 10648−10653. (16) Jäger, M., Nguyen, H., Crane, J. C., Kelly, J. W., and Gruebele, M. (2001) The folding mechanism of a β-sheet: The WW domain. J. Mol. Biol. 311, 373−393. (17) Ranganathan, R., Lu, K. P., Hunter, T., and Noel, J. P. (1997) Structural and functional analysis of the mitotic rotamase Pin1 suggests substrate recognition is phosphorylation dependent. Cell 89, 875−886. (18) Chen, W., Enck, S., Price, J. L., Powers, D. L., Powers, E. T., Wong, C.-H., Dyson, H. J., and Kelly, J. W. (2013) The structural and energetic basis of carbohydrate-aromatic packing interactions in proteins. J. Am. Chem. Soc. 135, 9877−9884. (19) Wang, X., Ji, C., and Zhang, J. (2012) Exploring the molecular mechanism of stabilization of the adhesion domains of human CD2 by N-glycosylation. J. Phys. Chem. B 116, 11570−11577. (20) Hang, H., and Bertozzi, C. (2001) Chemoselective approaches to glycoprotein assembly. Acc. Chem. Res. 34, 727−736. (21) Gamblin, D., Scanlan, E., and Davis, B. (2009) Glycoprotein synthesis: An update. Chem. Rev. 109, 131−163. (22) Rostovtsev, V. V., Green, L. G., Fokin, V. V., and Sharpless, K. B. (2002) A stepwise huisgen cycloaddition process: Copper(I)-catalyzed regioselective “ligation” of azides and terminal alkynes. Angew. Chem., Int. Ed. 41, 2596−2599. (23) Tornøe, C., Christensen, C., and Meldal, M. (2002) Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(i)-catalyzed 1,3-dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem. 67, 3057−3064. (24) Papst, S., Noisier, A., Brimble, M., Yang, Y., and Krissansen, G. (2012) Tyrosine modified analogues of the α4β7 integrin inhibitor biotin-R8ERY prepared via Click Chemistry: Synthesis and biological evaluation. Bioorg. Med. Chem. 20, 2638−2644. (25) Davis, N. J., and Flitsch, S. L. (1991) A novel method for the specific glycosylation of proteins. Tetrahedron Lett. 32, 6793−6796. (26) Swanwick, R., Daines, A., Tey, L.-H., Flitsch, S., and Allemann, R. (2005) Increased thermal stability of site-selectively glycosylated dihydrofolate reductase. ChemBioChem 6, 1338−1340.

(27) Hirano, K., Macmillan, D., Tezuka, K., Tsuji, T., and Kajihara, Y. (2009) Design and synthesis of a homogeneous erythropoietin analogue with two human complex-type sialyloligosaccharides: Combined use of chemical and bacterial protein expression methods. Angew. Chem., Int. Ed. 48, 9557−9560. (28) Tey, L.-H., Loveridge, E., Swanwick, R., Flitsch, S., and Allemann, R. (2010) Highly site-selective stability increases by glycosylation of dihydrofolate reductase. FEBS J. 277, 2171−2179. (29) Grogan, M., Pratt, M., Marcaurelle, L., and Bertozzi, C. (2002) Homogeneous glycopeptides and glycoproteins for biological investigation. Annu. Rev. Biochem. 71, 593−634. (30) Schmaltz, R., Hanson, S., and Wong, C.-H. (2011) Enzymes in the synthesis of glycoconjugates. Chem. Rev. 111, 4259−4307. (31) Rich, J., and Withers, S. (2009) Emerging methods for the production of homogeneous human glycoproteins. Nat. Chem. Biol. 5, 206−215.

2144

dx.doi.org/10.1021/cb4004496 | ACS Chem. Biol. 2013, 8, 2140−2144