Thiyl glycosylate of olefinic proteins: S-linked glycoconjugate synthesis

Article abstract of DOI:10.1002/anie.200903135

Tagged for thiolation: A novel glycoconjugation strategy utilizes a non-natural olefin-containing amino acid (homoallylglycine, Hag) as a "tag" for modification and a photoinitiated hydroglycothiolation reaction that is selective only for the Hag olefinic "tag". Application of this method to a number of model proteins allowed complete and precise site-selective glycosylate generating glycoconjugates that include, for example, virus-like particles displaying up to 180 glycans at preselected positions (see scheme).

Full text of DOI:10.1002/anie.200903135

Communications
DOI: 10.1002/anie.200903135
Protein Modification
Thiyl Glycosylation of Olefinic Proteins: S-Linked Glycoconjugate
Synthesis**
Nicola Floyd, Balakumar Vijayakrishnan, Julia R. Koeppe, and Benjamin G. Davis*
Over half of all proteins in nature are estimated to be
glycosylated,[1] and these biomolecules play key roles in
protein expression, folding and stability[2–7] and are funda-
mental to various biological processes.[8–13] In recent years,
synthetic homogenous glycoforms[14–20] of glycoproteins have
been one of the primary targets in glycobiology,[11,21] not only
to allow function determination, but also to create glycopro-
tein mimetics useful as, e.g., therapeutic agents. Beyond the
preference for the more abundant native O- and N-linked
glycoproteins, S-linked glycoproteins are also attractive
synthetic targets as a result of their enhanced chemical
stability and enzymatic resistance.[22] Following the discovery
of the first natural S-glycosidic linkage by Lote and Weiss in
1971,[23,24] methods have been developed for the synthesis of
S-linked glycopeptides[25] and more recently S-linked glyco-
proteins.[26,27]
radical addition hydrothiolation reactions, under conditions
mild enough to retain protein activity throughout (Scheme 1).
In this way, Hag functions as a new “tag” combined here with
a new modification as part of a general “tag–modify” strategy
for synthetic-protein construction.[30] Using this strategy
previously we have, for example, been able to demonstrate
the successful use of azide or alkyne tags.[31,32] Prior, elegant
conjugations have shown that radical-addition reactions may
be successfully applied to proteins.[33–35] Whilst 1-thioglyco-
side formation by the free-radical addition of 1-glycosyl thiols
to alkenes has been reported for the synthesis of small
molecules,[36–38] to date this method has not been applied to
the synthesis of S-linked glycoproteins or bioconjugates. The
unique reactivity profile of l-Hag, with an olefinic side-chain
compared to the natural amino acids characteristically found
in proteins, allows for a chemoselective chemical reaction.
Amongst these selectivity advantages is the inertness of Hag
to almost all common protein modification reactions, thereby
allowing the potential for orthogonal use in combined,
multireaction protein chemistry strategies.[31] This inertness
to other reagents would not be shared by strategies that would
use a protein-thiyl radical, perhaps derived from Cys,
although this “reverse” approach would allow the potential
for use of proteins containing only natural amino acids. In this
context, during the final stages of this work,[39] we became
aware of usefully complementary methods employed by
Dondoni, Massi and co-workers for glycosylation of protein-
thiyl radicals.[35]
Here, we describe the development of a convergent
approach for the synthesis of a novel class of S-linked
glyconjugate proteins through the site-specific ligation of 1-
glycosyl thiols to proteins (Scheme 1). The strategy exploits
non-natural amino acid incorporation[28,29] for the introduc-
tion of l-homoallylglycine (l-Hag) into a protein and free-
The free-radical reaction of glycosyl thiyls was investi-
gated and optimized initially on representative amino acid
model systems containing Hag (Table 1). As a prerequisite for
later protein stability, solubility and viability, the reaction was
developed to comply with aqueous solution chemistry.
Significantly, all prior protocols[36–38] for glycosyl thiyl gen-
eration have until now employed organic solvents. Since the
stability and reactivity of glycosyl thiyls can vary, and to
demonstrate substrate breadth and broad applicability of the
method, we utilized a wide range of 1-glycosyl thiols, in both
protected and unprotected form, as starting materials includ-
ing 1-thio-b-d-glucose (b-GlcSH), 1-thio-a-d-glucose (a-
GlcSH), 2-acetamido-2-deoxy-1-thio-b-d-glucose (GlcNAc-
SH), 1-thio-b-d-galactose (GalSH), 1-thio-b-d-mannose
(ManSH), and disaccharide 4-O(b-d-galactosyl)-1-thio-b-d-
glucose (Gal(b1,4)GlcSH). Together reactions (Table 1)
probed the effect of bulk, differing configuration, protecting
groups, and anomeric stereochemistry in the thiols; variation
around the N- and C-termini of Hag and varying conditions,
including alternative modes of initiation (using water soluble
initiator Vazo44 (VA044, 2,2’-azobis[2-(2-imidazolin-2-
Scheme 1. Summarized strategic approach.
[*] N. Floyd, Dr. B. Vijayakrishnan, Dr. J. R. Koeppe, Prof. B. G. Davis
Department of Chemistry, University of Oxford, Chemistry Research
Laboratory, Mansfield Road, Oxford, OX1 3TA (UK)
Fax: (+44)1865-285-002
E-mail: ben.davis@chem.ox.ac.uk
Homepage: http://users.ox.ac.uk/~dplb0149/
[**] This work was supported by the International AIDS Vaccine Initiative
(IAVI). We thank Profs. J. Blanchard and M. G. Finn for kindly
providing pET23a/Np276 and p75m/Qb plasmids, respectively, and
Dr. J. Errey for helpful discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200903135.
yl)propane]dihydrochloride)[40]
and/or
photochemical
7798
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7798 –7802

Angewandte
Chemie
Table 1: Amino acid glycoconjugation reactions.
hydroxythiols were good to excellent. For these more reactive
thiols, conversions were essentially complete even for larger
sugars such as disaccharide Gal(b1,4)GlcSH; the isolated
yields largely reflect difficulties handling amphiphilic prod-
ucts. Although disulfides have been suggested as putative
thiyl precursors previously,[42] use of disulfide (b-GlcS)2 gave
no product (entry 9). This suggests that not only thiols provide
an important hydride source in the conjugation reaction but
that the lower yields observed at higher pH (10) may be a
consequence of more rapid competitive oxidation to unreac-
tive disulfide. The chemoselectivity of the glycoconjugation
reaction for Hag over each of the 20 naturally occurring
natural amino acids was also investigated (see Supporting
Information): when b-GlcSH was exposed to each amino acid
under optimal conditions no conjugation products with any
amino acid other than Hag were observed.[43]
Next, the thiyl glycoconjugation procedure was applied to
a number of model protein systems: a TIM-barrel protein
(SsbG),[44] a “cuboid”, right-handed quadrilateral b-helix
protein (Np276),[45] and self-assembled homomultimer virus-
like bacteriophage particle Qb.[46–48] Non-natural amino acid
Hag was site-specifically introduced into these protein
systems as an olefinic “tag” through expression of corre-
sponding gene sequences in an auxotrophic strain of E. coli
(B834(DE3)).[29] Gene sequences were designed to create
proteins displaying olefin at a single site, the position of which
could simply be controlled by the “Met” triplet codon
ATG.[49]
Model protein substrate SsbG-Hag43, with an olefin
uniquely positioned at site 43, is formally a twelve-point
mutant of the wild-type Sulfolobus solfataricus b-glycosidase
in which all cysteine (Cys344Ser) and methionine (Met21Ile,
Met73Ile, Met148Ile, Met206Ile, Met236Ile, Met275Ile,
Met280Ile, Met383Ile, Met439Ile) residues occurring in the
wild-type protein, except Met43, have been mutated to serine
and isoleucine, respectively. Importantly, the elucidation of
key structural and sequence features associated with an
unique, serendipitous photocleavage reaction in certain TIM-
barrel proteins such as SsbG-WT[39,50] allowed us to also
introduce a key stabilizing Pro152Ala mutation, in order to
make the protein stable to UV light.
Use of SsbG-Hag43 as a substrate in reactions with b-
GlcSH allowed optimization (See Table 2 and Supporting
Information for time course) of the protein chemistry and
revealed conditions consistent with those determined in
amino acid systems: fastest, complete conversions (> 95%)
were seen at pH < 7 through the combined use of Vazo44 and
photoinitiation. Relatively high concentrations of thiol (up to
100 mm) also proved important for rapid reaction. Initial
results suggest that hydride abstraction may be rate limit-
ing;[51,52] unreacted thiol is readily recovered as its disulfide
post-reaction. Importantly, all glycoconjugated SsbG prod-
ucts were functionally (enzymatically) active, thereby con-
firming the benign nature of the reaction.
Entry 1-Glycosyl
thiol[a]
Olefin
Solvent
Yield[b]
[%]
H2O/MeOH
(1:1)
1
2
3
4
5
6
7
8
79
H2O/MeOH
(1:1)
71[c]
81
H2O
H2O/MeOH
(1:1)
57
pH 4
acetate
42
(ꢀ50[d])
pH 4
acetate
63
(>98[d])
pH 4
acetate
70
(>80[d])
pH 4
acetate
28
(ꢀ50[d])
pH 4
acetate
9
0
pH 4
acetate
56
10
11
(>98[d])
pH 4
acetate
91[b]
(>98[d])
[a] 1.2 Equivalents. [b] Yield of isolated compounds after 8 h reaction
time. [c] Mono-Boc formed as product olefin. Boc=tert-butoxycarbonyl.
[d] Conversion determined by crude 1H NMR spectroscopy.
(medium-pressure 125 W Hg lamp with borosilicate vessel
giving maximum emission at 365 nm)).
Optimization studies with b-GlcSH (Table S2) indicated
that both forms of initiation allowed successful conjugation;
interestingly, enhancements in rate and yields were observed
through photoinitiation in the presence of Vazo44. pH proved
important with lower pHs (4–7) proving superior to the use of
pH 10 CAPS buffer (N-cyclohexyl-3-aminopropanesulfonic
acid). Importantly, ambient conditions in aqueous solutions
potentially compatible with protein stability (without the
need for, e.g., degassing)[41] proved effective. Control reac-
tions with l-Hag and representative thiols b-GlcSH and
GlcNAcSH confirmed all to be individually stable (e.g., to
racemization/anomerization) under the reaction conditions
and, in particular, to UV irradiation. Good regioselectivity
(>98% e-thiolation) and retention of anomeric configuration
in product were observed. Use of extended reaction times or
excess thiol (3–10 equiv) allowed essentially complete con-
version.
Pleasingly, extension of the reaction to the full range of
glycosyl thiols revealed successful coupling in all cases
(Table 1). Notably, rates and yields for cis-1,2-hydroxythiols
b-ManSH and a-GlcSH were significantly lower (entries 5
(42%) and 8 (28%)), whilst those for less hindered trans-
Having established conditions for the successful site-
selective glycoconjugation to the single olefinic side-chain in
SsbG-Hag43 we explored extension to other glycans and
proteins (Table 3). These revealed that, although slightly
slower, glycoconjugations at pH 6 (entries 2, 6 and 9) were
Angew. Chem. Int. Ed. 2009, 48, 7798 –7802
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7799

Communications
Table 2: Optimization of protein glycoconjugations.[a]
Table 3: Protein glycoconjugation reactions.
Entry
Glycosyl
thiol[a]
Olefinic
protein
pH
4
t
[h]
Product
[%]
1
3
5
>95
>95
>95
0
Conditions
0.25m
acetate
(pH 4)
0.25m
0.25m CAPS
(pH 10)
sodium
phosphate
(pH 7)
2
3
4
6
4
6
hn
558C
>95% (5 h)
ꢀ10% (8 h)
>95% (3 h)
>95% (3 h)
26% (8 h)
NR (8 h)
45% (9 h)
25% (7 h)
NR (8 h)
NR (8 h)
25% (9 h)
NR (9 h)
Vazo44 (0.2 equiv), hn
Vazo44 (0.2 equiv), 558C
3
[a] NR=No reaction.
equally successful, yielding > 95% product in 5 h or less.
Although no degradation was seen here with any of the
proteins at pH 4, this more mild pH should allow glycoconju-
gation to even highly acid-sensitive proteins. Rates of reaction
for different thiols paralleled those seen in Table 1; for
example, b-GlcNAcSH (Table 3, entry 3) was slightly slower
than b-GlcSH (entry 1). Importantly, use of SsbG-Met43, as a
negative control substrate that does not contain Hag, showed
no coupling (entry 4), further confirming the chemoselectivity
of the reaction for the olefinic “tag” and the lack of reaction
with other amino acids in SsbG.
12
5
6
4
6
2
3
>95
>95
Cuboid protein Np276–Hag61 was also successfully gly-
coconjugated. Incorporation of Hag as the olefinic “tag” was
again achieved through the mutation of the native np276 gene
to position a single ATG triplet codon appropriately, here
allowing display of a single olefin at position 61. Reactions
under optimal conditions were even more rapid than for
SsbG-Hag43. Again, reactions at pH 4 and 6 gave > 95%;
those at pH 7 proved sluggish and gave only 40% conversion
after 3 h and did not reach completion.
Finally, we extended the glycoconjugation reaction to the
self-assembled multimeric, virus-like particle Qb (Table 3,
entries 8–10).[46] Creation of a protein (Qb-Hag16) with Hag
at position 16 and self-assembly and purification allowed the
creation of a virus-like particle displaying 180 olefins.
Glycoconjugation under optimal conditions at either pH 4
or 6 gave rapid formation of product with complete con-
sumption of starting protein suggesting complete glycoconju-
gation at all 180 sites (entries 8 and 9).[53–55] Again, use of a
negative control substrate that does not contain Hag, showed
no coupling (entry 10). Furthermore, reductive disassembly of
the Qb virus-like particle after reaction and alkylative
labelling (see Supporting Information) confirmed that struc-
turally key intramolecular disulfide bonds mediated by Cys75
and Cys81, which covalently link each assembled monomer,
remained untouched by the hydrothiolation glycoconjugation
reaction. This further confirmed the excellent chemoselectiv-
ity of the reaction for the Hag olefinic site, even with Cys-
containing proteins. The Qb particle is composed of 180
copies of a coat protein assembled into a icosahedral virion of
average diameter 27 nm.[56] To our knowledge this precise and
apparently complete site-selective glycoconjugation at multi-
7
7
4
6
4
3
2
3
8
40
>95
>95
0
8
9
10
[a] 20-100 mm thiol, 0.1–0.5 mgmLÀ1 protein, Vazo44 (0.2 equiv), hn,
room temperature.
ple sites in a “protein nanoparticle” is unprecedented,[57,58]
and we believe may find utility in the creation of precise
glycoconjugate vaccines in which high levels of loading may
be achieved whilst precisely controlling the position of
epitope display.[48,59,60]
In summary, we have developed an effective protein
glycoconjugation method based on hydrothiolation that
allows rapid and complete reaction under mild conditions
(ambient temperature, pH 4–6) that have been shown to be
compatible with a representative selection of different
proteins. Combined use of this reaction with an olefinic
“tag” that may be positioned using the non-natural amino
7800 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7798 –7802

Angewandte
Chemie
[15] B. G. Davis, M. A. T. Maughan, M. P. Green, A. Ullman, J. B.
acid Hag allows site-selective glycoconjugation at pre-deter-
mined sites using chosen sugars and proteins through a “tag–
modify”[30] approach (Hag “tag”–hydrothiolation “modify”).
The reaction is compatible with all other natural amino acids,
which are inert under these conditions; this chemoselectivity
should allow its use in orthogonal, multi-modification
approaches.[31] In addition, the glycosyl thiols needed as
glycoconjugation partners may be readily accessed in a
number of ways, even directly from unprotected reducing
sugars.[61] Here, the reaction has been applied to proteins with
different structures, which retain full function, for precise,
site-selective glycoconjugation at single and even high levels
of multiple sites. We are currently exploring the potential of
this method to create glycoconjugate immunogens as putative
vaccines.
Jones, Tetrahedron: Asymmetry 2000, 11, 245.
[16] R. C. Lloyd, B. G. Davis, J. B. Jones, Bioorg. Med. Chem. 2000, 8,
1537.
[17] H. Liu, L. Wang, A. Brock, C.-H. Wong, P. G. Schultz, J. Am.
Chem. Soc. 2003, 125, 1702.
[18] D. Macmillan, R. M. Bill, K. A. Sage, D. Fern, S. L. Flitsch,
Chem. Biol. 2001, 8, 133.
[19] Y. Shin, K. A. Winans, B. J. Backes, S. B. H. Kent, J. A. Ellman,
C. R. Bertozzi, J. Am. Chem. Soc. 1999, 121, 11684.
[20] K. Witte, P. Sears, C.-H. Wong, J. Am. Chem. Soc. 1997, 119,
2114.
[21] D. P. Gamblin, E. M. Scanlan, B. G. Davis, Chem. Rev. 2009, 109,
131.
[22] D. Horton, J. D. Wander, Carbohydrates: Chemistry and Bio-
chemistry, Vol. 4B, Academic Press, New York, 1990, p. 799.
[23] C. J. Lote, J. B. Weiss, FEBS Lett. 1971, 16, 81.
[24] C. J. Lote, J. B. Weiss, Biochem. J. 1971, 123, 25p.
[25] K. Pachamuthu, R. R. Schmidt, Chem. Rev. 2006, 106, 160.
[26] G. J. L. Bernardes, E. J. Grayson, S. Thompson, J. M. Chalker,
J. C. Errey, F. El Oualid, T. D. W. Claridge, B. G. Davis, Angew.
Chem. 2008, 120, 2276; Angew. Chem. Int. Ed. 2008, 47, 2244.
[27] G. J. L. Bernardes, J. M. Chalker, J. C. Errey, B. G. Davis, J. Am.
Chem. Soc. 2008, 130, 5052.
[28] J. C. M. van Hest, D. A. Tirrell, FEBS Lett. 1998, 428, 68.
[29] J. C. M. van Hest, K. L. Kiick, D. A. Tirrell, J. Am. Chem. Soc.
2000, 122, 1282.
[30] B. G. Davis, Pure Appl. Chem. 2009, 81, 285.
[31] S. I. van Kasteren, H. B. Kramer, H. H. Jensen, S. J. Campbell, J.
Kirkpatrick, N. J. Oldham, D. C. Anthony, B. G. Davis, Nature
2007, 446, 1105.
Experimental Section
Representative glycoconjugation reaction: To a solution of Qb-Hag16
(0.5 mgmLÀ1, 500 mL) in 0.25m sodium acetate buffer (pH 4) was
added 1-thio-b-d-glucose (100 mm) and Vazo44 (20 mm). The reac-
tion mixture was placed in a cuvette and irradiated with a medium-
pressure 125 W Hg-lamp with borosilicate filter. The progress of the
reaction was monitored by mass spectrometry (ESI-MS (ES + ) and
MALDI-TOF). After reaction virus-like particle samples (50 mL,
0.5 mgmLÀ1) were mixed with urea (75 mL, 8m) and DTT (dithio-
threitol) (7.5 mL, 1m), and incubated at 378C for 1 h to allow the
protein to denature prior to analysis by ESI-MS. The samples were
further incubated with iodoacetamide (7.5 mL, 1m) at 378C for up to
16 h to cap any free cysteines prior to analysis by ESI-MS. m/z for
monomer of Qb-Hag16-S-b-Glc: calcd. 14301 (14417 with iodoacet-
amide); found 14303 (14416 with iodoacetamide). After 2 h these
revealed essentially complete conversion (> 95%) of Qb-Hag16 into
Qb-(CH2)5-S-Glcb, which could be isolated through dialysis, sucrose
gradient and/or size-exclusion membrane purification.
[32] S. I. van Kasteren, H. B. Kramer, D. P. Gamblin, B. G. Davis,
Nat. Protoc. 2007, 2, 3185.
[33] S. Wittrock, T. Becker, H. Kunz, Angew. Chem. 2007, 119, 5319;
Angew. Chem. Int. Ed. 2007, 46, 5226.
[34] G. Triola, L. Brunsveld, H. Waldmann, J. Org. Chem. 2008, 73,
3646.
[35] A. Dondoni, A. Massi, P. Nanni, A. Roda, Chem. Eur. J., in press.
[36] J. M. Lacombe, N. Rakotomanomana, A. A. Pavia, Tetrahedron
Lett. 1988, 29, 4293.
Received: June 11, 2009
Published online: September 8, 2009
[37] C. Schips, T. Ziegler, J. Carbohydr. Chem. 2005, 24, 773.
[38] S. Knapp, D. S. Myers, J. Org. Chem. 2002, 67, 2995.
[39] N. Floyd, N. J. Oldham, C. J. Eyles, M. Brouard, B. G. Davis,
Abstracts of Papers, 237th ACS National Meeting, Salt Lake
City, USA, March 22 – 26, 2009, ORGN.
Keywords: glycoconjugates · glycoproteins · glycosyl thiols ·
hydrothiolation · photochemical reactions
.
[40] G. S. Hammond, R. C. Neuman, J. Am. Chem. Soc. 1963, 85,
1501.
[41] M. S. Kharasch, W. Nudenberg, G. J. Mantell, J. Org. Chem.
1951, 16, 524.
[42] M. S. Kharasch, W. Nudenberg, T. H. Meltzer, J. Org. Chem.
1953, 18, 1233.
[43] Under prolonged conditions in aerated solvent low levels of
disulfide bond formation between b-GlcSH and cysteine thiol
was the only reaction observed.
[44] C. F. Aguilar, I. Sanderson, M. Moracci, M. Ciaramella, R.
Nucci, M. Rossi, L. H. Pearl, J. Mol. Biol. 1997, 271, 789.
[45] M. W. Vetting, S. S. Hegd, K. Z. Hazleton, J. S. Blanchard,
Protein Sci. 2007, 16, 755.
[46] T. M. Kozlovska, I. Cielens, D. Dreilinna, A. Dislers, V.
Baumanis, V. Ose, P. Pumpens, Gene 1993, 137, 133.
[47] E. Strable, J. Duane E. Prasuhn, A. K. Udit, S. Brown, A. J. Link,
J. T. Ngo, G. Lander, J. Quispe, C. S. Potter, B. Carragher, D. A.
Tirrell, M. G. Finn, Bioconjugate Chem. 2008, 19, 866.
[48] E. Strable, M. G. Finn, Curr. Top. Microbiol. Immunol. 2009, 327,
1.
[1] R. Apweiler, H. Hermjakob, N. Sharon, Biochim. Biophys. Acta
Gen. Subj. 1999, 1473, 4.
[2] A. J. Parodi, Annu. Rev. Biochem. 2000, 69, 69.
[3] A. Helenius, Mol. Biol. Cell 1994, 5, 253.
[4] G. Opdenakker, P. M. Rudd, C. P. Ponting, R. A. Dwek, FASEB
J. 1993, 7, 1330.
[5] B. Imperiali, S. E. OꢀConnor, Curr. Opin. Chem. Biol. 1999, 3,
643.
[6] D. Shental-Bechor, Y. Levy, Proc. Natl. Acad. Sci. USA 2008,
105, 8256.
[7] V. Sriram, C. A. Willard, J. Liu, R. R. Brutkiewicz, Immunology
2008, 123, 272.
[8] R. A. Dwek, Chem. Rev. 1996, 96, 683.
[9] K. J. Doores, D. P. Gamblin, B. G. Davis, Chem. Eur. J. 2006, 12,
656.
[10] P. Sears, C. H. Wong, Cell. Mol. Life Sci. 1998, 54, 223.
[11] B. G. Davis, Chem. Rev. 2002, 102, 579.
[12] J. B. Lowe, J. D. Marth, Annu. Rev. Biochem. 2003, 72, 643.
[13] A. Varki, Glycobiology 1993, 3, 97.
[49] This reassignment of the Met codon prevents the use of proteins
that contain functionally essential methionines.
[14] B. G. Davis, R. C. Lloyd, J. B. Jones, J. Org. Chem. 1998, 63, 9614.
Angew. Chem. Int. Ed. 2009, 48, 7798 –7802
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7801

Communications
[50] N. Floyd, N. J. Oldham, C. J. Eyles, S. Taylor, D. A. Filatov, M.
[56] K. Vallegꢂrd, K. Fridborg, L. Liljas, Acta Crystallogr. Sect. D
1994, 50, 105 – 109.
Brouard, B. G. Davis, J. Am. Chem. Soc. 2009, 131, 12518 –
12519.
[51] We are currently also exploring alternative hydride donor
sources that may allow a productive addition cycle under these
conditions.
[52] S. Deprꢁle, J.-L. Montchamp, J. Org. Chem. 2001, 66, 6745.
[53] Consistent with previous observations[54,55] glycosylated Qb
showed reduced ionization efficiency allowing sensitive detec-
tion of any unreacted, unglycosylated starting material in
reaction mixtures. See Supporting Information for calibration
curves.
[54] A. L. Burlingame, Curr. Opin. Biotechnol. 1996, 7, 4.
[55] H. Geyer, R. Geyer, Biochim. Biophys. Acta Proteins Proteomics
2006, 1764, 1853.
[57] Recently, elegant work by the Finn group[58] has utilized
cycloaddition reactions to partially glycoconjugate alkynes
introduced by acylation of surface lysine side chains or through
the use of non-natural amino acid homopropargylglycine
(average 85 modified out of 180).
[58] E. Kaltgrad, M. K. OꢀReilly, L. Liao, S. Han, J. C. Paulson, M. G.
Finn, J. Am. Chem. Soc. 2008, 130, 4578.
[59] I. Vasiljeva, T. Kozlovska, I. Cielens, A. Strelnikova, A. Kazaks,
V. Ose, P. Pumpens, FEBS Lett. 1998, 431, 7.
[60] B. Chackerian, Expert Rev. Vaccines 2007, 6, 381.
[61] G. J. L. Bernardes, D. P. Gamblin, B. G. Davis, Angew. Chem.
2006, 118, 4111; Angew. Chem. Int. Ed. 2006, 45, 4007.
7802 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7798 –7802