Simple and Efficient Preparation of O- and S-GlcNAcylated Amino Acids through InBr3-Catalyzed Synthesis of β- N-Acetylglycosides from Commercially Available Reagents

Article abstract of DOI:10.1021/acs.orglett.8b02182

The facile synthesis of serine, threonine, and cysteine β-glycosides using commercially available peracetylated β-N-acetylglucosamine (β-Ac₄GlcNAc) and catalytic amounts of indium bromide (InBr₃) is described. This method involves only inexpensive reagents that require no further modification or special handling. The reagents are simply mixed, dissolved, and refluxed to afford the GlcNAcylated amino acids in great yields (70-80%). This operationally simple procedure should facilitate the study of O-GlcNAcylation without necessitating expertise in synthetic carbohydrate chemistry.

Full text of DOI:10.1021/acs.orglett.8b02182

Letter
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
pubs.acs.org/OrgLett
Simple and Efficient Preparation of O- and S‑GlcNAcylated Amino
Acids through InBr ‑Catalyzed Synthesis of β‑N‑Acetylglycosides
3
from Commercially Available Reagents
Cesar A. De Leon, Geoffrey Lang, Marcos I. Saavedra, and Matthew R. Pratt*
†
†
†
,†,‡
†
‡
Department of Chemistry and Department of Molecular and Computational Biology, University of Southern California, Los
Angeles, California 90089, United States
*
S Supporting Information
ABSTRACT: The facile synthesis of serine, threonine, and
cysteine β-glycosides using commercially available peracety-
lated β-N-acetylglucosamine (β-Ac GlcNAc) and catalytic
4
amounts of indium bromide (InBr ) is described. This method
3
involves only inexpensive reagents that require no further
modification or special handling. The reagents are simply
mixed, dissolved, and refluxed to afford the GlcNAcylated
amino acids in great yields (70−80%). This operationally
simple procedure should facilitate the study of O-GlcNAcy-
lation without necessitating expertise in synthetic carbohydrate chemistry.
-
GlcNAcylation is a dynamic post-translational mod-
ification (PTM) of intracellular proteins in which the
single monosaccharide N-acetylglucosamine (GlcNAc) is β-
Scheme 1. Previous Methods for β-Ac GlcNAc Amino Acids
3
O
1
,2
linked to serine or threonine residues of substrate proteins.
This highly abundant PTM is critical for several biological
processes, and its misregulation has important implications in
mammalian development, survival, and disease. Currently, the
only approach for deciphering the biochemical and biophysical
function of site-specific O-GlcNAcylation is through semi-
synthetic preparation of the glycosylated target protein, which
3
requires site-specifically modified glycopeptides. Given the
4
heterogeneity that arises from in vitro enzymatic glycosylation
and the racemization observed with post-translational muta-
5
,6
genesis, these glycopeptides must be prepared using solid-
phase peptide synthesis (SPPS). Additionally, the key O-
GlcNAcylated amino acids required for SPPS must also be
synthesized, rendering site-specific modification of proteins via
chemical ligation methods out of reach for many non-
specialists.
The majority of published routes for O-GlcNAcylated amino
acids require the prior synthetic preparation of complex
glycosyl halides, trichloroacetimidates, acetates, or thioglyco-
sides that may contain special protecting groups, such as
trichloroethoxycarbonyl (Troc), at the C-2 amino group of
7
−12
glucosamine (Scheme 1).
introduced in order to enable neighboring group participation
NGP), which results in the formation of an intermediate
These protecting groups are
quite challenging because of its poor reactivity and inclination
to form a stable 2-methyloxazoline that results from proton
(
13
abstraction (Scheme 2). As noted above, the acetyl group is
therefore typically replaced by the more reactive carbamates
that contain additional electronegative atoms. Even so, the
oxazolinium that restricts the nucleophile’s approach from the
α-face and thus favors the formation of β-glycosides.
Theoretically, the same stereochemical outcome can be
achieved with the use of the native acetyl group at the C-2
amino group of glucosamine, but experimentally the
glycosylation of this charged intermediate has proven to be
Received: July 12, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02182
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Postulating InBr -Mediated Glycosylation
3
major drawbacks to the use of these intricate donor sugars is
the need for stoichiometric amounts of toxic promoters (e.g., 3
equiv of BF ·OEt ), extremely dry conditions (e.g., molecular
Table 1. Screening Conditions for Glycosylation of Fmoc-
Ser-OH
a
3
2
sieves), and additional chemical transformations after the
glycosylation event (e.g., N-Troc deprotection/N-acetylation).
It is therefore imperative to develop an operationally simple
synthetic route to O-GlcNAc-modified amino acids in order to
expedite the generation of glycopeptides for both biological
3
and functional studies.
b
Recently, Polt and co-workers demonstrated the ability of
entry
equiv of donor
InBr
3
(mol %)
solvent
conv. (%)
indium bromide (InBr ) to mediate the activation of glucose
1
2
3
4
5
6
7
8
9
10
11
12
13
1
2
3
3
3
3
3
3
3
3
3
3
100
100
100
5
10
20
50
20
20
20
20
20
20
DCE
DCE
DCE
DCE
DCE
DCE
DCE
29
30
63
49
64
75
77
2
64
61
22
2
3
anomeric acetates and facilitate the formation of an
acyloxonium ion that results from NGP of the C-2
1
4−16
acetate.
This reactive intermediate was suggested to be
responsible for the exclusive formation of β-glycosides. Thus,
we hypothesized that InBr may also be able to mediate the
3
activation of 2-acetamido-2-deoxy-1,3,4,6-tetra-O-acetyl-β-D-
c
glucopyranose (β-Ac GlcNAc) and favor β-glycosides through
THF
toluene
4
NGP of the C-2 amide (Scheme 2, step 1). As previously
detailed, the only major concern was that the poor reactivity of
the 2-methyloxazolinium and the stability of its neutral
counterpart (2-methyloxazoline) that are expected to be
generated in situ could hinder the glycosylation event.
However, in support of our hypothesis, Braga and co-workers
demonstrated the ability of indium(III) salts to cleave simple
oxazolines, yielding β-selenoamides through coordination of
CH CN
3
c
CHCl3
dioxane
DCE
d
3
77
a
Unless otherwise noted, reactions were run at 80 °C and at 200 mM
concentration of the acceptor. Determined by HPLC. The reaction
was run at 60 °C. α-Ac4GlcNAc was used.
b
c
d
17
the nitrogen heteroatom. Thus, it seemed plausible that
InBr could also promote the ring opening of the oxazoline
Although the caramelization temperature is typically high, it is
well-known that strongly acidic or basic conditions can
3
derived from β-Ac GlcNAc (Scheme 2, step 2), which is
4
18,19
generally considered an unreactive byproduct of glycosylation
reactions.
accelerate the process.
With that in mind, we hypothesized
that lowering the amount of InBr could resolve the formation
3
Using Fmoc-Ser-OH as a model acceptor (Table 1), we
attempted the glycosylation reaction at an analytical scale with
of impurities. Excitingly, lowering the loading of InBr with
3
respect to the donor minimized the formation of impurities
(Figure S2) and also improved the product conversion (entries
4−9). At lower loadings, we found there to be a good
relationship between the amount of Lewis acid used and the
product conversion.
However, the conversion did not notably improve with
catalyst loadings above 20 mol %, indicating that catalytic
stoichiometric amounts of InBr3 and β-Ac GlcNAc in
4
dichloroethane (DCE) (entry 1). After the reaction vessel
was heated at 80 °C for 16 h, analysis of the crude mixture by
reversed-phase high-performance liquid chromatography (RP-
HPLC) revealed a very complex mixture of signals (Figure S1).
By the use of a previously synthesized standard, the product
signal for β-Ac GlcNAc serine was identified from the crude
amounts of InBr are optimal to achieve maximal conversion
and minimal side-product formation. The highest conversions
were obtained with dichloroethane (DCE) compared with
3
3
trace and verified using mass spectroscopy. Despite the very
low conversion of 29%, these results encouraged us to further
optimize the reaction through manipulation of the amounts of
sugar donor and Lewis acid, the concentration, and the solvent.
As indicated in Table 1, we found that increasing the amount
chloroform (CHCl ) and other non-halogenated solvents such
3
as toluene, acetonitrile (CH CN), tetrahydrofuran (THF), and
3
1,4-dioxane (entries 8−12). We also tested the reaction with
of β-Ac GlcNAc could improve the product conversion
the less-reactive α-Ac GlcNAc (entry 13). Interestingly, the
4
4
(
(
entries 1−3) but did not prevent the formation of impurities
two anomers display similar reactivities and result in nearly
identical product conversions (Figure S4). However, given its
Figure S1). We postulated that stoichiometric InBr , together
3
with the high temperature and acidity of the reaction mixture
commercial availability, we decided to use β-Ac GlcNAc for
4
(
pH ∼2), facilitated caramelization of the sugar donor. This
the remainder of the work described in this paper.
Having optimized the conditions to maximize the product
conversion, we proceeded to run the reaction at a larger scale
was supported by the presence of high-molecular-weight
products and the dark-brown color of the reaction mixtures.
B
DOI: 10.1021/acs.orglett.8b02182
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
with a refluxing temperature of 84 °C for 16 h (Table 2).
Scheme 3. In Situ Deprotection and Glycosylation
Following purification, Ac GlcNAc serine was isolated in 80%
3
Table 2. Evaluation of the Substrate Scope under the
Optimized Conditions
a
b
entry Fmoc-amino acid
product
yield (%)
α:β
1
2
3
4
Fmoc-Ser-OH
Fmoc-Thr-OH
Fmoc-Thr-OH
β-Ac GlcNAc serine
80
77
71
β
β
β
β
3
β-Ac GlcNAc threonine
3
c
c
β-Ac GlcNAc threonine
3
Fmoc-Cys-OH
β-Ac GlcNAc cysteine
50
3
a
b
1
c
formation of β-Ac GlcNAc cysteine Pfp ester in 73% yield
3
Isolated yields. Determined by H NMR analysis. Monohydrate.
(
Scheme 3).
In light of the exclusive formation of β-glycosides using
yield (entry 1) exclusively as the β-anomer as determined by
NMR spectroscopy (Figure S9). Importantly, we also did not
observe any racemization of the amino acid in the NMR
spectra, although we did not characterize this possibility
further. Extension of the same conditions to the secondary
various acceptors, we reasoned that NGP of the C-2 amide to
generate the 2-methyloxazoline must be occurring as
postulated in Scheme 2 (step 1). In support of this, we were
able to detect formation of the 2-methyloxazoline during the
course of the reactions by mass spectroscopy (Figure S31), but
its isolation from the crude mixtures proved to be very difficult.
alcohol of Fmoc-Thr-OH yielded β-Ac GlcNAc threonine in
3
7
7% yield (entry 2). Notably, these yields were obtained
To test whether InBr could also catalyze the ring opening of
3
without the need for dry solvents or special handling of the
catalyst, suggesting a resilience to moisture. In order to
investigate the effect of water on the catalyst, we purchased the
monohydrate of Fmoc-Thr-OH and submitted it to the
optimized conditions. Surprisingly, 1 equiv of water had only
the 2-methyloxazoline as hypothesized in Scheme 2 (step 2),
we synthesized the 2-methyloxazoline according to the
22
published literature procedure and submitted it to the
optimized conditions using Fmoc-Ser-OH as the amino acid.
HPLC analysis of the crude reaction mixture revealed a
chromatogram almost identical to that for the reaction with β-
a very small effect, as β-Ac GlcNAc threonine was isolated in
3
7
1% yield (entry 3). Thus, this highlighted the ability of the
Ac GlcNAc (Figure S5), and NMR analysis confirmed the
4
reaction to tolerate the presence of water. Notably, these
reactions can be easily scaled up to a gram with reproducible
yields that range between 70 and 80%. Furthermore, the β-
glycosides generated according to Table 2 can be readily
activated for SPPS as pentafluorophenyl (Pfp) esters in great
yields (see the Supporting Information).
exclusive formation of the β-anomer. Importantly, no product
formation was detected in the absence of InBr (Figure S27).
3
These data provide evidence for the 2-methyloxazoline as a key
intermediate and suggest that InBr is a dual catalyst in these
3
reactions with roles in anomeric activation and ring opening of
the 2-methyloxazoline. In order to evaluate other 2-acetamido
Given that we had previously demonstrated the enzymatic
donors, we synthesized α-Ac GalNAc and submitted it to the
stability of S-GlcNAc (β-Ac GlcNAc cysteine) and its potential
4
3
conditions developed in Table 2 using Fmoc-Ser-OH. These
conditions resulted in 69% product conversion and exclusive
formation of the β-glycoside as determined by NMR analysis
Figures S6, S21, and S30).
In summary, we have reported the facile synthesis of serine,
utility for in vivo studies, we decided to further extend the
20
reaction to Fmoc-Cys-OH. Unfortunately, attempts with the
commercial hydrate afforded β-Ac GlcNAc cysteine in 50%
3
(
yield (entry 4). We attributed the lower yield to the oxidation
of cysteine in the presence of stoichiometric amounts of water.
To overcome the relatively poor yield obtained from the
hydrate, we sought to protect the thiol with the acid-labile
triphenylmethyl (trityl) group. We reasoned that the acidic
conditions generated from the activation of the anomeric
acetate could facilitate the in situ deprotection, thereby
unmasking the thiol for nucleophilic attack (Scheme 3). To
prevent the addition of the trityl cation back onto to the thiol,
triisopropylsilane (TIPS) could be added to scavenge the
cation. However, given the ability of tertiary silanes to reduce
threonine, and cysteine β-glycosides using commercially
available β-Ac GlcNAc and catalytic amounts of InBr . The
4
3
present route yields exclusively the β-glycosides in yields
comparable to those for published methods without the need
for several chemical transformations. Unlike other Lewis acids
used in glycosylation reactions, indium can tolerate moderate
levels of water, as exemplified by the glycosylation of amino
acid hydrates. Lastly, we have provided evidence that supports
neighboring group participation of the C-2 amide and
21
carboxylic acids in the presence Lewis acids, we chose to
protect the C-terminal acid with a Pfp ester. Since it is a
common practice to activate the β-glycosides prior to SPPS,
the use of the Pfp ester would remove the need for additional
chemical transformations following the glycosylation. As
postulated, treatment of Fmoc-Cys(Trt)-OPfp with the
optimized reaction conditions and excess TIPS led to the
demonstrates the dual catalyst feature of InBr in the activation
3
and glycosylation event. Overall, this synthetic route represents
an excellent alternative to current procedures and does not
require extensive experience in carbohydrate chemistry. We
believe that these convenient features will make this synthetic
route the standard choice in the preparation of β-Ac GlcNAc
3
amino acids.
C
DOI: 10.1021/acs.orglett.8b02182
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(
17) Braga, A. L.; Vargas, F.; Galetto, F. Z.; Paixao
̃
, M. W.; Schwab,
ASSOCIATED CONTENT
Supporting Information
■
R. S.; Taube, P. S. Eur. J. Org. Chem. 2007, 2007, 5327.
18) Hrynets, Y.; Ndagijimana, M.; Betti, M. J. Agric. Food Chem.
2015, 63, 6249.
*
S
(
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02182.
(
(
19) Hong, P. K.; Betti, M. Food Chem. 2016, 212 (C), 234.
20) De Leon, C. A.; Levine, P. M.; Craven, T. W.; Pratt, M. R.
Biochemistry 2017, 56, 3507.
(
(
Full experimental procedures and characterization data
for all compounds, including NMR spectra (PDF)
21) Bezier, D.; Park, S.; Brookhart, M. Org. Lett. 2013, 15, 496.
22) Nakabayashi, S.; Warren, C. D.; Jeanloz, R. W. Carbohydr. Res.
1986, 150, c7.
AUTHOR INFORMATION
Corresponding Author
■
*Address: University of Southern California, 840 Downey
Way, LJS250, Los Angeles, CA 90089. Phone: (213) 740-3014.
E-mail: matthew.pratt@usc.edu.
ORCID
Matthew R. Pratt: 0000-0003-3205-5615
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
C.A.D.L. is a National Science Foundation Graduate Research
Fellow (DGE-0937362). This research was supported by the
National Institutes of Health (R01GM114537 to M.R.P.). The
authors thank Narek Darabedian (USC) for assistance with
figure preparation and compound characterization. C.A.D.L.
thanks his family for their unwavering support and Dr. Ricardo
Pacheco (USC) for insightful discussions.
REFERENCES
■
(
(
(
1) Bond, M. R.; Hanover, J. A. J. Cell Biol. 2015, 208, 869.
2) Yang, X.; Qian, K. Nat. Rev. Mol. Cell Biol. 2017, 18, 452.
3) Chuh, K. N.; Batt, A. R.; Pratt, M. R. Cell Chem. Biol. 2016, 23,
8
(
6.
4) Yuzwa, S. A.; Yadav, A. K.; Skorobogatko, Y.; Clark, T.;
Vosseller, K.; Vocadlo, D. J. Amino Acids 2011, 40, 857.
5) Wright, T. H.; Bower, B. J.; Chalker, J. M.; Bernardes, G. J. L.;
(
Wiewiora, R.; Ng, W. L.; Raj, R.; Faulkner, S.; Vallee, M. R. J.;
Phanumartwiwath, A.; Coleman, O. D.; Thezenas, M. L.; Khan, M.;
Galan, S. R. G.; Lercher, L.; Schombs, M. W.; Gerstberger, S.; Palm-
Espling, M. E.; Baldwin, A. J.; Kessler, B. M.; Claridge, T. D. W.;
Mohammed, S.; Davis, B. G. Science 2016, 354, aag1465.
(
6) Yang, A.; Ha, S.; Ahn, J.; Kim, R.; Kim, S.; Lee, Y.; Kim, J.; Soll,
D.; Lee, H. Y.; Park, H. S. Science 2016, 354, 623.
7) Enugala, R.; Carvalho, L. C. R.; Dias Pires, M. J.; Marques, M.
M. B. Chem. - Asian J. 2012, 7, 2482.
8) Meinjohanns, E.; Meldal, M.; Bock, K. Tetrahedron Lett. 1995,
6 (50), 9205.
9) Dullenkopf, W.; CastroPalomino, J.; Manzoni, L.; Schmidt, R.
Carbohydr. Res. 1996, 296, 135.
10) Saha, U. K.; Schmidt, R. R. J. Chem. Soc., Perkin Trans. 1 1997,
2, 1855.
11) Mitchell, S. A.; Pratt, M. R.; Hruby, V. J.; Polt, R. J. Org. Chem.
001, 66, 2327.
12) Marotta, N. P.; Cherwien, C. A.; Abeywardana, T.; Pratt, M. R.
ChemBioChem 2012, 13 (18), 2665.
13) Arsequell, G.; Krippner, L.; Dwek, R. A.; Wong, S. Y. C. J.
Chem. Soc., Chem. Commun. 1994, 20 (20), 2383.
14) Lefever, M. R.; Szabo, L. Z.; Anglin, B.; Ferracane, M.; Hogan,
J.; Cooney, L.; Polt, R. Carbohydr. Res. 2012, 351, 121.
15) Coss, C.; Carrocci, T.; Maier, R. M.; Pemberton, J. E.; Polt, R.
Helv. Chim. Acta 2012, 95, 2652.
16) Szabo, L. Z.; Hanrahan, D. J.; Jones, E. M.; Martin, E.;
Pemberton, J. E.; Polt, R. Carbohydr. Res. 2016, 422, 1.
(
(
3
(
(
1
(
2
(
(
(
̀
(
(
̀
D
DOI: 10.1021/acs.orglett.8b02182
Org. Lett. XXXX, XXX, XXX−XXX