Chemical N-glycosylation by asparagine under integrated microfluidic/batch conditions

Article abstract of DOI:10.1055/s-0029-1217343

An integrated microfluidic/batch system was applied to the chemical N-glycosylation by the asparagine amide group, a key glycosyl bond-formation reaction in the synthesis of N-glycopeptides. By applying the advantageous features of microfluidic conditions

Full text of DOI:10.1055/s-0029-1217343

LETTER
1571
Chemical N-Glycosylation by Asparagine under Integrated Microfluidic/
Batch Conditions
C
hemical N-Glyc
o
a
sylation by
t
A
spar
s
agine unori Tanaka, Takuya Miyagawa, Koichi Fukase*
Department of Chemistry, Graduate School of Science, Osaka University, Machikaneyama 1-1, Toyonaka, Osaka 560-0043, Japan
Fax +81(6)68505419; E-mail: koichi@chem.sci.osaka-u.ac.jp
Received 16 February 2009
investigated as more general solvents for the preparative-
scale synthesis of the N-glycopeptides fragments. When
1.5 equivalents of glycosyl donor 1a with respect to the
asparagine derivative were treated with 0.2 equivalent of
Abstract: An integrated microfluidic/batch system was applied to
the chemical N-glycosylation by the asparagine amide group, a key
glycosyl bond-formation reaction in the synthesis of N-glycopep-
tides. By applying the advantageous features of microfluidic condi-
tions, that is, efficient mixing and rapid heat transfer, the TMSOTf in dichloromethane at room temperature for 12
GlcNTrocbAsn and the Fuca(1–6)GlcNTrocbAsn fragments were
hours, desired N-glycoside 3a was obtained in 61% yield
(entry 1), which is comparable to the reported results.3 Be-
cause TLC analysis indicated that the donor decomposed
during the reaction, the donor solution was slowly added
to a premixed solution of the acceptor and TMSOTf, but
the reaction did not proceed (entry 2). Employing excess
efficiently prepared.
Key words: N-glycosylation, asparagine, microreactor, oligosac-
charide, N-glycopeptide
Chemical N-glycosylation by asparagine residues is a donor (3.0 equiv) and an activator (0.5 equiv) did not in-
challenging topic in N-glycopeptides synthesis1 due to the crease product formation (50%, entry 3), and the yield de-
inherently low nucleophilicity of the amide nitrogen to- creased when propionitrile was used as the solvent (27%,
ward glycosylation. Kahne and co-workers2a initially re- entry 4).
ported N-glycosylation through amide functions by
N-Glycosylation was less effective when more sterically
demanding disaccharide donor, Fuca(1–6)GlcNTroc N-
phenyltrifluoroacetimidate (1b), was glycosylated with
asparagine amide; 35% of N-glycoside 3b was produced
under the optimal conditions obtained for 1a (entry 5). Al-
though 2.0 equivalents of acceptor 2 relative to donor 1b
utilizing N-silylated acetamide to enhance the nucleophi-
licity of the amide nitrogen.2b Takahashi and co-workers
have recently reported a more direct and efficient N-gly-
cosylation of the glycosyl imidates via protected aspa-
ragine derivatives.3 Their protocol utilizes N-
phenyltrifluoroacetimidate4 as a leaving group, TMSOTf
resulted in an improved yield (47%, entry 6), the excess
as an activator, and nitromethane as a solvent; glucosyl,
acceptor (5.0 equiv) decreased the efficiency (32%, entry
glucosaminyl, mannosyl, and galactosyl imidates have
7).
been N-glycosylated in 68–99% yields, especially by the
Time-course TLC analyses of entry 6 provided informa-
protected monoasparagine residue, Z-Asn(OAll).
tion about the intermediates during the course of the reac-
tion (Figure 1). Namely, the addition of TMSOTf into a
mixture of the donor and acceptor at room temperature (5
min TLC) provided several intermediates, which presum-
ably consisted of the silylated acceptor, O-glycoside, N-
glycoside, hydrolyzed donor, etc. However, the mixture
of these products ultimately converged into desired N-gly-
coside 3b, decomposed donor, and unconsumed acceptor
Our research on the solid-phase synthesis of N-glycans
and N-glycopeptides requires a sufficient amount of the
GlcNTrocbAsn and Fuca(1–6)GlcNTrocbAsn motifs.5
Although Takahashi’s protocol is very attractive, ni-
tromethane, which is the optimum solvent for efficient N-
glycosylation, is explosive, and not suitable for large-
scale synthesis. Herein, we report a practical and high-
yielding N-glycosylation in dichloromethane using inte-
2 (12 h TLC). Although the initial intermediates detected
on the 5 min TLC were not examined in detail, we hypo-
grated microfluidic/batch conditions. This method is suc-
cessful on a few gram-scale N-glycosylation without
thesized that the heat generated during the initial mixing
nitromethane, and is readily applicable to N-glycans and/
might yield a mixture of the intermediates, resulting in the
decomposition of the donor and eventually the decreased
or N-glycopeptides synthesis.
We initially examined N-glycosylation of the suitably yield of N-glycoside 3b. Surprisingly, more of the accep-
protected N-Troc glucosaminyl N-phenyltrifluoroacetimi- tor was recovered upon slowly adding a Lewis acid at 0 °C
date (1a) with Z-L-Asn(OBn) (2) under batch conditions than the reaction performed at room temperature in entry
according to Takahashi’s procedure (Table 1). Besides 6 (less than 10% of 3b).
nitromethane, dichloromethane, and propionitrile were
Based on the observations in Table 1, which suggest that
efficient mixing and heat transfer may be responsible for
N-glycosylation, we decided to examine the microfluidic
conditions. A continuous flow microreactor, an innova-
tive technology, has been used to realize efficient mixing
SYNLETT 2009, No. 10, pp 1571–1574
Advanced online publication: 02.06.2009
DOI: 10.1055/s-0029-1217343; Art ID: U01809ST
x
x
.x
x
.2
0
0
9
© Georg Thieme Verlag Stuttgart · New York

1572
K. Tanaka et al.
LETTER
microfluidic b-mannosylation.8e Namely, the reaction so-
lution prepared in the micromixing system was subse-
quently inserted into a batch system, and conventionally
stirred in a flask for hours to complete the reaction
(Table 2). Small quantities of materials were used to de-
termine the optimal conditions, that is, substrate concen-
trations, mixing speed, temperature, and flow rate
(optimization factors 1–4, in Table 2). Thus, a dichlo-
romethane solution of glucosaminyl donors 1a,b and
asparagine acceptor 2 with various concentrations (opti-
mization factor 1) were mixed with a TMSOTf solution in
dichloromethane to determine the optimal concentration
(factor 2) at the appropriate temperature (factor 3) using
an IMM micromixer9 at various flow rates (factor 4). Un-
like the b-mannosylation case8e where micromixing oc-
curs at a very low temperature (–90 °C) using a Comet X-
01 micromixer10 with a channel width of ca. 500 mm, we
took advantage of the more delicate microstructure of
IMM (40 mm) and assumed that solution blockage might
not be a severe problem due to the relatively good solubil-
ity of both the donor and acceptor in dichloromethane
above room temperature. For the rapid optimization of the
microfluidic conditions, the product yields were initially
estimated by the TLC stain contrasts using ImageJ 1.40.
Table 1 N-Glycosylation with Asparagine Amidea under Batch
Conditions
OR
NHZ
O
NPh
CF3
OBn (or Allyl)
O
FmocO
BnO
+
H2N
O
O
TrocHN
1a,b
2
OR
O
H
O
FmocO
N
TMSOTf
BnO
OBn (or Allyl)
TrocHN
MS4A, r.t., 12 h
O
NHZ
3a,b
OAc
AcO
1a, 3a: R = Bn
1b, 3b: R =
BnO
O
Entry Donor
(equiv)
Acceptor 2 TMSOTf Solvent
Yield
(%)b
(equiv)
(equiv)
1
2c
3
4
5
6
7
1a (1.5)
1a (1.5)
1a (3.0)
1a (1.5)
1b (1.5)
1b (1.0)
1b (1.0)
1.0
0.2
CH2Cl2
CH2Cl2
CH2Cl2
EtCN
61
0
1.0
0.2
1.0
0.5
50
27
35
47
32
Table 2 shows representative data obtained by the microf-
luidic N-glycosylation of 1a and 1b with asparagine de-
rivative 2, when micromixing was performed at room
temperature, and the flow rate and the concentration of
TMSOTf were fixed at 1.0 mL/min and 43 mM, respec-
tively. All the entries in Table 2 using batch stirring at
room temperature required 12 hours to complete the N-
glycosylation. When the concentrations of monosaccha-
ride imidate 1a and acceptor 2 were adjusted to 86 mM
(1.0 equiv) and 172 mM (2.0 equiv), respectively, GlcN-
TrocbAsn fragment 3a was obtained in 60% (entry 1). A
slight excess of donor 1a relative to 2, that is, 130 mM of
1a (1.5 equiv) and 86 mM (1.0 equiv) of 2, gave similar
results (55%, entry 2). However, applying more of donor
1a (259 mM, 3.0 equiv) completely consumed the aspar-
agine acceptor, and desired N-glycoside 3a was produced
in 81% yield (entry 3).
1.0
0.2
1.0
0.2
CH2Cl2
CH2Cl2
CH2Cl2
2.0
0.2
5.0
0.2
a Benzyl ester was used for entries 1–4, and allyl ester was used for
entries 5–7.
b Isolated yields.
c Donor solution was added slowly to a premixed solution of the ac-
ceptor and TMSOTf.
and rapid heat transfer in organic syntheses.6,7 Once the
reaction conditions are optimized for a small-scale opera-
tion, the same conditions are directly applicable to large-
scale synthesis under the flow process. We have recently
applied a microfluidic system to cation-mediated
reactions8 to realize an improved a-sialylation,8b dehydra-
tion,8c and reductive opening of the benzylidene acetal
groups in sugar.8d Moreover, we have developed an inte-
grated microfluidic/batch system to manage reactions re-
quiring a long reaction time after micromixing. This
protocol has been applied to practical b-mannosylation.8e
Because the inefficiency of all these cation-mediated
reactions under the batch process are due to inefficient
mixing with the acid reagents, we envisioned that micro-
fluidic conditions might also lead N-glycosylation to take
place efficiently even in dichloromethane. Hence, this
protocol was used in the preparative synthesis of N-glyco-
sides 3a,b.
An efficient microfluidic N-glycosylation was also real-
ized using disaccharide imidate 1b as a donor; but the mi-
cromixing between 22 mM (1.0 equiv) of 1b and 43 mM
(2.0 equiv) of 2 provided N-glycoside 3b only in 27%
yield (entry 4). However, doubling the concentrations for
both the reactants dramatically enhanced the yield (84%,
entry 5). Further trials with higher concentrations of the
reactants, that is, 86 mM of the donor and 172 mM of the
acceptor, caused the microchannel to become blocked due
to the limited solubility of donor 1b. The batch and
microfluidic reactions displayed vastly different TLC be-
haviors (Figure 1). The batch reaction provided several
intermediates on the 5 min TLC, but gradually produced
N-glycoside 3b as well as decomposed materials. On the
other hand, the 5 min TLC under the microfluidic condi-
tions detected only a few spots with obvious structures, in-
cluding the starting donor, the acceptor, and N-glycoside;
Because the current N-glycosylation requires a long reac-
tion time (Table 1), a microfluidic apparatus was con-
structed based on our previous experiences with
Synlett 2009, No. 10, 1571–1574 © Thieme Stuttgart · New York

LETTER
Chemical N-Glycosylation by Asparagine
1573
Table 2 N-Glycosylation with Asparagine Amide under Microfluidic Conditionsa
donor and acceptor
Lewis acid
(factor 1: concn)
(factor 2: concn)
OR
TMSOTf in CH2Cl2
NPh
CF3
batch
O
FmocO
BnO
r.t.
12 h
O
TrocHN
(factor 3: temp.)
micromixer
1a,b
NHZ
O
r.t.
(factor 4: flow rate)
OAllyl
H2N
1.0 mL/min
O
2
in CH2Cl2
microfluidic
OR
OAc
AcO
1a, 3a: R = Bn
1b, 3b: R =
O
H
O
FmocO
N
BnO
BnO
O
OAllyl
TrocHN
3a,b
O
NHZ
Entry
Donor (mM)
1a (86)
Acceptor 2 (mM)
TMSOTf (mM)
Yield (%)b
1
2
3
4
5
172
86
43
43
43
43
43
60
55
81
27
84
1a (130)
1a (259)
1b (22)
86
43
1b (43)
86
a Solutions flowed through a Teflon tube (f = 1.0 mm) at a rate of 1.0 mL/min, and micromixing was performed at r.t. The Lewis acid concen-
tration was fixed at 43 mM.
b Yields were estimated based on TLC stain contrast detected by ImageJ 1.40 (see Figure 1).
as rapid heat transfer might be achieved for the current N-
glycosylation with the asparagine amide.
Preliminary optimized conditions in Table 2 were then ap-
plied to the preparative scale synthesis of N-glycosides 3a
and 3b; continuously pumping the stock solutions of the
substrates and the Lewis acid activator into the integrated
microfluidic/batch system gave isolated yields of 85% for
3a and 84% for 3b.11
In summary, we established an efficient N-glycosylation
of glycosyl N-phenyltrifluoroacetimidates with aspa-
ragine amide under integrated microfluidic/batch condi-
tions. The optimal conditions were rapidly determined
using a fluidic system and simple TLC monitoring. The
subsequent preparative-flow reaction led to the reproduc-
ible synthesis of GlcNTrocbAsn and the Fuca(1–6)GlcN-
TrocbAsn derivative in high yields, and is applicable to
our N-glycopeptides synthesis. The success of the proto-
col must be due to efficient mixing and rapid heat transfer
temperature control, which inhibit decomposition of the
reactants and/or the formation of complicated glycosyl in-
termediates during the batch mixing between the donor,
Figure 1 TLC analysis of batch and microfluidic N-glycosylation
of disaccharide 1b with asparagine 2 (eluent, hexane–EtOAc = 1:1)
N-glycoside was then gradually and cleanly produced
over 12 hours. Thus, efficient microfluidic mixing as well acceptor, and activator.
Synlett 2009, No. 10, 1571–1574 © Thieme Stuttgart · New York

1574
K. Tanaka et al.
LETTER
Codee, J. D. C.; Seeberger, P. H. Org. Lett. 2007, 9, 2285.
(h) Usutani, H.; Tomida, Y.; Nagaki, A.; Okamoto, H.;
Nokami, T.; Yoshida, J. J. Am. Chem. Soc. 2007, 129, 3046.
(i) Tanaka, K.; Motomatsu, S.; Koyama, K.; Fukase, K.
Tetrahedron Lett. 2008, 49, 2010.
Acknowledgment
This work was partly supported by Grants-in-Aids for Scientific Re-
search No. 19681024 and 19651095 from the Japan Society for the
Promotion of Science, Collaborative Development of Innovative
Seeds from Japan Science and Technology Agency (JST), New
Energy and Industrial Technology Development Organization
(NEDO, project ID: 07A01014a), Research Grants from Yamada
Science Foundation as well as the Molecular Imaging Research Pro-
gram, and Grants-in-Aid for Scientific Research from the Ministry
of Education, Culture, Sports, Science and Technology (MEXT) of
Japan.
(8) (a) Fukase, K.; Takashina, M.; Hori, Y.; Tanaka, D.; Tanaka,
K.; Kusumoto, S. Synlett 2005, 2342. (b) Tanaka, S.; Goi,
T.; Tanaka, K.; Fukase, K. J. Carbohydr. Chem. 2007, 26,
369. (c) Tanaka, K.; Motomatsu, S.; Koyama, K.; Tanaka,
S.; Fukase, K. Org. Lett. 2007, 9, 299. (d) Tanaka, K.;
Fukase, K. Synlett 2007, 164. (e) Tanaka, K.; Mori, Y.;
Fukase, K. J. Carbohydr. Chem. 2009, 28, 1.
(9) IMM micromixer: http://www.imm-main2.de/
(10) Comet X-01 micromixer: http://homepage3.nifty.com/
techno-applications/ or e-mail: yukio-matsubara@nifty.com.
(11) Procedure of N-Glycosylation Using an Integrated
Microfluidic/Batch System
References and Notes
(1) (a) Introduction to Glycoscience Synthesis of
Carbohydrates, In Comprehensive Glycoscience, from
Chemistry to Systems Biology, Vol 1; Kamerling, J. P.;
Boons, G.-J.; Lee, Y. C.; Suzuki, A.; Taniguchi, N.;
Voragen, A. G. J., Eds.; Elsevier: Oxford, 2007.
(b) Tanaka, K.; Fukase, K. Polymer-Supported and Tag-
Assisted Methods in Oligosaccharide Synthesis in
Glycoscience: Chemistry and Chemical Biology, 2nd ed.,
Vol. I-III; Fraser-Reid, B. O.; Tatsuta, K.; Thiem, J., Eds.;
Springer: New York, 2009.
(2) (a) Kahne, D.; Walker, S.; Cheng, Y.; Engen, D. V. J. Am.
Chem. Soc. 1989, 111, 6881. (b) Garcia, B. A.; Gin, D. Y.
J. Am. Chem. Soc. 2000, 122, 4269.
(3) Tanaka, H.; Iwata, Y.; Takahashi, D.; Adachi, M.;
Takahashi, T. J. Am. Chem. Soc. 2005, 127, 1630.
(4) (a) Yu, B.; Tao, H. Tetrahedron Lett. 2001, 42, 2405.
(b) Adinolfi, M.; Barone, G.; Iadonisi, A.; Schiattarella, M.
Synlett 2002, 269. (c) Cai, S.; Yu, B. Org. Lett. 2003, 5,
3827.
(5) Tanaka, K.; Fujii, Y.; Tokimoto, H.; Mori, Y.; Tanaka, S.;
Bao, G.-m.; Siwu, E. R. O.; Nakayabu, A.; Fukase, K. Chem.
Asian J. 2009, 4, 574; based on the established solid-
supported protocol, N-glycosylation strategy with the
optimal protection groups on 1a,b and 2 will be applied
to the future glycopeptide synthesis.
(6) For representative reviews, see: (a) Microreaction
Technology; Ehrfeld, W., Ed.; Springer: Berlin, 1998.
(b) Microsystem Technology in Chemistry and Life Sciences;
Manz, A.; Becker, H., Eds.; Springer: Berlin, 1998.
(c) Ehrfeld, W.; Hessel, V.; Lowe, H. Microreactors; Wiley-
VCH: Weinheim, 2000. (d) Hessel, V.; Hardt, S.; Lowe, H.
Chemical Micro Process Engineering; Wiley-VCH:
Weinheim, 2004. (e) Yoshida, J.-I.; Suga, S.; Nagaki, A.
J. Synth. Org. Chem. Jpn. 2005, 63, 511.
(7) For recent applications, especially to oligosaccharides
synthesis, see: (a) Pennemann, H.; Hessel, V.; Loewe, H.
Chem. Eng. Sci. 2004, 59, 4789. (b) Jahnisch, K.; Hessel,
V.; Loewe, H.; Baerns, M. Angew. Chem. Int. Ed. 2004, 43,
406. (c) Nagaki, A.; Togai, M.; Suga, S.; Aoki, N.; Mae, K.;
Yoshida, J.-I. J. Am. Chem. Soc. 2005, 127, 11666.
(d) Ratner, D. M.; Murphy, E. R.; Jhunjhunwala, M.;
Snyder, D. A.; Jensen, K. F.; Seeberger, P. H. Chem.
Commun. 2005, 578. (e) Flogel, O.; Codee, J. D. C.;
Seebach, D.; Seeberger, P. H. Angew. Chem. Int. Ed. 2006,
45, 7000. (f) Geyer, K.; Seeberger, P. H. Helv. Chim. Acta
2007, 90, 395. (g) Carrel, F. R.; Geyer, K.; Jeroen, D. C.;
A solution of TMSOTf (33 mL, 180 mmol, 43 mM) in CH2Cl2
(4.2 mL) was injected, in advance, into the micromixer by a
syringe pump at a flow rate of 1.0 mL/min. Then a solution
of donor 1a (1.0 g, 1.1 mmol, 260 mM) and acceptor 2 (110
mg, 360 mmol, 86 mM) dissolved in CH2Cl2 (4.2 mL) was
injected into the IMM micromixer by another syringe pump
at a flow rate of 1.0 mL/min. The reaction was mixed at r.t.
After the reaction mixture was allowed to flow at r.t. for an
additional 94 s through a Teflon tube reactor (F = 1.0 mm,
l = 1.0 m), the mixture was introduced into a flask, and
stirred for 12 h at this temperature. Then the mixture was
quenched by an aq NaHCO3 solution. The resulting mixture
was extracted with EtOAc, washed with brine, dried over
Na2SO4, filtered, and concentrated in vacuo to give the
crude product. The residue was purified by column
chromatography on silica gel (from 25–33% EtOAc in
hexane) to give N-glycoside 3a as a white solid (376 mg,
85%). ESI-MS: m/z calcd for C53H53Cl3N3O13 [M + H]+:
1044.3; found: 1044.2. 1H NMR (500 MHz, CDCl3): d =
7.76 (d, J = 7.6 Hz, 1 H), 7.73 (d, J = 7.5 Hz, 1 H), 7.59 (d,
J = 7.6 Hz, 1 H), 7.55 (d, J = 7.4 Hz, 1 H), 7.41–7.17 (m, 19
H), 6.80 (d, J = 8.5 Hz, 1 H), 5.91 (d, J = 9.2 Hz, 1 H), 5.88–
5.80 (m, 1 H), 5.27 (dd, J = 17.2, 1.3 Hz, 1 H), 5.19 (dd,
J = 10.5, 1.3 Hz, 1 H), 5.14–5.04 (m, 4 H), 4.89 (dd, J = 9.2,
9.2 Hz, 1 H), 4.75 (d, J = 12.1 Hz, 1 H), 4.69 (d, J = 12.0 Hz,
1 H), 4.68–4.58 (m, 3 H), 4.52–4.35 (m, 6 H), 4.15 (dd,
J = 6.9, 6.9 Hz, 1 H), 3.65–3.61 (m, 3 H), 3.56–3.49 (m, 2
H), 2.86 (dd, J = 16.7, 3.8 Hz, 1 H), 2,68 (dd, J = 16.4, 4.2
Hz, 1 H).
Data for 3b
ESI-MS: m/z calcd for C63H67Cl3N3O19 [M + H]+: 1274.3;
found: 1274.2. 1H NMR (500 MHz, CDCl3): d = 7.76 (d,
J = 7.6 Hz, 1 H), 7.73 (d, J = 7.5 Hz, 1 H), 7.61 (d, J = 7.6
Hz, 1 H), 7.60 (d, J = 7.3 Hz, 1 H), 7.41–7.19 (m, 19 H), 6.89
(d, J = 8.2 Hz, 1 H), 5.94 (d, J = 8.6 Hz, 1 H), 5.86–5.79 (m,
1 H), 5.31–5.25 (m, 3 H), 5.18 (d, J = 11.6 Hz, 1 H), 5.13 (d,
J = 12.2 Hz, 1 H), 5.06 (d, J = 12.4 Hz, 1 H), 4.99 (d, J = 3.5
Hz, 1 H), 4.90–4.87 (m, 2 H), 4.76 (d, J = 12.2 Hz, 1 H), 4.68
(dd, J = 12.1, 4.0 Hz, 2 H), 4.62–4.52 (m, 10 H), 4.37 (d,
J = 11.8 Hz, 1 H), 4.23 (dd, J = 6.7, 6.7 Hz, 1 H), 3.80 (dd,
J = 10.2, 3.5 Hz,1 H), 3.71 (dd, J = 11.8, 2.2 Hz, 1 H), 3.67–
3.52 (m, 3 H), 2.85 (dd, J = 16.5, 3.6 Hz, 1 H), 2.69 (dd,
J = 16.5, 3.9 Hz, 1 H), 2.09 (s, 3 H), 1.96 (s, 3 H), 1.06 (d,
J = 6.5 Hz, 3 H).
Synlett 2009, No. 10, 1571–1574 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.