Practical heavy fluorous tag for carbohydrate synthesis with minimal chromatographic purification

Article abstract of DOI:10.1016/j.tetlet.2010.10.016

A practical heavy fluorous tag 5 bound to a benzylic linker was prepared and applied to carbohydrate synthesis. The fluorous tag 5 was readily introduced to the desired hydroxyl group and carboxyl group by using various methods. Synthesis of the oligosacc

Full text of DOI:10.1016/j.tetlet.2010.10.016

Tetrahedron Letters 51 (2010) 6539–6541
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Practical heavy fluorous tag for carbohydrate synthesis with minimal
chromatographic purification
⇑
Kohtaro Goto, Mamoru Mizuno
Laboratory of Glyco-organic Chemistry, The Noguchi Institute, 1-8-1, Kaga, Itabashi-ku, Tokyo, 173-0003, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A practical heavy fluorous tag 5 bound to a benzylic linker was prepared and applied to carbohydrate syn-
thesis. The fluorous tag 5 was readily introduced to the desired hydroxyl group and carboxyl group by
using various methods. Synthesis of the oligosaccharide, which included the terminal structure of class
III mucin, was achieved with single-column chromatographic purification. In addition, because of the
symmetrical structure of 5, each fluorous synthetic intermediate could be analyzed much easier by
NMR spectroscopy than in the case of the fluorous compounds connecting our previous fluorous tags.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 9 September 2010
Revised 4 October 2010
Accepted 5 October 2010
Available online 13 October 2010
Since the seminal articles of fluorous biphasic catalyst tech-
niques in 19941 and the heavy fluorous tag method in 19972, fluor-
ous chemistry has found a wide range of applications3, such as in
catalytic chemistry and synthetic chemistry. A major advantage
of the heavy fluorous tag method is that a non-column chromato-
graphic purification system can be used owing to the easy separa-
tion of the fluorous compounds from the general non-fluorous
compounds. This separation can be readily achieved in a straight-
forward manner with fluorous–organic partitioning. We have also
reported the synthesis of oligosaccharides and peptides by using
the heavy fluorous tag method with ‘polyamide-type’ fluorous
tags4 and the more useful ‘polyether-type’ fluorous tag.5 The de-
sired properties of fluorous tags are (1) high introduction yields
and cleavage yields; (2) easily recyclable; (3) simple analysis of
the NMR spectrum; (4) introducing to desired positions; and (5)
stability under various reaction conditions. However, even with
our previous polyether-type tag5 it was difficult to analyze the
NMR spectra of these fluorous compounds, because the peaks of
the fluorous tags frequently overlap with those of the sugar ring
protons. Furthermore, in oligosaccharide synthesis, the introduc-
tion of these heavy fluorous tags is virtually limited to the anomer-
ic position of the carbohydrate. A more user-friendly fluorous tag,
which would resolve these intractable problems, is essential for
practical application of the heavy fluorous tag method. Herein,
we describe the synthesis of carbohydrates by using more a prac-
tical fluorous tag bound to a benzylic linker.
ation yields, complexity of the NMR spectra, and difficulty in recy-
cling the fluorous tag after the reaction. In order to overcome these
problems, we designed and synthesized a novel fluorous tag 5
(Scheme 1).
A triphenylmethyl group was introduced to a primary alcohol of
pentaerythritol7 (1) to give triol 28, which was coupled with fluor-
ous tosylate9; subsequent removal of the triphenylmethyl group
afforded fluorous alcohol 3 connecting three fluorous chains in
78% yield. Finally,
a,
a0-m-dibromoxylene (4) was introduced as a
benzylic linker to give a fluorous tag 510 in 83% yield.
Next, we investigated the conditions of introduction of the flu-
orous tag 5 to compounds 6, 8, and 10, which are commercially
available (Scheme 2). Method 1 is Williamson ether synthesis, a
conventional and commonly used method of forming ether bonds.
a
b
c
TrO
C
OH
HO
C
OH
3
3
1
2
Br
Br
4
d
HO
C
O
O
C8F17
3
3
The benzyl group is one of the most important and useful pro-
tecting groups in oligosaccharides synthesis, and some benzyl-type
fluorous tags have already been exploited in oligosaccharide syn-
thesis.6 However, these benzyl type tags have some problems that
need to be resolved, such as low introduction yields and glycosyl-
C8F17
O
C
3
5 =
F-tag
Br
Br
Scheme 1. Preparation of benzyl-type fluorous tag 5. Reaction conditions: (a) TrCl,
DMAP, Pyr-DMF, 100 °C, 3 h, 59%; (b) TsO(CH2)3C8F17, NaH, 15-Crown-5, THF, rt,
17 h; (c) CSA, MeOH–CHCl3, rt, 19 h, two steps 78%; (d) NaH, 15-Crown-5, PhCF3,
10 °C, 24 h, 83%.
⇑
Corresponding author.
E-mail address: mmizuno@noguchi.or.jp (M. Mizuno).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.10.016

6540
K. Goto, M. Mizuno / Tetrahedron Letters 51 (2010) 6539–6541
Method 1
O
O
Ph
OH
5
O
12 ; R = H
O
d
O
13 ; R = Bn
O
O
a
b
c
O
O
F-tag
O
O
OMp
9
8
O
O
g
a
6
t
RO
OAc
-
F
5
O
O
OR1
a
7
HO
O
O
AcO
AcO
Method 2
HO
O
5
NH
CCl3
OMp
OH
OH
N3
e
g
HO
a
t
BnO
16
O
-
F
c
b
O
O
; R1 = H
14
15
OMp
O
HO
OMp
f
g
; R1 = Bz
g
a
t
HO
9
HO
8
-
F
OR2
O
3
fluorous
layer
Method 3
R2O
R2O
94% (8 steps)
5
OR3
i
N
3O
OH
O
e
d
Boc
OH
Asp(tBu)
Asp(tBu)
F-tag
Boc
O
HO
HO
O
OMp
O
g
OH
organic
layer
a
t
BnO
AcNH
HO
10
11
-
O
F
O
HO
17 ; R2 = Ac, R3 =Bz
OMp
Scheme 2. Introduction conditions. Reagent and conditions: (a) NaH, 15-Crown-5,
THF, rt, 22 h, quant; (b) n-Bu2SnO, MeOH, reflux, 4 h; (c) TBAB, THF, reflux, 16 h
90%; (d) Cs2CO3, EtOH-H2O, rt, 10 min; (e) THF–DMF, rt, 16 h, quant.
h
; R2 = H, R3 = H
18
19 (8 steps 55% from 5)
1 column chromatography
In this case, 15-Crown-5 is essential for constructing the ether
bond smoothly.5a It seems that 15-Crown-5 can dramatically en-
hance the reactivity between the corresponding alkoxide and the
fluorous compound. Method 2 is the selective introduction by using
n-Bu2SnO, and Method 3 is the introduction to the amino acid
derivative 10 with Cs2CO3.11 The results showed that this fluorous
tag 5 could be easily introduced not only to the desired hydroxyl
group, but also to the carboxyl group, in excellent yields.
Scheme 3. Disaccharide synthesis by using
a fluorous tag 5. Reagent and
conditions: (a) n-Bu2SnO, MeOH, reflux, 4 h; (b) TBAB, THF, reflux, 16 h; (c)
PhCH(OMe)2, CSA, MeCN–HFE710012, rt, 1 h; (d) BnBr, NaH, 15-Crown-5, THF, rt,
1 h; (e) CSA, MeOH–HFE7100, rt, 7 h; (f) BzCl, Et3N, THF, ꢀ20 °C, 14 h; (g) TMSOTf,
CH2Cl2–ether, rt, 2 h; (h) NaOMe, MeOH-HFE7100, rt, 2 h; (i) H2 gas, Pd(OH)2, Ac2O,
AcOH, EtOH–HFE7100, rt, 20 h.
position was conducted as in Scheme 2 to give 9. The formation of
benzaldehyde dimethylacetal produced 12 and subsequent benzy-
lation afforded 13. Cleavage of the benzylidene acetal of 13 gave a
O-4,6 diol 14. Selective benzoylation of 14 to primary alcohol then
This fluorous tag 5 was then applied to the synthesis of disaccha-
ride 19, which included the terminal structure of a class III mucin13
(Scheme 3). The introduction of fluorous tag 5 to compound 8 at O-3
Fluorous tag’s signals are overlapped with
sugar ring protons (H-2,3,4,5,5’,6a’,6b’)
OAc
AcO
ClAcO
OBz
O
O
C8F17
C8F17
O
O
O
BnO
O
O
AcO
BnO
O
C8F17
20
OAc
O
AcO
AcO
OBz
N
3
O
O
O
C
O
C F
8
17
O
OMp
3
BnO
17
Fluorous tag’s
signals only
sugar ring protons
Figure 1. NMR spectra of compounds 17 and 20.5a

K. Goto, M. Mizuno / Tetrahedron Letters 51 (2010) 6539–6541
6541
Waragai, H.; Matsumoto, H.; Ohmae, M.; Ishida, H.-k.; Satoh, A.; Inazu, T. J. Org.
Chem. 2004, 69, 5348; (g) Miura, T.; Satoh, A.; Goto, K.; Murakami, Y.; Imai, N.;
Inazu, T. Tetrahedron: Asymmetry 2005, 16, 3; (h) Mizuno, M.; Matsumoto, H.;
Goto, K.; Hamasaki, K. Tetrahedron Lett. 2006, 47, 8831; (i) Mizuno, M.; Goto, K.;
Miura, T.; Inazu, T. QSAR Comb. Sci. 2006, 25, 742.
produced glycosyl acceptor 15. Compound 15 and glycosyl donor
1614 were coupled by the Schmidt method15 with TMSOTf to give
a fluorous disaccharide 17.16 All acyl groups of 17 were removed
by NaOMe in MeOH–HFE7100 to afford 18, this was followed by
hydrogenation in the presence of Pd(OH)2 to give the crude product
of the desired disaccharide 19. In this synthesis, each compound
bound to the fluorous tag (12–15, 17, and 18) was partitioned into
the fluorous layer,17 and the desired compound 19 was partitioned
into the organic layer.18 No further purification (e.g., by silica-gel
column chromatography) of the fluorous intermediates was con-
ducted. After a single-column chromatographic purification of 19,
pure 1919 was obtained in 58% yield. The fluorous alcohol 320 was
recovered from the fluorous layer in 94% yield and was recyclable.
Finally, a comparison of 1H NMR spectra with fluorous com-
pound 17, which has a pentaerythritol (1) scaffold, and a previ-
ously reported fluorous compound 205a bound to an old type of
fluorous tag, which has a meso-erythritol scaffold, is shown in Fig-
ure 1. In the spectrum of 20 we observed complicated peaks
around 3.5–4.0 ppm (Fig. 1, top panel). These peaks are overlap-
ping signals of the fluorous tag moiety and some sugar ring pro-
tons. In general, many peaks of the sugar ring protons are also
observed around the same range. This result indicates that the pre-
vious tags are not appropriate for efficient oligosaccharide synthe-
sis because of the difficulty in analyzing their NMR spectra. In
contrast, because of the more symmetrical structure of the fluorous
part, the spectrum of 17 was much simpler than that of 20 and the
signals of the fluorous moiety barely influenced the analysis of the
sugar ring protons (Fig. 1, bottom panel).
In conclusion, we developed a more practical fluorous tag 5 that
could be easily introduced not only to various hydroxyl groups of
monosaccharides but also to carboxyl groups, in excellent yields.
Effective application of 5 in oligosaccharide synthesis could be
achieved by single-column chromatographic synthesis of 19 in
good yields. Each fluorous synthetic intermediate could be ob-
tained in a straightforward manner by a simple fluorous–organic
solvent partitioning and easily analyzed via its NMR spectrum ow-
ing to its symmetrical structure. Furthermore, the fluorous alcohol
3 was easily recyclable from the fluorous layer, in excellent yields.
We are currently investigating the possibility of the total synthesis
of a wide variety of oligosaccharides and glycoconjugates by the
fluorous tag method.
5. (a) Goto, K.; Mizuno, M. Tetrahedron Lett. 2007, 48, 5605; (b) Mizuno, M.;
Kitzawa, S.; Goto, K. J. Fluorine Chem. 2008, 129, 955; (c) Kawakami, H.; Goto, K.;
Mizuno, M. Chem. Lett. 2009, 38, 906.
6. (a) Curran, D. P.; Ferritto, R.; Hua, Y. Tetrahedron Lett. 1998, 39, 4937; (b) Goto,
K.; Miura, T.; Mizuno, M.; Takaki, H.; Imai, N.; Murakami, Y.; Inazu, T. Synlett
2004, 2221; (c) Goto, K.; Miura, T.; Mizuno, M. Tetrahedron Lett. 2005, 46, 8293.
7. Pentaerythritol (1) is commercially available (Tokyo Kasei Kogyo Co., Ltd).
8. Tanaka, K.; Tanaka, T.; Hasegawa, T.; Shionoya, M. Chem. Lett. 2008, 37, 440.
9. Campo, F. D.; Lastécouères, D.; Vincent, J.-M.; Verlhac, J.-B. J. Org. Chem. 1999,
64, 4969.
10. Compound 5: 1H NMR (600 MHz, CDCl3) d = 1.79–1.87 (m, 6H), 2.06–2.18 (m,
6H), 3.40 (s, 6H), 3.44 (t, J = 6.2 Hz, 6H), 3.44 (s, 2H), 7.20–7.33 (m, 4H); 13C
NMR (150 MHz, CDCl3) d = 20.65 (br s, –CH2CH2CH2–C8F17), 27.84 (t,
2JCF = 21.7 Hz, –CH2CH2CH2–C8F17), 33.25, 45.46, 69.37, 69.52, 69.59, 69.66,
72.95, 106.13–120.65 (complex signals of –CF2– and –CF3), 127.31, 127.89,
128.10, 128.73, 137.91, 139.52; (MALDI-TOF MS): calcd for C46H34BrF51O4Na
m/z [M+Na]+: 1722.6, found: 1723.5.
11. Wang, S. S.; Grisin, B. F.; Winter, D. P.; Makofske, R.; Kulesha, D.; Tzougraki, C.;
Meienhofer, J. J. Org. Chem. 1977, 42, 1286.
12. HFE7100 is a commercially available fluorocarbon solvent (3 M, Tokyo), which
consists of perfluorobutylmethyl ether (C4F9OMe) isomers. It is called Novec™
HFE7100 and is miscible in common organic solvents and fluorous solvents.
13. Kawakubo, M.; Ito, Y.; Okimura, Y.; Kobayashi, M.; Sakura, K.; Kasama, S.;
Fukuda, M. N.; Fukuda, M.; Katsuyama, T.; Nakayama, J. Science 2004, 305,
1003.
14. Rele, S. M.; Iyer, S. S.; Baskaran, S.; Chaikof, E. L. J. Org. Chem. 2004, 69, . 9159.
15. Schmidt method: Schmidt, R. R.; Michel, J.; Roos, M. Liebigs Ann. Chem. 1984,
1343.
16. Compound 17: 1H NMR (600 MHz, CDCl3) d = 1.77–1.85 (m, 6H), 2.01 (s, 3H),
2.04 (s, 3H), 2.05–2.16 (m, 9H), 3.38 (s, 6H), 3.41 (t, J = 6.7 Hz, 6H), 3.42 (s, 2H),
3.44 (dd, J = 3.4, 10.3, Hz, 1H), 3.58–3.63 (m, 2H), 3.75 (s, 3H), 3.81 (t, J = 6.7 Hz,
1H), 3.94 (dd, J = 2.7, 12.4 Hz, 1H), 4.00 (dd, J = 7.6, 9.6 Hz, 1H), 4.09 (d,
J = 2.7 Hz, 1H), 4.40(s, 2H), 4.43 (d, J = 10.3 Hz, 1H), 4.64–4.73 (m, 3H), 4.81 (d,
J = 11.6 Hz, 1H), 4.84 (d, J = 7.6 Hz, 1H), 4.92 (d, J = 11.0 Hz, 1H), 5.03–5.10 (m,
3H), 5.51 (t, J = 9.6 Hz, 1H), 6.74 (d, J = 8.9 Hz, 2H), 7.02 (d, J = 8.9 Hz, 2H), 7.19
(s, 1H), 7.25–7.34 (m, 6H), 7.39 (d, J = 6.9 Hz, 2H), 7.47 (t, J = 7.6 Hz, 2H), 7.61 (t,
J = 7.6 Hz, 1H), 8.03 (d, J = 7.6 Hz, 2H); 13C NMR (150 MHz, CDCl3) d = 20.67,
2
20.78, 27.83 (t, JCF = 21.6 Hz, –CH2CH2CH2–C8F17), 45.46, 55.63, 61.07, 61.92,
62.58, 67.94, 68.23, 69.49, 69.56, 69.66, 71.42, 72.39, 73.17, 73.64, 75.15, 75.78,
78.52, 79.65, 98.95, 103.22, 106.02–120.68 (complex signals of –CF2– and –
CF3), 114.51, 118.55, 128.59, 126.83, 126.89, 127.81, 128.38, 128.45, 128.55,
128.59, 129.66, 129.79, 133.44, 137.81, 138.29, 139.21, 151.46, 155.40, 166.10,
169.83, 169.97, 170.47; (MALDI-TOF MS): calcd for
C85H76F51O19Na m/z
[M+Na]+: 2435.4, found: 2434.6.
17. Product mixtures containing the fluorous compounds 12–15, 17, and 18 were
partitioned between fluorous mixed solvent (HFE7100: FC7221 = 4:1) and 95%
aq MeCN. None of the fluorous compounds was detected by TLC of the organic
layer after three extractions with fluorous mixed solvent; this shows that these
compounds were quantitatively extracted with the fluorous layer.
18. Compound 19 was partitioned between FC-72 and MeOH and extracted with
the MeOH layer.
References and notes
19. Compound 19: 1H NMR (600 MHz, CD3OD) d = 2.01 (s, 3H), 3.45 (t, J = 10.3 Hz,
1H), 3.67–3.62 (m, 2H), 3.84–3.70 (m, 8H), 3.82 (dd, J = 2.7, 11.7 Hz, 1H), 3.95
(dd, J = 4.1, 10.3 Hz, 1H), 4.03 (d, J = 2.7 Hz, 1H), 4.28 (ddd, J = 2.1, 2.7, 10.3 Hz,
1H), 4.82 (d, J = 7.6 Hz, 1H), 4.93 (d, J = 4.1 Hz, 1H), 6.85 (d, J = 8.9 Hz, 2H), 7.07
(d, J = 8.9 Hz, 2H); 13C NMR (150 MHz, CD3OD) d = 22.67, 55.67, 56.07, 60.76,
62.33, 72.04, 72.45, 72.65, 73.69, 74.35, 76.86, 77.71, 100.32, 104.03, 115.55,
119.08, 153.00, 156.78, 173.74; HRMS(ESI-TOF MS): calcd for C21H32NO12 m/z
[M+H]+: 490.1919, found: 490.1891. The correlation between anomeric
protons of N-acetyl glucosamine residue and 4-position carbons of galactose
residue was observed by HMQC and HMBC of NMR spectroscopic analysis.
20. Compound 3: 1H NMR (600 MHz, CDCl3) d = 1.83–1.90 (m, 6H), 2.08–2.20 (m,
6H), 2.58 (t, J = 6.2 Hz, 1H), 3.44 (s, 6H), 3.47 (t, J = 6.2 Hz, 6H), 3.68 (d,
J = 5.5 Hz, 2H); 13C NMR (150 MHz, CDCl3) d = 20.70 (br s, –CH2CH2CH2–C8F17),
27.85 (t, 2JCF = 21.6 Hz, –CH2CH2CH2–C8F17), 45.09, 65.68, 69.99, 71.01, 105.84–
120.79 (complex signals of –CF2– and –CF3); (MALDI-TOF MS): calcd for
1. Horváth, I. T.; Rábai, J. Science 1994, 266, 72.
2. Studer, A.; Hadida, S.; Ferritto, R.; Kim, S.-Y.; Jeger, P.; Wipf, P.; Curran, D. P.
Science 1997, 275, 823.
3. (a)Handbook of Fluorous Chemistry; Gladysz, J. A., Curran, D. P., Horváth, I. T.,
Eds.; Wiley-VHC: Weinheim, 2004; (b) Zhang, W. Chem. Rev. 2009, 109, 749; (c)
Sato, S.; Iida, J.; Suzuki, K.; Kawano, M.; Ozeki, T.; Fujita, M. Science 2006, 313,
1273; (d) Bakkari, M. E.; Vincent, J.-M. Org. Lett. 2004, 6, 2765; (e) Bejot, R.;
Fowler, T.; Carroll, L.; Boldon, S.; Moore, J. E.; Declerck, J.; Gouverneur, V.
Angew. Chem., Int. Ed. 2009, 48, 586; (f) Kojima, M.; Nakamura, Y.; Takeuchi, S.;
Ito, S. Tetrahedron Lett. 2009, 50, 6143; (g) Homma, T.; Harano, K.; Isobe, H.;
Nakamura, E. Angew. Chem., Int. Ed. 2010, 49, 1665.
4. (a) Miura, T.; Hirose, Y.; Ohmae, M.; Inazu, T. Org. Lett. 2001, 3, 3947; (b) Miura,
T.; Goto, K.; Hosaka, D.; Inazu, T. Angew. Chem., Int. Ed. 2003, 42, 2047; (c)
Mizuno, M.; Goto, K.; Miura, T.; Hosaka, D.; Inazu, T. Chem. Commun. 2003, 972;
(d) Goto, K.; Miura, T.; Hosaka, D.; Matsumoto, H.; Mizuno, M.; Ishida, H.-k.;
Inazu, T. Tetrahedron 2004, 60, 8845; (e) Mizuno, M.; Goto, K.; Miura, T.;
Matsuura, T.; Inazu, T. Tetrahedron Lett. 2004, 45, 3425; (f) Miura, T.; Goto, K.;
C
38H27F51O4Na m/z [M+Na]+: 1539.5, found: 1539.4.
21. FC72 is commercially available fluorocarbon solvent, which consists of
perfluorohexane (C6F14) isomers and is called Fluorinert™ FC-72.
a