Design and synthesis of immobilized Tamiflu analog on resin for affinity chromatography

Article abstract of DOI:10.1016/j.tetlet.2009.01.142

A resin linked with the Tamiflu core was synthesized by modifying our original synthetic route of oseltamivir phosphate (Tamiflu). The prepared resin bound to the influenza virus enzyme neuraminidase, the main target of Tamiflu. The immobilized Tamiflu an

Full text of DOI:10.1016/j.tetlet.2009.01.142

Tetrahedron Letters 50 (2009) 3205–3208
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Design and synthesis of immobilized Tamiflu analog on resin
for affinity chromatography
b
Yasuaki Kimura a, Kenzo Yamatsugu a, Motomu Kanai a, , Noriko Echigo ,
*
Takashi Kuzuhara b, Masakatsu Shibasaki a,
*
a Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
b Faculty of Pharmaceutical Sciences, Tokushima Bunri University, Yamashiro-cho, Tokushima 770-8514, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A resin linked with the Tamiflu core was synthesized by modifying our original synthetic route of osel-
tamivir phosphate (Tamiflu). The prepared resin bound to the influenza virus enzyme neuraminidase, the
main target of Tamiflu. The immobilized Tamiflu analog will be useful for isolating and identifying pre-
sumed endogenous vertebrate proteins that interact with Tamiflu, which might relate to the abnormal
behavior exhibited by some influenza patients treated with Tamiflu.
Received 11 January 2009
Revised 27 January 2009
Accepted 29 January 2009
Available online 1 February 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Tamiflu
Immobilization
Affinity chromatography
Abnormal behavior
Neuraminidase
Oseltamivir phosphate (1: TamifluÒ, Fig. 1)1 exhibits potent
antiviral activity against influenza viruses type A and B by selec-
tively inhibiting neuraminidase (NA), an essential enzyme for the
release of virions from infected cells to expand infection. Tamiflu
is a prodrug, and its active form is the corresponding carboxylic
acid Ro 64-0802 (2).
brate proteins is similar to that for binding virus NA. Analysis of
the X-ray crystal structures of the NA–2 complex1a,3 and the re-
ported structure–activity relationship1a led to the identification
of the following two major bases for the molecular design: (1)
modifications of the C3 pentyloxy part do not completely disrupt
the binding to NA, although modifications of the C1 carboxyl
group, the C4 acetamide moiety, and the C5 amino functionality
do disrupt binding to NA. Therefore, the C3 ether part was selected
to connect the oseltamivir core to the resin. (2) The binding site of
2 is positioned close to the surface of NA,3a suggesting that a rela-
tively short spacer (five carbons) between the oseltamivir core and
the resin would be sufficient for binding to NA. Based on these con-
siderations, we designed 3 immobilized to Affi-Gel 10Ò resin
(Fig. 2).
Tamiflu is a very effective drug for treating influenza patients at
the early stage of infection. In quite rare cases, abnormal behaviors
(such as hallucinations and impulsive behavior) have been re-
ported in Japanese patients, especially those under age of 20, after
taking Tamiflu. Several research groups, including ours, are inves-
tigating whether there is any molecular-level correlation between
Tamiflu medication and the abnormal behaviors.2 Based on our
previous studies,2b–d we hypothesized that an endogenous human
protein is specifically affected by Tamiflu (possibly, in the central
nervous system). Affinity chromatography is a direct method for
detecting the existence of biomolecules that interact with a partic-
ular organic molecule of interest. Here, we report the design, syn-
thesis, and functional assessment of an immobilized Tamiflu
analog on resin, which might be useful for affinity chromatography
to identify possible endogenous vertebrate biomolecular targets of
Tamiflu.
The synthesis of 3 should be straightforward by extending the
previously established synthetic route of Tamiflu.1n Briefly, this
O
O
NHAc
NH2
NHAc
3
1
5
Our design of the immobilized Tamiflu analog is based on the
assumption that the binding mode of 2 to potential target verte-
HO2C
EtO2C
NH2• H3PO4
Tamiflu® (1)
Ro 64-0802 (2)
* Corresponding authors. Fax: +81 3 5684 5206 (M.S.).
E-mail address: mshibasa@mol.f.u-tokyo.ac.jp (M. Shibasaki).
Figure 1. Structure of TamifluÒ and Ro 64-0802.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.01.142

3206
Y. Kimura et al. / Tetrahedron Letters 50 (2009) 3205–3208
synthesis. An Fmoc group was selected as P2 because it can be
affinity-chromatographic resin
(Affi-Gel 10®)
spacer moiety
cleaved under mild conditions without affecting the agarose skel-
eton of the resin. Moreover, the amount of immobilized oseltamivir
core can be quantified based on the amount of released fluorenyl
derivative (22). P1 should be removed in the presence of the Fmoc
group, and therefore we selected an allyl group for P1 (12b).
Our synthesis of 3 started from known epoxide 11, which was
synthesized following the previously developed procedures shown
in Scheme 1.1n Treatment of 11 with potassium carbonate in allyl
alcohol yielded allyl ester 14 via alcoholysis of the acetoxydicy-
anomethyl moiety and the subsequent eliminative epoxide open-
O
O
H
O
N
N
H
N
H
5 O
O
NHAc
HO2C
NH2
3
ing (Scheme 2). Inversion of
a-alcohol 14 to b-isomer 15 was
Tamiflu core
realized via SN2 attack of p-nitrobenzoic acid under Mitsunobu
conditions, followed by selective cleavage of the p-nitrobenzoate.
DBU was used as a base in the alcoholysis step because previously
utilized LiOHÁH2O1n was not effective due to its low solubility in al-
lyl alcohol. Mitsunobu aziridine formation using dimethylphenyl
phosphine and DIAD in the presence of a catalytic amount of tri-
ethylamine1n afforded 12b in high yield. Next, the aziridine open-
ing reaction with 5-azido-1-pentanol required some optimization.
The use of a large excess of azido alcohol4,5 was important in this
reaction: ether 16 was obtained in 78% yield in the presence of
1.5 equiv of BF3ÁOEt2 and 30 equiv of 5-azido-1-pentanol. The ex-
cess azido alcohol was recovered for reuse in more than 70% yield
after chromatographic separation.
With ether 16 in hand, the next key intermediate was amino
acid 20a or 20b. Conversion of the protecting group at the C5 ami-
no group from Boc to Fmoc, reduction of the azide group, and
cleavage of C1 allyl ester were required, but the order of the three
transformations had to be optimized (Scheme 3). After converting
the protecting group from Boc to Fmoc, which produced 17, reduc-
tion of the azide group was attempted. Staudinger reduction using
various phosphines did not afford target compound 18. On the
other hand, reduction proceeded in moderate yield in AcOH sol-
vent using zinc powder pre-activated with 1 M HCl. Subsequent
optimization regarding the method of zinc activation, the proton
source, and the solvent led us to identify the optimum conditions:
zinc powder pre-activated with dibromoethane and TFA/EtOH sol-
vent. Amine 18 was isolated in 73% yield under these optimized
conditions. Subsequent allyl ester cleavage using Pd catalysts,
however, was not successful.6
Figure 2. The structure of the designed immobilized Tamiflu analog.
Catalytic-Asymmetric
Diels-Alder-Type Reaction
Ba-6 (2.5 mol %)
OH
OTMS
CsF (2.5 mol %);
HCl aq.
CO2Me
CO2Me
CO2Me
+
Ph
Ph
MeO2C
P
O
4
5
7
HO
O
F
F
y. 91%, 95% ee
6
endo : exo = 5 : 1
HO
O
Curtius
Rearrangement
Allylic
O
Substitution
NHAc
NH
AcO
NC
NHBoc
NHBoc
9
CN
CN
CN
10
8
AcO
O
NAc
NHAc
AcO
NC
P1O2C
NHP2
NHBoc
12
CN
11
12a: P1 = Et, P2 = Boc
12b: P1 = Allyl, P2 = Fmoc
3-pentanol
Tamiflu (1)
K2CO3
OH
AllylOH
0 ºC
O
ref 1n
NHAc
NHAc
H2N 5 O
immobilization;
AcO
NC
NHAc
NHP2
87%
NHBoc
AllylO2C
NHBoc
3
this work
deprotection
CN
HO2C
11
14
13
Scheme 1. Brief summary of the previously developed synthetic route of Tamiflu
(1) and synthetic plan for the immobilized Tamiflu analog 3.
p−nitrobenzoic acid
DEAD, PPh3;
DIAD
Me2PPh
NEt3
OH
DBU, AllylOH
NHAc
NHBoc
route involves an asymmetric Diels–Alder-type reaction between
siloxy diene 4 and dimethyl fumarate (5) catalyzed by a barium–
FujiCAPO (6, CAPO = CAtechol Phosphine Oxide) complex, a Curtius
rearrangement, and Pd-catalyzed allylic substitution with MAC re-
agent 9 (MAC = Masked Acyl Cyanide) as key steps (Scheme 1).
Introduction of 5-aminopentan-1-ol (or its equivalent) through a
ring-opening reaction of an aziridine (12) would produce 13 con-
taining an oseltamivir core with a spacer, which should be immo-
bilized to Affi-Gel 10 via simple amide formation. The proper
selection of the protecting groups at the C1 carboxylic acid moiety
(P1) and the C5 amino group (P2) is crucial for the success of the
90%
80%
AllylO2C
15
BF3•OEt2
5-azido pentanol
N3 5 O
NAc
NHAc
AllylO2C
NHBoc
78%
AllylO2C
NHBoc
12b
16
Scheme 2. Synthesis of ether 16.

Y. Kimura et al. / Tetrahedron Letters 50 (2009) 3205–3208
3207
1) TFA/CH2Cl2
N3 5 O
2) FmocCl, NaHCO3
NHAc
16
5
80%
2 steps
AllylO2C
NHFmoc
17
N-methylaniline
Pd(PPh3)4 (10 mol%)
92%
Zn
TFA/EtOH
73%
H2N 5 O
N3 5 O
NHAc
NHAc
Figure 3. Binding of neuraminidase (rvH1N1NA) to resin-conjugated Tamiflu
analog 3. Binding assay of rvH1N1NA was performed for 3. The eluted protein
was analyzed by SDS–PAGE and was visualized by silver staining. A molecular
weight of rvH1N1NA is 48 kDa, and it migrates as a 57 kDa band in SDS–PAGE. The
position of rvH1N1NA was marked by the arrowhead. rvH1N1NA (lanes 3 and 5) or
buffer (lanes 2 and 4) was applied for control resin (lanes 2 and 3) or 3 (lanes 4 and
AllylO2C
NHFmoc
HO2C
NHFmoc
19
18
Zn
5). M, molecular weight marker; lane 1, rvH1N1NA standard (0.5 lg).
TFA/EtOH
95% (NMR y.)
Complete removal of the Fmoc group was confirmed by quantita-
tive NMR analysis of 22, a byproduct in the Fmoc cleavage reaction,
which was recovered in the solution phase.
H2N 5 O
H N 5 O
CF3CO2•
3
NHAc
NHAc
We then assessed the binding affinity of synthesized conjugate
3 to neuraminidase. The criterion in this preliminary binding study
was whether NA binds to the resin. A control resin was prepared by
reacting piperidine, instead of 20b, with Affi-Gel 10Ò. Recombinant
HO2C
NHFmoc
HO2C
NHFmoc
20a
20b
Scheme 3. Preparation of the Tamiflu core.
Influenza A virus H1N1 neuraminidase (rvH1N1NA, 0.5
chased from R&D Systems, Inc.) and 50 L of each 40% resin (3 or
control resin) were mixed and incubated in 250 L of Tamiflu bind-
lg, pur-
l
Therefore, allyl ester cleavage of 17 was next investigated
(Scheme 3). The conversion proceeded in high yield (92%) in the
presence of 10 mol % of Pd(PPh3)4 and N-methylaniline. Reduction
of the azide in 19 under the above-mentioned optimized condi-
tions afforded 20b in excellent yield (95%). Because 20b is water
soluble and highly polar, 20b was only partially purified by filtra-
tion through Celite to eliminate the excess zinc and resulting zinc
salts.
The final step of the synthesis was linking 20b to the chromato-
graphic resin, Affi-Gel 10Ò (21; Scheme 4). The coupling reaction
was performed using 20b (20b: the activated ester residues of
Affi-Gel 10 = 1:6) under slightly basic conditions (pH 8) in the pres-
ence of ca. 50 equiv Et3N in MeOH at room temperature for 15 h.
After the coupling reaction, the resin was separated by filtration
and washed with MeOH.7 Finally, removal of the Fmoc group and
blocking of the remaining activated ester on the resin were con-
ducted simultaneously using excess piperidine (in a volume equal
to that of the undried resin) in DMF at room temperature for 24 h.8
l
ing buffer (50 mM Tris–HCl, 100 mM NaCl, 5 mM CaCl2; pH 7.5) at
4 °C for 2 h. After washing the resin with 1 mL of Tamiflu-binding
buffer five times, the resin was resuspended in SDS-sample buffer
[3.3% SDS (sodium dodecyl sulfate), 16.7% glycerol, 133 mM
Tris–HCl, pH 6.8, 2.5% b-ME (2-mercapoethanol), 0.03% BPB (bro-
mophenol blue)] and was boiled for 2–3 min. The supernatant
was subjected to electrophoresis on SDS–10% polyacrylamide gel,
and was visualized with silver-staining. rvH1N1NA was detected
in the bound fraction at the expected position on the gel when
rvH1N1NA was mixed with resin-conjugated Tamiflu analog 3
(Fig. 3, lane 5). rvH1N1NA was not detected in other lanes on the
gel (Fig. 3, lanes 2–4). The control result revealed that Tamiflu ana-
log 3 has a binding activity to rvH1N1NA.
In conclusion, we synthesized an immobilized Tamiflu analog
on resin by modifying our original synthetic route of Tamiflu.1n
This study demonstrated the highly flexible nature of our synthetic
route. The prepared resin bound to NA, the main target of Tamiflu.
Studies are ongoing to identify Tamiflu’s possible endogenous tar-
get biomolecules in vertebrates using affinity chromatography of
the synthesized resin and/or its derivatives.
agarose
O
O
O
H
N
O
N
H
N
20b
Acknowledgments
O
O
O
Affi-Gel 10® (21)
Financial support was provided by The Uehara Memorial Foun-
dation and Grant-in-Aid for Scientific Research (S) from JSPS. Mr.
Liang Yin is deeply acknowledged for synthesizing 7.
1) 21, NEt3
MeOH
2) piperidine
DMF
N
References and notes
22
1. (a) Kim, C. U.; Lew, W.; Williams, M. A.; Liu, H.; Zhang, L.; Swaminathan, S.;
Bischofberger, N.; Chen, M. S.; Mendel, D. B.; Tai, C. Y.; Laver, W. G.; Stevens, R. C.
J. Am. Chem. Soc. 1997, 119, 681; For studies on the synthesis of Tamiflu, see: (b)
Yueng, Y.-Y.; Hong, S.; Corey, E. J. J. Am. Chem. Soc. 2006, 128, 6310; (c) Fukuta,
Y.; Mita, T.; Fukuda, N.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2006, 128,
6312; (d) Cong, X.; Yao, Z.-J. J. Org. Chem. 2006, 71, 5365; (e) Mita, T.; Fukuda, N.;
Kanai, M.; Shibasaki, M. Org. Lett. 2007, 9, 259; (f) Yamatsugu, K.; Kamijo, S.;
Suto, Y.; Kanai, M.; Shibasaki, M. Tetrahedron Lett. 2007, 48, 1403; (g) Bromfield,
K. M.; Graden, H.; Hagberg, D. P.; Olsson, T.; Kann, N. Chem. Commun. 2007,
3183; (h) Satoh, N.; Akiba, T.; Yokoshima, S.; Fukuyama, T. Angew. Chem., Int. Ed.
2007, 46, 5734; (i) Shie, J.-J.; Fang, J.-M.; Wang, S.-Y.; Tsai, K.-C.; Cheng, Y.-S. E.;
O
O
H
N
O
N
H
N
H
5 O
O
NHAc
NH2
HO2C
3
Scheme 4. Coupling of Tamiflu core 20b and Affi-Gel 10Ò
.

3208
Y. Kimura et al. / Tetrahedron Letters 50 (2009) 3205–3208
Yang, A.-S.; Su, C.-Y.; Wong, C.-H. J. Am. Chem. Soc. 2007, 129, 11892; (j) Kipassa,
relationship between Tamiflu administration and abnormal behaviors based on
animal experiments performed by the Chugai Pharmaceutical Company, which
sells Tamiflu in Japan. For details, see http://www.mhlw.go.jp/shingi/2007/12/
s1210-6.html, http://www.chugai-pharm.co.jp/html/info/071210.html.
3. (a) Rusell, R. J.; Haire, L. F.; Stevens, D. J.; Collins, P. J.; Lin, Y. P.; Blackburn, G. M.;
Hay, A. J.; Gamblin, S. J.; Skehel, J. J. Nature 2006, 443, 45; (b) Collins, P. J.; Haire,
L. F.; Lin, Y. P.; Liu, J.; Rusell, R. J.; Walker, P. A.; Skehel, J. J.; Martin, S. R.; Hay, A.
J.; Gamblin, S. J. Nature 2008, 453, 1258.
4. For the synthesis of 5-azido-1-pentanol, see: Gracias, V.; Frank, K. E.; Milligan, G.
L.; Aube, J. Tetrahedron 1997, 53, 16241.
5. The reaction with 4 equiv or 10 equiv of the alcohol gave low yields
(approximately 20% and 40%, respectively) of the ether 16.
6. The allyl ester removal from 18 using catalytic Pd(PPh3)4 afforded a complex
mixture.
7. Approximately, 40% of unreacted 20b was recovered based on NMR analysis of
the combined solution phases.
8. Ullrich, T.; Ghobrial, M.; Peters, C.; Billich, A.; Guerini, D.; Nussbaumer, P. Chem.
Med. Chem. 2007, 3, 356.
N. T.; Okamura, H.; Kina, K.; Hamasa, T.; Iwagawa, T. Org. Lett. 2008, 10, 815; (k)
Trost, B. M.; Zhang, T. Angew. Chem., Int. Ed. 2008, 47, 3759; (l) Zutter, U.; Iding,
H.; Spurr, P.; Wirz, B. J. Org. Chem. 2008, 73, 4895; (m) Shie, J.-J.; Fang, J.-M.;
Wong, C.-H. Angew. Chem., Int. Ed. 2008, 47, 5788; (n) Yamatsugu, K.; Yin, L.;
Kamijo, S.; Kimura, Y.; Kanai, M.; Shibasaki, M. Angew. Chem., Int. Ed. 2009, 48,
1070.; (o) Matveenko, M.; Willis, A. C.; Banwell, M. G. Tetrahedron Lett. 2008, 49,
7018; (p) Ishikawa, H.; Suzuki, T.; Hayashi, Y. Angew. Chem., Int. Ed. 2009, 48,
1304.; For reviews on the synthesis of Tamiflu, see: (q) Farina, N.; Brown, J. D.
Angew. Chem., Int. Ed. 2006, 45, 7300; (r) Shibasaki, M.; Kanai, M. Eur. J. Org.
Chem. 2008, 1839.
2. (a) Izumi, Y.; Tokuda, K.; O’Dell, K. A.; Zorumski, C. F.; Narahashi, T. Neurosci. Lett.
2007, 426, 54; (b) Ishii, K.; Hamamoto, H.; Sasaki, T.; Ikegaya, Y.; Yamatsugu, K.;
Kanai, M.; Shibasaki, M.; Sekimizu, K. Drug Discov. Ther. 2008, 2, 24; (c) Morita,
M.; Sone, T.; Yamatsugu, K.; Sohotome, Y.; Matsunaga, S.; Kanai, M.; Watanabe,
Y.; Shibasaki, M. Bioorg. Med. Chem. Lett. 2008, 18, 600; (d) Usami, A.; Sasaki, T.;
Satoh, N.; Akiba, S.; Yokoshima, S.; Fukuyama, T.; Yamatsugu, K.; Kanai, M.;
Shibasaki, M.; Matsuki, N.; Ikegaya, Y. J. Pharamacol. Sci. 2008, 106, 659; (e)
According to the Ministry of Health, Labor, and Welfare of Japan, there is no