Expeditious access to (-)-shikimic acid derivatives for Tamiflu synthesis

Article abstract of DOI:10.1016/j.tetlet.2011.09.035

A three-step, one-pot process has been developed for the synthesis of 5-epi-shikimic acid derivative 5 using inexpensive and abundant D-ribose as the starting material. Based on this pivotal intermediate, the syntheses of two key shikimic acid derivatives

Full text of DOI:10.1016/j.tetlet.2011.09.035

Tetrahedron Letters 52 (2011) 6352–6354
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Expeditious access to (ꢀ)-shikimic acid derivatives for Tamiflu synthesis
Hiroshi Osato a, Ian L. Jones b, Huini Goh c, Christina L.L. Chai b, Anqi Chen b,
⇑
a Shionogi & Co., Ltd, 1-3, Kuise Terajima 2-chome, Amagasaki, Hyogo 660-0813, Japan
b Institute of Chemical and Engineering Sciences (ICES), Agency for Science, Technology and Research (A⁄STAR), 8 Biomedical Grove, Neuros #07-01, Singapore 138665, Singapore
c Nanyang Technological University, School of Physical & Mathematical Sciences, Nanyang Technological University, SPMS-CBC-02-01, 21 Nanyang Link, Singapore 637371, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 July 2011
Revised 26 August 2011
Accepted 7 September 2011
Available online 10 September 2011
A three-step, one-pot process has been developed for the synthesis of 5-epi-shikimic acid derivative 5
using inexpensive and abundant D-ribose as the starting material. Based on this pivotal intermediate,
the syntheses of two key shikimic acid derivatives 6 and 7 for Tamiflu have been achieved in a concise
and practical fashion.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
D-Ribose
Tamiflu
Shikimic acid
Ring-closing olefin metathesis
Bernet–Vasella reaction
Tamiflu (Oseltamivir phosphate, 1, Fig. 1) is one of the most
widely prescribed drugs for the treatment and prevention of vari-
ous types of influenza including avian flu H1N1 and H5N1 which
caused worldwide pandemics in recent years. This anti-influenza
drug was initially discovered by Gilead Sciences and subsequently
licensed to and marketed by Hoffmann-La Roche Ltd. The commer-
cial manufacturing process of Tamiflu uses (ꢀ)-shikimic acid (2), a
natural product isolated from the star anise fruit, as the raw mate-
rial.1 Due to seasonal and geographical constraints, the supply of
shikimic acid was of great concern when Tamiflu was in high de-
mand during the avian flu epidemic period.2 Consequently, syn-
thetic chemists worldwide have devised many ingenious
alternative syntheses3 to this seemingly simple yet challenging
molecule.
this early stage intermediate. This would be advantageous in the
event of urgent need for the drug as it would not require any signif-
icant changes to most of the current manufacturing process. Amide
7, on the other hand, has been obtained from 4-amino-shikimic acid
(3) produced by fermentation of glucose using genetically engi-
neered amino shikimate dehydrogenase.5 This compound has been
used in a patented route (7 steps only) for Tamiflu synthesis.6 Herein
we describe two concise routes to shikimic acid derivatives 6 and 7
using compound 5 as a common intermediate.
The common intermediate 5 was prepared on a multigram scale
based on our previously reported route3b,16 with significant modifi-
With an aim to provide a solution to the raw material supply is-
sue, we recently reported a synthetic route to 1 using inexpensive
CO2H
CO2Et
H3PO4
HO
HO
O
and abundant D
-ribose as the starting material.3b This new route
CO2Et
O
led to an efficient synthesis of Tamiflu via a late-stage aziridine
intermediate 4 derived from 5-epi-shikimic acid derivative 5. To fur-
ther capitalise on the easily accessible intermediate 5 in alternative
routes to 1, our attention was attracted to two shikimic acid deriva-
tives, ketal 64 and amide 7,5 which are the key intermediates for
Tamiflu syntheses. Ketal 6 has been obtained from (ꢀ)-shikimic acid
or (ꢀ)-quinic acid and is an early stage intermediate used in the
Roche process for Tamiflu manufacturing.4 Hence, if 5 could be con-
veniently converted into 6, it would provide an alternative source for
AcHN
R
NH
HN
2 (R = OH)
(R = NH2)
Oseltamivir phosphate
4
3
1
(Tamiflu,
)
CO2Et
CO2Et
HO
CO2Et
O
O
O
HO
O
NHAc
OH
OH
6
7
5
⇑
Corresponding author. Tel.: +65 67998530; fax: +65 64642102.
E-mail address: chen_anqi@ices.a-star.edu.sg (A. Chen).
Figure 1. Structures of Tamiflu (1) and compounds 2–7.
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.09.035

H. Osato et al. / Tetrahedron Letters 52 (2011) 6352–6354
6353
HO
CO2Et
Conditions
O
O
O
CO2Et
OMe
O
O
b
a
76%
R
CO2Et
elimination
products
5
+
96%
O
O
OR
(R=H)
O
O
OH
5
9
6
11 (R=Ms)
12 (R=Tf)
8a (R = OH)
8b
(R = I)
N
N
Mes
Mes
CHO
Cl
Cl
Ru
O
Table 1
Investigation of the conditions for inversion of the 5-hydroxy group
O
O
8c
Entry Substrate Conditions Result
Ph3P, DIAD, PhCO2H, 0 °C to rt Elimination9
10
1
2
3
4
5
6
5
Scheme 1. Improved synthesis of 5: (a) (i) I2, imid., PPh3, THF, reflux, 20 min; (ii) Zn
dust, H2O, reflux, 30 min; (iii) ethyl 2-(bromomethyl)acrylate, reflux, 40 min; (b)
1,2-dichloroethane, 10 (2 mol %), reflux, 4 h.
11
11
11
12
12
CaCO3, H2O/THF (1:1), reflux
CsOAc, DMF, rt
CsOAc, DMF, 80 °C
CsOAc, DMF, rt
NaNO2, DMF, 0 °C–rt
Recovered 5
Recovered 5
Elimination9
Elimination9
6 (75%) + elimination9
cation (Scheme 1). The conversion of the known ketal 8a into diene 9
was previously carried out in two separate steps via the formation of
iodide 8b, followed by a one-pot Bernet–Vasella reaction. In this pro-
cess, the intermediate iodide 8b had to be isolated. After optimisa-
tion, it was found that this series of three-step transformations
could be carried out in a more efficient one-pot operation. Thus,
alcohol 8a was converted into the corresponding iodide 8b (Ph3P/
I2/imidazole) in refluxing THF. The resultant iodide then underwent
the Bernet–Vasella reaction7 to form the intermediate aldehyde 8c
in the presence of zinc dust. Aldehyde 8c was further reacted with
ethyl 2-(bromomethyl)acrylate in the presence of excess zinc dust
(present from the previous step), providing the desired diene alcohol
9 in 76% yield after chromatographic separation of the minor diaste-
reomer. This improved procedure, without the need to isolate the
intermediates, greatly simplifies the procedure for scalable opera-
tion. The diene 9b was then cyclised, as described previously,3b via
ring-closing olefin metathesis using the Grubbs-Hoveyda second
generation catalyst (10) to provide 5-epi-shikimic acid derivative 5
in multigram quantity in 96% yield.
With the establishment of a scalable route to 5, we set out to
investigate the inversion of 5-hydroxy group. Since conformational
analysis8 revealed that 5-hydroxy group adopts an axial orienta-
tion, (Fig. 2) we envisioned that its inversion would be favourable
although b-elimination reactions might occur.
The first attempted inversion via the Mitsunobu reaction re-
sulted in the formation of elimination products9 instead of the re-
quired inversion product (Table 1, entry 1). An alternative
displacement via mesylate 11 was then investigated. Whilst reac-
tions of 11 with either CaCO3/H2O–THF10 at reflux (entry 2) or
CsOAc11 in DMF at room temperature (entry 3) returned the start-
ing material, an increase in the reaction temperature to 80 °C in the
latter case led to the formation of elimination products. However,
it was reasoned that the use of triflate as a better leaving group
might favour the displacement. Nevertheless, treatment of triflate
12 with CsOAc also resulted in the formation of elimination prod-
ucts. At this point our attention was turned to a less known meth-
od for the inversion of triflate with sodium nitrite.12 Gratifyingly,
treatment of 12 with sodium nitrite at room temperature in DMF
led to successful inversion, providing alcohol 613 in 75% yield with
<10% of elimination products being formed. This direct inversion is
more convenient than the Mitsunobu conditions or using CsOAc as
it negates the ester hydrolysis step.
Having achieved the synthesis of 6, we turned our attention to
the transformation of 5 into 7 (Scheme 2). Nucleophilic displace-
ment of triflate 12 with sodium azide proceeded smoothly to pro-
vide 13. It is worth noting that control of the temperature was
crucial for the reaction. While the reaction proceeded sluggishly
at <ꢀ20 °C, significant amounts of elimination products were
formed when it was raised to ambient temperature (ꢁ25 °C). The
optimal temperature range for this reaction was found to be be-
tween ꢀ10 and 0 °C at which 13 was obtained in 82% yield. The
azide 13 was subsequently reduced to the corresponding amine
at room temperature under atmospheric pressure of hydrogen
using Lindlar catalyst. These conditions left the less reactive double
bond intact.14 Without isolation, the crude amine was acetylated to
provide amide 1415 in 89% yield over two steps. Final removal of
the ketal-protecting group (1.0 M aq HCl in methanol) furnished
the desired amide 7 in 62% overall yield from 5.
In summary, we have developed a scalable, three-step, one-pot
process for the synthesis of 5-epi-shikimic acid derivative 5 using
inexpensive and abundant D-ribose as the starting material. Using
5 as a common intermediate, the syntheses of two key shikimic
CO2Et
O
O
CO2Et
O
O
a
b
quant.
82%
OH
OTf
12
5
CO2Et
O
O
CO2Et
CO2Et
O
O
HO
HO
c
d
86%
89%
N3
13
NHAc
14
NHAc
7
Scheme 2. Synthesis of 7: (a) Tf2O, Py; (b) NaN3, DMF, ꢀ10 to 0 °C; (c) (i) H2
Figure 2. Energy minimised conformation of 5.
(1 atm), Lindlar catalyst (5 mol % Pd); (ii) Ac2O, Et3N; (d) 1.0 M aq HCl, MeOH, rt.

6354
H. Osato et al. / Tetrahedron Letters 52 (2011) 6352–6354
Hoffmann, U.; Huber, G.; Janatsch, G.; Lauper, S.; Rockel-Stabler, O.;
Trussardi, R.; Zwahlen, A. G. Org. Process Res. Dev. 1999, 3, 266.
5. Guo, J.; Frost, J. W. Org. Lett. 2004, 6, 1585.
acid derivatives 6 and 7 for Tamiflu syntheses have been achieved,
providing alternative routes for the syntheses of this antiviral drug.
6. Guo, J.; Frost, J. W. US 2007/0190621A1, 16/08/2007.; Chem. Abstr. 2007, 147,
275841.
Acknowledgements
7. (a) Bernet, B.; Vasella, A. Helv. Chim. Acta 1979, 62, 1990; (b) Nakane, M.;
Hutchinson, C. R.; Gollman, H. Tetrahedron Lett. 1980, 21, 1213; (c) Fürstner, A.;
Jumbam, D.; Teslic, J.; Weidmann, H. J. Org. Chem. 1991, 56, 2213.
8. The energy minimized conformation was obtained using the MMFF94 force
field, SPARTAN’04 version for Windows, Wavefunction, Inc.
We thank Dr. Paul H. Bernardo (Institute of Chemical and Engi-
neering Sciences, Agency for Science, Technology and Research) for
helping with the molecular modelling, Shionogi & Co., Ltd (H. Osa-
to) and Agency for Science, Technology and Research (A⁄STAR), Sin-
gapore for supporting this work.
9. An inseparable mixture of two products
A
and
B
identified by 1H NMR
spectroscopy and GC–MS analysis was obtained.
CO2Et
CO2Et
O
O
O
O
Supplementary data
A
B
10. Carnell, A. J.; Head, R.; Bassett, D.; Schneider, M. Tetrahedron: Asymmetry 2004,
15, 821.
Supplementary data (complete experimental procedures, char-
acterization and spectral data of all key compounds) associated with
this article can be found, in the online version, at doi:10.1016/j.
tetlet.2011.09.035.
11. Dijkstra, G.; Kruizinga, W. H.; Kellogg, R. M. J. Org. Chem. 1987, 52, 4230.
12. (a) Albert, R.; Dax, K.; Link, R. W.; Stütz, A. E. Carbohydr. Res. 1983, 118, 118; (b)
Guo, H.; O’Doherty, G. A. J. Org. Chem. 2008, 73, 5211; (c) Balskus, E. P.;
Jacobsen, E. N. J. Am. Chem. Soc. 2006, 128, 6810.
13. Shikimic acid derivative 6: colorless oil; Rf = 0.43 (EtOAc/PE = 2:3); ½a D24
ꢀ74.92
ꢂ
(c 1.08, CHCl3); mmax (EF)/cmꢀ1 3460, 2975, 2938, 1713, 1653, 1462, 1364,
1248, 1172, 1067, 1037, 934, 873, 751, 726, 644, 519; 1H NMR (400 MHz,
CDCl3) d 6.94–6.93 (m, 1H) 4.78–4.73 (m, 1H), 4.22 (q, J = 8.0 Hz, 2H), 4.09 (t,
J = 8.0 Hz, 1H), 3.94–3.87 (m, 1H), 2.81–2.76 (m, 1H), 2.28–2.21 (m, 1H), 2.04
(br s, 1H), 1.70–1.64 (m, 4H), 1.30 (t, J = 8.0 Hz, 3H), 0.92 (t, J = 8.0 Hz, 3H), 0.89
(t, J = 8.0 Hz, 3H); 13C NMR (100 MHz, CDCl3) d 166.2, 134.0, 130.4, 113.6, 77.9,
72.2, 68.9, 61.1, 29.7, 29.3, 29.1, 14.2, 8.6, 7.9; HRMS (ESI-TOF): m/z found
293.1359 [M+Na]+; C14H22NaO5 requires 293.1365.
References and notes
1. The Roche Group, Tamiflu Factsheet, 17 November, 2006.
2. (a) Enserink, M. Science 2006, 312, 382; (b) Farina, V.; Brown, J. D. Angew. Chem.,
Int. Ed. 2006, 45, 7330.
3. For selected references, see: (a) Magano, J. Chem. Rev. 2009, 109, 4398; (b)
Osato, H.; Jones, I. L.; Chen, A.; Chai, C. L. L. Org. Lett. 2010, 12, 60. and references
therein; (c) Werner, L.; Machara, A.; Hudlicky, T. Adv. Synth. Catal. 2010, 352,
195; (d) Ma, J.; Zhao, Y.; Ng, S.; Zhang, J.; Zeng, J.; Than, A.; Chen, P.; Liu, X.-W.
Chem. Eur. J. 2010, 16, 4533; (e) Ko, J. S.; Keum, J. E.; Ko, S. Y. J. Org. Chem. 2010,
75, 7006; (f) Weng, J.; Li, Y.-B.; Wang, R.-B.; Li, F.-Q.; Liu, C.; Chan, A. S. C.; Lu, G.
J. Org. Chem. 2010, 75, 3125; (g) Kamimura, A.; Nakano, T. J. Org. Chem. 2010, 75,
3133; (h) Zhu, S.; Yu, S.; Wang, Y.; Ma, D. Angew. Chem., Int. Ed. 2010, 49, 4656;
(i) Pawinee, W.; Sunisa, A.; Ngampong, K.; Boonsong, K. Tetrahedron Lett. 2010,
51, 3208; (j) Sadagopan, R.; Babu, S. V. Tetrahedron 2011, 67, 2044; (k) Trost, B.
M.; Zhang, T. Chem. Eur. J. 2011, 17, 3630.
14. Nie, L.-D.; Shi, X.-X. Tetrahedron: Asymmetry 2009, 20, 124.
15. Amide 7: amorphous solid; Rf = 57 (EtOAc/PE = 9:1);
½ ꢂ ꢀ103.3 (c 1.2,
a 2D4
MeOH);
m
max (EF)/cmꢀ1 3289, 2976, 2941, 1716, 1654, 1550, 1371, 1172, 1073,
1059, 927, 732; 1H NMR (400 MHz, CDCl3) d 6.87 (m, 1H), 4.31 (t, J = 4.2 Hz,
1H), 4.25–4.18 (m, 3H), 3.74 (dd, J = 9.0, 4.1 Hz, 1H), 2.82 (dd, J = 18.1, 5.2 Hz,
1H), 2.23–2.08 (m, 1H), 2.00 (s, 3H), 1.32 (t, J = 7.1 Hz, 3H); 13C NMR (100 MHz,
CDCl3) 174.5, 168.7, 139.2, 132.2, 72.2, 67.9, 62.8, 49.0, 31.1, 23.5, 15.3; HRMS
(ESI-TOF): m/z found 244.1812 [M+H]+; C11H18NO5 requires 244.1815.
16. A similar route was reported independently: Kancharla, P. K.; Doddi, V. R.;
Kokatla, H.; Vankar, Y. D. Tetrahedron Lett. 2009, 50, 6951.
4. Federspiel, M.; Fisher, R.; Henning, M.; Mair, H. J.; Oberhauser, T.; Rimmler, G.;
Albiez, T.; Bruhin, J.; Estermann, H.; Gandert, C.; Gockel, V.; Gotzo, S.;