A concise synthetic approach toward tamiflu (oseltamivir phosphate): cis-aziridine as the key synthon and RCM

Abstract

The key synthon cis-aziridine has been efficiently utilised for the synthesis of oseltamivir phosphate, using Wittig olefination, Barbier addition, Mitsunobu reaction and ring closing metathesis (RCM) as key essentials.

---

The constant and continuous world-wide threat of seasonal influenza to human and animal health is a cause for concern due to the mortality and mobility of the viral strains.¹ Neuraminidase inhibition is necessary for stopping the spread of the viral infection from the infected host cell. These surface glycoproteins cleave the sialic acid complex, which invade the lead of the viron in new cells. The molecules which were investigated for blocking the neuraminidase active site, help the inhibition of the viral infection and avoid the viron aggregation. Two of these molecules such as zanamivir (Relenza®) and oseltamivir phosphate (Tamiflu®) are the drugs available in the market for the treatment and prevention of the seasonal influenza caused by mutant viral strains (Fig. 1).

Fig. 1 Neuraminidase inhibitors.

The zanamivir has low bioavailability and is administrated by inhalation which could be problematic in case of patients suffering from the respiratory disease. Whereas the oseltamivir phosphate (tamiflu) is given in the form of an oral prodrug and it has high bioavailability.

Tamiflu is effective against both H5N1 and H1N1 viral strains but its dose should be administered within 36–48 h after influenza symptoms detection.² The first report for tamiflu synthesis by Gilead science³ and Roche's⁴ processes from the (−)-shikimic acid is utilized for industrial scale synthesis of tamiflu.

One of the major factors which restrict the large scale production of neuraminidase inhibitor drugs is the shortage of the synthetic precursor (−)-shikimic acid, whose natural source is limited. The 30 kg of dried star anise plant provides 1 kg (−)-shikimic acid through extraction method. Whereas 30% of its total requirement supplied by fermentation process.⁵ Recently many synthetic groups focused on the alternative synthetic strategies from the cheap and more readily available starting materials which resulted in the different synthetic routes for the synthesis of the tamiflu.⁶

Shibasaki and co-workers reported different approaches based on asymmetric aziridine opening, Diels–Alder reaction and ring closing metathesis.⁷ Whereas Corey et al. reported the synthesis of tamiflu involving a Diels–Alder reaction⁸ while Hayashi's report on three “one-pot” operational strategy for tamiflu synthesis is the highest yielding protocol so far fro the tamiflu.⁹ There are also other synthetic routes reported for the tamiflu synthesis which involve L-serine, D-xylose, D-ribose, D-mannitol, L-methionine and D-tartarate chiral starting materials as alternatives to the (−)-shikimic acid.¹⁰ However these synthetic strategies suffer from several limitations, which include the utilization of expensive starting materials, hazardous chemicals, tedious reaction conditions, azide intermediates and low yielding reaction steps. Due to this there is still a need to design and develop new synthetic routes which overcome the drawbacks of the reported strategies. The main task is to substitute the raw material of current manufacturing process. The increasing demand of the tamiflu on worldwide seasonal pandemics of influenza puts increasing pressure to fulfill this need.

Since its discovery by Gabriel in 1888, aziridines are synthetically attractive intermediates and the building blocks for the synthesis of natural products.¹¹ They have been extensively explored for the construction of stereogenic centers containing nitrogen compounds. Tamiflu contains ether, vicinal trans diamine as three contiguous chiral centers in the cyclohexene structure. Hence, the well-organized and accurate placement of these groups is essential for its synthesis. The aziridine ring can be used to access the trans diamine. We herewith report the synthetic approach towards oseltamivir phosphate from the D-mannitol as an inexpensive and abundant chiral starting material involving aziridine as key precursor.

Retrosynthetic analysis is shown in Scheme 1. Aziridine 2 was considered a key precursor, which could be obtained by intramolecular RCM of bis-olefin 3. Bis olefin 3 in turn could be obtained from the aziridine 4 by consecutive DIBALH reduction, one carbon Wittig olefination, acetonide deprotection, oxidative cleavage of diol and Barbier addition. The aziridine 4 is the key synthon which could be easily prepared from D-mannitol.¹²

Scheme 1 Retrosynthetic analysis for tamiflu 1.

According to retrosynthetic plan, the synthesis began with cis-aziridine 4, which was easily prepared from the D-mannitol and benzhydril amine in good yield using the reported protocol.¹² The aziridine ester 4 was reduced to the corresponding aldehyde 5 using DIBAL-H at −78 °C (Scheme 2). The resultant aldehyde 5 without purification was directly subjected to one carbon Wittig homologation using KO^tBu in toluene to furnish olefin 6 in 65% yield over two steps. Our next task was to obtain the RCM precursor 3a from the olefin 6 by appropriate chemical transformation.

Scheme 2 Synthesis of aziridine 2.

Accordingly, acetonide deprotection of olefin 6 was carried out with TMSOTf in DCM to afford diol 7 in 85% yield. Diol 7 was subjected to oxidative cleavage using NaIO₄ in DCM to afford aldehyde 8. The crude aldehyde 8 was directly subjected to the Barbier addition of ethyl 2-(bromomethyl)-acrylate/Zn in THF/aq NH₄Cl to furnish the diastereomeric mixture of 3a : 3b in 3 : 2 ratio in 94% yield. The diastereomers 3a and 3b were separated by careful flash chromatography using pet.ether–ethyl acetate (90 : 10) as eluent. The undesired stereoisomer 3a was readily converted to the desired 3b by Mitsunobu inversion followed by ester hydrolysis with NaOEt/EtOH to afford 3b in 69% yield.¹³

The relative stereochemistry of diastereomer 3a and 3b was confirmed by transforming the compound 3b to 10 by ring closing metathesis using Grubbs' II generation catalyst in refluxing DCM to furnish cyclohexene aziridine 9, which on mesylation furnished the corresponding mesylate 10. The data of mesylate 10 was found to be in good agreement with the documented data for 10 synthesized by different route.¹⁴ From this the stereochemistry of compound 3b as well as 3a as assigned. According to our retrosynthetic plan, remaining job was to convert the aziridine 10 to intermediate 2, which was carried out by the treatment with 3-pentanol in the presence of BF₃·Et₂O, which resulted into the regio and stereospecific ring opening of aziridine 10 and in situ re-aziridination to afford intermediate 2 in 80% yield (Scheme 2). Since the conversion of intermediate 2 to aziridine 11 and from aziridine 11 to tamiflu 1 are reported in the literature,^4a,b,14 this constitutes for the formal synthesis of tamiflu 1.

We would like to stress that although synthesis of tamiflu has been earlier reported from D-mannitol,^10d,e it required 16 and 18 steps respectively. Our synthesis requires only 13 purification steps. It may be further pointed out that the previous two syntheses in addition to chiral pool strategy employ catalytic asymmetric dihydroxylation to install additional chirality. Our synthesis on the contrary exploits the existing chiral center of mannitol and utilizes it for introducing the desired chirality without resorting any further asymmetric catalysis. In addition to this our synthesis does not generate undesired isomers, as is the case with other two syntheses, except 3a which in turn can be readily transformed to the desired isomer 3b. The aziridine intermediate 11 can be readily transformed to tamiflu by azide free route following the procedure described in ref. 4a, in 5.5% overall yield which is better than the one reported by Mandai et al.^10d

Conclusions

In conclusion a novel, practical and formal synthesis of tamiflu has been accomplished starting with inexpensive and abundant chiral material viz. D-mannitol in 5.5% overall yields. The key building block cis-aziridine is utilised in the synthesis to fix the desired stereocenters of the neuraminidase inhibitor drug oseltamivir phosphate. This notable synthetic attempt involved Wittig olefination and Barbier reaction to access acyclic diene precursor and which was converted to cyclohexene, the core skeleton of tamiflu by ring closing metathesis (RCM). The undesired diastereomer of Barbier reaction was fruitfully converted into the desired isomer using Mitsunobu conditions. The synthetic route is concise, involving inexpensive reagents throughout the synthesis and involves high yielding reaction steps.

Acknowledgements

PNC & LBK thank to CSIR New Delhi, India for SRF. We also thank U. R. Kalkote & H. B. Borate for helpful discussion and SPC thanks CSIR, INDIA and ORIGIN for research funding.

Notes and references

1. R. G. Webster, A. S. Monto, T. J. Braciale and R. A. Lamb, Textbook of Influenza, Oxford, UK: Blackwell Science, 1998, pp. 219–264.
2. A. Graul, P. A. Leeson and J. Castañer, Drugs Future, 1999, 24, 1189.
3. (a) C. U. Kim, W. Lew, M. A. Williams, H. Liu, L. Zhang, S. Swaminathan, N. Bischofberger, M. S. Chen, D. B. Mendel, C. Y. Tai, W. G. Laver and R. C. Stevens, J. Am. Chem. Soc., 1997, 119, 681; (b) C. U. Kim, W. Lew, M. A. Williams, H. Wu, L. Zhang, X. Chen, P. A. Escarpe, D. B. Mendel, W. G. Laver and R. C. Stevens, J. Med. Chem., 1998, 41, 2451.
4. (a) M. Karpf and R. Trussardi, J. Org. Chem., 2001, 66, 2044; (b) J. C. Rohloff, K. M. Kent, M. J. Postich, M. W. Becker, H. H. Chapman, D. E. Kelly, W. Lew, M. S. Louie, L. R. McGee, E. J. Prisbe, L. M. Schultze, R. H. Yu and L. Zhang, J. Org. Chem., 1998, 63, 4545; (c) S. Abrecht, P. Harrington, H. Iding, M. Karpf, R. Trussardi, B. Wirz and U. Zutter, Chimia, 2004, 58, 621.
5. G. Flores, January 4, 2006, http://www.the-scientist.com/?articles.view/articleNo/23583/title/Getting-around-flu-drug-shortage/.
6. (a) J. Magano, Chem. Rev., 2009, 109, 4398; (b) J. Magano, Tetrahedron, 2011, 67, 7875.
7. (a) Y. Fukuta, T. Mita, N. Fukuda, M. Kanai and M. Shibasaki, J. Am. Chem. Soc., 2006, 128, 6312; (b) T. Mita, N. Fukuda, F. X. Roca, M. Kanai and M. Shibasaki, Org. Lett., 2007, 9, 259; (c) K. Yamatsugu, S. Kamijo, Y. Suto, M. Kanai and M. Shibasaki, Tetrahedron Lett., 2007, 48, 1403; (d) K. Alagiri, M. Furutachi, K. Yamatsugu, N. Kumagai, T. Watanabe and M. Shibasaki, J. Org. Chem., 2013, 78, 4019.
8. Y. Y. Yeung, S. Hong and E. J. Corey, J. Am. Chem. Soc., 2006, 128, 6310.
9. H. Ishikawa, T. Suzuki and Y. Hayashi, Angew. Chem., Int. Ed., 2009, 48, 1304.
10. (a) X. Cong and Z. J. Yao, J. Org. Chem., 2006, 71, 5365; (b) J. J. Shie, J. M. Fang, S. Y. Wang, K. C. Tsai, Y. S. E. Cheng, A. S. Yang, S. C. Hsiao, C. Y. Su and C. H. Wong, J. Am. Chem. Soc., 2007, 129, 11892; (c) H. Osato, I. L. Jones, A. Chen and C. L. L. Chai, Org. Lett., 2010, 12, 60; (d) T. Mandai and T. Oshitari, Synlett, 2009, 783; (e) J. S. Ko, J. E. Keum and S. Y. Ko, J. Org. Chem., 2010, 75, 7006; (f) T. Oshitari and T. Mandai, Synlett, 2009, 787; (g) J. Weng, Y. B. Li, R. B. Wang, F. Q. Li, C. Liu, A. S. C. Chan and G. Lu, J. Org. Chem., 2010, 75, 3125.
11. S. Gabriel, Chem. Ber., 1888, 21, 1049.
12. A. L. Williams and J. N. Johnston, J. Am. Chem. Soc., 2004, 126, 1612.
13. P. Tapolcsányi, J. Wölfling, E. Mernyák and G. Schneider, Monatshefte fur Chemie, 2004, 135, 1129.
14. K. Ishiwata, S. Terashima and K. Ujita, EP1229022 B1, or EP1229022, A4 or EP1229022 A1, Aug 7, 2002.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedure, characterization data and copies of ¹H and ¹³C-NMR spectra for the compounds. See DOI: 10.1039/c3ra47210h

---