Synthesis of EFdA via a diastereoselective aldol reaction of a protected 3-keto furanose

Article abstract of DOI:10.1021/ol5036535

An efficient enantioselective total synthesis of EFdA, a remarkably potent anti-HIV nucleoside analogue with various favorable pharmacological profiles, has been achieved in 37% overall yield from diacetone-d-glucose by a 14-step sequence that features a

Full text of DOI:10.1021/ol5036535

Letter
pubs.acs.org/OrgLett
Synthesis of EFdA via a Diastereoselective Aldol Reaction of a
Protected 3‑Keto Furanose
Kei Fukuyama,† Hiroshi Ohrui,*,‡ and Shigefumi Kuwahara*,†
†Laboratory of Applied Bioorganic Chemistry, Graduate School of Agricultural Science, Tohoku University,
Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
‡Yokohama University of Pharmacy, 601 Matano-cho, Totsuka-ku, Yokohama 245-0066, Japan
S
* Supporting Information
ABSTRACT: An efficient enantioselective total synthesis of
EFdA, a remarkably potent anti-HIV nucleoside analogue with
various favorable pharmacological profiles, has been achieved in
37% overall yield from diacetone-D-glucose by a 14-step sequence
that features a highly diastereoselective installation of the
tetrasubstituted stereogenic center at the C4′ position, direct
oxidative cleavage of an acetonide-protected diol derivative to an
aldehyde, and one-pot 2′-deoxygenation of a ribonucleoside intermediate.
4′-Ethynyl-2-fluoro-2′-deoxyadenosine (EFdA, 1) is a nucleo-
side reverse transcriptase inhibitor (NRTI) created by
collaborative studies among the Ohrui group, Mitsuya group,
and Yamasa Corp. (Figure 1).1 The structure of 1 differs
in progress aimed at developing 1 as a therapeutic agent for
HIV infection.6
As for the synthesis of 1, there have been only two methods
developed: one by Ohrui and co-workers and the other by us.
The first synthesis of 1 reported by Ohrui et al.,1a,b which led to
the discovery of 1 as a potent NRTI, required a lengthy 18-step
sequence from the expensive starting material 2-amino-2′-
deoxyadenosine, which resulted in a modest overall yield of
2.5%. The second synthesis accomplished by our group
succeeded in reducing the number of steps and enhancing
the overall yield [18% over 12 steps from (R)-glyceraldehyde
acetonide], but unfortunately, it necessitated careful chromato-
graphic separation of diastereomeric intermediates at an early
and a late stage of the synthesis.7 The problems posed in the
two previous syntheses prompted our efforts at developing a
more efficient synthesis of 1 that could supply 1 more easily.
We describe herein a new practical synthesis of 1 in 37% overall
yield from diacetone-D-glucose by a 14-step sequence which
requires only four chromatographic purifications and involves
an exclusively diastereoselective construction of the tetrasub-
stituted asymmetric center at the C4′ position.
Scheme 1 outlines our retrosynthetic analysis of 1. To install
the nucleobase unit of 1 in a stereoselective manner, we chose a
2′-acetoxyfuranose derivative 3 as a key intermediate; its N-
glycosidation with 2-fluoroadenine should proceed in a highly
β-selective manner based on the neighboring group partic-
ipation of the 2′-acetoxy substituent,8 and subsequent 2′-
deacetylation of the resulting product would lead to ribonucleo-
side derivative 2. Deoxygenation at the C2′ position of 2
followed by deprotection would then furnish the target
molecule 1. The acetylenic intermediate 3 would be derived
from aldehyde 4 by an appropriate alkyne-forming reaction.
Figure 1. Structure of EFdA.
substantially from other NRTIs clinically approved for the
treatment of human immunodeficiency virus (HIV) infection,
since it retains the 3′-OH functionality of 2′-deoxyadenosine as
well as the presence of the 2-fluoro substituent.2 The
installation of an ethynyl group and a fluoro substituent at
the C4′ and C2 positions of the parent natural nucleoside,
respectively, while maintaining the 3′-hydroxy group, endowed
1 with highly promising pharmacological profiles as an anti-
HIV agent including (1) exceptionally potent antiviral activity
against both wild-type and multidrug-resistant HIV-1 strains
(ED50, low nM to pM level),3 (2) no acute toxicity in ICR mice
at a dose of 100 mg/kg,1a,4 and (3) longer intracellular half-life
(t1/2, 17.2 h) of its active form (EFdA-5′-triphosphate) than
that of zidovudine triphosphate (AZT-5′-triphosphate, t1/2, 2.8
h),3a which may enable a once- or twice-daily regimen and
thereby improve the quality of life (QOL) of people suffering
from AIDS. Because of these favorable characteristics of 1 in
terms of potency, safety, and pharmacokinetics, detailed
mechanistic as well as clinical studies have been carried out
on 1. There is particular interest in establishing why resistance
to 1 occurs less frequently,5 and these investigations are actively
Received: December 18, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/ol5036535
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
conversion could be best performed by treating 7 with NaOCl
and KBr in CH2Cl2/aq NaHCO3 in the presence of a
surprisingly small amount (0.005 mol %) of Nor-AZADO
(8).15 This reaction went to completion within 30 min at 0 °C
and could be successfully conducted on a multigram scale (at
least up to 100 g). The new oxidation system for the conversion
of 7 into 6 seems to be superior to previous ones in terms of
scalability, operational easiness, and also greenness.16 The 4′-
hydroxymethylation of 6, one of the key steps in our synthesis,
proceeded as expected by simply treating 6 with paraformalde-
hyde and K2CO3 in THF, affording a mixture of aldol 5 and
cyclic hemiacetal 9 (ca. 3:1), the latter of which must have been
formed by 5′-O-hydroxymethylation of 5 with excess form-
aldehyde followed by cyclic hemiacetal formation between the
newly formed hydroxyl group and the 3′-keto functionality.
The structure of 9, which could be isolated as a white powder
from the mixture of 5 and 9, was fully characterized by
spectroscopic analyses including COSY, DEPT, and HMBC
experiments.17,18 To the best of our knowledge, this type of
hydroxymethylation at the C4 position of 3-keto furanose
derivatives has no precedent in the literature; a one-pot aldol−
Cannizzaro reaction of a protected 5-oxo pentofuranose with
formaldehyde has been known to hydroxymethylate the C4
position of the furanose derivative,19 but the formation of two
primary hydroxyl groups in the product brought about a
bothersome problem regarding the selective protection of the
1,3-diol.19b,d The raw reaction mixture of 5 and 9, obtained
after the aldol reaction, was filtered, diluted with water, and
treated directly with NaBH4 to give 10 as a white solid in an
excellent overall yield of 93% from 7 after purification by simple
trituration. Protection of the 5′-hydroxy group of 10 as its
benzyl ether 11 was then followed by chemoselective
deprotection of the terminal acetonide group (I2 in MeCN)20
and concomitant oxidative cleavage of the resulting diol
intermediate (NaIO4) in one pot to give known aldehyde 12
(nearly quantitatively).21
Scheme 1. Retrosynthetic Analysis of 1
The aldehyde 4 would be obtainable by diastereoselective
reduction of ketone 5 from the convex face of the bicyclic fused
ring system and subsequent chemoselective deprotection of the
terminal acetonide followed by oxidative cleavage of the vicinal
diol intermediate produced. For the diastereoselective for-
mation of 5 bearing a tetrasubstituted asymmetric center at the
C4′ position, we envisaged the use of the aldol reaction of
protected 3-keto furanose 6 with formaldehyde. Based on a
literature precedent,9 treatment of 6 under suitable basic
conditions was anticipated to produce an enolate intermediate
with a double bond at the C3′−C4′ position. The electrophilic
approach of formaldehyde to the enolate from the less hindered
β-face would then give 5 in a diastereoselective manner.
Compound 6 is commercially available or can be readily
prepared by oxidation of the cheap starting material diacetone-
D-glucose (7).
Our synthesis of 1 commenced with the oxidation of 7 to
ketone 6 (Scheme 2). This transformation has previously been
effected by the Dess−Martin oxidation,10 the Swern oxida-
tion,11 or by reagent systems such as PDC/Ac2O,12 PDC/
AcOH,13 and TEMPO (0.5 mol %)/NaOCl/KBr.14 After
examination of various oxidation conditions, we found that the
The aldehyde 12 was then converted into glycosyl acetate 16
(Scheme 3) via acetylenic derivative 14 according to our
Scheme 3. Conversion of Aldehyde 12 into Ribonucleoside
Derivative 19
Scheme 2. Preparation of Aldehyde Intermediate 12 Bearing
a 4′-Tetrasubstituted Stereocenter
B
DOI: 10.1021/ol5036535
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
previously reported procedures except that the transformation
of dibromoolefin 13 to acetylenic intermediate 14 was effected
in a single operation by directly trapping an acetylide
intermediate with TESCl.21,22 Removal of the acetonide
group of 14 under acidic conditions followed by bis-acetylation
of the resulting hydroxyl hemiacetal 15 furnished 16. The N-
glycosylation of 16 with 2-fluoroadenine (17) was best
performed by treating 16 with TMSOTf and DBU in MeCN
to give 18,23 the deacetylation of which then afforded 19 as a
single diastereomer. Each step of the reaction sequence from 10
to 19 proceeded in a good-to-excellent yield, providing 14 in
88% overall yield from 10 and 19 in 75% overall yield from 14.
The final stage of our synthesis of EFdA (1) is shown in
Scheme 4. The adenosine derivative 19 was subjected to a one-
diastereomers (6 → 5, 5 → 10, and 16 → 18) proceeded
with virtually perfect stereoselection. From these favorable
features, as well as the use of 7 as an inexpensive starting
material, the new synthesis described herein is considered to be
more efficient and practical than previous syntheses.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, characterization data, and copies of
NMR spectra for new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail: h.ohrui@hamayaku.ac.jp.
*E-mail: skuwahar@biochem.tohoku.ac.jp.
Scheme 4. Completion of the Synthesis of EFdA (1)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful to Prof. Iwabuchi and Dr. Sasano (Tohoku
University) for kindly providing Nor-AZADO.
■
REFERENCES
■
(1) (a) Kohgo, S.; Ohrui, H.; Kodama, E.; Matsuoka, M.; Mitsuya, H.
Can. Patent CA 2502109, 2005. (b) Kohgo, S.; Yamada, K.; Kitano, K.;
Iwai, Y.; Sakata, S.; Ashida, N.; Hayakawa, H.; Nameki, D.; Kodama,
E.; Matsuoka, M.; Mitsuya, H.; Ohrui, H. Nucleosides Nucleotides
Nucleic Acids 2004, 23, 671−690. (c) Ohrui, H. Chem. Rec. 2006, 6,
133−143. (d) Ohrui, H. Proc. Jpn. Acad., Ser. B 2011, 87, 53−65.
(e) Ohrui, H. J. Antivir. Antiretrovir. 2014, 6, 32−39.
(2) (a) Sarafianos, S. G.; Marchand, B.; Das, K.; Himmel, D. M.;
Parniak, M. A.; Hughes, S. H.; Arnold, E. J. Mol. Biol. 2009, 385, 693−
713. (b) Nikolenko, G. N.; Kotelkin, A. T.; Oreshkova, S. F.; Ilyichev,
A. A. Mol. Biol. 2011, 45, 93−109.
(3) (a) Nakata, H.; Amano, M.; Koh, Y.; Kodama, E.; Yang, G.;
Bailey, C. M.; Kohgo, S.; Hayakawa, H.; Matsuoka, M.; Anderson, K.
S.; Cheng, Y.-C.; Mitsuya, H. Antimicrob. Agents Chemother. 2007, 51,
2701−2708. (b) Kawamoto, A.; Kodama, E.; Sarafianos, S. G.;
Sakagami, Y.; Kohgo, S.; Kitano, K.; Ashida, N.; Iwai, Y.; Hayakawa,
H.; Nakata, H.; Mitsuya, H.; Arnold, E.; Matsuoka, M. Int. J. Biochem.
Cell Biol. 2008, 40, 2410−2420. (c) Michailidis, E.; Marchand, B.;
Kodama, E. N.; Singh, K.; Matsuoka, M.; Kirby, K. A.; Ryan, E. M.;
Sawani, A. M.; Nagy, E.; Ashida, N.; Mitsuya, H.; Parniak, M. A.;
Sarafianos, S. G. J. Biol. Chem. 2009, 284, 35681−35691.
pot process comprising its derivatization into 21 with 1,1′-
(thiocarbonyl)diimidazole 20 and subsequent reduction of 21
under Barton−McCombie conditions,24 providing 22 in an
excellent yield of 98%. Although 2′-deoxygenation of
ribonucleosides has previously been carried out in two separate
steps,25 we found that the two-step sequence could be effected
in one pot by using toluene as the same solvent for both
reactions.26 Finally, the 2′-deoxyadenosine derivative 22 was
exposed to TBAF in THF and then to Na in liquid ammonia to
remove the TES and benzyl protecting groups, respectively,
furnishing EFdA (1) in 61% yield for the two steps. The 1H and
13C NMR spectra of 1 were identical with those of an authentic
(4) Hattori, S.; Ide, K.; Nakata, H.; Harada, H.; Suzu, S.; Ashida, N.;
Kohgo, S.; Hayakawa, H.; Mitsuya, H.; Okada, S. Antimicrob. Agents
Chemother. 2009, 53, 3887−3893.
(5) For recent publications, see: (a) Muftuoglu, Y.; Sohl, C. D.;
Mislak, A. C.; Mitsuya, H.; Sarafianos, S. G.; Anderson, K. S. Antiviral
Res. 2014, 106, 1−4. (b) Zhang, W.; Parniak, M. A.; Sarafianos, S. G.;
Empey, P. E.; Rohan, L. C. Eur. J. Pharmacol. 2014, 732, 86−95.
(c) Maeda, K.; Desai, D. V.; Aoki, M.; Nakata, H.; Kodama, E. N.;
Mitsuya, H. Antivir. Ther. 2014, 19, 179−189. (d) Matsuzawa, T.;
Kawamura, T.; Ogawa, Y.; Maeda, K.; Nakata, H.; Moriishi, K.;
Koyanagi, Y.; Gatanaga, H.; Shimada, S.; Mitsuya, H. J. Invest.
Dermatol. 2014, 134, 1158−1161. (e) Michailidis, E.; Huber, A. D.;
Ryan, E. M.; Ong, Y. T.; Leslie, M. D.; Matzek, K. B.; Singh, K.;
Marchand, B.; Hagedorn, A. N.; Kirby, K. A.; Rohan, L. C.; Kodama,
E. N.; Mitsuya, H.; Parniak, M. A.; Sarafianos, S. G. J. Biol. Chem. 2014,
289, 24533−24548.
material, and the specific rotation of 1 [[α]27 +12.9 (c 1.00,
D
MeOH)] showed good agreement with a reported value
[[α]25 +12.4 (c 0.97, MeOH)].7
D
In conclusion, an enantioselective total synthesis of EFdA (1)
has been accomplished in an excellent overall yield of 37% from
diacetone-D-glucose 7 by a 14-step sequence that features the
diastereoselective installation of the tetrasubstitured stereo-
center at the C4′ position (6 → 5), chemoselective acetonide
hydrolysis of 11, and concomitant oxidative cleavage of the
resulting diol to form aldehyde 12. A one-pot 2′-deoxygenation
of 19 provided 22. Of value is the fact that the present synthesis
requires only four chromatographic purifications, mainly
because every reaction that could potentially produce
(6) Merck Home Page. http://www.merck.com/licensing/our-
partnership/Merck-chimerix-partnership.html (accessed Dec 2014).
(7) (a) Kageyama, M.; Nagasawa, T.; Yoshida, M.; Ohrui, H.;
Kuwahara, S. Org. Lett. 2011, 13, 5264−5266. (b) Kageyama, M.;
C
DOI: 10.1021/ol5036535
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Miyagi, T.; Yoshida, M.; Nagasawa, T.; Ohrui, H.; Kuwahara, S. Biosci.
Biotechnol. Biochem. 2012, 76, 1219−1225.
(8) Conventional atom numbering for nucleosides is used for all
compounds in this paper.
(9) For an example of the conversion of a protected 3-keto furanose
into the corresponding enol acetate with a double bond at the C3−C4
position, see: Hanaya, T.; Baba, H.; Kanemoto, M.; Yamamoto, H.
Heterocycles 2008, 76, 635−644.
(10) Johnson, D. C., II; Wildlanski, T. S. J. Org. Chem. 2003, 68,
5300−5309.
(11) Chun, M. W.; Choi, S. W.; Kang, T. K.; Choi, W. J.; Kim, H. O.;
Gao, Z.-G.; Jacobson, K. A.; Jeong, L. S. Nucleosides Nucleotides Nucleic
Acids 2008, 27, 408−420.
(12) Andersson, F.; Samuelsson, B. Carbohydr. Res. 1984, 129, C1−
C3.
(13) Shing, T. K. M.; Wong, C.-H.; Yip, T. Tetrahedron: Asymmetry
1996, 7, 1323−1340.
(14) Turks, M.; Rodins, V.; Rolava, E.; Ostrovskis, P.; Belyakov, S.
Carbohydr. Res. 2013, 375, 5−15.
(15) Hayashi, M.; Sasano, Y.; Nagasawa, S.; Shibuya, M.; Iwabuchi, Y.
Chem. Pharm. Bull. 2011, 59, 1570−1573.
(16) See the Supporting Information for comparison of some
oxidation conditions to convert 7 into 6.
(17) For examples of analogous cyclic hemiacetals, see: (a) Tronchet,
J. M. J.; Neeser, J.-R.; Charollais, E. J. Helv. Chim. Acta 1977, 60, 243−
246. (b) Kodama, T.; Shuto, S.; Ichikawa, S.; Matsuda, A. J. Org. Chem.
2002, 67, 7706−7715.
(18) For an example of analogous 5′-O-hydroxymethylation, see:
Haraguchi, K.; Shimada, H.; Kimura, K.; Akutsu, G.; Tanaka, H.; Abe,
H.; Hamasaki, T.; Baba, M.; Gullen, E. A.; Dutschman, G. E.; Cheng,
Y.-C.; Balzarini, J. ACS Med. Chem. Lett. 2011, 2, 692−697.
(19) (a) Youssefyeh, R. D.; Verheyden, J. P. H.; Moffatt, J. G. J. Org.
Chem. 1979, 44, 1301−1309. (b) Waga, T.; Nishizaki, T.; Miyakawa,
I.; Ohrui, H.; Meguro, H. Biosci. Biotechnol. Biochem. 1993, 57, 1433−
1438. (c) Koshikin, A. A.; Singh, S. K.; Nielsen, P.; Rajwanshi, V. K.;
Kumar, R.; Meldgaard, M.; Olsen, C. E.; Wengel, J. Tetrahedron 1998,
54, 3607−3630. (d) Sharma, V. K.; Kumar, M.; Olsen, C. E.; Prasad, A.
K. J. Org. Chem. 2014, 79, 6336−6341.
(20) Yadav, J. S.; Satyanarayana, M.; Raghavendra, S.; Balanarsaiah, E.
Tetrahedron Lett. 2005, 46, 8745−8748.
(21) Ohrui, H.; Kohgo, S.; Kitano, K.; Sakata, S.; Kodama, E.;
Yoshimura, K.; Matsuoka, M.; Shigeta, S.; Mitsuya, H. J. Med. Chem.
2000, 43, 4516−4525.
(22) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 13, 3769−
3772.
(23) (a) Kristinsson, H.; Nebel, K.; O’Sullivan, A. C.; Struber, F.;
Winkler, T.; Yamaguchi, Y. Tetrahedron 1994, 50, 6825−6838.
(b) Moreau, C.; Kirchberger, T.; Swarbick, J. M.; Bartlett, S. J.;
Fliegert, R.; Yorgan, T.; Bauche, A.; Harneit, A.; Guse, A. H.; Potter, B.
V. L J. Med. Chem. 2013, 56, 10079−10102.
(24) Crich, D.; Quintero, L. Chem. Rev. 1989, 89, 1413−1432.
́
(25) For examples, see: (a) Velazquez, S.; Jimeno, M. L.; Huss, S.;
Balzarini, J.; Camarasa, M.-J. J. Org. Chem. 1994, 59, 7661−7670.
(b) Ravan, J.; Freitag, M.; Nielsen, P. Org. Biomol. Chem. 2003, 1,
811−816. (c) Stauffiger, A.; Leumann, C. J. Eur. J. Org. Chem. 2009,
1153−1162.
(26) The one-pot deoxygenation from 19 to 22 was conducted by
using 60 mg of 19 at most, and the scalability of this reaction was not
examined in the present study.
D
DOI: 10.1021/ol5036535
Org. Lett. XXXX, XXX, XXX−XXX