Rapid continuous synthesis of 5′-deoxyribonucleosides in flow via Br?nsted acid catalyzed glycosylation

Article abstract of DOI:10.1021/ol301324g

A general, green, and efficient Br?nsted acid-catalyzed glycosylation serves as a key step in the one-flow, multistep syntheses of several important 5′-deoxyribonucleoside pharmaceuticals.

Full text of DOI:10.1021/ol301324g

ORGANIC
LETTERS
2012
Vol. 14, No. 13
3348–3351
Rapid Continuous Synthesis of
50-Deoxyribonucleosides in Flow via
Brønsted Acid Catalyzed Glycosylation
Bo Shen and Timothy F. Jamison*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139, United States
tfj@mit.edu
Received May 12, 2012
ABSTRACT
A general, green, and efficient Brønsted acid-catalyzed glycosylation serves as a key step in the one-flow, multistep syntheses of several
important 50-deoxyribonucleoside pharmaceuticals.
Compelling biological activities are associated with
many deoxynucleosides.1 Among them, 50-deoxyribonu-
cleosides, including doxifluridine (1), galocitabine (2), and
capecitabine (3) (Figure 1), demonstrate potent antiviral
and antitumor effects. In particular, capecitabine is an
important drug used for the treatment of breast and
colorectal cancers.2 Syntheses of some specific 50-deoxy-
ribonucleoside targets have been reported,3,4 and in all cases
This method joins a protected sugar (e.g., 4) with silylated
nitrogenous bases in the presence of a Lewis acid. How-
ever, the reaction generally requires several hours and a
stoichiometric amount or excess of a Lewis acid, with
SnCl4 being the most commonly used. Aqueous workup
followed by extraction and separation is required, gener-
ating copious amounts of waste. A greener and more
€
the key glycosylation step is effected using the Vorbruggen
(3) Syntheses via glycosylation: (a) Moon, B. S.; Shim, A. Y.; Lee,
K. C.; Lee, H. J.; Lee, B. S.; An, G. I.; Yang, S. D.; Chi, D. Y.; Choi,
C. W.; Lim, S. M.; Chun, K. S. Bull. Korean Chem. Soc. 2005, 26, 1865.
(b) Moon, B. S.; Lee, K. C.; Lee, H. J.; An, G. I.; Yang, S. D.; Chi, D. Y.;
Chun, K. S. J. Labelled Compd. Radiopharm. 2005, 48, S201. (c) Fei,
X. S.; Wang, J. Q.; Miller, K. D.; Sledge, G. W.; Hutchins, G. D.; Zheng,
Q. H. Nucl. Med. Biol. 2004, 31, 1033. (d) Shimma, N.; Umeda, I.;
Arasaki, M.; Murasaki, C.; Masubuchi, K.; Kohchi, Y.; Miwa, M.; Ura,
M.; Sawada, N.; Tahara, H.; Kuruma, I.; Horii, I.; Ishitsuka, H. Bioorg.
Med. Chem. 2000, 8, 1697. (e) Li, J. J., Johnson, D. S., Eds. Modern Drug
Synthesis; John Wiley & Sons: Hoboken, 2010; pp 57ꢀ71 and references
cited therein. Selected patents:(f)Brinkman, H. R.;Kalaritis, P.; Morrissey,
J. F. US Pat. 5476932, Dec 19, 1995. (g) Hu, T.-C.; Huang, H.-T. PCT Int.
Appl. 2011010967, Jan 27, 2011. (h) Kamiya, T.; Ishiduka, M.; Nakajima, H.
Eur. Pat. Appl. 602478, Jun 22, 1994. (i) Fujiu, M.; Ishitsuka, H.; Miwa, M.;
Umeda, I.; Yokose, K. Eur. Pat. Appl. 316704, May 24, 1989. (j) Li, J.; He, B.;
Shao, L.; Wang, L. PCT Int. Appl. 2005080351, Sep 1, 2005. (k) Bertolini,
G.; Frigerio, M. PCT Int. Appl. 2005040184, May 6, 2005.
modification of the silylꢀHilbertꢀJohnson reaction.5,6
(1) (a) Ichikawa, E.; Kato, K. Curr. Med. Chem. 2001, 8, 385. (b) Chu,
C. K., Baker, D. C., Eds. Nucleosides and Nucleotides as Antitumor and
Antiviral Agents; Plenum Press: New York, 1993. (c) Peters, G. J. Deox-
ynucleoside Analogs in Cancer Therapy; Humana Press: Totowa, NJ, 2006.
(d) Herdwijn, P., Modified Nucleosides: in Biochemistry, Biotechnology
and Medicine; Wiley-VCH: Weinheim, 2008. (e) Perigaud, C.; Gosselin, G.;
Imbach, J. L. Nucleosides Nucleotides 1992, 11, 903.
(2) For reviews on capecitabine, see: (a) Walko, C. M.; Lindley, C.
Clin. Ther. 2005, 27, 23. (b) Wagstaff, A. J.; Ibbotson, T.; Goa, K. L.
Drugs 2003, 63, 217. (c) McGavin, J. K.; Goa, K. L. Drugs 2001, 61,
2309. (d) Koukourakis, G. V.; Kouloulias, V.; Koukourakis, M. J.;
Zacharias, G. A.; Zabatis, H.; Kouvaris, J. Molecules 2008, 13, 1897.
(e) Malet-Martino, M.; Martino, R. Oncologist 2002, 7, 288. (f) Ishitsuka,
H.; Shimma, N.; Horii, I. J. Pharmaceut. Soc. Jap. 1999, 119, 881.
r
10.1021/ol301324g
Published on Web 06/13/2012
2012 American Chemical Society

efficient process for the synthesis of 50-deoxyribonucleo-
sides is highly desired. Herein we report an efficient syn-
thesis of these valuable 50-deoxyribonucleosides, a one-flow
method that combines two to three steps into a single,
telescoped procedure and affords the desired compounds
rapidly (typically <10 min) and in high yields (typically
>90%).
Table 1. Brønsted Acid Catalyzed Glycosylation Reaction in
Flow
Figure 1. 50-Deoxyribonucleoside drugs.
Continuous flow synthetic chemistry has emerged as a
promising technology in the past decade, offering several
advantages overtraditional batchprocesses.7,8 Werecently
reported the first organocatalytic method for nucleobase
glycosylation, a Brønsted acid-catalyzedCꢀN bond-form-
ing reaction in flow that provides ribonucleosides in high
yields.9 Pyridinium triflates, e.g., 5, were found to be the
superior catalysts.
We thus began our present study with the glycosyla-
tion of silylated thymine 6a by 5-deoxyribose donor 4.
A syringe pump was used to deliver the solution of the base
(4) Syntheses via modification of ribonucleosides: (a) Bavaro, T.;
Rocchietti, S.; Ubiali, D.; Filice, M.; Terreni, M.; Pregnolato, M. Eur.
J. Org. Chem. 2009, 1967. (b) Cotticelli, G.; De Meglio, G.; Monciardini,
S.; Ordanini, G. PCT Int. Appl., 9852960, Nov 26, 1998. (c) Cook, A. F.
Patentschrift (Switz.), 633810, Dec 31, 1982.
€
(5) (a) Niedball, U.; Vorbruggen, H. Angew. Chem., Int. Ed. 1970, 9,
€
461. (b) Vorbruggen, H.; Krolikiewicz, K. Angew. Chem., Int. Ed. 1975,
€
14, 421. (c) Vorbruggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber.
1981, 114, 1234.
(6) For a review on the synthesis of nucleosides, see: Vorbruggen, H.;
€
Ruh-Polenz, C. Org. React. 2000, 55, 1.
a All reactions were performed using 1 equiv of sugar 4, 1.1ꢀ2 equiv
of silylated base 6, and 5 mol % of catalyst 5, unless otherwise noted. See
the Supporting Information for complete details. b Isolated yields.
Bisglycosylation product yields in parentheses. c 10 mol % of catalyst
5 was used. d Mixture of isomers N9/N7 = 5:1.
(7) For recent general reviews on continuous flow chemistry, see:
(a) Wiles, C.; Watts, P. Chem. Commun. 2011, 47, 6512. (b) Wegner, J.;
Ceylan, S.; Kirschning, A. Chem. Commun. 2011, 47, 4583. (c) Hartman,
R. L.; McMullen, J. P.; Jensen, K. F. Angew. Chem., Int. Ed. 2011, 50,
7502. (d) Yoshida, J. I.; Kim, H.; Nagaki, A. Chemsuschem 2011, 4, 331.
(e) Illg, T.; Lob, P.; Hessel, V. Bioorg. Med. Chem. 2010, 18, 3707.
(f) Hartman, R. L.; Jensen, K. F. Lab Chip 2009, 9, 2495. (g) Geyer, K.;
Gustafsson, T.; Seeberger, P. H. Synlett 2009, 2382. (h) Wirth, T.,
Microreactors in Organic Synthesis and Catalysis; Wiley-VCH:
Weinheim, 2008. (i) Fukuyama, T.; Rahman, T.; Sato, M.; Ryu, I. Synlett 2008,
151. (j) Mason, B. P.; Price, K. E.; Steinbacher, J. L.; Bogdan, A. R.;
McQuade, D. T. Chem. Rev. 2007, 107, 2300. (k) Watts, P.; Wiles, C.
Chem. Commun. 2007, 443. (l) Kirschning, A.; Solodenko, W.; Mennecke,
K. Chem.;Eur. J. 2006, 12, 5972. (m) Jahnisch, K.; Hessel, V.; Lowe,
H.; Baerns, M. Angew. Chem., Int. Ed. 2004, 43, 406.
and the sugar with the catalyst in two syringes. They were
mixed in a T-mixer and reacted at an elevated temperature
in PFA tubing coils (120 μL) as they flowed. A back
pressure regulator (100 psi) was employed to allow reac-
tion temperatures above the boiling point of solvent
CH3CN. We quickly identified that in the presence of
5 mol % of catalyst 5, the glycosylation of 6a with 5-deoxy-
ribose 4 was complete at 120 °C in only 5 min, affording
50-deoxyribonucleoside 7a in 99% yield and >98:2 dr
(limit of detection) after purification via column chroma-
tography (Table 1, entry 1). It is worth noting that no
(8) For reviews on the syntheses of pharmaceticals by flow chemistry,
see: (a) Malet-Sanz, L.; Susanne, F. J. Med. Chem. 2012, 55, 4062.
(b) Proctor, L.; Dunn, P. J.; Hawkins, J. M., Wells, A. S.; Williams,
M. T. Continuous Processing in the Pharmaceutical Industry. In Green
Chemistry in the Pharmaceutical Industry; Wiley-VCH: Weinheim, 2010;
pp 221ꢀ242.
(9) Sniady, A.; Bedore, M. W.; Jamison, T. F. Angew. Chem., Int. Ed.
2011, 50, 2155.
Org. Lett., Vol. 14, No. 13, 2012
3349

Scheme 1. One-Flow, Two-Step Synthesis of 50-Deoxyribonu-
cleosides
Scheme 2. Synthesis of Doxifluridine
Scheme 3. Synthesis of Galocitabine and Capecitabine
aqueous workup was necessary, reducing labor, cost, and
waste.
Other pyrimidine derivatives were examined and indivi-
dually optimized, rapidly (1ꢀ12 min) affording desired
50-deoxynucleosides 7 in high yields (Table 1, entries 3ꢀ7).
An important trend was observed that merits discussion:
nucleophiles bearing electron-withdrawing groups were
more reactive, requiring less time and lower temperature
(cf. entries 1 vs 7 and 3 vs 4), consistent with our previous
study wherein a ribose derivative was used as the glycosyl
donor.9 Smaller nucleophiles tended to give small amounts
of N1,N3-bisglycosylation products (entries 3, 4, 6, and 8).
Nevertheless, the formation of the desired monosubstitu-
tion product was favored at a lower temperature with
extended reaction time (cf. entries 2 vs 3, 55% vs 92%
yield, respectively).
This glycosylation method encompasses other silylated
bases, such as derivatives of naturally occurring cytosine
(6g), adenosine (6h) and guanine (6i), as well as the
nonnatural imidazole derivative 6j (entries 8ꢀ11). All were
smoothly converted to the corresponding deoxynucleo-
sides in high yield and, as above, >98:2 dr. In the case of
guanine derivative 6i, a mixture of regioisomers (N9/N7 = 5:1)
was obtained, similar to our previous observations.9
For a reaction with 0.4 M in sugar 4 and residence time
of 5 min, using a 120 uL tubing reactor, a throughput of
0.288 mmol/h was obtained (115 mg/h for MW of 400
based on 100% yield). Higher through-put can generally
be achieved using reactors with a larger volume and faster
flow rate under the same reaction conditions.
and isolating intermediate products.10 The very efficient
heat transfer of the continuous flow format also allows
drastic changes in reaction conditions from one step to the
next; very rapid heating/cooling of the small reaction
volumes is easily accomplished. We examined the one-
flow, two-step synthesis of fully deprotected 50-deoxyribo-
nucleosides by introducing a solution of NaOH in metha-
nol and water to the exiting stream of the glycosylation
reaction (Scheme 1). The deprotection was complete in
only 2 min at room temperature, affording 8a and 8f in
excellent yields.11 The use of methanol/water mixture as
the solvent is critical as it maintains the homogeneity of the
reaction mixture. When only methanol was used, the
product precipitated and clogged the tubing; when only
water was used, the reaction mixture became biphasic,
significantly slowing the reaction. Deprotection using
methanolic ammonia (7 N) was also examined but in
comparison was much more sluggish at room temperature.
(10) For reviews, see: (a) Webb, D.; Jamison, T. F. Chem. Sci. 2010,
1, 675. (b) Wegner, J.; Ceylan, S.; Kirschning, A. Adv. Synth. Catal.
2012, 354, 17.
(11) The product crude mixture was neutralized, concentrated, and
purified using column chromatography. See the Supporting Informa-
tion for experimental details.
Multistep synthesis in flow has emerged as a very
effective strategy as it circumvents the need for purifying
3350
Org. Lett., Vol. 14, No. 13, 2012

Scheme 4. One-Flow, Three-Step Synthesis of Capecitabine
We then investigated the one-flow, two-step synthesis of
doxifluridine. Because of its poor solubility in methanol
(even the presence of water could not prevent precipitation
in this case), the streams of the glycosylation reaction
and NaOMe were simultaneously introduced to a round-
bottom flask, where the deprotection occurred (Scheme 2).
We used NaOMe instead of NaOH for deprotection because
the byproduct (methyl acetate) was easier to remove than
acetic acid. Thus, doxifluridine (1) was obtained in 89% yield
in only 10 min.
In order to synthesize galocitabine and capecitabine,
silylated 5-fluorocytosine derivatives 6k and 6l were pre-
pared and immediately used. Being much less reactive,
these substrates required higher reaction temperature and
10 mol % catalyst 5. The glycosylation of 6k occurred
smoothly at 140 °C and upon deprotection provided
galocitabine (2) in 89% yield (Scheme 3).
immediately; otherwise, decomposition occurred. With
10 mol % of pyridinium triflate 5, this three-step-in-
one-flow protocol (glycosylationꢀcarbamate formationꢀ
deprotection) afforded capecitabine (3) in 72% yield in less
than 1 h and is clearly greener and more efficient than any
other previously reported processes.3aꢀj
Insummary, wehavedeveloped a generalsynthesisof50-
deoxyribonucleosides by way of an organocatalytic glyco-
sylation reaction in continuous flow. The use of a Brønsted
acid as a catalyst renders this method greener and more
efficient than those requiring stoichiometric amounts
(or more) of Lewis acids. The one-flow, multistep sequence
circumvents purification of the intermediate products
and produces unprotected 50-deoxyribonucleosides in a
streamlined manner, as demonstrated by the syntheses of
the important compounds doxifluridine, galocitabine, and
capecitabine.
However, the glycosylation of 6l was accompanied with
significant decomposition, resulting in only 50% yield of
the corresponding product. An alternative sequence that
began with glycosylation of 6m instead of carbamate 6l
provided a solution to this problem (Scheme 4). The more
electron-rich and thus less reactive 6m was glycosy-
lated in 20 min at 130 °C. The subsequent carbamate
formation required 30 min at ambient temperature.
Attempts to heat this stage in order to improve efficiency
proved unfruitful. Critical are both mixing the chlorofor-
mate and pyridine at 0 °C and using the resulting mixture
Acknowledgment. We thank the NovartisꢀMIT Center
for Continuous Manufacturing for their continued gener-
osity and support.
Supporting Information Available. Complete experi-
mental procedures and characterization data of novel
compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 13, 2012
3351