The first chemical synthesis of pyrazofurin 5′-triphosphate

Article abstract of DOI:10.1016/j.tetlet.2018.08.008

As an archetype C-nucleoside, pyrazofurin possesses broad-spectrum antiviral and antitumor activities. However, the presence of the acidic enol in the nucleobase of pyrazofurin poses a huge challenge to the conventional NTP synthetic methods. On the basis

Full text of DOI:10.1016/j.tetlet.2018.08.008

Accepted Manuscript
The first chemical synthesis of pyrazofurin 5′-triphosphate
Hua-Shan Huang, Rui Wang, Wei-Jie Chen, Ji-Zong Chen, Shan-Shan Gong,
Qi Sun
PII:
S0040-4039(18)30973-0
https://doi.org/10.1016/j.tetlet.2018.08.008
TETL 50183
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
21 June 2018
27 July 2018
2 August 2018
Please cite this article as: Huang, H-S., Wang, R., Chen, W-J., Chen, J-Z., Gong, S-S., Sun, Q., The first chemical
synthesis of pyrazofurin 5′-triphosphate, Tetrahedron Letters (2018), doi: https://doi.org/10.1016/j.tetlet.
2018.08.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Tetrahedron Letters
journal homepage: www.elsevier.com
The first chemical synthesis of pyrazofurin 5′-triphosphate
Hua-Shan Huang, Rui Wang, Wei-Jie Chen, Ji-Zong Chen, Shan-Shan Gong* and Qi Sun
Jiangxi Key Laboratory of Organic Chemistry, Jiangxi Science & Technology Normal University, 605 Fenglin Avenue, Nanchang, 330013, China
———
ARTICLE INFO
ABSTRACT
 Corresponding author. Tel.: +86-791-8380-5183; fax: +86-791-8382-6894; e-mail: gongshanshan@jxstnu.edu.cn; sunqi@jxstnu.edu.cn.
Article history:
Received
Received in revised form
Accepted
As an archetype C-nucleoside, pyrazofurin possesses broad-spectrum antiviral and antitumor
activities. However, the presence of the acidic enol in the nucleobase of pyrazofurin poses a
huge challenge to the conventional NTP synthetic methods. On the basis of a selective
phosphoylation method and the P(V)-N activation strategy, the first chemical synthesis of
pyrazofurin 5′-triphosphate (PTP) was accomplished, which will greatly facilitate the
investigation on the interactions of viral RNA polymerases and PTP, an important cellular
metabolite of pyrazofurin.
Available online
Keywords:
Pyrazofurin
triphosphate
2009 Elsevier Ltd. All rights reserved.
phosphorylation
P(V)-N activation
Pyrazofurin, a naturally occurring C-nucleoside first isolated
from the fermentation of a strain of Streptomyces candidus in
1969,1 is a close structural analog of a natural imidazole
nucleoside, bredinin,2 and a synthetic triazole nucleoside drug,
ribavirin (Figure 1).3 As a microbial metabolite, pyrazofurin
possesses broad-spectrum antiviral activities against both RNA
and DNA viruses4 and antitumor properties.5
synthesis of pyrazofurin 5′-triphosphate (PTP) has always been
an unsolved problem for phosphorus chemists. But it can be
envisioned that the availability of synthetic PTP will help
virologists and medicinal chemists elucidate the potential roles of
PTP in halting virus replication and develop effective antiviral
strategies.
In recent years, our research group reported the P(V)-N
activation strategy for the synthesis of nucleoside
polyphosphates,14-15 nucleoside diphosphate sugars,16 and
dinucleoside polyphosphates.17 The high coupling efficacy and
tolerance of diverse functional groups on nucleobases offer an
ideal approach to PTP. We report herein the first chemical
synthesis of PTP from the corresponding pyrazofurin 5′-
monophosphate, which was efficiently prepared from
Figure 1. Structures of pyrazofurin and its analogs
isopropylidene-protected
pyrazofurin
via
selective
phosphorylation without enol protection.
In cellular metabolism, both pyrazofurin and ribavirin are
converted to their monophosphates and triphosphates.6-7 Early
research on the mechanism of action of the two antiviral agents
showed that pyrazofurin 5′-monophosphate (PMP) and ribavirin
5′-monophosphate (RMP) inhibit the orotidine 5′-monophosphate
(OMP) decarboxylase8 and inosine 5′-monophosphate (IMP)
dehydrogenase,9 respectively, and thus block the de novo
pyrimidine and purine biosynthesis. In 2000, with the assistance
of synthetic ribavirin 5′-triphosphate (RTP), Cameron and his
coworker were able to investigate the interactions of RTP with
viral RNA polymerases on the basis of a novel primer-extension
assay and found that non-specific incorporation of ribavirin into
viral genome leads to lethal mutagenesis of virus population.10-12
Similarly, in the mechanistic investigation of T-705, a pyrazine
nucleoside drug (favipiravir), synthetic T-705 5′-triphosphate (T-
705TP) helped Furuta and coworkers reveal that the drug’s
antiviral activity comes from the inhibition of influenza virus
RNA polymerase by T-705TP.13 In contrast, due to the unique
enol-containing nucleobase in pyrazofurin, the chemical
Due to the poor commercial availability of pyrazofurin, we
first synthesized the isopropylidene-protected pyrazofurin (10)
according to a known synthetic route.18-20 The detailed synthetic
methods of certain steps were optimized to improve the
practicability (Scheme 1). First, D-ribose (1) was treated with
catalytic amount of p-TsOH in 2,2-dimethoxypropane and
acetone. But the yield of 2,3-O-isopropylidene-D-ribose (2) was
only around 70%.21 Instead, the application of 5 mol‰ Hf(OTf)4
as the catalyst afforded 2 in 93% yield within 30 min.22 Direct
tritylation of the 5-OH of 2 with TrCl in refluxing pyridine

2
Tetrahedron
Scheme 1. Optimized synthesis of isopropylidene-protected pyrazofurin (10)
was sluggish (45% yield, 24 h).21 We found that addition of 4
isopropylidene-protected PMP in very low yield (<10%), and the
contaminated inorganic phosphate was hard to separate. The less
reactive dibenzylphosphoryl chloride27 was highly prone to
hydrolysis and failed to generated any phosphorylated product in
this case.
equiv of DMAP significantly shortened the reaction time to 4 h
and increase the yield of 3 to 85%.23 In the Wittig reaction, 3 and
phosphorane 4 were initially refluxed in acetonitrile as described
in literature.18 However, the reaction was extremely slow and C-
nucleoside 5 was obtained in only 60% yield after 7 days. When
the reaction of 3 and 4 was conducted in benzene,19 catalytic
amount of benzoic acid (5 mol%) resulted in moderate
improvement (66% yield, 5 d). Therefore, we gradually increased
the amount of benzoic acid from 5 mol% to 50 mol%, and found
that 20 mol% acid catalyst resulted in the best outcome (82%
yield, 2 d). The α/β ratio was determined as 1.4:1. Then, the
Then, we switched to more reactive phosphoramidite reagents.
It was anticipated that if excess P(III) reagent was used, both 5′-
OH and enol should be phosphitylated. However, the phosphite
triester on enol should be much less stable than the one at 5′-O
position. Therefore, upon in situ oxidation with aqueous H2O2,
the phosphite triester might be cleaved by hydrolysis to afford
5′-OH selectively phosphorylated product. In our experiments,
the hindered di-tert-butyl N,N-diisopropylphosphoramidite28 was
not reactive at all. In contrast, excess dibenzyl N,N-
diisopropylphosphoramidite29-30 reacted with both 5′-OH and
enol as we speculated. In the 31P NMR spectrum of the
diphosphite intermediate (13), two peaks of equal intensity
showed up at 139.4 ppm and 137.5 ppm corresponding to the two
phosphite triesters at 5′-O and 4-O positions. After H2O2 was
added, the major product was isolated in 68% yield and
characterized as the 5′-O-dibenzyl phosphate (14), indicating that
the phosphite triester on enol was hydrolytically removed upon
addition of H2O2. Sequential removal of benzyl esters and
isopropylidene afforded the desired PMP (15) in excellent yield
(Scheme 2). To our surprise, when the phosphate
diazotization of
5 with tosyl azide/Et3N and subsequent
cyclization of 6 with NaH in DME were conducted strictly
according to literature procedures which afforded the diazo
compound 6 and anomeric mixtures of pyrazole 7 and 8 in
reproducible yields. 7 and 8 were separated by column
chromatography and obtained in 38% and 17% yields,
respectively. In order to reduce the risk of potential explosion
resulting from the aminolysis of 7 with NH3/MeOH in a sealed
tube at 90–95 ºC, concentrated ammonia was used instead in our
method. The aminolysis was performed at 80 ºC for 12 h and
gave the amide 9 in comparable yield. Finally, the trityl was
cleaved with 1% TFA in CH2Cl2 to afford 10 in 80% yield.
Alternatively, treatment of 9 with TFA/H2O (1:1) at room
temperature efficiently yielded equal amounts of pyrazofurin (11)
and its epimer, pyrazofurin B (12). This result was in agreement
with previous reports that pyrazofurin is prone to epimerize in
acidic aqueous solution.1,24
To apply the P(V)-N activation strategy to the synthesis of
PTP, efficient preparation of PMP is quite essential. In a previous
report, Revankar et al. prepared PMP from 4-O-benzyl
pyrazofurin with POCl3. While the benzylation/ debenzylation of
enol took extra steps, phosphorylation with POCl3 was low-
yielding. In addition, a pressurized hydrogenator was required for
debenzylation of the 4-O-benzyl ether.25 Though direct
phosphorylation of pyrazofurin with POCl3 has been reported,
however, the very low yield and tedious purification procedures
prohibited its application in our synthesis.26 Therefore, on the
basis of the different reactivity of 5′-OH and enol, we attempted
to selectively phosphorylate 5′-OH of 10 without enol protection.
A series of P(V) and P(III) reagents were tested. Our first trial
with POCl3 as the phosphorylating reagent yielded the desired
Scheme 2. A selective phosphorylation method of 10 for synthesis of PMP (15).

intermediate was treated with TFA/H2O (1:1) at room
temperature, only ~10% of 15 epimerized to α-anomer as seen on
1H NMR. Further decreasing the deprotection temperature to 0
ºC effectively inhibited the epimerization of 15.
upon extended stay in aqueous solution for several days, ~10%
epimerized pyrazofurin B 5′-triphosphate was observed by H
NMR.
1
In the following research, 15 was converted to the
corresponding phosphoropiperidate (16) according to the redox
condensation method described in a previous paper15 (Scheme 3).
The treatment of 15 with 2,2-dithiodi-aniline/PPh3 and
piperidine in DMSO afforded 16 smoothly over 6 h. Addition of
acetone solution of NaI precipitated 16 as sodium salt. Our
attempt to purify 16 with LH-20 size-exclusion chromatography
caused partial decomposition. Therefore, 16 was directly
transformed into triethyl-ammonium salt form by passing
through ion exchange resin and used in the next step without
further purification (86% yield, purity >95%).
Figure 3. HPLC (A) and HRMS (B) analysis of synthetic PTP (17).
In summary, for the first time, pyrazofurin 5′-triphosphate (17)
was synthesized via the P(V)-N activation strategy, which
exhibited excellent coupling efficacy and tolerance of acidic enol
functional group on nucleobase. Moreover, certain procedures for
the synthesis of isoproylidene-protected pyrazofurin (10) in the
Wittig reaction-based route were optimized and a selective
phosphorylation method for the preparation of pyrazofurin 5′-
monophosphate (15) was also developed. The current research
will be useful for virological and medicinal investigations of the
role of pyrazofurin 5′-triphosphate (17), an important cellular
metabolite of pyrazofurin, for the marked antiviral activity of
pyrazofurin, which may be helpful for the discovery of novel
antiviral strategies.
Scheme 3. The redox condensation method for the synthesis of pyrazofurin 5′-
phosphoropiperidate (16).
According to the P(V)-N activation method,14 16 was treated
with 2 equiv of pyrophosphate in the presence of 6 equiv of 4,5-
dicyanoimidazole (DCI) as the activator. 31P NMR tracing of the
reaction showed that the coupling of pyrophosphate with 16
proceeded efficiently and was not affected by the enol on the
nucleobase (Figure 2). Upon completion, 31P NMR spectrum of
the precipitated crude triphosphate as sodium salt form (Figure 2
inset) indicated that conversion ratio from 16 to triphosphate 17
was over 90%.
Acknowledgments
We thank the National Natural Science Foundation of China
(21562021), Natural Science Foundation (20143ACB21014),
Fellowship for Young Scientists (2015BCB23009), and Sci &
Tech Project from Dept of Education (GJJ160763) of Jiangxi
Province, and Innovation fund from JXSTNU (YC2017-X27) for
financial support.
References and notes
1.
2.
Gutowski GE, Sweeney MJ, DeLong DC, et al. Ann N Y Acad Sci.
1975;255:544551.
Witkowski JT, Robins RK, Sidwell RW, et al.
J Med Chem.
1972;15:11501154.
3.
4.
5.
6.
Mizuno K, Tsujino M, Takada M, et al. J Antibiot. 1974;27:775782.
De Clercq E. Med Res Rev. 2010;29:611645.
De Clercq E. J Med Chem. 2016;59:2301–2311.
Dix DE, Lehman CP, Jakubowski A, et al. Cancer Res. 1979;39:4485
4490.
Figure 2. 31P NMR tracing of the P(V)-N activation strategy for the synthesis of
PTP (17).
7.
8.
9.
Zimmerman TP, Deeprose RD. Biochem Pharmacol. 1978;27:709716.
Charette M, Gray MW. IUBMB Life. 2000;49:341351.
Muller WE, Maidhot A, taschner H, et al. Biochem Pharmacol.
1977;26:10711075.
The crude product was purified by ion exchange
chromatography. Passage through an ion exchange resin (Na+
form) and lyopholization afforded the target PTP 17 in 65% yield
as a white solid. The HPLC analysis of 17 (for detailed
conditions, see SI) determined that the triphosphate prepared by
this method was of high purity (>98%). On the high resolution
mass spectrum of 17 (ESI–), both the molecular ion peak of 17
and fragment peaks corresponding to pyrazofurin 5′-diphosphate
and 5′-monophosphate were observed. It is worth to mention that
10. Arnold JJ, Cameron CE. J Bio Chem. 2000;275:53295336.
11. Graci JD, Cameron CE. Virology. 2002;298:175182.
12. Crotty S, Cameron CE, Andino R. PNAS. 2001;98:68956900.
13. Furuta Y, Takahashi K, Kuno-Maekawa M, et al. Antimicrob Agents
Chemother. 2005;49:981986.
14. Sun Q, Gong SS, Sun J, et al. J Org Chem. 2013;78:84178426.
15. Sun Q, Gong SS, Sun J, et al. Tetrahedron Lett. 2014;55:21142118.

4
Tetrahedron
16. Sun Q, Li XJ, Sun J, et al. Tetrahedron. 2014;70:294300.
To create your abstract, type over the instructions in the
17. Sun Q, Gong SS, Liu S, et al. Tetrahedron. 2014;70:45004506.
18. Karagiri N, Takashima K, Haneda T, et al. J Chem Soc Perkin Trans 1.
1984;553560.
template box below.
19. Herrera FJL, Baelo CU. Carbohydr Res. 1985;143:161174.
20. Chen X, Schneller SW. J Med Chem. 1993;36:37273730.
21. Fan GT, Pan YS, Lu KC, et al. Tetrahedron. 2005;61:18551862.
22. Unpublished results.
23. Cho JH, Bernard DL, Sidwell RW, et al. J Med Chem. 2006;49:1140
1148.
24. De Bernardo S, Weigele M. J Org Chem. 1976;41:287290.
25. Petrie III CR, Revankar GR, Dalley NK, et al.
J Med Chem.
1986;29:268278.
26. Meza-Avina ME, Wei LH, Liu Y, et al. Bioorg Med Chem.
2010;18:40324041.
27. Hall RH, Khorana HG. J Am Chem Soc. 1954;76:50565060.
28. Talukdar A, Morgunova E, Duan JX, et al. Bioorg Med Chem.
2010;18:35183534.
29. Korboukh I, Hull-Ryde EA, Rittiner JE, et al.
J Med Chem.
2012;55:64676477;
30. Lambrecht MJ, Brichacek M, Barkauskaite E, et al. J Am Chem Soc.
2015;137:35583564.
Supplementary Material
Supplementary
data
(experimental
procedures
and
characterization data) associated with this article can be found at
 Pyrazofurin triphosphate was first synthesized
via P(V)-N activation strategy.
 A selective phosphorylation method was
developed for pyrazofurin 5′-phosphate.
 Certain synthetic procedures for isoproylidene-
protected pyrazofurin were modified.
Graphical Abstract
The first chemical synthesis of pyrazofurin
Leave this area blank for abstract info.
5′-triphosphate
Hua-Shan Huang, Rui Wang, Wei-Jie Chen, Ji-Zong Chen, Shan-Shan Gong and Qi Sun