Phosphorus deprotonation by a non-nucleophilic base is critical for the synthesis of nucleoside α-P-boranodiphosphates via a phosphoramidite approach

Article abstract of DOI:10.1016/j.tetlet.2014.08.047

Here we report the synthesis of nucleoside α-P-boranodiphosphate analogs via our improved phosphoramidite approach, where the addition of a non-nucleophilic base, TEA, to deprotonate phosphorus is considered to be critical to improve final product yields. Protonation on trivalent phosphorus might not only have blocked Lewis acid borane modification, but also have promoted byproduct formation through a trans-nucleophile process. Here, based on ³¹P NMR data, mechanisms are proposed for P(V)P(III) bond cleavage via phosphorus protonation-promoted nucleophile-exchange, and TEA deprotonation of phosphorus in promoting product formation in anhydrous synthetic reaction mixtures.

Full text of DOI:10.1016/j.tetlet.2014.08.047

Tetrahedron Letters 55 (2014) 5605–5608
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Phosphorus deprotonation by a non-nucleophilic base is critical for
the synthesis of nucleoside
phosphoramidite approach
a-P-boranodiphosphates via a
⇑
⇑
Zhihong Xu , Barbara Ramsay Shaw
Department of Chemistry, Box 90346, Duke University, Durham, NC 27708, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Here we report the synthesis of nucleoside a-P-boranodiphosphate analogs via our improved phosphor-
Received 14 May 2014
Revised 11 August 2014
Accepted 13 August 2014
Available online 22 August 2014
amidite approach, where the addition of a non-nucleophilic base, TEA, to deprotonate phosphorus is
considered to be critical to improve final product yields. Protonation on trivalent phosphorus might not
only have blocked Lewis acid borane modification, but also have promoted byproduct formation through
a trans-nucleophile process. Here, based on 31P NMR data, mechanisms are proposed for P(V)AP(III) bond
cleavage via phosphorus protonation-promoted nucleophile-exchange, and TEA deprotonation of
phosphorus in promoting product formation in anhydrous synthetic reaction mixtures.
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
P(V)AP(III)-cleavage
Nucleophile-exchange
Phosphorus-deprotonation
Nucleoside boranodiphosphate
Borane phosphate
31P NMR-monitored synthesis
Many ribo- and deoxyribonucleoside diphosphate (NDP and
dNDP) analogs are essential in biological systems. In addition to
their phosphorylation to triphosphates (d)NTPs, which play key
roles in DNA replication and transcription, the (d)NDPs also have
important, wide-ranging roles in fundamental cellular metabolic
pathways such as second messengers, energy donors, substrates,
and cofactors for various enzymes. Chemical synthesis of (d)NDP
analogs has been an active area of phosphate chemistry. The
monophosphate on an activated nucleoside pentavalent mono-
phosphate.6 The availability of new NDP analogs and the develop-
ment of efficient synthetic approaches to obtain such compounds
remain challenging. Pyrophosphate bond formation is usually a
bottleneck key step. To obtain (d)NDPaB, several chemical syn-
thetic approaches were initiated by our group. In the first reported
phosphoramidate approach using base-unprotected nucleosides as
starting materials to synthesize ADPaB and GDPa aB-
B,7a and ADP
a-P-boranodiphosphate ((d)NDPaB) 5, wherein the
a-phosphate
glucose,7b ꢀ7 h were needed to obtain the monophosphoramidate
intermediates, and the diphosphate bond formation involved heat.
In the second reported method,8a,b the inconvenient distillation of
the reagent 2-chloro-1,3,2-oxathiaphospholane and the release of
toxic ethylene episulfide in Stec’s oxathiaphospholane approach
detracted from practical application of the method. Our report on
is a borane derivative of the natural functional group, provides a
unique class of diphosphate analogs that are isoelectronic and
isolobal with normal phosphate esters.1 Upon activation of
a-P-
borane-substituted dideoxynucleoside diphosphate (ddNDP
a
B) to
its corresponding triphosphate (ddNTPaB), a viral chain could be
better terminated upon incorporation of the Rp isomer.2 Further,
the synthesis of a
-P-modified NDPs with ethylenediamine9 pro-
nucleoside diphosphate kinase (NDK) showed improved phosphor-
vided a convenient protocol with higher yields; however, a number
of impurities including the corresponding boranotriphosphates
were observed in final reaction mixtures.
ylation for the Rp isomer of ddNDPa
Bs.3 The presence of an
a-P-borano group resulted in a 10-fold enhancement of the cata-
lytic efficiency of the reaction relative to AZTDP or d4TDP.3b
NDP analogs are typically accessed in a number of ways,4
mainly including nucleoside phosphorylation by pyrophosphate
in the presence of an activator,5 or attack by a (modified)
To overcome these challenges, we reinvestigated the phosphor-
amidite approach10 that has been used extensively in oligonucleo-
tide synthesis where 1H-tetrazole, an acid and also a nucleophile,
is one of the most used activators in the coupling reaction. Our ini-
tial attempts to obtain NDPa
B via that procedure10 by simply
replacing the reported nucleophiles with an inorganic phosphate
and using boranation reagent borane dimethyl sulfur complex
(Me2S:BH3) instead of oxidizing reagents were unsuccessful. Later,
⇑
Corresponding authors. Fax: +1 919 660 1605.
E-mail addresses: zhihong.xu@duke.edu (Z. Xu), barbara.r.shaw@duke.edu
(B.R. Shaw).
http://dx.doi.org/10.1016/j.tetlet.2014.08.047
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

5606
Z. Xu, B.R. Shaw / Tetrahedron Letters 55 (2014) 5605–5608
we modified the approach and synthesized the modified pyrophos-
phate bond via a phosphate-phosphite intermediate,11 where both
tetrazole and 4,5-dicyanoimidazole were used as activators to
promote the coupling of an inorganic tetrabutylammonium dihy-
drogen phosphate (TBADP) with an activated nucleoside phosphor-
amidite. Compared to the recent published coupling procedure of a
glycosyl natural P@O phosphate with a nucleoside phosphorami-
dite by Gold et al.,12 the procedures were quite similar in most
steps, but the order of activator and reagent addition, and reaction
time length differed. In addition, our experiments showed that the
ammonium hydroxide (concd NH4OH) produced title compounds
5 in good overall yields. After work-up and purification, the iso-
lated yields (21–40%) were determined by UV. The addition of
TEA in the boranation step was critical. Without TEA, a large quan-
tity of monophosphate byproduct formed, and we propose that this
byproduct resulted from the a-phosphorus protonation of 3 based
on our 31P NMR data of reaction mixtures.
By monitoring the intermediate and product formations with
31P NMR at 162 MHz, the extent of each reaction step could be esti-
mated according to the percentages of phosphorus peak areas.
Route A from 1 to its phosphoramidite 2 is fairly standard and for-
mation of phosphate-phosphite intermediate 3 is straightforward
(ꢀ100% yield based on 31P NMR). While tetrazole was very effec-
tive in catalyzing the cross-coupling to form 3, oxidation may
successful attachment of a borano group on the
a-P was not
achieved if the oxidizer tert-butyl hydroperoxide in Gold’s report
was simply replaced by Me2S:BH3.
Here we report the improved phosphoramidite approach for the
synthesis of a boranopyrophosphate bond in detail. This is a one-
pot strategy from protected nucleoside 1 to final product 5 through
occur on the
a-P of 3, presumably by O2 in the solvent; thus a
20 min limit for the step from 2 to 3 and the use of freshly opened
solvent are recommended. Direct boranation of 3 with ꢀ6 equiv
Me2S:BH3 was initially tried. After stirring for 15–20 min, the
a P(V)AP(III)-intermediate 3 to compound 4 resulting in (d)NDPaB
5, as shown in Scheme 1. Our goal was to use protected normal
nucleosides (a deoxyribonucleoside, i.e., thymidine (T), a ribonu-
cleoside, i.e., uridine (U), and a deoxyribonucleoside that needs
base protection, i.e., deoxyadenosine (dA)) as model starting mate-
rials to develop a synthetic approach applicable to the synthesis of
clinically useful electrophile or Lewis acid (such as borane) modi-
fied pyrophosphate analogues. One equivalent 2-cyanoethyl N,N-
diisopropylchlorophosphoramidite and triethylamine (TEA) were
added into a septum-sealed round bottom flask with protected
nucleoside 1 in anhydrous acetonitrile. After stirring for 10 min
at room temperature, the formation of phosphoramidite diastereo-
mers 2 was completed. TBADP (2 equiv) in acetonitrile was then
added to form the nucleoside phosphite-phosphate diastereomers
3 in the presence of tetrazole (2 equiv). This coupling step usually
took 15–20 min to complete. Then ꢀ6 equiv Me2S:BH3 (with
ꢀ4 equiv TEA) were used to introduce the borane group onto the
P(III) transformation of
a-P of 3 (d 131.5–133 ppm) to P(IV) (a-P
of 4) with a broad signal at around d 106–108 ppm indicated the
completion of the boranation step. However, in addition to the
phosphorus peaks of 4, a large quantity of a broad peak with dp
at around 117 ppm was also observed (Route A), indicating the
formation of a byproduct from P(V)AP(III) bond cleavage of 3.
Notably, ꢀ4 equiv non-nucleophilic base TEA addition effectively
reduced the formation of the byproduct (dp 117–121 ppm). Depro-
tection of 4 using DBU afforded b-elimination to give the diphos-
phate analog with a pentavalent borano-
a
-P (dp ꢀ79 ppm, br,
Fig. 1) within several minutes; the addition of non-nucleophilic
base TEA in the boranation step was critical to improve our product
yields.
For comparison, compound 6 (dp ꢀ118 ppm) was synthesized
via Route B of Scheme 1 by direct boranation of 2, and the 31P
NMR signal of 6 matched the byproduct signal at dp 117–
121 ppm in the reaction mixture of 4. However, upon adding tetra-
zole followed by TBADP into the reaction mixture of 6, diphosphate
bond formation was not observed by 31P NMR. As expected, base
treatment by DBU converted the byproduct peak in the reaction
mixture of 4 (base: T) to a broader peak at d 92–94 ppm (Fig. 1,
and Supporting information (SI) Fig. S-1), which was in agreement
with the report that concd NH4OH treatment of 6 gave monophos-
phoramidate 7 (dp ꢀ93 ppm, see structure in Scheme 2).13
a-P of 3 to form 4. Deprotection of 4 with DBU and concentrated
CN
O
CN
O
P
O
P
Base
Base
R2
, TEA
Cl
N
HO
O
N
O
R1
Route A
R1
2
R2
1
Route B
Me2S:BH3
Tetrazole, TBADP
CN
CN
O
P
O
P
O
T
N
O
Base
O
P
O
HO
O
O
BH3
OH
6
AcO
H
R2
R1
3
Tetrazole, TBADP
TEA, Me2S:BH3
CN
O
P
O
O
O
P
1. DBU
2. NH4OH
Base
Base
R3
-
-
O
P
O
O
O
P
O
O
-
O
O
OH
OH
BH3
BH3
R1
R2
4
5
HO
O
P
OH
O
P
BH3
O
P
OH
O
P
BH3
-
-
Base
Base
O
O
O
O
O
O
-
O
O
-
5a
Rp isomer
5b
Sp isomer
HO
R3
HO
R3
Bz
1
5
1
4
5
R1=OAc; R2=OAc(H); R3=OH(H); Base: T, U in to ; A in to , A in
Scheme 1. One-pot synthesis of (d)NDP
aB 5 by an improved phosphoramidite
Figure 1. 31P NMR monitoring (in D2O) of DBU treatment of reaction mixture of 4
approach.
(base: T).

Z. Xu, B.R. Shaw / Tetrahedron Letters 55 (2014) 5605–5608
5607
Table 1
CN
O
P
Isolated yields and LC–MS profile of (d)NDPaBs
O
Base
(d)NDPaB
Overall isolated yielda
HPLCb tR (min) [area %]
HO
P
O
O
O
OH
CH3CN (%)
Rp isomer
Sp isomer
OAc(H)
3
AcO
dADP
TDP
UDP
a
B
B
21
39
40
2.5
2
1
4.9 [ꢀ41]
3.9 [ꢀ48]
6.0 [ꢀ46]
5.9 [ꢀ59]
5.2 [ꢀ52]
7.2 [ꢀ54]
a
Tetrazole/H+
CN
aB
a
Yield calculated in percentage by UV.
The Rp isomer elutes before Sp isomer. Flow rate: 40
condition of 10 mM TEAA (pH 7) and acetonitrile on
150 Â 1.0 mm column. tR: retention time.
O
b
O
lL/min, with isocratic
Base
+
a Polaris 3 C-18 A
HO
P
O
P
H
O
O
OH
8
OAc(H)
AcO
TEA
Me2S:BH3
isomer 5b) of each NDPaB 5 were separated by RP-HPLC, and their
i-Pr2NH
LC profiles and NMR were compared with the same compounds
produced by the ethylenediamine method we reported earlier.9
The overall isolated yields and LC–MS profiles of the individual
(d)NDPaB (i.e., dADPaB, TDPaB, and UDPaB) Rp and Sp isomers,
synthesized by our improved phosphoramidite approach, are listed
CN
O
CN
No Boranation
O
P
O
O
i-Pr
i-Pr
Base
Base
-
O
P
O
O
N
P O
O
OH
OAc(H)
2
OAc(H)
in Table 1. Formation of the Sp isomer is slightly favored. The syn-
AcO
AcO
thesis of NDPaB in our phosphoramidite approach can be finished
in one pot in much less time than other methods reported. The
Tetrazole
TBADP
Me2S:BH3
Me2S:BH3
final reaction mixture was relatively clean (SI Fig. S-2(1)) and
CN
3
CN
O
(d)NDPaBs were easily isolated.
O
P
O
O
i-Pr
i-Pr
Base
-
Base
To summarize, phosphorus protonation of
3
promotes
P
O
O
O
O
N
P O
P(V)AP(III) bond cleavage via a nucleophile-exchange in anhydrous
reaction mixtures. This Letter is the first report that boranodi-
phosphates could be effectively synthesized via the nucleophilic
attack of an inorganic phosphate on activated trivalent phosphorus
followed by boranation, wherein phosphorus deprotonation by a
non-nucleophilic base is a key step involved. This approach might
be valuable for the synthesis of other electrophile- or Lewis acid-
modified pyrophosphate bonds.
OH
BH3
BH3
OAc(H)
4
AcO
OAc(H)
AcO
Byproduct 6
1. DBU
2. NH4OH
1. DBU
2. NH4OH
O
O
O
P
i-Pr
i-Pr
Base
Base
-
O
P
O
P
O
-
N
O
-
O
O
OH
BH3
BH3
OH(H)
HO
Product 5
OH(H)
HO
Byproduct 7
Acknowledgments
This work was supported by NIH Grants to B.R.S. (R01-AI-52061
and 2R01-GM-57693), and Duke University. Z. Xu designed and
conducted the research, and analyzed the experimental data.
Acknowledgement should be given to Duke University NMR Center
and Chemistry Department for NMR and LC–MS. We are grateful to
Drs. George R. Dubay and Anthony A. Ribeiro for their training on
LC–MS and NMR experiments.
Scheme 2. Proposed mechanisms of (1) phosphorus protonation-promoted
P(V)AP(III) bond cleavage, via nucleophile-exchange, and (2) TEA deprotonation
of phosphorus in promoting product formation.
Based on our 31P NMR data, we propose that
a-phosphorus
protonation of 3 (as shown in Scheme 2, structure 8) promotes
nucleophile-exchange, which results in P(V)AP(III) bond cleavage
to regenerate 2 in an equilibration mixture, and the generation of
2 is pushed to a maximum by the subsequent boranation step to
give 6 (Scheme 2). Similar to this trans-nucleophile process,
transesterifications14 involving trialkyl phosphites have been
reported to be catalyzed by p-toluenesulfonic acid, phosphoric acid,
diethyl hydrophosphonate, aluminum chloride, tetrazole/tetrazolide,
Supplementary data
Supplementary data (detailed procedures for synthesis and iso-
lation of (d)NDPaB, with isomer structures and spectroscopic data)
associated with this article can be found, in the online version, at
etc. Further, we propose that once the a-P of 3 is protonated, the bor-
http://dx.doi.org/10.1016/j.tetlet.2014.08.047.
anation is hindered; instead, nucleophilic attack by diisopropyl
amine (i-Pr2NH, generated from step 2 to 3 in Scheme 1) reproduces
2, and the boranation of 2 in the reaction mixture generates
byproduct 6 with the broad peak at d 117–121 ppm. Subsequent base
treatment converts 6 to byproduct 7 in the reaction mixture. By
contrast, non-nucleophilic base addition in the boranation step
effectively reduced the formation of monophosphate byproducts in
our experiments; and this procedure might be useful to any acid
catalyzed diphosphate bond synthesis where an electrophile or a
Lewis acid modification of a P(III) is involved. In addition, the success-
ful synthesis of final product 5 is also dependent on the molar equiv-
alents of activator and reagents; the use of 2 equiv each of tetrazole
and TBADP in the step from 2 to 3 is recommended (SI Fig. S-2).
In Route A, concd NH4OH was finally used for the deprotection
of hydroxyl group(s) of the sugar ring and amino group on the
adenosine base ring. Two stereoisomers (Rp isomer 5a, and Sp
References and notes
1. (a) Sood, A.; Shaw, B. R.; Spielvogel, B. F. J. Am. Chem. Soc. 1990, 112, 9000–9001;
(b) Shaw, B. R.; Dobrikov, M.; Wang, X.; Wan, J.; He, K.; Lin, J. L.; Li, P.; Rait, V.;
Sergueeva, Z. A.; Sergueev, D. Ann. N.Y. Acad. Sci. 2003, 1002, 12–29; (c) Li, P.;
Sergueeva, Z. A.; Dobrikov, M.; Shaw, B. R. Chem. Rev. 2007, 107, 4746–4796.
2. (a) Selmi, B.; Boretto, J.; Sarfati, S. R.; Guerreiro, C.; Canard, B. J. Biol. Chem. 2001,
276, 48466–48472; (b) Dobrikov, M. I.; Sergueeva, Z. A.; Shaw, B. R. Nucleosides
Nucleotides Nucl. 2003, 22, 1651–1655.
3. (a) Meyer, P.; Schneider, B.; Sarfati, S.; Deville-Bonne, D.; Guerreiro, C.; Boretto,
J.; Janin, J.; Veron, M.; Canard, B. EMBO 2000, 19, 3520–3529; (b) Schneider, B.;
Meyer, P.; Sarfati, S.; Mulard, L.; Guerreiro, C.; Boretto, J.; Janin, J.; Veron, M.;
Deville-Bonne, D.; Canard, B. Nucleosides Nucleotides Nucl. 2001, 20, 297–306.
4. Johnson, D. C.; Widlanski, T. S. In Curr. Protoc. Nucleic Acid Chem; Beaucage, S.
L. et al., Eds.; John Wiley: New York, 2004; Chapter 13.
5. (a) Davisson, V. J.; Davis, D. R.; Dixit, V. M.; Poulter, C. D. J. Org. Chem. 1987, 52,
1794–1801; (b) Vincent, S.; Mioskowski, C.; Lebeau, L. Tetrahedron Lett. 1998,
39, 2321–2324; (c) Rezende, C. M.; Vallejos, G.; Osorio-Olivares, M.; Sepulveda-
Boza, S. Synth. Commun. 2001, 31, 3699–3705.

5608
Z. Xu, B.R. Shaw / Tetrahedron Letters 55 (2014) 5605–5608
6. (a) Hoard, D. E.; Ott, D. G. J. Am. Chem. Soc. 1965, 87, 1785–1788; (b) Shimazu,
M.; Shinozuka, K.; Sawai, H. Tetrahedron Lett. 1990, 31, 235–238; (c) Goody, R.
S.; Eckstein, F. J. Am. Chem. Soc. 1971, 93, 6252–6257; (d) Wendicke, S.;
Warnecke, S.; Meier, C. Angew. Chem. Int. Ed. 2008, 47, 1500–1502.
7. (a) Lin, J.; He, K.; Shaw, B. R. Helv. Chim. Acta 2000, 83, 1392–1397; (b) Lin, J.;
Shaw, B. R. Tetrahedron Lett. 2000, 41, 6701–6704.
8. (a) Okruszek, A.; Sierzchala, A.; Zmudzka, K.; Stec, W. J. Nucleosides Nucleotides
Nucl. 2001, 20, 1843–1849; (b) Li, P.; Shaw, B. R. J. Org. Chem. 2004, 69, 7051–7057.
9. Li, P.; Xu, Z.; Liu, H.; Wennefors, C. K.; Dobrikov, M. I.; Ludwig, J.; Shaw, B. R. J.
Am. Chem. Soc. 2005, 127, 16782–16783.
11. (a) Xu, Z.; Shaw, B. R. 56th SRMACS, RTP, NC, USA, Nov 10, 2004, GEN-146.; (b)
Xu, Z.; Shaw, B. R. 230th ACS Meeting, Washington, DC, USA, Aug 28, 2005,
MEDI-157.; (c) Xu, Z.; Shaw, B. R. 231st ACS Meeting, Atlanta, GA, USA, March
26, 2006, MEDI-341.; (d) Xu, Z. Ph.D. Dissertation, Duke University, 2007,
Chapter 3.
12. Gold, H.; van Delft, P.; Meeuwenoord, N.; Codee, J. D. C.; Filippov, D. V.; Eggink,
G.; Overkleeft, H. S.; van der Marel, G. A. J. Org. Chem. 2008, 73, 9458–
9460.
13. Tomasz, J.; Shaw, B. R.; Porter, K.; Spielvogel, B. F.; Sood, A. Angew. Chem., Int.
Ed. Engl. 1992, 31, 1373–1375.
10. (a) Caruthers, M. H.; Barone, A. D.; Beaucage, S. L.; Dodds, D. R.; Fisher, E. F.;
McBride, L. J.; Matteucci, M.; Stabinsky, Z.; Tang, J. Y. Methods Enzymol. 1987,
154, 287–313; (b) Jin, Y.; Just, G. Tetrahedron Lett. 1998, 39, 6429–6432.
14. (a) Hoffmann, F. W.; Ess, R. J.; Usinger, J. R. P. J. Am. Chem. Soc. 1956, 78, 5817–
5821; (b) Nitta, Y.; Arakawa, Y.; Ueyama, N. Chem. Pharm. Bull. 1986, 34, 2710–
2718; (c) Watanabe, Y.; Maehara, S.-I. Heterocycles 2000, 52, 799–810.