Base-dependent regioselective and P-stereocontrolled hydrolysis of nucleoside 3'-O-(O-2,4,6-trimethylbenzoyl methanephosphonothioate)s

Full text of DOI:10.1021/jo981304v

J . Org. Chem. 1998, 63, 9109-9112
9109
Sch em e 1
Ba se-Dep en d en t Regioselective a n d
P -Ster eocon tr olled Hyd r olysis of
Nu cleosid e 3′-O-(O-2,4,6-Tr im eth ylben zoyl
Meth a n ep h osp h on oth ioa te)s
Lucyna A.Woz´niak, Arkadiusz Chworos´,
J arosław Pyzowski, and Wojciech J . Stec*
Polish Academy of Sciences, Centre of Molecular and
Macromolecular Studies, Department of Bioorganic
Chemistry, Sienkiewicza 112, 90-363 Ło´dz´, Poland
Received J uly 6, 1998
An early observation that oligonucleotides containing
methanephosphonate linkages of (RP) configuration at
internucleotide positions possess higher affinity toward
complementary DNA or RNA1 facilitated studies on the
synthesis of the chimeric oligonucleotide constructs with
incorporated (RP)-dinucleoside methanephosphonates (1).2
Therefore, an approach to the stereoselective synthesis
of 13 appeared as the prerequisite for construction of the
second generation of antisense molecules.4 Recently, we
have described a novel approach to the stereoselective
synthesis of dinucleoside (3′,5′)-methanephosphonates (1)
based upon the synthesis of the appropriately protected
nucleoside 3′-O-methanephosphonoanilidothioates, their
separation into diastereomeric species, and sequential
stereospecific conversion of each single [(RP) or (SP)]
diastereomeric methanephosphonoanilidothioate into (RP)-
5′-O-DMT-(N-protected)-nucleoside 3′-O-(S-alkyl metha-
nephosphonothiolate) (3), followed by substitution of the
thioalkyl group in (RP)-3 with the 5′-OH group of N-
protected 3′-O-acetyl (or tert-butyldimethylsilyl) nucleo-
side (2).5 To the best of our knowledge, this has been
the first method allowing for the synthesis of one required
diastereomer of dinucleoside (3′,5′)-methanephosphonates
from either of the separated diastereomeric precursors.
Chemoselective S-alkylation of (RP)-6 results in the
formation of (RP)-5′-O-DMT-(N-protected, except for thym-
ine) nucleoside 3′-O-(S-alkyl methanephosphonothiolate)s
(3) which, as depicted in Scheme 1, are used as the
substrates for synthesis of the desired (RP)-1. The
opposite diastereomers, (SP)-6, are treated with sterically
hindered 2,4,6-trimethylbenzoyl chloride. The process of
acylation is completely regioselective (exclusive O-ben-
zoylation) and occurs with retention of configuration at
the phosphorus atom, providing in high yield of (RP)-5′-
O-DMT-(N-substituted, except thymine) nucleoside 3′-O-
(2,4,6-trimethylbenzoyl methanephosphonothioate)s (7).
The hydrolysis of the mixed anhydrides (RP)-7 in the
presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) oc-
curs smoothly within 30-40 min giving (RP)-methane-
phosphonothioates 6 in high (over 90%) preparative
yields.9 This process is fully stereospecific and occurs
with inversion of configuration at the phosphorus center.10
Therefore, the benzoylation of (SP)-6 followed by the
DBU-assisted hydrolysis of the resulting (RP)-7 allows
for the stereoconversion of (SP)-6 into (RP)-6 and the use
of both separated diastereomers 6 for the stereoselective
synthesis of slow-(RP)-311 (Data are collected in Table 1).
In this paper we describe an alternative route leading
exclusively to (RP)-3. A corollary of this accomplishment
is the simple synthesis of suitably protected nucleoside
3′-O-(O-alkyl methanephosphonothioate)s (5) (R ) -CH3,
or preferably -CH2CH2CN). After their separation into
diastereomers and subsequent O-dealkylation,6 ammo-
nium salts of diastereomerically pure (RP)- and (SP)-5′-
O-DMT-nucleoside methanephosphonothioates (6) are
obtained7,8 (Scheme 1).
* Author to whom correspondence should be addressed. Fax: (48-
42) 681 97 44. E-mail: wjstec@bio.cbmm.lodz.pl.
(1) (a) Miller, P. S.; Annan, N. D.; McParland; K. B.; Pulford, S. M.
Biochemistry 1982, 21, 2507-2512. (b) Les´nikowski, Z.; J aworska, M.;
Stec, W. J . Nucleic Acids Res. 1990, 18, 2109-2115.
(2) Reynolds, M. A.; Hogrefe, R. I.; J aeger, J . A.; Schwartz, D. A.;
Riley, T. A.; Marvin, W. B.; Daily, W. J .; Vaghefi, M. M.; Beck, T. A.;.
Knowles, S. K.; Klem, R. E.; Arnold, L. J ., J r. Nucleic Acids Res. 1996,
24, 4584-4591.
(3) (a) Kaji, A.; Higuchi, H., Endo, T. Biochemistry 1990, 29, 8747-
8753. (b) Loschner, T.; Engels, J . Tetrahedron Lett. 1989, 30, 5587-
5590.
(4) (a) Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 544-584. (b)
Englisch, U.; Gauss, D. H. Angew. Chem., Int. Ed. Engl. 1991, 30, 613-
629.
(7) (RP)- and (SP) assignments are in accordance with Cahn-Ingold-
Prelog formalism,10 but O-benzoylation changes the priorities of ligands
surrounding phosphorus in compounds 6 and 7.
(8) Cahn, R. S.; Ingold, C. K.; Prelog, V. Angew. Chem., Int. Ed. Engl.
1966, 5, 385-414.
(5) Stec, W. J .; Woz´niak, L. A.; Pyzowski, J .; Niewiarowski, W.
Antisense Nucleic Acid Drug Dev. 1997, 7, 383-397.
(6) Gray, M. D. M.; Smith, D. J . H. Tetrahedron Lett. 1980, 21, 859-
862.
10.1021/jo981304v CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/10/1998

9110 J . Org. Chem., Vol. 63, No. 24, 1998
Notes
Ta ble 1. Ster eocon ver sion of Nu cleosid e 3′-O-Meth a n ep h osp h on oth ioic Acid s
slow (RP)-7 fast (SP)-7
B′
(SP)-6 31P NMRa
31P NMR
1H NMRb
(RP)-6 31P NMR
31P NMR
1H NMRb
FAB MS (calcd)
Thy
CBz
ABz
GiBu
77.63
75.77
77.46
76.82
91.41
91.50
91.06
91.29
2.09 (15.9)
2.08 (15.9)
2.14 (15.9)
2.14 (15.9)
77.45
75.66
77.27
77.01
91.77
91.57
91.49
91.85
2.04 (15.9)
2.05 (15.9)
2.07 (15.9)
2.09 (15.9)
784.2 (784.2583)
873.2849 (867.3067)
896.9 (897.2961)
879.10 (879.3067)
a
31P NMR in CDCl3, as triethylammonium salt. b 1H NMR in CDCl3 (2J P-CH in Hz).
3
Interestingly, if the hydrolysis of mixed anhydrides
(RP)-7 is performed in the presence of 4-(dimethylamino)-
pyridine DMAP, (SP)-nucleoside methanephosphonothio-
ate 6 is recovered. The hydrolysis of the opposite
diastereomer (SP)-7 in the presence of DMAP occurs with
the same base-dependent stereochemistry, so this reac-
tion proceeds with retention of configuration at the
phosphorus atom.
phosphonothioic acid 6. The isotope chemical shift effect
measured for 6 was ∆δ ) 3.7 Hz. The product of
hydrolysis was then S-alkylated and the value of the
isotope effect was ∆δ ) 4.1 Hz for ester 3. Similar
substitution-directing effects have also been observed for
the DBU- or DMAP-assisted alcoholysis (details not
given).
The effectiveness of our new stereoselective synthesis
of (RP)-nucleoside 3′-O-(S-methyl methanephosphonothi-
olate)s 3 is summarized in Table 2.
The observation of variable stereochemistry of the
base-promoted hydrolysis of mixed anhydrides 7 indicates
that this reaction most probably proceeds via two differ-
ent mechanisms and is dependent upon the nature of the
catalyst. Strongly basic DBU (pKa ) 11.6)12 promotes
an attack of the hydroxyl oxygen at the phosphorus atom,
while the weaker base, DMAP (pKa ) 9.70)13 causes
exclusive attack of water at the carbonyl function.14 An
influence of the base strength upon the site of attack of
the water molecule on ambident electrophiles containing
both carbonyl and methanephosphonothioyl centers (as
in compounds 7) has also been confirmed in the studies
involving stereoselective incorporation of [18O] into 5′-O-
DMT-thymidine 3′-O-methanephosphonothioate 6. In
separate experiments, the hydrolysis of 5′-O-DMT-thy-
midine 3′-O-(2,4,6-trimethylbenzoyl methanephosphono-
thioate) (7) was performed in H218O (95% enriched), in
the presence of DBU or DMAP. The reaction progress
was monitored by TLC (5% MeOH in CHCl3). After
hydrolysis was complete (about 1 h), the reaction mixture
was analyzed by both 31P NMR and mass spectrometry.
It was found that, in the DBU-assisted hydrolysis, oxygen
[18O] originating from H218O was almost exclusively
incorporated into 5′-O-DMT-thymidine 3′-O-methane-
Analysis of the data included in Tables 1 and 2
indicates that the 31P NMR chemical shift values are not
always reliable parameters for distinguishing between
the (RP) and (SP) diastereomers of 5, 6, and 3. The
apparent distinction between diastereomers is only pos-
sible via HPLC (or HP TLC) comparison of the given
single diastereomer with the mixture of both diastereo-
mers. The earlier assignments17 of the absolute config-
uration at the phosphorus atom in both diastereomers
of compound 6 (B ) Thy) are, however, in agreement with
our recent results.
Besides the practical application of the base-directed
solvolysis of mixed anhydrides 7 for the cost-effective
stereoselective synthesis of 1, we wish to emphasize that
the obtained results allow us to demonstrate a new
Walden cycle in phosphorus chemistry18 (Scheme 2).
Mechanistic elucidation of the observed base-depend-
ent regioselectivity of hydrolysis of mixed anhydrides 7
deserves further study, but the results presented here
imply that DBU,19 in contrast to DMAP,20 exerts in the
hydrolysis of 7 a function of the base strong enough to
generate hydroxyl ions in aqueous solutions, which are
then able to attack the phosphorus atom, leading to
inversion of configuration at this center. DMAP, which
is known to be less basic than DBU but much more
nucleophilic, most probably attacks the carbonyl center
with the formation of an adduct,20 which is subsequently
(9) It has been known that mixed phosphorus/carboxylic anhydrides
may exercise both acylating and phosphorylating properties, as
demonstrated in the following: (a) Clark, V. M.; Hutchinson, D. V.;
Kirby, A. J .; Warren, S. G. Angew. Chem. 1964, 76, 704-711; Angew.
Chem., Int. Ed. Engl. 1964, 3, 678-685. (b) Lambie, A. J . Tetrahedron
Lett. 1966, 33, 3709-3712. (c) Sokolova, N. I.; Nosova, V. V.; Velmoga,
I. S.; Shabarova, Z. A. Zh. Obshch. Khim. 1975, 45, 1197. (d) J ackson,
A. G.; Kenner, G. W.; Moore, G. A.; Ramage, R.; Thorpe, W. D.
Tetrahedron Lett. 1976, 40, 3627-3630. (e) Drutsa, V. L.; Zarytova, Z.
F.; Knorre, D. G.; Lebedev, A. V.; Sokolova, N. I.; Shabarova, Z. A.
Dokl. Akad. Nauk SSSR 1977, 223, 595-597. (f) J acobsen, E.; Bartlett,
P. A. J . Am. Chem. Soc. 1983, 105, 1619-1624.
(15) Wasiak, J .; Helin´ski, J .; Da¸bkowski, W.; Skrzypczyn´ski, Z.;
Michalski, J . Pol. J . Chem. 1995, 69, 1027-1032.
(16) Woz´niak, L. A.; Pyzowski, J .; Wieczorek, M.; Stec, W. J . J . Org.
Chem. 1994, 59, 5843-5846.
(10) The assignment of the absolute configurations of compounds 6
by means of X-ray analysis of their derivatives 3 was independently
supported by the stereoselective enzymatic digestion of fully depro-
tected 6 with Nuclease P1.5
(11) It has to be mentioned that other direct methods of synthesis
of 3 do not allow for diastereoconversion of separated isomers, for
example: Brill, W. K.-D.; Caruthers, M. H. Tetrahedron Lett. 1987,
28, 3205-3208. Brill, W. K.-D. Tetrahedron Lett. 1995, 36, 703-706.
(12) Nakatani, K.; Hashimoto, S. Soc. Synth. Org. Chem. 1975, 33,
925.
(17) (a) Niewiarowski, W.; Les´nikowski, Z.; Wilk, A.; Guga, P.;
Okruszek, A.; Uznan´ski, B.; Stec, W. J . Acta Biochim. Pol. 1987, 34,
217-231. (b) Lebedev, A. V.; Rife, P. J .; Seligsohn, H. W.; Wenzinger,
G. R.; Wickstrom, E. Tetrahedron Lett. 1990, 31, 855-858. (c) Engels,
J . W.; Loschner, T. Nucleic Acid Res. 1990, 18, 5083-5088.
(18) Mikołajczyk, M.; Michalski, J .; Omelan´czuk, J . Tetrahedron
Lett. 1965, 23, 1779-1784.
(19) The function of DBU as a “strong non-nucleophilic base” has
been demonstrated. See: (a) Seebach, D.; Thaler, A.; Blaser, D.; Ko,
S. Y. Helv. Chim. Acta 1991, 74, 1102-1118. (b) Muathen, H. A. J .
Org. Chem. 1992, 57, 2740-2741. (c) Tawfik, D. S.; Eshhar, Z.;
Bentolila, A.; Green, B. S. Synthesis 1993, 968-972. (d) Bennua-
Skalmowski, B.; Vorbruggen, H. Tetrahedron Lett. 1995, 36, 2611-
2614. (e) Soloshonok, V. A.; Ono, T. J . Org. Chem. 1997, 62, 3030-
3032.
(13) Chakrabarty, M. R.; Handlosed, C. S.; Mosher, M. W. J . Chem.
Soc., Perkin Trans. 2 1973, 938-942.
(14) The regioselectivity of alcoholysis of mixed phosphilic/carboxylic
anhydrides is still unpredictable and depends not only upon the kind
of the base but also upon the nature of substituents at both phosphorus
and the carbonyl center, as well as upon the solvolytic medium. For
example, O,O-diethyl-O-trifluoroacetyl phosphate reacts with ethanol
as an acetylating agent as promptly noticed by Wasiak et al.,15
correcting our erroneous data.16
(20) Hofle, G.; Steglich, W.; Vorbruggen, H. Angew. Chem. 1978,
90, 602-616. Angew. Chem., Int. Ed. Engl. 1978, 17, 569-583.
(21) Oligonucleotide Synthesis. A Practical Approach; Gait, M. J .,
Ed.; IRL Press: Oxford, 1984.

Notes
J . Org. Chem., Vol. 63, No. 24, 1998 9111
Ta ble 2. Sp ectr a l Ch a r a cter istics for 5′-O-DMT-n u cleosid e Meth a n ep h osp h on oth ioa tes (O-a lk yl 5), (S-a lk yl 3), a n d 6
5′-O-DMT-nucleoside 3′-O-(O-alkyl
methanephosphonothioate)
(R ) CH2CH2CN) 5
5′-O-DMT-nucleoside 3′-O-(S-alkyl
methanephosphonothiolate)
(R ) CH3) 3
thioacid 6
31P
NMR
(δ, ppm)
1H NMR
(δ, ppm; J PCH3
in Hz)
31P
NMR
(yield, %)c (δ, ppm)d (yield, %)e (δ, ppm)
HR FAB
MS
1H NMR
config
config
31P NMR
(δ, ppm; J P-H,
nucleosidea
configb
(RP)-5
fast (0.58)
(SP)-5
[M-H]f
in Hz)g
thymidine
(80)
98.85
98.34
98.90
98.31
98.32
98.47
98.47
98.80
1.89; 15.54h
1.84; 15.58
1.88; 15.52
1.83; 15.60
1.91; 15.57
1.87; 15.54
1.85; 15.52
1.83; 15.61
(RP)-6 (86)
(SP)-6 (87)
(RP)-6 (70)
(SP)-6 (65)
(RP)-6 (85)
(SP)-6 (81)
(RP)-6 (70)
(SP)-6 (75)
72.67
72.34
72.97
72.58
72.93
72.70
72.06
72.22
(RP)-3 (95)
slow
(SP)-3 (95)
fast
(RP)-3 (90)
slow
(SP)-3 (86)
fast
(RP)-3 (80)
slow
(SP)-3 (75)
fast
(RP)-3 (85)
slow
(SP)-3 (82)
fast
57.22
57.21
56.96
57.01
57.32
57.41
57.34
57.47
637.178
1.74; 15.78 (CH3P, d)
2.19, 11.6 (CH3S, d)
(637.177)i 1.82; 15.67 (CH3P, d)
2.18; 13.18 (CH3S, d)
740.209
slow (0.48)
(RP)-5
fast (0.55)
(SP)-5
slow (0.43)
(RP)-5
fast (0.58)
(SP)-5
slow (0.48)
(RP)-5
fast (0.30)
(SP)-5
cytidine
(67)
1.81; 15.63 (CH3P, d)
2.21; 13.30 (CH3S, d)
(740.219) 1.83; 15.6 (CH3P, d)
2.27; 13.3 (CH3S, d)
764.4
adenosine
(63)
1.85; 13.6 (CH3P, d)
2.2; 12.08 (CH3S, d)
(764.261) 1.88; 15.07 (CH3P, d)
2.1; 11.50 (CH3S, d)
746.235
guanosine
(51)
1.81; 20.38 (CH3P, d)
2.10; 13.35 (CH3S, d)
(746.241) 1.82; 15.55 (CH3P, d)
2.32; 13.19 (CH3S, d)
slow (0.20)
a
b
In parentheses yields of compounds 5 are given. Relative mobilities of diastereomers on HP TLC analytical plates (E. Merck).
Developing system: CHCl3/MeOH (96:4 v/v, twice). c Yields of conversion 5 f 6 (after purification). Spectra registered in DMF/C6D6
d
(sodium salt, precipitated from petroleum ether). e In parenthesis, yields of conversions 6 f 3 are given. f HR MS analysis was performed
g
for 1:1 mixture of diastereomers 3. Calculated molecular weights are given in parentheses. Data are given for P-CH3 protons, in CDCl3.
Data are given for P-CH3 and P-SCH3 protons, in CDCl3, appropriately. [M-CH3] peak was observed.
h
i
Gen er a l P r oced u r e for P r ep a r a tion of 5-O-DMT-(N-
Sch em e 2
P r otected ) Nu cleosid e 3′-O-(O-Alk yl Meth a n ep h osp h on o-
th ioa te)s (5) fr om MeP (S)Cl2. To a solution of MeP(S)Cl2 (2
mmol) in dry pyridine (15 mL), a solution of 5′-O-DMT-(N-
protected, except thymine) nucleoside (1 mmol), synthesized
according to the literature21 in pyridine (5 mL), was added. The
reaction mixture was stirred for 15 min at 55 °C and cooled to
ambient temperature, followed by an addition of 4-6 mmol of
alcohol. After 30 min, the reaction was completed, the reaction
mixture was concentrated under reduced pressure to about 1/3
of the original volume, and the oily residue was dissolved in
CHCI3, (30 mL) and washed 3 times with citric acid (0.1 M, 15
mL). The organic layer was dried over anhydrous MgSO4,
concentrated, and purified by means of silica gel column chro-
matography [elution with chloroform containing petroleum ether
(20-5%)]. Appropriate fractions of products fast-5 (isomer faster
migrating during a silica gel chromatography) and slow-5
(isomer slower migrating during a silica gel chromatography)-
(R ) NCCH2CH2) were collected and concentrated to give pure
diastereomers as colorless foams. Their characteristics are given
in Table 2.
hydrolyzed, restoring parent 6 with unchanged configu-
Separation of esters 5 (R ) OMe) was more difficult, so they
were often collected as a mixture of diastereomers and further
separated for analytical reasons. Alternatively, separation can
be directly achieved for methanephosphonothioates 6 (silica gel
column chromatography, silica gel 240-400 mesh, gradient
3-6% MeOH in CHCl3, with 0.5% Et3N).
ration at phosphorus.
Exp er im en ta l Section
Gen er a l. NMR spectra were recorded at 300.13 (1H) and
121.47 (31P) MHz. Chemical shifts are reported (δ) relative to
1
Fast(Rp)-5 (B ) T, R ) OMe): 31P NMR δ 98.91 (CDCl3); H
TMS (1H) and 80% H3PO4 31P) as external standards. 2D 1H-
(
1H NMR (NOESY) correlations were applied for identification
of signals in NMR spectra. Positive chemical shift values were
assigned for compounds resonating at lower fields than the
standards. Mass spectra were recorded on a mass spectrometer
with a Cs+ gun operating at 13 keV using a 3-nitrobenzyl alcohol
matrix. High-resolution mass spectra were recorded for 1:1
mixtures of diastereomers. Column chromatography and HP
TLC analyses were performed on silica gel (240-400 mesh) and
precoated F254 silica gel plates, respectively. Solvents and
reagents were purified according to the standard laboratory
techniques and stored under argon. MeCN was distilled from
CaH2 directly to the reaction vessels. LiCl was recrystallized
from MeOH and dried under vacuum (150 °C/0.05 mmHg) for
several days. Reactions involving air- or moisture-sensitive
compounds were carried out in glassware that was dried either
in an oven or under high vacuum and then placed under a
positive pressure of argon.
NMR (CDCI3) δ 1.77 (9.68, d, 3H, PCH3), 3.74 (8.22, d, 3H,
POCH3), HR FAB [M - H] 651.186 (calcd 651.193). Slow(Sp)-5
(B ) Thy, R ) OMe): 31P NMR δ 99,00 (CDCI3); 1H NMR
(CDC13) δ 1.85 (10.63, 3H, PCH3), 3.70 (7.35, d, 3H, POCH3).
1
Fast(Rp)-5 (C ) CBz, R ) OMe): 31P NMR δ 98.77 (CDCl3); H
NMR (CDCl3) δ 1.76 (11.05, d, 3H, PCH3), 3.72 (14.00 d, 3H,
POCH3); HR FAB [M - H] 740.215 (calcd 740.219). Slow(Sp)-5
(B ) CBz, R ) OMe): 31P NMR δ 99.37 (CDCI3); 1H NMR (CDCI3)
δ 1.84 (14.01, 3H, PCH3), 3.70 (7.35, d, 3H, POCH3). Fast(Rp)-5
(B ) ABz, R ) OMe): 31P NMR δ 98.91 (CDCl3); 1H NMR (CDCl3)
δ 1.81 (15.49, d, 3H, PCH3), 3.74 (14.07,d, 3H, POCH3); HR FAB
[M - H] 764.238 (calcd 764.261). Slow(Sp)-5 (R ) ABz, R )
OMe): 31P NMR δ 99.00 (CDCI3); 1H NMR (CDCl3) δ 1.89 (15.53,
3H, PCH3), 3.62 (13.99, d, 3H, POCH3). Fast(RP)-5 (B ) Gibu, R
) OMe): 31P NMR δ 99.23 (CDCI3); 1H NMR (CDCI3) δ 1.76
(15.49, d, 3H, PCH3), 3.51 (14.01, d, 3H, POCH3); HR FAB [M
- H]: 746.238 (calcd 746.241).

9112 J . Org. Chem., Vol. 63, No. 24, 1998
Notes
Gen er a l P r oced u r e for P r ep a r a tion of 5′-O-DMT-(N-
P r otected Excep t Th ym in e) Nu cleosid e 3′-O-(O-2,4,6-Tr i-
m eth ylben zoyl Meth a n ep h osp h on a te)s (7). N-Methyl-N-
tert-butylammonium salts of (SP)-5′-O-DMT-nucleoside meth-
anephosphonothioates (6) (0.5 mmol) obtained from 5 according
to methodology described earlier,8 were twice coevaporated with
pyridine, and left overnight on high vacuum, and redissolved in
dry pyridine (8 mL) with Et3N (2 mmol), and to this solution
2,4,6-trimethylbenzoyl chloride (1-1.5 mmol) was added in one
portion. After 30 min, the reaction mixture was concentrated
to 1/4 of the original volume, diluted with CHCl3 (15 mL), and
washed with 0.05 M citric acid. The organic layer was dried
over MgSO4 and concentrated to dryness. 31P NMR spectra
confirmed complete conversion to 7. Compounds 7 were purified
by means of silica gel column chromatography. Data are
collected in Table 1.
means of MS [found FAB MS [M - 1] 166.1 (calcd 166.2)]. The
reaction product, (SP)-6, was treated with MeI (0.1 mL), and the
isolated (SP)-3 was identical to a genuine sample prepared
independently.5 The diastereomeric purity of slow (SP)-3 was
95% [31P NMR (CDCl3) δ 58.6].
St er eocon ver sion of t h e Nu cleosid e 3′-O-Met h a n e-
p h osp h on oth ioa te (Sp )-6 in to (Rp )-6 via Mixed An h yd r id e
(Rp )-7. The triethylammonium salt of (SP)-nucleoside 3′-O-
methanephosphonothioate (6) was coevaporated with pyridine
and dried overnight on high vacuum. To a pyridine solution,
2,4,6-trimethylbenzoyl chloride (1.2 equiv) was added in one
portion. The reaction progress was controlled by TLC (5%
MeOH/CHCl3). After the reaction was complete (about 30 min),
the reaction mixture was concentrated and the crude viscous
liquid was diluted with CHCl3 and washed with 0.05 M citric
acid. The combined organic layers were dried with MgSO4 and
concentrated. Product (RP)-7 was purified by silica gel column
chromatography (chloroform as eluent). The pure mixed anhy-
dride (RP)-7 was dissolved in MeCN, followed by addition of DBU
(20 equiv) and water (100 equiv). After 2 h, the reaction mixture
was concentrated, dissolved with chloroform, and washed with
0.05 M citric acid. The obtained (RP)-7 was purified by means
of flash silica gel column chromatography (4% EtOH/CHCl3) or
used without purification for S-alkylation (data are collected in
Table 1).
Hyd r olysis of Mixed An h yd r id es 7 w ith H218O a n d DBU.
Mixed anhydride (RP)-7 (B ) Thy, R′ ) 2,4,6-trimethylphenyl)
(0.05 mmol) was dissolved in MeCN (1 mL) with 0.05 mL of H2-
18O (95% ie) and DBU (0.2 mmol). After 4 h, a 31P NMR
spectrum of the reaction mixture revealed the presence of a
single product (RP)-6 [31P NMR (MeCN-d3) δ 74.6 ppm]. MS
analysis of the mixture of unseparated products confirmed the
presence of a species with a molecular ion [found FABMS [M -
H] 639.3 (calcd 639.2)], indicating the incorporation of [18O] into
(RP)-6. The molecular ion of 2,4,6-trimethylbenzoic acid [found
FAB MS [M - 1] 163.2 (calcd 163.2)], which was not enriched
in [18O] isotope, was also detected. The reaction product was
treated with MeI (0.1 mL), and the isolated (RP)-3 was identical
to a genuine sample prepared independently.5 The diastereo-
meric purity of slow-(RP)-3 was 95% [31P NMR (CDCl3) δ 58.6].
Hyd r olysis of Mixed An h yd r id es 7 w it h H 218O a n d
DMAP . Mixed anhydride (RP)-7 (B ) Thy, R′ ) 2,4,6-trimeth-
ylphenyl) (0.05 mmol) was dissolved in MeCN (1 mL) with 0.05
mL of H218O (95% ie) and DMAP (0.2 mmol). After 2 h, a 31P
NMR spectrum of the reaction mixture revealed the presence
of a single product of hydrolysis, (SP)-6 [31P NMR (MeCN-d3) δ
74.3; yield +99%; found FAB MS [M - H] 637.3 (calcd 637.2)].
The incorporation of [18O] into benzoic acid was confirmed by
Ack n ow led gm en t. Presented results were obtained
within the project financially assisted by the State
Committee for Scientific Research (KBN, Grant No. 3
TO9A 031 011, for L.A.W.)
1
Su p p or tin g In for m a tion Ava ila ble: Copies of H NMR
spectra and high-resolution mass spectra (FAB) of two mixed
anhydrides (10 pages). This material is contained in libraries
on microfiche, immediately follows this article in microfilm
version of the journal, and can be ordered from ACS; see any
current masthead page for ordering information.
J O981304V