Mild and room temperature C-C bond forming reactions of nucleoside C-6 arylsulfonates

Article abstract of DOI:10.1021/jo0513764

Palladium catalyzed cross coupling of nucleoside arylsulfonates and arylboronic acids has been accomplished under mild conditions and at room temperature. Among three structurally similar ligands that differ in their steric and electronic properties, one

Full text of DOI:10.1021/jo0513764

Mild and Room Temperature C-C Bond Forming Reactions of
Nucleoside C-6 Arylsulfonates
Mahesh K. Lakshman,* Padmaja Gunda, and Padmanava Pradhan†
Department of Chemistry, The City College and The City University of New York,
138th Street at Convent Avenue, New York, New York 10031-9198
lakshman@sci.ccny.cuny.edu
Received July 3, 2005
Palladium catalyzed cross coupling of nucleoside arylsulfonates and arylboronic acids has been
accomplished under mild conditions and at room temperature. Among three structurally similar
ligands that differ in their steric and electronic properties, one yielded an effective catalyst in
conjunction with Pd(OAc)2. Of the nucleoside arylsulfonates evaluated, the O6-(2,4,6-trimethylphe-
nyl)sulfonate proved optimal, but other alkyl and alkoxy derivatives were also reasonably reactive.
On the other hand, a 2-nitrophenyl and a 2-thienyl derivative were ineffective substrates. PhMe
and THP were suitable as solvents, yielding good results in several cases, although reactions of
some arylboronic acids were faster in PhMe. In contrast, reactions of arylboronic acids bearing
strongly electron-withdrawing groups proceeded more successfully in THP. Interplay between
several factors that include substituents on the nucleoside arylsulfonate, ligand substituents, and
solvent is responsible for successful cross coupling. Using 31P NMR, an initial investigation has
been conducted to study the interaction of Pd(OAc)2 with the ligand. At a 1:1 stoichiometry of ligand
and Pd(OAc)2, a predominant species, likely a cyclopalladation product, was obtained. At a 2:1
ratio of ligand and Pd(OAc)2, a different species bearing chemically distinct phosphine ligands was
observed. Both complexes display catalytic activity, although the 2:1 species may be superior.
Introduction
aryl nucleosides, wherein the hydrogen-bonding amino
group is replaced by hydrophobic aryl moieties and some
C-6 aryl nucleosides so derived have shown cytostatic
activity against cancer cell lines.6 Among nonhalo nucleo-
side substrates, we have demonstrated that the easily
prepared O6-(2,4,6-trimethylphenyl)sulfonyl derivative of
2′-deoxyguanosine could be converted to an assortment
of C-6 aryl nucleoside analogues via Pd-catalyzed C-C
bond formation at elevated temperatures.7 In that study,
The use of aryl triflates for C-C bond formation with
boronic acids has become a standard in organic synthe-
sis.1 On the other hand, aryl arenesulfonates that are
generally cheaper to prepare have not been perceived as
general substrates for Pd- or Ni-catalyzed Suzuki-
Miyaura cross-coupling until recently.2,3 Among nucleo-
side derivatives, we4 and others5,6 have shown that
halopurine nucleosides can be efficiently converted to C-6
* Corresponding author. Tel.: (212) 650-7835; fax: (212) 650-6107.
† P. P. NMR Facility Manager, CCNY.
(4) Lakshman, M. K.; Hilmer, J. H.; Martin, J. Q.; Keeler, J. C.;
Dinh, Y. Q. V.; Ngassa, F. N.; Russon, L. M. J. Am. Chem. Soc. 2001,
123, 7779-7787.
(5) (a) Havelkova´, M.; Hocek, M.; Cˇ esnek, M.; Dvorˇa´k, D. Synlett
1999, 1145-1147. (b) Hocek, M.; Holy´, A.; Votruba, I.; Dvorˇa´kova´, H.
Collect. Czech. Chem. Commun. 2000, 65, 1683-1697. (c) Hocek, M.;
Holy´, A.; Votruba, I.; Dvorˇa´kova´, H. Collect. Czech. Chem. Commun.
2001, 66, 483-499.
(1) (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457-2483.
(b) Stanforth, S. P. Tetrahedron 1998, 54, 263-303. (c) Suzuki, A. J.
Organomet. Chem. 1999, 576, 147-168. (d) Kotha, S.; Lahiri, K.;
Kashinath, D. Tetrahedron 2002, 58, 9633-9695.
(2) Pd-catalyzed: Nguyen, H. N.; Huang, X.; Buchwald, S. L. J. Am.
Chem. Soc. 2003, 125, 11818-11819.
(3) Ni-catalyzed: (a) Zim, D.; Lando, V. R.; Dupont, J.; Monteiro,
A. L. Org. Lett. 2001, 3, 3049-3051. (b) Tang. Z.-Y.; Hu, Q.-S. J. Am.
Chem. Soc. 2004, 126, 3058-3059. (c) Percec, V.; Golding, G. M.;
Smidrkal, J.; Weichold, O. J. Org. Chem. 2004, 69, 3447-3452.
(6) (a) Hocek, M.; Holy´, A.; Votruba, I.; Dvorˇa´kova´, H. J. Med. Chem.
2000, 43, 1817-1825. (b) Hocek, M.; Nausˇ, P.; Pohl, R.; Votruba, I.;
Furman, P. A.; Tharnish, P. M.; Otto, M. J. J. Med. Chem. 2005, 48,
5869-5873.
10.1021/jo0513764 CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/12/2005
J. Org. Chem. 2005, 70, 10329-10335
10329

Lakshman et al.
TABLE 1. Influence of Solvent on the Cross Coupling of
1a with PhB(OH)2 at Room Temperaturea
entry
solvent
time (h)
% 2a formedb
1
2
tert-BuOH
1,2-DME
1,4-dioxane
THF
21
21
21
21
21
21
21
19
21
15.5
19
49
3
55
4
79
FIGURE 1. Structures of the nucleoside arylsulfonates and
biphenyl ligands evaluated.
5
THF/H2Oc
THF/iso-PrOHd
tetra-Me-THF
THP
70
6
40
7
5
8
100e
46
although catalysis was seen to occur at room tempera-
ture, conversions were largely incomplete.
9
PhCF3
PhMe
10
100e
Compared with smaller aryl systems, nucleosides are
more complex in their reactivities, and data obtained
from reactions of simpler aromatics are not always
directly applicable to nucleosides. Furthermore, little is
known about factors that influence Pd catalyzed reactions
of nucleosides in general. For these reasons, we became
interested in understanding whether C-C cross coupling
of nucleoside arylsulfonates could be accomplished at
room temperature and in gaining further insight into
these reactions. To our knowledge, room temperature
reactions at the C-6 position of nucleosides have not been
reported to date. In the broader context, accomplishing
such reactions under ambient conditions generally rep-
resents a nontrivial undertaking. In addition, we were
interested in evaluating the interactions between the
ligand and the metal that lead to catalyst activity.
a Sulfonate concentration 0.084 M. b Reactions were monitored
by TLC, and the ratio of 1a/2a was determined by integration of
the H-1′ signals in the crude reaction products (H-1′ appears at δ
6.28 ppm in 1a and δ 6.43 ppm in 2a). c H2O (10 µL) was added to
the reaction mixture. d iso-PrOH (96 µL) was added to the reaction
mixture. e H-1′ of 1a was not observed by NMR.
At the outset of this series of experimentation, influ-
ence of the solvent was considered an important factor.
The coupling of 1a and PhB(OH)2 with 10 mol %
Pd(OAc)2/20 mol % L-1/2 molar equiv of K3PO4 was used
as a model for analyzing solvent influence. Since we have
observed faster reactions with increased concentrations
of arylboronic acids in some instances,4,7 2 molar equiv
of the boronic acids was used in the present study to
offset the decrease in reaction rate due to temperature.8
The data gathered from this initial assessment are
summarized in Table 1.
Results and Discussion
From these initial experiments, both tetrahydropyran
(THP) and PhMe emerged as optimal solvents (entries 8
and 10). THF alone (entry 4) was better than combina-
tions with H2O or iso-PrOH (entries 5 and 6). 2,2,5,5-
Tetramethyltetrahydrofuran (entry 7) was substantially
inferior as compared to THF, as was PhCF3 (entry 9) as
compared to PhMe.
The effectiveness of catalytic systems derived from L-2
and L-3 was also analyzed for the cross coupling of 1a
with PhB(OH)2 under the conditions stated previously.
In PhMe, the L-2 derived catalyst was superior to that
from L-3, with a 72% yield of 2a obtained from the former
and incomplete reaction observed with the latter. Thus,
among the three related ligands, L-1 yielded a catalytic
system that was by far the most effective. With Pd(PPh3)4
(10 mol %) and K3PO4 (2 molar equiv), the coupling of
PhB(OH)2 (2 molar equiv) with 1a in PhMe at room
temperature showed no progress. In contrast, this cata-
lytic system was effective at 80-100 °C, clearly demon-
strating the superior nature of the biphenyl-based ligands.9
Next, we reasoned that oxidative addition of the metal
to the C-O bond of the nucleoside arylsulfonate could
be perhaps manipulated through substituents on the
In a recent preliminary communication, we had shown
that a single O6-arylsulfonyl derivative of 2′-deoxygua-
nosine was an effective substrate for C-C cross coupling
at elevated temperatures (80 °C in 1,4-dioxane).7 Al-
though good yields were obtained and the method was
generally applicable, a detailed analysis of the cross
coupling was not performed in that study. Influence of
the arylsulfonate substituents, solvents, and other factors
on the course of these reactions had not been analyzed.
Since each of these can substantially impact the efficiency
of the cross coupling, we were interested in a more
detailed evaluation of the reactions. In fact, even among
reactions of simpler aryl arenesulfonates, these factors
do not seem to have received attention thus far.
For this study, a variety of O6-arylsulfonyl 2′-deoxy-
guanosine derivatives (Figure 1) was selected that not
only varied in electronic properties but also differed in
steric requirements around the reactive center. Addition-
ally, three related ligands were chosen for analysis
(Figure 1). We have shown that L-1 is a good ligand for
C-C cross coupling,4,7 but the commercial availability of
newer ligands L-2 and L-3, which differed from L-1 in
their steric and electronic properties, begged a compari-
son of the relative reactivities of catalytic systems derived
from them.
(8) Ambient air temperature was measured during the entire course
of the experimentation and was found to be 23-28 °C.
(9) Gunda, P.; Russon, L. M.; Lakshman, M. K. Angew. Chem., Int.
Ed. 2004, 43, 6372-6377; also see corrigendum: Gunda, P.; Russon,
L. M.; Lakshman, M. K. Angew. Chem., Int. Ed. 2005, 44, 1154.
(7) Lakshman, M. K.; Thomson, P. F.; Nuqui, M. A.; Hilmer, J. H.;
Sevova, N.; Boggess, B. Org. Lett. 2002, 4, 1479-1482.
10330 J. Org. Chem., Vol. 70, No. 25, 2005

C-C Bond Forming Reactions of C-6 Arylsulfonates
TABLE 2. Cross Coupling of Various O6-Arylsulfonates
of 2′-Deoxyguanosine with PhB(OH)2 at Room
Temperaturea
TABLE 3. Generality of the Cross Coupling, Solvent
Effects, and Yields of the C-6 Aryl Nucleoside
Derivatives
entry
sulfonate
time (h)
result, yield of 2ab
1
2
3
4
5
6
1a
1b
1c
1d
1e
1f
15.5
15.5
15.5
15.5
15.5
15.5
complete, 90%
no progress
complete, 85%
complete, 82%
complete, 85%
incomplete, 21%c
a Sulfonate concentration 0.084 M in PhMe. b Reactions were
monitored by TLC for completion; yield is of isolated, purified
product. c Assessed by integration of the H-1′ signals in the crude
reaction product (H-1′ appears at δ 6.30 ppm in 1f and at δ 6.43
ppm in 2a).
arylsulfonate moiety. Among simpler aryl systems, elec-
tron deficient ones are activated toward oxidative addi-
tion.10,11 Attempts were therefore directed toward ren-
dering the nucleoside-sulfonate bond more labile. Thus,
2-nitrophenylsulfonate 1b (Figure 1) was synthesized via
known procedures.12 Reaction of 1b with 2 molar equiv
of PhB(OH)2 using 10 mol % Pd(OAc)2/20 mol % L-1/2
molar equiv of K3PO4 in PhMe showed little progress.
This therefore prompted the screening of several other
arylsulfonates listed in Figure 1, the results of which are
tabulated in Table 2.
The ineffectiveness of 1b was somewhat surprising
(entry 2). In contrast, all alkyl-substituted phenylsul-
fonates were reactive. Interestingly, the (2,4,6-triisopro-
pylphenyl)sulfonate 1d, which had produced diminished
C-C coupling yields at 80 °C,7 provided a significant yield
improvement at room temperature. It is therefore plau-
sible that a competing reaction pathway at elevated
temperature is a contributing factor to the diminished
yield. 4-Methoxyphenylsulfonate 1e was also effectively
activated for the reaction.
At this stage, the generality of the room temperature
cross coupling was explored, and arylboronic acids with
varying electronic properties were selected for this pur-
pose. A continued comparison of the arylsulfonate sub-
strates as well as PhMe and THP as solvents was
considered. The results of these experiments are compiled
in Table 3. For comparative purposes, prior results from
reactions performed in 1,4-dioxane at 80 °C are also
presented in this table.
As evident from Table 3, several arylboronic acids
underwent cross coupling at room temperature. A slightly
elevated temperature (45 °C) was necessary for effecting
complete cross coupling of the electron deficient arylbo-
ronic acids (m-nitro, p-acetyl, and m-carboethoxy) and the
hindered o-ethoxyphenylboronic acid. PhMe was a gener-
a Reported in ref 7 (reactions were conducted in 1,4-dioxane at
80 °C). b Conducted at 45 °C. c Reaction was incomplete.
ally better solvent as compared to THP, in many cases
producing shorter reaction times. The solvent effect was
most pronounced in the reactions of p-phenoxyphenyl-
boronic acid (entry 17 vs 18) and p-(diphenylamino)-
phenylboronic acid (entry 22 vs 23). When strongly
electron-depleting substituents were present (m-nitro and
p-acetyl), THP was superior to PhMe (entry 24 vs 25 and
(10) Amatore, C.; Pfluger, F. Organometallics 1990, 9, 2276-2282.
(11) Jutand, A.; Mosleh, A. Organometallics 1995, 14, 1810-
1817.
(12) (a) Hayakawa, Y.; Hirose, M.; Noyori, R. J. Org. Chem. 1993,
58, 5551-5555. (b) Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda,
M.; Sasaki, S. Tetrahedron 1997, 53, 3035-3044.
J. Org. Chem, Vol. 70, No. 25, 2005 10331

Lakshman et al.
29 vs 30). This was also the case with 3,4-(methylene-
dioxy)phenylboronic acid (entries 19 and 20).
the disappearance of free phosphine within 10 min. An
ABquart at 24-32 ppm, a small singlet at 31.71 ppm, two
sharp singlets at 42.49 and 42.74 ppm, and a broad signal
at 47.27 ppm appeared. After 1.5 h at room temperature
(Figure 2A), the singlet at 31.71 diminished, but the
ABquart, two sharp singlets, and the broad resonance
remained. After 24 h (Figure 2B), the intensity of the
broad signal at 47.27 ppm increased slightly, whereas
the ABquart and the two sharp singlets decreased. A
singlet appeared at 50.21 ppm, and the singlet at 31.71
ppm disappeared. After 3 days, the ABquart completely
disappeared.
Preparation of the same 1:1 complex as stated previ-
ously and warming to 40 °C resulted in the disappearance
of the free phosphine signal within 10 min and the
formation of two sharp singlets (42.50 and 42.75 ppm)
along with a broad resonance (46.55 ppm) and a trace of
an ABquart (24-32 ppm). After 1 h at 40 °C, there was an
increase in the broad signal with a concomitant decrease
in intensity of the two sharp singlets. A trace of the
ABquart remained, and a small sharp singlet appeared at
50.28 ppm. After 24 h at 40 °C, the singlets (42.50 and
42.75 ppm) were significantly smaller, and the broad
resonance (46.55 ppm) was predominant (>90%). There
was no evidence of the ABquart, the signal at 50.28 ppm
increased, and a trace of a signal at 44.09 ppm became
evident (see Supporting Information for NMR traces).
There was no change after 29 h. In benzene-d6, these
patterns were essentially identical except that progress
seemed faster.
(b) 2:1 Complex of L-1 and Pd(OAc)2. When a 2:1
mixture of L-1/Pd(OAc)2 in toluene-d8 was analyzed by
31P{1H} NMR, interesting differences emerged. After 2
h at room temperature, 31P NMR showed a broad signal
at -13.33 ppm (perhaps free phosphine), and the pre-
dominant signal was the ABquart (24-32 ppm). A singlet
at 31.71 ppm, a minor broad resonance (47.27 ppm), and
a singlet at 50.21 ppm appeared. After 21 h, the broad
signal at -13.33 ppm persisted, and the ABquart remained
as the major component. The singlet at 31.71 ppm
diminished, the broad resonance (46.77 ppm) remained
minor, and the singlet at 50.21 ppm increased in intensity
slightly.
Among the arylsulfonates, 1a appeared to be a more
general substrate (entry 1 as compared to 4), although
in several cases (entries 8, 12, and 16), 1c showed good
utility. Entries 24 and 25 as well as 27 and 28 provide
additionally interesting comparisons. The more hindered
sulfonate 1d is a less effective substrate as compared to
1a, but the solvent effect was identical in each, with
reactions in THP providing better product yields. When
comparing the results from the cross coupling reactions
performed in 1,4-dioxane at 80 °C to those at room
temperature, the latter are significantly superior in most
cases. Comparable results were observed with p-nitro-
phenylboronic acid and o-ethoxyphenylboronic acid. It is
noteworthy, however, that reactions in entries 15 and 36
were conducted with 2.5 and 3 molar equiv of the
arylboronic acids, respectively, whereas the room tem-
perature reactions were conducted with 2 molar equiv.
Toward gleaning more insight into the cross coupling,
1a was allowed to react with phenylboronic acid, and the
reaction was stopped when TLC indicated complete
consumption of the sulfonate (7.5 h). The yield of chro-
matographically purified 2a in this case was 82%. The
absence of 1a in this reaction and a lowered product yield
(by ∼10%) may indicate that consumption of starting
material alone may not signify complete product forma-
tion and that conversion of intermediates may require
longer periods. Since the reactions of m-nitro and p-
acetylphenylboronic acids with 1a in PhMe were lower
yielding, crude products from these reactions were ana-
lyzed by 1H NMR. In these cases TLC had indicated
consumption of 1a. Correspondingly, evaluation of the
NMR spectra of the reaction mixtures showed no pseudo-
triplet corresponding to the anomeric H-1′ of 1a. How-
ever, in addition to the H-1′ resonance of 2h and 2i, other
smaller pseudo-triplets were observed in each case. In
contrast, reactions performed in THP, which afforded
higher product yields (entries 19, 24, 27, and 29 in Table
3), did not show discernible additional pseudo-triplets in
the H-1′ region.
Similarly, reactions of 1a with PhB(OH)2 catalyzed by
the Pd/L-2 and Pd/L-3 complexes were evaluated. The
reaction mixture obtained with the Pd/L-2 complex
showed no H-1′ corresponding to 1a, but a resonance from
2a and smaller pseudo-triplets were observed. The reac-
tion with the Pd/L-3 complex that was incomplete showed
H-1′ resonances of 1a and 2a as well as additional minor
pseudo-triplets. Although these data are not evidence of
reaction intermediates, minor H-1′ resonances other than
that of product were observed in reactions that were
lower yielding. More detailed studies of the mechanistic
aspects of the cross coupling are currently being consid-
ered.
When the 2:1 system was analyzed at 40 °C, after 10
min, the broadened signal at -13.33 ppm was visible.
The ABquart (24-32 ppm) was the predominant signal,
and a singlet appeared at 32.15 ppm. Two sharp singlets
at 42.50 and 42.75 ppm and a minor broad signal at 46.55
ppm were also evident. After 3 h, the broad signal at
-13.33 ppm was still present, but the ABquart remained
significant. The singlets at 32.15, 42.50, and 42.75 ppm
decreased significantly. The broad signal at 46.55 ppm
increased slightly, and a singlet at 50.28 ppm appeared.
After 24 h, the broad signal at -13.33 ppm remained,
and the ABquart was still the predominant resonance. The
singlet at 32.15 ppm completely disappeared, the broad
resonance (46.55 ppm) was still present, the singlet at
50.28 ppm increased, and a singlet at 44.33 ppm ap-
peared (see Supporting Information for NMR traces).
There were no major changes when this system was
reanalyzed after 28 h at room temperature.
31P NMR Data of L-1/Pd(OAc)2 Complexes. To
obtain some understanding of the interactions of L-1 with
Pd(OAc)2, preliminary 31P{1H} NMR studies were un-
dertaken. In toluene-d8, relative to 85% H3PO4 as exter-
nal standard, the 31P resonance of the free phosphine L-1
appeared at -13.36 ppm and the corresponding phos-
phine oxide at 46.14 ppm.
(a) 1:1 Complex of L-1 and Pd(OAc)2. The addition
of an equimolar amount of Pd(OAc)2 to L-1 [1:1 ratio of
L-1/Pd(OAc)2] in toluene-d8 at room temperature led to
There are striking differences in the 31P{1H} NMR
spectra of the 1:1 and 2:1 complexes, although the overall
phenomena appear similar within each type at room
10332 J. Org. Chem., Vol. 70, No. 25, 2005

C-C Bond Forming Reactions of C-6 Arylsulfonates
FIGURE 2. 31P{1H} NMR spectra of 1:1 and 2:1 L-1/Pd(OAc)2 complexes in toluene-d8. (A) The 1:1 complex after 1.5 h at room
temperature. (B) The 1:1 complex after 24 h at room temperature. (C) The 2:1 complex after 2 h at room temperature. (D) The 2:1
complex after 21 h at room temperature.
SCHEME 1. Plausible Complexes that Can Be Formed from L-1 and Pd(OAc)2
temperature and at 40 °C. The data indicate that there
are discrete entities possible depending upon the L-1/
Pd(OAc)2 stoichiometry (Scheme 1). Complex A, a 1:1
species, produces the broad resonance at 46-47 ppm and
is the dominant species formed within 24 h at 40 °C. The
1H NMR spectrum of this complex indicates 8 aromatic
and 25 aliphatic protons. This would be consistent with
a cyclopalladated species such as that shown in Scheme
1. Although current experiments are not indicative of the
oxidation state of Pd in the species, it is most likely Pd-
(II) with coordinated acetate and this is also consistent
with the integration in the aliphatic region of the 1H
NMR. Additionally, the signal at ∼50 ppm in the 31P
NMR is consistent with the phosphine oxide as deter-
mined by a spiking experiment. Since the phosphine
oxide was formed slowly and was a minor component, it
is possible that phosphine mediated reduction of Pd(II)
to Pd(0) is not very rapid, and this is also indicative that
A is possibly a Pd(II) complex. A similar cyclopalladated
Pd(II) species from di(tert-butylphosphino)biphenyl is
known where acetate is coordinated with the metal.13
Complex B, on the other hand, is an unsymmetrical
2:1 complex based upon the chemically distinct phospho-
rus resonances (ABquart δA ) 25.74 ppm, δB ) 30.60 ppm,
JAB ) 370 Hz). The magnitude of the coupling constant
is suggestive of a trans relationship between the phos-
phorus atoms. As observed by NMR, during the formation
of the 1:1 complex, some 2:1 product is also initially
formed (appearance of the ABquart even at 40 °C). This
accounts for the absence of free phosphine in this case.
Since the ABquart diminishes and disappears, one possible
explanation is that the 2:1 complex converts to the 1:1
complex irreversibly, over time. During the formation of
the 2:1 complex, some 1:1 product is formed as a minor
component accounting for a signal at -13.33 ppm and
the broad resonance at 46.55 ppm. However, the 2:1
complex appears quite stable in solution through the
(13) Zim, D.; Buchwald, S. L. Org. Lett. 2003, 5, 2413-2415.
J. Org. Chem, Vol. 70, No. 25, 2005 10333

Lakshman et al.
continued presence of the ABquart as the major 31P
resonance at both room temperature and at 40 °C.
Since reactions were conducted at a 2:1 L-1/Pd(OAc)2
ratio, complex B is a species of interest. Again, although
present results do not provide information on the oxida-
tion state of Pd, slow formation of the phosphine oxide
may indicate B to be a Pd(II) complex with coordinated
acetate. Since Pd(0) species with η1 as well as ipso carbon
coordination are known in the literature,14,15 entities such
as B-1 and B-2 could be plausible candidates for the 2:1
complex as these possess chemically distinct, trans
coordinated phosphines that could display an ABquart
pattern in the 31P NMR as seen in this study.16
reported aqueous phase C-C cross coupling reactions at
the C-8 position of purine nucleosides,19 there are no
other reports of such cross coupling reactions under am-
bient conditions. In the present case, aqueous conditions
are precluded due to the hydrolytic lability of the nucleo-
side arylsulfonates. The activation of nucleoside arylsul-
fonates under mild catalysis conditions represents an
advantage in that metal catalyzed transformations can
be conducted in a relatively simple and facile manner,
without the need for more sophisticated precursors.20
Experimental Procedures
Thin-layer chromatography was performed on 250 µm silica
plates, and column chromatographic purifications were per-
formed on 200-300 mesh silica gel. The ligands, Pd(OAc)2 and
all other reagents were obtained from commercial sources and
used without further purification. Prior to each reaction,
toluene was freshly distilled from sodium. All nucleoside
arylsulfonates were prepared based upon reported methods.12
1H NMR spectra (500 MHz) and 13C data (125 MHz) were
recorded in deacidified CDCl3 (prepared by percolating the
solvent through a bed of solid NaHCO3 and basic alumina).
Spectra were referenced to residual protonated solvent. 31P
NMR spectra (202.3 MHz) were acquired in toluene-d8 and
were referenced to 85% H3PO4 as an external standard.
Chemical shifts are reported in δ ppm, and coupling constants
are in hertz. Sugar protons are numbered 1′-5′ beginning at
the anomeric carbon and proceeding via the carbon chain to
the primary carbinol carbon. Where possible, proton assign-
ments have been made based upon analogy to known com-
In additional 31P NMR experiments, the 2:1 complex
B was exposed to a stoichiometric amount of PhB(OH)2
and K3PO4. Upon shaking the NMR tube and data
acquisition, changes observed included the formation of
several new species as well as a significant entity that
produced a broad signal at 31 ppm. A similar broad
resonance has been reported for a 2:1 L-1/Pd(0) com-
plex.14
To determine the catalytic viability of various ligand-
palladium complexes that were identified, C-C cross
coupling reactions of 1a were conducted using the 1:1
complex A isolated by forming that species at 40 °C.
Using this complex, the reaction with phenylboronic acid
in PhMe was complete within 15 h, affording 2a in 86%
yield, and that with m-nitrophenylboronic acid in THP
was complete in 24 h at 45 °C, producing 2h in 66% yield.
Since we have been unable to isolate the 2:1 complex,
this was generated as solutions in PhMe and THP for
reactions with phenylboronic and m-nitrophenylboronic
acids, respectively. Use of the preformed 2:1 species for
the coupling of 1a with phenylboronic and m-nitrophe-
nylboronic acids, in PhMe and THP, respectively, led to
complete reaction in each case with isolated yields of 90%
for 2a (15 h at room temperature) and 78% for 2h (18 h
at 45 °C). These experiments seem to suggest that
although the 1:1 complex could have reasonable catalytic
activity, a 2:1 L-1/Pd(OAc)2 stoichiometry may be impor-
tant for higher activity. Results of experiments aimed at
greater mechanistic understanding of these reactions and
detailed structural evaluations of Pd complexes, as well
as other applications of the room temperature cross
coupling, will be forthcoming.
1
pounds and H-1H COSY data of representative examples.
3′,5′-Bis-O-(tert-butyldimethylsilyl)-O6-[(4-methylphe-
nyl)sulfonyl]-2′-deoxyguanosine (1c).12b (Typical Proce-
dure for Synthesis of Nucleoside Arylsulfonates). In an
oven-dried, round-bottomed flask equipped with a stir bar was
placed 3′,5′-bis-O-(tert-butyldimethylsilyl)-2′-deoxyguanosine
(500 mg, 1.008 mmol). CH2Cl2 (16 mL) was added, and the
mixture was cooled to 0 °C. Et3N (0.3 mL, 2 molar equiv),
DMAP (30 mg, 24.5 µmol), and p-TsCl (400 mg, 2.098 mmol)
were added. The mixture was brought to room temperature
and left stirring at room temperature under nitrogen. Upon
completion, the reaction mixture was diluted with additional
CH2Cl2 and washed with saturated aq NaHCO3 and brine.
Evaporation of the filtrate provided the crude product that was
triturated with 1:1 ether-hexanes to yield 1c as a white
powder (72%). Data for this compound have been reported in
the literature.12b Compounds that were homogeneous by 1H
NMR were routinely obtained by this method.
3′,5′-Bis-O-(tert-butyldimethylsilyl)-O6-[(2,4,6-trimeth-
ylphenyl)sulfonyl]-2′-deoxyguanosine (1a).12a White pow-
In recent years, metal mediated C-C and C-N bond
forming reactions have generated much interest due
to the high biological value of modified nucleosides. Via
this approach, our own research17 as well as those of
others18 has begun to provide facile access to novel
structural entities. With the exception of the recently
1
der (84%), Rf (10% EtOAc in CH2Cl2) ) 0.65. H NMR: 7.98
(s, 1H, Ar-H), 6.97 (s, 2H, Ar-H), 6.28 (t, 1H, 1′, J ) 6.4),
4.82 (br s, 2H, NH2), 4.57 (m, 1H, 3′), 3.97 (app q, 1H, 4′, J ∼
3.4), 3.79 (dd, 1H, 5′, J ) 4.4, 11.2), 3.74 (dd, 1H, 5′, J ) 2.9,
11.2), 2.75 (s, 6H, CH3), 2.53 (app quint, 1H, 2′, J ) 6.8, 6.8,
13.2), 2.34 (ddd, 1H, 2′, J ) 3.4, 6.4, 13.2), 2.31 (s, 3H, CH3),
0.91, 0.90 (2s, 18H, SiC(CH3)3), 0.096, 0.070, 0.061 (3s, 12H,
SiCH3). 13C NMR: 159.1, 156.3, 155.6, 144.5, 141.0, 140.7,
133.2, 132.3, 117.3, 88.5, 84.5, 72.6, 63.5, 41.5, 26.6, 26.4, 23.4,
21.8, 19.1, 18.7, -4.0, -4.1, -4.7, -4.8. HRMS calcd for
C31H52N5O6SSi2 (M+ + H) 678.3177, found 678.3182.
(14) Reid, S. M.; Boyle, R. C.; Mague, J. T.; Fink, M. J. J. Am. Chem.
Soc. 2003, 125, 7816-7817.
(15) (a) Kocˇovsky´, P.; Vyskocˇil, S.; C´ısarˇova´, I.; Sejbal, J.; Tisˇle-
rova´,I.; Smrcˇina, M.; Lloyd-Jones, G. C.; Stephen, S. C.; Butts, C. P.;
Murray, M.; Langer, V. J. Am. Chem. Soc. 1999, 121, 7714-7715. (b)
Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2004, 43, 1871-1876. (c) Barder, T. E.; Walker, S. D.;
Martinelli, J. R.; Buchwald, S. L. J. Am. Chem. Soc. 2005, 127, 4685-
4696.
(18) For other reviews, please see: (a) Hocek, M. Eur. J. Org. Chem.
2003, 245-254. (b) Agrofoglio, L. A.; Gillaizeau, I.; Saito, Y. Chem.
Rev. 2003, 103, 1875-1916.
(16) The reviewers are thanked for their insightful comments. In
particular, one reviewer pointed out details regarding the structures
shown in Scheme 1 as well as matters that helped the discussion of
the complexes.
(17) Reviewed in: (a) Lakshman, M. K. J. Organomet. Chem. 2002,
653, 234-251. (b) Lakshman, M. K. Curr. Org. Synth. 2005, 2, 83-
112.
(19) Western, E. C.; Daft, J. R.; Johnson, E. M., II; Gannett, P. M.;
Shaughnessy, K. M. J. Org. Chem. 2003, 68, 6767-6774.
(20) For some examples of cross coupling reactions involving iodo,
azolyl, fluoro, alkylsulfanyl, and alkylsufonyl nucleoside derivatives,
please see: (a) Liu, J.; Janeba, Z.; Robins, M. J. Org. Lett. 2004, 6,
2917-2919. (b) Liu, J.; Robins, M. J. Org. Lett. 2004, 6, 3421-3423.
(c) Liu, J.; Robins, M. J. Org. Lett. 2005, 7, 1149-1151.
10334 J. Org. Chem., Vol. 70, No. 25, 2005

C-C Bond Forming Reactions of C-6 Arylsulfonates
3′,5′-Bis-O-(tert-butyldimethylsilyl)-O6-[(2-nitrophenyl)
sulfonyl]-2′-deoxyguanosine (1b). Yellow powder (48%), Rf
purified by column chromatography on silica gel using CH2-
Cl2 followed by 3% acetone in CH2Cl2.
Characterization data for 2a-d, 2f, 2h, 2i, and 2k have
been described previously,7,9 and data for new compounds are
listed.
1
(10% EtOAc in CH2Cl2) ) 0.51. H NMR: 8.32 (dd, 1H, Ar-
H, J ) 1.0, 7.8), 8.02 (s, 1H, Ar-H), 7.85-7.75 (m, 3H, Ar-
H), 6.29 (t, 1H, 1′, J ) 6.6), 5.03 (br s, 2H, NH2), 4.57 (m, 1H,
3′), 3.98 (app q, 1H, 4′, J ∼ 3.5), 3.80 (dd, 1H, 5′, J ) 3.9, 11.2),
3.74 (dd, 1H, 5′, J ) 3.4, 11.2), 2.54 (app quint, 1H, 2′, J )
5.9, 5.9, 13.2), 2.36 (ddd, 1H, 2′, J ) 3.9, 6.4, 13.2), 0.91, 0.90
2-Amino-6-[4-(phenoxy)phenyl]-9-[3,5-bis-O-(tert-bu-
tyldimethylsilyl)-â-D-erythro-pentofuranosyl]purine (2e).
Pale cream-colored foam, Rf (4% MeOH/16% CH2Cl2/80%
(2s, 18H, SiC(CH3)3), 0.098, 0.074, 0.066 (3s, 12H, SiCH3). 13
C
hexanes) ) 0.29. H NMR: 8.67 (d, 2H, Ar-H, J ) 8.8), 8.05
1
NMR: 158.4, 156.2, 154.0, 148.7, 140.7, 135.1, 132.1, 132.0,
130.8, 124.8, 116.7, 87.9, 84.0, 71.9, 62.8, 41.0, 26.0, 25.8, 18.4,
18.0, -4.7, -4.8, -5.4, -5.5. HRMS calcd for C28H45N6O8SSi2
(M+ + H) 681.2558, found 681.2537.
(s, 1H, Ar-H), 7.38-7.35 (m, 2H, Ar-H), 7.16-7.07 (m, 5H,
Ar-H), 6.40 (t, 1H, 1′, J ) 6.6), 4.98 (br s, 2H, NH2), 4.62 (m,
1H, 3′), 4.01 (app q, 1H, 4′, J ∼ 3.6), 3.82 (dd, 1H, 5′, J ) 4.4,
11.2), 3.77 (dd, 1H, 5′, J ) 3.4, 11.2), 2.65 (ddd, 1H, 2′, J )
5.9, 6.8, 12.7), 2.39 (ddd, 1H, 2′, J ) 3.4, 5.9, 12.7), 0.93, 0.91
(2s, 18H, SiC(CH3)3), 0.12, 0.083, 0.080 (3s, 12H, SiCH3). 13C
NMR: 159.6, 159.4, 156.6, 155.2, 153.9, 139.7, 131.4, 130.8,
129.8, 125.8, 123.7, 119.5, 118.3, 87.8, 83.6, 72.1, 62.9, 40.7,
26.0, 25.8, 18.4, 18.0, -4.7, -4.8, -5.3, -5.5. HRMS calcd for
C34H50N5O4Si2 (M+ + H) 648.3401, found 648.3404.
2-Amino-6-[4-(diphenylamino)phenyl]-9-[3,5-bis-O-(tert-
butyldimethylsilyl)-â-D-erythro-pentofuranosyl]purine
(2g). Yellow foam, Rf (5% EtOAc in CH2Cl2) ) 0.33. 1H NMR:
8.51 (d, 2H, Ar-H, J ) 8.8), 8.02 (s, 1H, Ar-H), 7.29-7.28
(m, 4H, Ar-H), 7.17-7.14 (m, 6H, Ar-H), 7.07 (t, 2H, Ar-H,
J ) 7.5), 6.39 (t, 1H, 1′, J ) 6.6), 4.95 (br s, 2H, NH2), 4.62 (m,
1H, 3′), 4.00 (app q, 1H, 4′, J ∼ 3.6), 3.82 (dd, 1H, 5′, J ) 4.4,
11.2), 3.77 (dd, 1H, 5′, J ) 3.4, 11.2), 2.65 (ddd, 1H, 2′, J )
5.9, 6.8, 13.2), 2.39 (ddd, 1H, 2′, J ) 3.4, 5.9, 13.2), 0.93, 0.91
(2s, 18H, SiC(CH3)3), 0.12, 0.079, 0.076 (3s, 12H, SiCH3). 13C
NMR: 159.4, 155.6, 153.8, 150.1, 147.2, 139.4, 130.6, 129.3,
129.1, 125.7, 125.2, 123.6, 122.0, 87.7, 83.6, 72.1, 62.9, 40.6,
26.0, 25.8, 18.4, 18.0, -4.7, -4.8, -5.3, -5.5. HRMS calcd for
C40H55N6O3Si2 (M+ + H) 723.3874, found 723.3881.
2-Amino-6-[3-(ethoxycarbonyl)phenyl]-9-[3,5-bis-O-(tert-
butyldimethylsilyl )-â-D-erythro-pentofuranosyl]purine
(2j). White foam, Rf (4% MeOH/16% CH2Cl2/80% hexanes) )
0.24. 1H NMR: 9.25 (t, 1H, Ar-H, J ) 1.5), 8.92 (td, 1H, Ar-
H, J ) 1.5, 6.4), 8.15 (td, 1H, Ar-H, J ) 1.5, 7.8), 8.10 (s, 1H,
Ar-H), 7.60 (t, 1H, Ar-H, J ) 7.8), 6.40 (t, 1H, 1′, J ) 6.6),
5.04 (br s, 2H, NH2), 4.63 (m, 1H, 3′), 4.43 (q, 2H, J ) 7.3),
4.01 (app q, 1H, 4′, J ∼ 3.7), 3.83 (dd, 1H, 5′, J ) 4.4, 11.2),
3.77 (dd, 1H, 5′, J ) 3.4, 11.2), 2.65 (ddd, 1H, 2′, J ) 5.9, 6.8,
13.2), 2.40 (ddd, 1H, 2′, J ) 3.4, 5.9, 13.2), 1.43 (t, 3H, CH3, J
) 7.3), 0.93, 0.92 (2s, 18H, SiC(CH3)3), 0.12, 0.088, 0.083 (3s,
12H, SiCH3). 13C NMR: 166.5, 159.5, 154.7, 154.1, 140.2, 136.3,
134.2, 131.4, 130.9, 130.3, 128.5, 126.2, 87.8, 83.6, 72.1, 62.9,
61.1, 40.7, 26.0, 25.8, 18.4, 18.0, 14.4, -4.7, -4.8, -5.4, -5.5.
HRMS calcd for C31H50N5O5Si2 (M+ + H) 628.3351, found
628.3344.
3′,5′-Bis-O-(tert-butyldimethylsilyl)-O6-[(2,4,6-triiso-
propylphenyl)sulfonyl]-2′-deoxyguanosine (1d). White
foam (60%), Rf (10% EtOAc in CH2Cl2) ) 0.51. 1H NMR: 7.97
(s, 1H, Ar-H), 7.19 (s, 2H, Ar-H), 6.29 (t, 1H, 1′, J ) 6.6),
4.85 (br s, 2H, NH2), 4.57 (m, 1H, 3′), 4.31 (sept, 2H, CH, J )
6.7), 3.98 (app q, 1H, 4′, J ∼ 3.3), 3.80 (dd, 1H, 5′, J ) 3.9,
11.2), 3.74 (dd, 1H, 5′, J ) 2.9, 11.2), 2.91 (sept, 1H, CH, J )
6.8), 2.55 (app quint, 1H, 2′, J ) 5.9, 5.9, 13.2), 2.34 (ddd, 1H,
2′, J ) 3.4, 5.9, 13.2), 1.28-1.25 (overlapping d, 18H, (CH3)2),
0.91, 0.89 (2s, 18H, SiC(CH3)3), 0.096, 0.066, 0.054 (3s, 12H,
SiCH3). 13C NMR: 158.4, 155.6, 155.2, 154.2, 150.9, 139.9,
131.5, 123.8, 116.7, 87.9, 83.9, 72.0, 62.8, 40.9, 34.3, 29.8, 25.9,
25.8, 24.6, 23.5, 18.4, 18.0, -4.7, -4.8, -5.4, -5.5. HRMS calcd
for C37H64N5O6SSi2 (M+ + H) 762.4116, found 762.4083.
3′,5′-Bis-O-(tert-butyldimethylsilyl)-O6-[(4-methoxyphe-
nyl)sulfonyl]-2′-deoxyguanosine (1e). White powder (64%),
Rf (10% EtOAc in CH2Cl2) ) 0.32. 1H NMR: 8.08 (d, 2H, Ar-
H, J ) 9.1), 7.99 (s, 1H, Ar-H), 7.01 (d, 2H, Ar-H, J ) 9.1),
6.29 (t, 1H, 1′, J ) 6.6), 4.98 (br s, 2H, NH2), 4.57 (m, 1H, 3′),
3.97 (app q, 1H, 4′, J ∼ 3.4), 3.89 (s, 3H, OCH3), 3.80 (dd, 1H,
5′, J ) 4.4, 11.2), 3.74 (dd, 1H, 5′, J ) 3.4, 11.2), 2.54 (ddd,
1H, 2′, J ) 5.9, 6.8, 13.2), 2.34 (ddd, 1H, 2′, J ) 3.4, 5.9, 13.2),
0.91, 0.89 (2s, 18H, SiC(CH3)3), 0.099, 0.072, 0.064 (3s, 12H,
SiCH3). 13C NMR: 164.5, 158.9, 156.0, 154.8, 140.4, 131.6,
128.3, 117.2, 114.4, 88.1, 84.1, 72.2, 63.0, 56.0, 41.1, 26.2, 26.0,
18.7, 18.2, -4.4, -4.6, -5.2, -5.3. HRMS calcd for C29H48N5O7-
SSi2 (M+ + H) 666.2813, found 666.2821.
3′,5′-Bis-O-(tert-butyldimethylsilyl)-O6-[(2-thienyl)sul-
fonyl]-2′-deoxyguanosine (1f). White powder (62%), Rf (10%
1
EtOAc in CH2Cl2) ) 0.37. H NMR: 8.01 (s, 1H, Ar-H), 7.97
(dd, 1H, Ar-H, J ) 1.5, 3.9), 7.75 (dd, 1H, Ar-H, J ) 1.5,
4.9), 7.14 (dd, 1H, Ar-H, J ) 3.9, 4.9), 6.30 (t, 1H, 1′, J )
6.6), 5.03 (br s, 2H, NH2), 4.57 (m, 1H, 3′), 3.98 (app q, 1H, 4′,
J ∼ 3.4), 3.80 (dd, 1H, 5′, J ) 4.4, 11.2), 3.75 (dd, 1H, 5′, J )
2.9, 11.2), 2.54 (app quint, 1H, 2′, J ) 6.4, 6.4, 13.2), 2.36 (ddd,
1H, 2′, J ) 3.7, 6.4, 13.2), 0.91, 0.90 (2s, 18H, SiC(CH3)3), 0.099,
0.076, 0.068 (3s, 12H, SiCH3). 13C NMR: 158.4, 155.9, 154.3,
140.4, 136.3, 136.1, 135.0, 127.3, 116.8, 87.9, 83.9, 71.9, 62.8,
41.0, 25.9, 25.7, 18.4, 18.0, -4.7, -4.8, -5.4, -5.5. HRMS calcd
for C26H44N5O6S2Si2 (M+ + H) 642.2272, found 642.2286.
Typical Procedure for the Cross Coupling of 1a with
Arylboronic Acids. In an oven-dried, screw-cap vial equipped
with a stir bar were placed Pd(OAc)2 (1.3 mg, 5.8 µmol), L-1
(4.1 mg, 11.7 µmol), K3PO4 (25 mg, 117 µmol), the nucleoside
arylsulfonate 1a (40 mg, 58.9 µmol), and the arylboronic acid
(2 molar equiv). THP or PhMe (0.7 mL) was added, the vial
was flushed with nitrogen and sealed with a Teflon-lined cap,
and the mixture was allowed to stir at room temperature.
Upon completion, the reaction mixture was filtered through
Celite, and the residue was washed with CH2Cl2. Evaporation
of the filtrate provided the crude product mixtures that were
Acknowledgment. Support for this work by NSF
Grant CHE-0314326 and a PSC-CUNY 36 award is
gratefully acknowledged. Acquisition of a 500 MHz
NMR spectrometer was funded by NSF Grant CHE-
0210295. We thank Prof. Kevin Shaughnessy (Univer-
sity of Alabama) for helpful discussions.
Supporting Information Available: Proton NMR spectra
of 1a, 1b, 1d, 1e, 1f, 2e, 2g, and 2j and 31P{1H} NMR spectra
of 1:1 and 2:1 L-1/Pd(OAc)2 complexes obtained at 40 °C. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO0513764
J. Org. Chem, Vol. 70, No. 25, 2005 10335