15N double-labeled guanosine from inosine through ring-opening-ring-closing and one-pot Pd-catalyzed C-O and C-N cross-coupling reactions

Article abstract of DOI:10.1021/jo100808w

(Figure Presented) [N,1-¹⁵N₂]-Guanosine, or [1,NH₂-¹⁵N₂]-guanosine, and derivatives were prepared from tri-O-acetylinosine, via N-nitration and reaction with ¹⁵NH₂OH, followed by conversion of the ¹⁵N-labeled 1-hydroxyinosine to the corresponding 2,6-dichloropurine riboside. The sequential one-pot C-O and C-N key couplings of this dichloro derivative with PhCH₂OH and PhCO¹⁵NH₂ or ⁱPrCO¹⁵NH₂ was achieved in good overall yields, with Pd(0)-Xantphos as the best choice of five different catalytic systems examined.

Full text of DOI:10.1021/jo100808w

pubs.acs.org/joc
ANRORC processes.3 The process could be applied to
15N Double-Labeled Guanosine from Inosine
through Ring-Opening-Ring-Closing and
One-Pot Pd-Catalyzed C-O and C-N
Cross-Coupling Reactions
inosines,4 from which we also obtained 15N double-labeled
adenosines. When N-nitration failed (dI series), the 2-nitro-
benzenesulfonyl group (Ns) was the best alternative.5,6 The
overall process is summarized in Scheme 1, where asterisks
on N (N* for internal labels, Nf for amino groups) mean
g98% of 15N at the indicated position(s). The pros of this
general procedure are the following: all internal labels were
introduced at room temperature (rt); only 1.1 equiv of 15N-
labeled reagents was employed in all labeling steps; and
sugar-modified nucleosides and appropriately C-substituted
nucleobases may be amenable to it. The con is that we were
unable to convert [1-15N]-inosines to the double-labeled
[N,1-15N2]-guanosines efficiently, in a short number of steps
(not involving the degradation of inosine or ex-novo syn-
thesis7). Due to this, G*f and dG*f are lacking in Scheme 1.
Joaquim Caner and Jaume Vilarrasa*
´
ꢀ
´
Departament de Quımica Organica, Facultat de Quımica,
Universitat de Barcelona, Av. Diagonal 647, 08028 Barcelona,
Catalonia, Spain
jvilarrasa@ub.edu
Received April 25, 2010
SCHEME 1. 15N Labeling of Nucleosides by RORC2,4,5
[N,1-15N2]-Guanosine, or [1,NH2-15N2]-guanosine, and
derivatives were prepared from tri-O-acetylinosine, via
N-nitration and reaction with 15NH2OH, followed by
conversion of the 15N-labeled 1-hydroxyinosine to the
corresponding 2,6-dichloropurine riboside. The sequen-
tial one-pot C-O and C-N key couplings of this
dichloro derivative with PhCH2OH and PhCO15NH2 or
iPrCO15NH2 was achieved in good overall yields, with
Pd(0)-Xantphos as the best choice of five different
catalytic systems examined.
Jones et al.,8 by means of a Dimroth rearrangement from
1-alkoxy adenosines, prepared a series of 15N and 15N,13C-
multilabeled guanosines. In connection with our objective of
achieving all the 15N2-natural nucleosides in the most effi-
cient way, we report here an alternative approach to guano-
sines labeled both at N1 and on amino N, based on our
Nucleosides labeled with 15N at specific positions and,
thus, selectively labeled oligonucleotides, DNAs, and RNAs
have provided key information (such as distinguishing the
15N-H protons and characterizing 15N-H 15N hydrogen
3 3 3
bonds by 1H and 15N NMR) on nucleic acid structures and
nucleic acid-protein interactions.1
(4) (a) Ariza, X. Ph.D. Thesis, Universitat de Barcelona, 1995. (b) Ariza, X.;
Bou, V.; Vilarrasa, J. J. Am. Chem. Soc. 1995, 117, 3665. (c) Terrazas, M.; Ariza,
ꢀ
X.; Farras, J.; Guisado-Yang, J. M.; Vilarrasa, J. J. Org. Chem. 2004, 69, 5473.
In the 1990s, taking advantage of the unexpectedly easy
N-nitration of pyrimidine nucleosides, we discovered a new
reaction in which the addition of the nucleophile 15NH3 was
followed by ring-opening-ring-closing (RORC) steps,2
which showed some parallelism with other SNRORC or
(5) (a) Terrazas, M.; Ariza, X.; Vilarrasa, J. Org. Lett. 2005, 7, 2477.
(b) Terrazas, M.; Ariza, X.; Vilarrasa, J. Tetrahedron Lett. 2005, 46, 5127.
ꢀ
(c) For a related work, see: Terrazas, M.; Ariza, X.; Farras, J.; Vilarrasa, J.
Chem. Commun. 2005, 3968.
(6) Use of N-2,4-dinitrophenyl dI (Scheme 1, EWG = DNP) for the same
purpose has been reported: (a) De Napoli, L.; Messere, A.; Montesarchio,
D.; Piccialli, G. J. Org. Chem. 1995, 60, 2251. (b) De Napoli, L.; Messere, A.;
Montesarchio, D.; Piccialli, G.; Varra, M. J. Chem. Soc., Perkin Trans. 1
1997, 2079. (c) Catalanotti, B.; De Napoli, L.; Galeone, A.; Mayol, L.;
Oliviero, G.; Piccialli, G.; Varra, M. Eur. J. Org. Chem. 1999, 2235. We also
examined the case where EWG = COOtBu in the inosine series, but
nucleophiles tend to attack at the C atom of the carboxyl group (with
deprotection) rather than the C2 (or C4) carbon atoms (no RORC process).
(7) For example, from 15N-labeled AICA-riboside: (a) Bleasdale, C.;
Ellwood, S. B.; Golding, B. T.; Slaich, P. K.; Taylor, O. J.; Watson, W. P.
J. Chem. Soc., Perkin Trans. 1 1994, 2859. (b) Abad, J. L.; Gaffney, B. L.;
Jones, R. A. J. Org. Chem. 1999, 64, 6575.
(1) For recent, representative applications, see: (a) Bdour, H. M.; Kao,
J. L.-F.; Taylor, J.-S. J. Org. Chem. 2006, 71, 1640. (b) Dingley, A. J.; Nisius,
L.; Cordier, F.; Grzesiek, S. Nat. Protoc. 2008, 3, 242. (c) Wang, W.; Zhao, J.;
Han, Q.; Wang, G.; Yang, G.; Shallop, A. J.; Liu, J.; Gaffney, B. L.; Jones,
R. A. Nucleosides, Nucleotides Nucleic Acids 2009, 28, 424. (d) Baral, B.;
Kumar, P.; Anderson, B. A.; Ostergaard, M. E.; Sharma, P. K.; Hrdlicka,
P. J. Tetrahedron Lett. 2009, 50, 5850.
(2) (a) Bou, V. Ph.D. Thesis, Universitat de Barcelona, 1992. (b) Ariza, X.;
Bou, V.; Vilarrasa, J.; Tereshko, V.; Campos, J. L. Angew. Chem., Int. Ed. 1994,
ꢀ
33, 2454. (c) Ariza, X.; Farras, J.; Serra, C.; Vilarrasa, J. J. Org. Chem. 1997, 62,
(8) (a) Pagano, A. R.; Zhao, H.; Shallop, A. J.; Jones, R. A. J. Org. Chem.
1998, 63, 3213. (b) Shallop, A. J.; Gaffney, B. L.; Jones, R. A. J. Org. Chem.
2003, 68, 8657 and references cited therein.
1547. (d) Ariza, X.; Vilarrasa, J. J. Org. Chem. 2000, 65, 2827.
(3) Review: van der Plas, H. C. Adv. Heterocycl. Chem. 1999, 74, 1.
4880 J. Org. Chem. 2010, 75, 4880–4883
Published on Web 06/15/2010
DOI: 10.1021/jo100808w
r
2010 American Chemical Society

Caner and Vilarrasa
JOCNote
SCHEME 2. From N-Nitroinosines to N-Hydroxyinosines and
to 2,6-Dichloro Derivative 3
TABLE 1. Pd-Catalyzed Couplings of 3 with PhCH2OHa
entry
Pd source (mol %)
none
Pd2dba3 CHCl3, 5
ligand (mol %) time (h) 4, yieldb
1
2
3
4
none
dppf, 15
4.0
2.0
1.0
0.5
0
66
81
85
above-mentioned RORC procedure and on Pd-catalyzed
C-O and C-N bond forming reactions.
3
Pd2dba3 CHCl3, 2.5
Pd2dba3 CHCl3, 5
Xantphos, 7.5
Xantphos, 15
3
3
First, triacetylated N-nitroinosine 1 was prepared in 80%
yield from tri-O-acetylinosine by our N-nitration protocol4,5
(with a large excess of CF3COONO2 at -40 °C). Treatment
of 1 with HONH3þCl- (1.1 equiv) and NaOAc (2.2 equiv) in
1:1 CH3CN-H2O at rt afforded N-hydroxy derivative 2 in
98% yield (Scheme 2), via a polar intermediate (an open
species according to 1H NMR) that disappeared to give the
desired compound within 4 h. The fact that this RORC step
could be carried out with only 1.1 equiv of hydroxylamine
was instrumental in using it as an N1-labeling procedure,9
as otherwise scaling up of the overall process would be
economically unpractical. This indirect N-hydroxylation
procedure is general since it could be applied to other
protected inosines (tri-O-TBS and 50-O-TBS-20,30-O-isopro-
pylidene) in ca. 80% overall yields.
aAt 0.1 M concentrations in toluene at 80 °C. bIsolated yield, in %.
We envisaged replacing both chloride ions of 3 consecu-
tively, by means of appropriate Pd-catalyzed couplings.
Outstanding studies in the nucleoside field have been pub-
lished recently, mainly involving C-N couplings (Buchwald-
Hartwig reactions) of bromo- and iodopurines.12 Although
C-Cl bonds are much less reactive in Pd-catalyzed cou-
plings, new ligands have been developed to deal with aro-
matic chlorides.13
With this background some approaches were soon ruled
out.14 Moreover, when 3 was heated to 80 °C in toluene, with
1.1 equiv of PhCONH2, small amounts of Pd2dba3 CHCl3,
3
diphosphines of different bite angles (dppf,15 DPEphos,16a
or Xantphos16), and Cs2CO3, only the product substituted
on C6 was formed, confirming that position 6 is intrinsically
more reactive. Thus, the substitution at C6 has to be carried
out before that at C2. The SNAr-like replacement of 6-Cl by
different amounts of benzyl alcohol and strong bases gave
the desired 4, although in poor yields; this was mainly due
to the workup difficulties, when an excess of benzyl alcohol
was used, and to the byproducts from double addition and
deacylation reactions.
A sample of the 15N-labeled compound 2* was prepared
by this protocol, that is, from 1 and only 1.05 equiv of
HO15NH3þCl- (obtained by us from reduction of Na15NO2,10
but also commercially available with 98% 15N).
We took advantage of a known reaction of purine N-
oxides11 for the conversion of N-hydroxyinosine 2 to the
corresponding 2,6-dichloropurine nucleoside, 3 (Scheme 2).
This involves heating with a large excess of POCl3 and 2,6-
lutidine or 2-picoline.12 The same reaction did not work with
POBr3 (and only the bromination of position 6 took place
with POBr3 and N,N-diethylaniline in toluene). We applied
this procedure to the conversion of a sample of 2* to 3*.
In this context, we examined the following protocol
(see Table 1): 3 was mixed in toluene with PhCH2OH,
(13) For a review, see: Surry, D. S.; Buchwald, S. L. Angew. Chem., Int.
Ed. 2008, 47, 6338.
(9) Identical treatments of the N-Ns derivative relating to 1(withNsinsteadof
NO2 as EWG) afforded complex mixtures (containing N-deprotected compounds
as well as hydrolyzed and pyrimidine ring-cleaved species) with insignificant
amounts of the desired N-hydroxy compound, 2. With DNP instead of NO2 as
EWG, we needed 5-10 equiv of NH2OH and heating to convert 1 to 2.
(10) (a) Rajendran, G.; Van Ettern, R. L. Inorg. Chem. 1986, 25, 876.
(b) Tao, T.; Alemany, L. B.; Parry, R. J. Org. Lett. 2003, 5, 1213.
(11) (a) Kawashima, H.; Kumashiro, I. Bull. Chem. Soc. Jpn. 1967, 40,
639. (b) Robins, M. J.; Uznanski, B. Can. J. Chem. 1981, 59, 2601. (c) We
could also have arrived at 2* from [1-15N]-adenosine, through its N-oxide,
followed by a deamination-hydroxylation reaction according to the procedure
of Robins and Uznanski. However, it was preferable to enter the internal label as
late as possible. Also, we were interested in examining the performance of the
reaction of N-nitro nucleosides with NH2OH.
(12) (a) Piguel, S.; Legraverend, M. J. Org. Chem. 2007, 72, 7026 and
references cited therein. (b) Vandromme, L.; Legraverend, M.; Kreimerman,
S.; Lozach, O.; Meijer, L.; Grierson, D. S. Bioorg. Med. Chem. 2007, 15, 130.
(c) Li, X.; Vince, R. Bioorg. Med. Chem. 2006, 14, 5742. For C-N couplings
of nucleoside arylsulfonates with aromatic amines, see: (d) Gunda, P.;
Russon, L. M.; Lakshman, M. K. Angew. Chem., Int. Ed. 2004, 43, 6372.
For alternative routes via SNAr reactions, see: (e) Liu, J.; Robins, M. J. J. Am.
Chem. Soc. 2007, 129, 5962. (f) Kamaike, K.; Kayama, Y.; Isobe, M.;
Kawashima, E. Nucleosides, Nucleotides Nucleic Acids 2006, 25, 29 and
references cited therein.
(14) (a) Substitution of OH for the more reactive 6-Cl by means of a simple
treatment with KOH (1 M in CH3CN-H2O) at rt afforded fully deprotected
2-chloroinosine. This compound can be converted to guanosine, but under very
harsh conditions, probably because the anion is formed and is very reluctant to
undergo a second substitution reaction. (b) For an example of a substitution, with a
big excess of NH3/MeOH in a sealed tube at 145 °C for 72 h, see: Nord, L. D.;
Dalley, N. K.; McKernan, P. A.; Robins, R. K. J. Med. Chem. 1987, 30, 1044.
(15) Recent review: Fihri, A.; Meunier, P.; Hierso, J.-C. Coord. Chem.
Rev. 2007, 251, 2017.
(16) For a very recent review on diphosphines, see: (a) Birkholz, M.-N.;
Freixa, Z.; van Leeuwen, P. W. N. M. Chem. Soc. Rev. 2009, 38, 1099.
Xantphos preparation: (b) Hillebrand, S.; Bruckmann, J.; Krueger, C.;
Haenel, M. W. Tetrahedron Lett. 1995, 36, 75. (c) Kranenburg, M.; van
der Burgt, Y. E. M.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Goubitz, K.;
Fraanje, J. Organometallics 1995, 14, 3081. Pd2dba3 CHCl3 and Xantphos
3
in a 1:3 molar ratio, and 140 mol % of Cs2CO3 (in 1,4-dioxane at 100 °C),
were recommended for intermolecular amidations of ArBr/ArOTf/ArI:
(d) Yin, J.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 6043. (e) Yin, J.;
Buchwald, S. L. Org. Lett. 2000, 2, 1101. For the effect of bidentate ligands,
see: (f) Fujita, K.; Yamashita, M.; Puschmann, F.; Alvarez-Falcon, M. M.;
Incarvito, C. D.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128, 9044. Also see
ref 12a. Pd(0)/Xantphos was also shown to be useful for the N-arylation of
amino groups of nucleosides: (g) Ngassa, F. N.; DeKorver, K. A.; Melistas,
T. S.; Yeh, E. A.-H.; Lakshman, M. K. Org. Lett. 2006, 8, 4613.
J. Org. Chem. Vol. 75, No. 14, 2010 4881

Caner and Vilarrasa
SCHEME 3. One-Pot Conversion of 3 to 5a and 5b and
JOCNote
a
TABLE 2. Pd-Catalyzed Couplings of 4 with PhCONH2
Cleavage of the O-CH2Ph Bond To Give 6a and 6b
iPrCONH2) via a sequential one-pot reaction (see Scheme 3).18
We added 4 and 105 mol % of PhCH2OH to a reaction flask
containing the Pd(0)/Xantphos/Cs2CO3 mixture in toluene,
which was heated to 80 °C. After ca. 1 h (when TLC indicated
that 3 had disappeared), 110 mol % of PhCONH2 and a
second charge of Pd/Xantphos/Cs2CO3 were added, and a
gentle reflux was then maintained for 1 h. We noted that this
protocol afforded higher conversions and fewer byproduct
than in experiments with the full amount of catalyst/ligand/
base introduced from the beginning. We also saved Pd and
diphosphine. Yields of 70-73% of 5a were isolated by
column chromatography.19 With iPrCONH2 instead of
PhCONH2, 5b was similarly obtained.
entry
Pd source (mol %)
none
Pd2dba3 CHCl3, 5
ligand (mol %)
none
dppf, 15
DPEphos, 15
Xantphos, 15
time 5a, yieldb
1
2
3
4
5
6
7
4 h
4 h
0
0
3
Pd2dba3 CHCl3, 5
Pd2dba3 CHCl3, 5
80 min 64 (74)c
60 min 68 (77)c
3
3
Pd2dba3 CHCl3, 2 ꢀ 2.5 Xantphos, 2 ꢀ 2.5 40 min 72
3
Pd2dba3 CHCl3, 5
X-Phos:palladacycle, 10
X-Phos, 15
80 min 60
90 min 54
3
aAt 0.1 M concentrations in refluxing toluene. bIsolated yield, in %.
cYields with 10 mol % of Pd2dba3 CHCl3 are in parentheses.
3
Pd2dba3 CHCl3, dppf or Xantphos, and Cs2CO3 and heated
3
The removal of the benzyl group of 5a and 5b (H2 balloon,
Scheme 3) gave 6a and 6b, respectively, in nearly 70% yields
(overall for three steps).20
to 80 °C. With Cs2CO3 but without the catalyst (entry 1)
no substitution took place. With all the additives, the 6-O-
CH2Ph derivative, 4, was formed quite rapidly with both
diphosphines (entries 2-4). The conditions of choice turned
out to be those of entry 3 (2.5 mol % of catalyst, 0.05 equiv of
Pd), since the addition of twice the amount of catalyst and
ligand (entry 4) improved the yield only slightly. DPEphos
gave similar results to Xantphos, while reactions with
With all these “preliminary” studies and results in hand,
we carried out the reactions of interest. From 3 or 3*, we
achieved the series of mono- and double-labeled 5a and 5b,
6a and 6b, and guanosines (7) shown in Scheme 4, by using
PhCO15NH2 (prepared from PhCOCl, 15NH4Cl, and KOH
in CH3CN-H2O in 97% yield) and iPrCO15NH2, similarly
prepared, in the C-N cross-coupling step. In short, 15N-
labeled guanosines with the standard amino protecting
group for oligonucleotide synthesis (iPrCO, see the 6b series)
can be prepared efficiently as well.
X-Phos (7.5 mol %, with 2.5 mol % of Pd2dba3 CHCl3)
3
were too slow (data not included in Table 1 to save space).
We then investigated the best phosphine ligand for the
replacement of the second Cl (the Cl of 4) by benzamide
(PhCONH2) as an ammonia synthetic equivalent. Table 2
summarizes the main results, in refluxing toluene, since the
couplings did not progress at 80 °C. Refluxing 1,4-dioxane
and heating for longer times were counter-indicated with our
substrate; both increased the percentages of decomposition
and deacylation byproducts. Most of the experiments were
carried out three times, ensuring that dppf was inappropriate
(entry 2),17 while DPEphos (entry 3) and especially Xant-
phos (entry 4) afforded the highest yields of 5a. When the
catalyst and ligand were added in two portions (the second
one after 20 min of reaction) the disappearance of 4 was
faster and the yield improved (entry 5). The monodentate
X-Phos (entry 6) and the palladacycle (entry 7) were less
efficient. Thus, Xantphos became our preferred ligand for
these C-O/N couplings, but the low cost of DPEphos should
be taken into account for larger scale reactions.
In summary, we synthesized the desired 15N-labeled gua-
nosines 7* (G*) and 7*f (G*f) in seven steps from natural
inosine (four steps after introducing the first label). Only
inorganic sources of 15N (Na15NO2 or HO15NH3þCl-,
15NH4þCl-) were used and only in nearly equivalent amounts.
G*f has >98% of 15N at both the relevant positions for
examining Watson-Crick interactions. Although the introduc-
tion of internal 15N labels in nucleosides (and in heterocyclic
compounds in general) is usually much more complicated than
in the amino groups, and the substitution of 15NH2OH for
NH2NO2 took place in a remarkable 98% yield with only 1.05
equiv of labeling source (!), we focused our attention on the
optimization of the C-O and mainly the C-N bond formation
reactions, as we had to convert an expensive 15N-labeled inosine
derivative into single and double 15N-labeled guanosines in few
steps and good yields. The sequential, or consecutive, one-pot
Pd-catalyzed C-O and C-N couplings from a heteroaryl
dichloride, reported here for the first time to the best of our
knowledge, allowed us to reduce the amounts of Pd and of our
We were ready to carry out the preparation of 5a (reaction
with PhCONH2) and 5b (coupling with isobutyramide,
(17) However, in our lab, the analogous 2-Br derivative was replaced
efficiently by PhCONH2 with Pd/dppf at 80 °C (Terrazas, M. Ph.D.
Dissertation, Universitat de Barcelona, 2006).
(18) For examples of related one-pot intermolecular C-C couplings, see:
(a) Molander, G. A.; Yokoyama, Y. J. Org. Chem. 2006, 71, 2493. (b) Zhang,
X.; Liu, A.; Chen, W. Org. Lett. 2008, 10, 3849 and references cited therein.
(c) Beaumard, F.; Dauban, P.; Dodd, R. H. Org. Lett. 2009, 11, 1801.
Tandem couplings using two different metallic catalysts are more common.
For combinations of SNAr and Pd-catalyzed reactions, see: (d) Tikad, A.;
Routier, S.; Akssira, M.; Guilllaumet, G. Org. Biomol. Chem. 2009, 7, 5113.
(19) In one-pot C-O and C-N coupling experiments with K3PO4 instead
of Cs2CO3 the coupling reactions did work, but more slowly. With DIPEA no
coupling took place.
(20) We isolated 5a and 5b and subjected them to simple hydrogenolysis
when we might have examined the cleavage of the O-CH2Ph bond in the
same flask (three steps in one pot), but the isolation of 6a and 6b from the
final mixture was judged to be more cumbersome.
4882 J. Org. Chem. Vol. 75, No. 14, 2010

Caner and Vilarrasa
JOCNote
SCHEME 4. Synthesis of 15N-Labeled 5-7
toluene (1.2 mL) and benzyl alcohol (28 μL, 29.3 mg, 0.27
mmol) were added via syringe. The reaction mixture was
magnetically stirred at 80 °C under nitrogen, until the reaction
was complete, as shown by TLC (45 to 75 min depending on the
batch). Afterward, Pd2dba3 CHCl3 (6.5 mg, 6.3 μmol), Xant-
3
phos (11.2 mg, 18.9 μmol), Cs2CO3 (115 mg, 0.35 mmol), and
[15N]-benzamide (33.5 mg, 0.27 mmol) were added. The reac-
tion mixture was then heated at 110 °C under nitrogen, until the
second step was complete (40 to 80 min depending on the batch
of Pd). The heating bath was removed. The flask content was
diluted with 20 mL of CH2Cl2 and filtered through a pad of
Celite. The filtrate was concentrated under vacuum. The
residue was purified by flash chromatography (7:3 CH2Cl2/
EtOAc) to afford 110 mg (73%) of 5a*f as a pale white foam:
1H NMR (CDCl3, 400 MHz) δ 2.04 (s, 3H), 2.10 (s, 3H), 2.15 (s,
3H), 4.43-4.55 (m, 3H), 5.66 (s, 2H), 5.98-6.00 (m, 2H), 6.10
(m, 1H), 7.29-7.38 (m, 3H), 7.49-7.60 (m, 5H), 7.93 (s, 1H),
7.98 (m, 2H), 8.66 (dd, JH-Nf = 88.9 Hz, JH-N* = 2.0 Hz, 1
H); 13C NMR (CDCl3, 100.6 MHz) δ 20.4, 20.5, 20.7, 63.3, 68.9
(d, JC-N* = 3.8 Hz), 70.7, 73.3, 80.1, 87.0, 119.2 (d, JC-N* =
1.5 Hz, C5), 127.5, 128.2, 128.4, 128.6, 128.7, 132.2, 134.6 (d,
JC-Nf = 9.2 Hz), 135.8, 140.3, 152.2 (dd, JC-Nf = 25.2 Hz,
J
C-N* = 5.9 Hz, C2), 152.3 (dd, JC-Nf = 3.4 Hz, JC-N* = 0.7
Hz, C6), 160.9 (dd, JC-N* = 9.1 Hz, JC-Nf = 2.8 Hz, C4), 164.6
(dd, JC-Nf = 13.9 Hz, JC-N* = 1.3 Hz, CO), 169.4, 169.6,
=
170.5; 15N NMR (CDCl3, 40.5 MHz) δ -126.9 (dd, JN*-Nf
6.4 Hz, JN*-H = 2.0 Hz, N*), -208.8 (dd, JNf-H = 88.9 Hz,
JN*-Nf = 6.4 Hz, Nf); HRMS (ESI) calcd for C30H30N3-
15N2O9þ (M þ H)þ 606.1979, found 606.1981.
preferred ligand to date (Xantphos) and to increase the overall
yields.
Acknowledgment. J.C. is grateful for a studentship from
the Universitat de Barcelona (2003-2006), a stipend from
TRIoH (FP7, European Union, contract LHSB-CT-2003-
503480, during 2007), and a current fellowship from the
Experimental Section
[1-15N]-20,30,50-Tri-O-acetyl-1-hydroxyinosine (2*). A solu-
tion of 15NH2OH HCl (116 mg, 1.65 mmol) and NaOAc (271,
3
3.30 mmol) in water (12 mL) was added to a solution of 1
(660 mg, 1.50 mmol) in acetonitrile (12 mL). The reaction
mixture was stirred at rt for 4 h (the pale yellow color faded
and a suspension was formed of a more polar intermediate, as
indicated by TLC, which slowly disappeared). The solution was
partially concentrated (to remove acetonitrile) and CH2Cl2
and brine were added. The layers were separated. The organic
layer was washed with water and dried over Na2SO4. Filtration
and evaporation to dryness gave chromatographically pure
2* (599 mg, 96%) as a yellowish-white foam (spectra in the
Supporting Information).
ꢁ
Fundacio Privada Cellex of Barcelona. The Ministerio de
ꢁ
Educacion y Ciencia provided additional support (SAF2005-
24643-E). Ionela Cialicu carried out experiments with DNP
derivatives of inosines and NH2OH during her DEA studies
ꢀ
(2007). The advice given by Dr. Jaume Farras and Dr. Xavier
Ariza (of our Department) to J.C. during his Ph.D. period is
also acknowledged. This work is dedicated to Prof. Henk C.
van der Plas (Emeritus Prof., University of Wageningen) and
Prof. Piet van Leeuwen (ICIQ, Tarragona).
Representative One-Pot Reaction. A reaction flask was
charged with Pd2dba3 CHCl3 (6.5 mg, 6.3 μmol), Xantphos
Supporting Information Available: Experimental details
and copies of 1H, 13C, and 15N NMR spectra of the new com-
pounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
3
(11.2 mg, 18.9 μmol), [1-15N]-20,30,50-tri-O-acetyl-2,6-chloro-
purine (3*, 112 mg, 0.25 mmol), and Cs2CO3 (115 mg, 0.35
mmol). The flask was purged with nitrogen. Anhydrous
J. Org. Chem. Vol. 75, No. 14, 2010 4883