New ruthenium-based protocol for cleavage of terminal olefins to primary alcohols: Improved synthesis of a bicyclic nucleoside

Article abstract of DOI:10.1021/jo0491861

A new protocol for the oxidative cleavage of terminal alkenes to give exclusively primary alcohols in high yields is introduced. The protocol is based on RuO₄-mediated dihydroxylation, NaIO₄-mediated diol cleavage, and NaBH₄-mediated reduction, but the introduction of a reducing step before the diol cleavage removes the formation of byproducts and improves the yield significantly. The new protocol has been developed and used for the improved preparation of a [3.2.0]bicycloarabinonucleoside with important potential in antisense and antigene technology.

Full text of DOI:10.1021/jo0491861

New Ru th en iu m -Ba sed P r otocol for
Clea va ge of Ter m in a l Olefin s to P r im a r y
Alcoh ols: Im p r oved Syn th esis of a Bicyclic
Nu cleosid e
Pawan K. Sharma‡ and Poul Nielsen*
Nucleic Acid Center,† Department of Chemistry, University
of Southern Denmark, 5230 Odense M, Denmark
pon@chem.sdu.dk
Received May 14, 2004
Abstr a ct: A new protocol for the oxidative cleavage of
terminal alkenes to give exclusively primary alcohols in high
yields is introduced. The protocol is based on RuO4-mediated
dihydroxylation, NaIO4-mediated diol cleavage, and NaBH4-
mediated reduction, but the introduction of a reducing step
before the diol cleavage removes the formation of byproducts
and improves the yield significantly. The new protocol has
been developed and used for the improved preparation of a
[3.2.0]bicycloarabinonucleoside with important potential in
antisense and antigene technology.
F IGURE 1. Original preparation of the bicyclic nucleoside
1.3
stereoisomer, and the double bond was oxidatively cleaved
to give 4. A selective mesylation and ring-closure afforded
after debenzylation 1 in 8.5% yield over the 14 steps.3 A
careful perusal of the reaction sequence reveals that the
major bottleneck was the oxidative cleavage of the vinyl
group to a hydroxymethyl group using a two-step OsO4-
NaIO4/NaBH4 protocol. This resulted in a poor isolated
yield (36%) of the desired product along with the inter-
mediate diol, the starting alkene, and products in which
the pyrimidine double bond has been dihydroxylated, as
seen with other similar substrates.6
To find a better synthetic route toward 1, we decided
to perform the problematic oxidative cleavage step in the
beginning of the reaction sequence with the conveniently
obtained compound 23 as the substrate. Thus, dihydroxy-
lation of the pyrimidine would be avoided, and better
overall yields of the final target molecule were envisioned.
Using the standard OsO4-NaIO4/NaBH4 protocol7 on 2
resulted in a mixture of the required hydroxymethyl
product 5, the diol 6, and the starting material 2
(approximate ratio 5:6:3) in 60% yield (Scheme 1). Noting
the presence of diol in the product mixture, we thought
of first converting 2 into diol 6, which could subsequently
be degraded to 5. However, the reaction of 2 with OsO4-
NMO8 did not complete even after multiple additions of
OsO4 over a period of 8 days. Efforts to achieve the
desired molecule 5 by an ozonolysis/reduction protocol9
resulted in complex mixtures in very low yields. The
failures of these established procedures forced us to
search for an alternative protocol for the conversion of
monosubstituted olefins to primary alcohols.
The search for therapeutically efficient antisense and
antigene oligonucleotides has motivated the synthesis
and evaluation of a high number of chemically modified
nucleic acid analogues.1 Bicyclic and thereby conforma-
tionally restricted nucleosides have received special
attention in the design of oligonucleotides with high-
affinity DNA and RNA recognition.2 As a promising
example, the [3.2.0]bicycloarabinonucleoside 1 (Figure 1)
is the hitherto most efficient mimic of the interesting
E-type conformation, and oligodeoxynucleotides contain-
ing one or several incorporations of 1 have displayed
improved affinity toward both complementary single-
stranded DNA and RNA.3,4 In our continuous search for
antisense and antigene compounds and in the investiga-
tions of the crucial enzyme RNase H, the nucleoside 1
is, therefore, needed for further examination. However,
the preparation of larger amounts of 1 is hampered by
its complicated synthesis (Figure 1). Thus, from a D-
xylose derivative, the dibenzylated 3′-C-vinyl compound
2 was obtained via a stereoselective Grignard reaction
and used in the preparation of the ribo-nucleoside 3 (R
) H). By using the pyrimidine moiety in a 2-2′-anhydro-
approach,5 3 was converted to its arabino-configured
‡ On leave from Kurukshetra University, Kurukshetra 136119,
India.
† Nucleic Acid Center is funded by the Danish National Research
Foundation for studies on nucleic acid chemical biology.
(1) (a) Uhlmann, E. Curr. Opin. Drug Discuss. Dev. 2000, 3, 202-
213. (b) Buchini, S.; Leumann C. J . Curr. Opin. Chem. Biol. 2003, 7,
717-726.
(2) (a) Herdewijn, P. Biochim. Biophys. Acta 1999, 1489, 167-179.
(b) Meldgaard, M: Wengel, J . J . Chem. Soc., Perkin Trans. 1 2000,
3539-3554. (c) Leumann, C. J . Bioorg. Med. Chem. 2002, 10, 841-
854.
(3) Christensen, N. K.; Petersen, M.; Nielsen, P.; J acobsen, J . P.;
Olsen, C. E.; Wengel, J . J . Am. Chem. Soc. 1998, 120, 5458-5463.
(4) Tømmerholt H. V.; Christensen, N. K.; Nielsen, P.; Wengel, J .;
Stein, P. C.; J acobsen, J . P.; Petersen, M. Org. Biomol. Chem. 2003, 1,
1790-1797.
(5) Yung, N. C.; Fox, J . J . J . Am. Chem. Soc. 1961, 83, 3060-3066.
(6) (a) Ravn, J .; Freitag, M.; Nielsen, P. Org. Biomol. Chem. 2003,
1, 811-816. (b) Christensen, N. K.; Andersen, A. K. L.; Schultz, T. R.;
Nielsen, P. Org. Biomol. Chem. 2003, 1, 3738-3748.
(7) Pappo, R.; Allen D. S.; Lemieux, R. U.; J ohnson, W. S. J . Org.
Chem. 1956, 21, 478-479.
(8) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,
17, 1973-1976.
(9) Sousa, J . A.; Bluhm, A. L. J . Org. Chem. 1960, 25, 108-111.
10.1021/jo0491861 CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/17/2004
5742
J . Org. Chem. 2004, 69, 5742-5745

SCHEME 1. Oxid a tive Clea va ge of Alk en e 2
TLC after the treatment of NaIO4 indicated a completed
diol cleavage, we purified the crude mixture obtained
after the ruthenium treatment of 2 and isolated some
R-ketol 8, aldehyde 7, and the expected diol 6 (ap-
proximate ratio 1:2:6) (Scheme 1). We envisioned that the
R-ketol 8, formed on ruthenium treatment, remained
unchanged on NaIO4 treatment and was reconverted into
the diol 6 on further reaction with NaBH4. Moreover, we
found that it is almost impossible to completely remove
ruthenium from the crude reaction mixture even after
passing the solution through Celite or treating the
mixture with sodium thiosulfate. Therefore, the little
ruthenium carried forward in the reaction sequence
might be responsible for oxidizing some of the diol 6 to
the R-ketol 8 prior to periodate cleavage as well as
overoxidizing the aldehyde 7 to carboxylic acid, thus
reducing the overall yield of the reaction protocol.13
On the basis of these observations, we thought of
subjecting the crude mixture, obtained after the ruthe-
nium oxidation, to NaBH4 reduction, predicting that it
should reduce any active ruthenium species present as
well as reduce the R-ketol 8 to the diol 6 and hence
eliminate the hazards of overoxidation on further perio-
date treatment. The experiments performed on these
lines proved our assumption to be correct, as we are able
to obtain consistently high yields of 5 (71-72% from 2)
adopting this reaction sequence (Table 1). Another indi-
rect proof was obtained by carrying out a blank experi-
ment where no olefin was added. When the residue
obtained after the aqueous workup was treated with
NaBH4, a flocculent black powder (presumably inactive
ruthenium species) was separated, appearing similar to
what we obtained in the experiments using olefins.
The developed protocol worked equally well on a larger
scale, and we got consistent yields with 19 mmol of alkene
2. The remarkable success of this reaction protocol
provoked us to apply it to some other carbohydrate
substrates (including allyl and vinyl substituents) as well
as simple substrates such as styrene and 1-decene (Table
1). The protocol is so predictive that we carried out the
reaction sequence on 1-decene without any analytical
support like TLC, NMR, etc. and obtained 1-nonanol in
80% yield. This result (Table 1) can only be explained by
taking into account that all the R-ketol formed in the
dihydroxylation step was recycled back to diol and no
more R-ketol was generated in the reaction sequence.
Thus, 1-decene has been reported to yield only 67% diol
on treatment with RuO4.12 We did not examine a series
of different protecting groups, as RuO4 is known to accept
a vide range of these,11,12 whereas the scopes and limita-
tions of NaBH4 reduction is well-known.
Ruthenium tetraoxide (RuO4) (usually made in situ
from RuCl3‚xH2O and a cooxidant, normally NaIO4) has
long been known for being a much more vigorous oxidant
than its osmium analogue.10 However, its use is mainly
limited to the degradation of unsaturated organic com-
pounds to carboxylic acids.11 Although modified reaction
conditions have been reported recently making cis-diols
as the main product, overoxidation to yield carboxylic
acids is usually observed.12 The problem of overoxidation
is especially acute in monosubstituted olefins, and 67%
is the best yield of diol reported from any monosubsti-
tuted alkene using a catalytic ruthenium procedure.12 We
found that the reaction of 2 under modified conditions12
(0.07 mol equiv of RuCl3‚xH2O and 1.5 equiv of NaIO4 at
0 °C in a biphasic solvent system of ethyl acetate,
acetonitrile, and water (in a ratio of 3:3:1)) for 1.5 min
leads to complete disappearance of the starting material
(Scheme 1). The crude residue obtained after the aqueous
workup was treated with NaIO4 in THF-H2O until TLC
indicated no starting diol. After aqueous workup, the
residue was redissolved in THF-H2O and treated with
NaBH4 to yield 5. Surprisingly, HR-MALDI-MS of the
crude product showed the presence of some diol 6 in
addition to the required hydroxymethyl product 5. There-
fore, the crude product was subjected once again to NaIO4
followed by NaBH4 treatment after which no traces of 6
were detected and 5 was obtained in 26% overall yield
from 2 after column chromatography.
Next, we turned our attention to using this protocol
on nucleosides themselves (exemplified by 3 (R ) Ac)),
but as expected, we found that the double bond of the
pyrimidine ring undergoes facile dihydroxylation, render-
ing this protocol unusable on pyrimidine nucleosides.
Having achieved the desired goal of efficient conversion
of 2 to 5, we attempted the synthesis of the target
nucleoside 1 as depicted in Scheme 2. Benzoylation of
To explain the presence of diol 6 after the complete
reaction sequence of RuO4-NaIO4-NaBH4 even though
(10) ) Djerassi, C.; Engle, R. R. J . Am. Chem. Soc. 1953, 75, 3838-
3840.
(11) (a) Yang, D.; Chen, F.; Dong, Z.-M.; Zhang, D.-W. J . Org. Chem.
2004, 69, 2221-2223. (b) Griffith, W. P.; Kwong, E. Synth. Commun.
2003, 33, 2945-2951. (c) Rossiter, B. E.; Katsuki, T.; Sharpless, K. B.
J . Am. Chem. Soc. 1981, 103, 464-465. (d) Carlsen, P. H. J .; Katsuki,
T.; Martin, V. S.; Sharpless, K. B. J . Org. Chem. 1981, 46, 3936-3938.
(12) (a) Shing, T. K. M.; Tai, V. W.-F.; Tam, E. K. W. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 2312-2313. (b) Shing, T. K. M.; Tam, E. K.
W.; Tai, V. W. F.; Chung, I. H. F.; J iang, Q. Chem. Eur. J . 1996, 2,
50-57. (c) Plietker, B.; Niggemann, M. Org. Lett. 2003, 5, 3353-3356.
(13) After our work was finished, a report appeared indicating the
formation of R-ketols as side products on treating olefins with RuO4/
NaIO4, complementing our presumption; Plietker, B. Org. Lett. 2004,
6, 289-291.
J . Org. Chem, Vol. 69, No. 17, 2004 5743

TABLE 1. Oxid a tive Clea va ge of Ter m in a l Olefin sa
derivative 10. Conversion of 10 directly to the desired
bicyclic nucleoside 11 proceeded very satisfactory (53%
yield from 9) in a refluxing mixture of 2 M aqueous NaOH
and aqueous ethanol. This conversion involves inversion
of stereochemistry of the 2′-carbon atom, subsequent
cyclization in a one-pot reaction cascade involving the
2-2′-anhydro intermediate,5 and hydrolysis of this inter-
mediate to give the arabino-configurated nucleoside,
which is preorganized for the subsequent formation of a
four-membered oxetane ring. A similar tandem reaction
has been shown before to form a bicyclic nucleoside with
a five-membered ring.15 Recently, also a bicyclic nucleo-
side with an oxetane ring has been prepared using
reaction conditions that indicate a similar reaction
cascade.16 Finally, debenzylation of 11 afforded 1 in 98%
yield.
In conclusion, the use of our new ruthenium protocol
for the oxidative cleavage of a double bond to a primary
alcohol and the tandem reaction cascade for the ring
closure allowed us to develop a significantly improved
synthetic sequence for the bicyclic nucleoside 1. The
apparent advantages of the new synthesis are (i) the
overall yield of 1 from 2 is 25.8% as compared to
12.5% by the earlier route;3 (ii) only three steps in the
present sequence require column chromatographic sepa-
ration as compared to eight steps in the earlier route;3
and (iii) no selective mesylation (and concomitant forma-
tion of a dimesyl derative) is required. The preparation
of larger amounts of 1 as well as its analogues with other
pyrimidine nucleobases is hereby possible, and we are
now exploring oligonucleotides containing 1 in biological
studies.
a
b
See General Procedure in Experimental Section. Overall
reaction time. c Isolated yield after column purification. Yield at
d
19 mmol scale. e Isolated crude yield.
In this paper, we have described the first ruthenium-
catalyzed protocol (Table 1) for the oxidative cleavage of
monosubstituted alkenes to yield exclusively primary
alcohols, a process similar to the OsO4-NaIO4/NaBH4
protocol or the alternative ozonolysis/NaBH4 protocol,
however, with several advantages: (i) the ruthenium
reagent is far less expensive as compared to osmium
tetraoxide; (ii) RuCl3 is easy to handle, as it is solid,
nonvolatile, and stable at room temperature; (iii) the
speed of reaction is increased significantly; (iv) the
workup is very easy, and even a 19 mmol-scale reaction
can easily be completed in 1 day; (v) all of the steps are
under aqueous conditions, thus eliminating the need for
anhydrous solvents; and (vi) RuO4 is neither explosive
nor poisonous.17 Obviously, this reaction protocol repre-
sents an efficient and intriguing alternative to the
osmium- or ozone-based protocols, and we expect it to
find wide general application in synthetic chemistry.
SCHEME 2. New Syn th esis of Nu cleosid e 1a
a
Reagents and conditions: (a) BzCl, pyridine; (b) 80% AcOH,
then Ac2O, pyridine; (c) thymine, N,O-bis(trimethylsilyl)acetamide,
CH3CN, TMS-triflate (69%, three steps); (d) NaOCH3, MeOH; (e)
MsCl, pyridine; (f) NaOH, EtOH, H2O (53%, three steps); (g) H2,
Pd(OH)2/C, EtOH (98%).
Exp er im en ta l Section
Gen er a l P r oced u r e for Oxid a tive Clea va ge of Olefin s.
NaIO4 (1.5 mmol) was stirred in water (2 mL), and the solution
was cooled to 0 °C. RuCl3‚xH2O (40.6% Ru, obtained from
ChemPure, Germany) (0.07 mmol) was added followed by the
addition of ethyl acetate (6 mL) and acetonitrile (6 mL), and
the mixture was stirred for 5 min at 0 °C. The olefin (1 mmol)
the free primary alcohol, acetolysis, acetylation, and
coupling with thymine using the silyl Hilbert-J ohnson/
Birkofer method as modified by Vorbru¨ggen et al.14
yielded the nucleoside 9 in 69% yield from 5. Deprotection
of the acetyl and benzoyl groups in a single step followed
by complete mesylation afforded the di-O-mesylated
(15) Håkansson, A. E.; Koshkin, A. A.; Sørensen, M. D.; Wengel, J .
J . Org. Chem. 2000, 65, 5161-5166.
(16) Paramashivippa, R.; Kumar, P. P.; Rao, P. V. S.; Rao, A. S.
Tetrahedron Lett. 2003, 44, 1003-1005.
(14) (a) Vorbru¨ggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber.
1981, 114, 1234-1255. (b) Vorbru¨ggen, H.; Ho¨fle, G. Chem. Ber. 1981,
114, 1256-1268.
(17) Frunzke, J .; Loschen, C.; Frenking, G. J . Am. Chem. Soc. 2004,
126, 3642-3652.
5744 J . Org. Chem., Vol. 69, No. 17, 2004

was added, and the slurry was stirred for 90 s. The reaction was
quenched by the addition of a saturated aqueous solution of
Na2S2O3 (10 mL). Phases were separated, and the aqueous layer
was extracted with ethyl acetate (3 × 10 mL). The combined
organic phase was dried (Na2SO4) and concentrated under
reduced pressure. The residue was dissolved in a mixture of THF
(2.5 mL) and water (2.5 mL). NaBH4 (1 mmol) was added, and
a flocculent black powder began to separate after 5 min. The
mixture was stirred for 20 min at room temperature, water (5
mL) was added, and the mixture was extracted with dichlo-
romethane (3 × 10 mL). The combined organic phase was
washed with a saturated aqueous solution of NaHCO3 (3 × 5
mL), dried (Na2SO4), and concentrated under reduced pressure.
The residue was dissolved in THF (5 mL) and water (5 mL) at
0-5 °C. NaIO4 (2 mmol) was added in small portions, and the
solution was stirred for 20 min (followed by TLC, wherever
possible) at room temperature. Ehylene glycol (35 µL) was added,
and the reaction mixture was diluted with water (7 mL). The
mixture was extracted with ethyl acetate (3 × 5 mL). The
combined organic phase was dried (Na2SO4) and concentrated
under reduced pressure. The residue was redissolved in a
mixture of THF (4 mL) and water (4 mL), and NaBH4 (3 mmol)
was added. The reaction mixture was stirred at room temper-
ature for 1 h; water (5 mL) was added, and the mixture was
extracted with dichloromethane (3 × 10 mL). The combined
organic phase was washed with a saturated aqueous solution
of NaHCO3 (3 × 5 mL) and dried (Na2SO4). After concentration
under reduced pressure, the residue was purified by flash
chromatography.
3,5-Di-O-b en zyl-1,2-d i-O-isop r op ylid en e-3-C-h yd r oxy-
m eth yl-r-D-r ibofu r a n ose (5). Prepared from 3,5-di-O-benzyl-
1,2-di-O-isopropylidene-3-C-vinyl-R-D-ribofuranose (2)3 (2.68 g,
6.76 mmol) as starting material using the general procedure.
Column chromatography (eluent: gradient of methanol (0-0.4%)
in dichloromethane). The reaction yielded 1.93 g (72%) of a
colorless viscous oil: Rf 0.65 (19:1 dichloromethane/methanol);
1H NMR (300 MHz, CDCl3) δ 7.39-7.23 (m, 10H, Bn), 5.83
(d, 1H, J H1-H2 ) 3.9 Hz, H-1), 4.71 (d, 1H, J gem ) 10.5 Hz,
3-OCH2Ph), 4.64 (d, 1H, J gem ) 11.7 Hz, 5-OCH2Ph), 4.62 (d,
1H, J gem ) 10.5 Hz, 3-OCH2Ph), 4.56 (d, 1H, J gem ) 11.7 Hz,
5-OCH2Ph), 4.50 (d, 1H, J H1-H2 ) 3.9 Hz, H-2), 4.22 (m, 1H, H-4),
3.88-3.71 (m, 4H, 2 × H-5, 2 × H-1′), 3.37 (dd, 1H, J ) 6.3 Hz,
8.1 Hz, OH), 1.60 (s, 3H, CH3), 1.37 (s, 3H, CH3); 13C NMR (75
MHz, CDCl3) δ 138.6, 137.2, 128.6, 128.4, 128.1, 128.1, 127.9,
127.7 (Bn), 112.6 (C(CH3)2), 104.6 (C-1), 84.0 (C-3), 81.1 (C-2),
80.1 (C-4), 74.3 (5-OCH2Ph), 67.5 (3-OCH2Ph), 66.7 (C-5), 62.0
(C-1′), 27.0 (CH3), 26.7 (CH3); HR-MALDI MS m/z 423.1762 ([M
+ Na]+, C23H28O6Na+ calcd 423.1778). Anal. Calcd for C23H28O6
+ (1/2)H2O: C, 67.46; H, 7.13. Found: C, 67.37; H, 6.78.
Ack n ow led gm en t. The Danish National Research
Foundation is thanked for financial support.
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal section, including procedures for the preparation of the
compounds in Table 1 and compounds 1 and 9-11, as well as
NMR spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O0491861
J . Org. Chem, Vol. 69, No. 17, 2004 5745