Conformationally locked bicyclo[4.3.0]nonane carbanucleosides: Synthesis and bio-evaluation

Article abstract of DOI:10.1039/c4ob01763c

d-Ribose was converted into 3 novel carbobicyclic nucleosides bearing a bicyclo[4.3.0]nonane framework in 16-19 steps with 5-12% overall yields involving a Wittig olefination and an intramolecular Diels-Alder reaction as the key steps. The present synthesis also provides an efficient entry for chiral hydrindenones. The conformation studies of these carbanucleosides and their bio-evaluation as potential antiviral agents are reported.

Full text of DOI:10.1039/c4ob01763c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Conformationally locked bicyclo[4.3.0]nonane
carbanucleosides: synthesis and bio-evaluation†
Cite this: Org. Biomol. Chem., 2014,
2, 9439
1
a
a
b
b
c
Tony K. M. Shing,* Anthony W. H. Wong, Huiyan Li, Z. F. Liu and Paul K. S. Chan
D-Ribose was converted into 3 novel carbobicyclic nucleosides bearing a bicyclo[4.3.0]nonane framework
in 16–19 steps with 5–12% overall yields involving a Wittig olefination and an intramolecular Diels–Alder
reaction as the key steps. The present synthesis also provides an efficient entry for chiral hydrindenones.
The conformation studies of these carbanucleosides and their bio-evaluation as potential antiviral agents
are reported.
Received 18th August 2014,
Accepted 25th September 2014
DOI: 10.1039/c4ob01763c
www.rsc.org/obc
Introduction
In recent years, carbocyclic nucleosides or carbanucleosides
have warranted tremendous attention from both synthetic and
1
,2
medicinal chemists. The replacement of the oxygen atom in
the furanose ring by a methylene unit provides both enzymatic
and chemical stability, attributable to the loss of a true glycosi- Fig. 1 Naturally occurring carbocyclic nucleoside aristeromycin.
3
dic linkage. However, there is a significant change in the con-
formation due to the loss of the gauche and anomeric effects
which exist in conventional nucleosides. These effects main-
tain the sugar moiety of the nucleoside in either a C3′-endo
4
(
North) or a C2′-endo (South) conformation. In the absence of
these effects, the conformation of carbocyclic nucleosides is
mainly governed by the steric bulk of the nucleobase, which
prefers to occupy the equatorial position and adopts an atypi-
cal C1′-exo conformation. The conformation of nucleosides is
intriguing since any structure–activity relationship (SAR) study
of antiretroviral nucleosides is complicated by the complexity
of the anabolic process of activation. It involves three enzy-
Fig. 2 Conformationally locked nucleosides with a bicyclo[3.1.0]hexane
system.
matic steps to transform the nucleoside into its 5′-triphosphate RT. The route to NTP anabolite, however, involves different
NTP), plus the final interaction of the NTP with the target kinases, all of which are highly dependent on the nature of the
(
5
enzyme, the reverse transcriptase (RT). Therefore, the prefer- heterocyclic base.
ences of conformation exhibited by the nucleoside, or its
Aristeromycin is a representative and its conformation is
nucleotides, must be identified at each intervening enzymatic illustrated in Fig. 1. It has been reported to be a potent inhibi-
step. The only step which is invariant and common to all tor of S-adenosyl-L-homocysteine (AdoHcy) hydrolase, which
nucleoside RT inhibitors is the final interaction of NTP with plays an important regulatory role in the S-adenosyl-L-methion-
6
7
ine (AdoMet)-dependent methylation process. The methyl-
ation process is important as most plant and animal viruses
require a methylated cap structure at the 5′-terminus of their
a
Department of Chemistry and Center of Novel Functional Molecules, The Chinese
University of Hong Kong, Shatin, Hong Kong, China
b
mRNA for viral replication. However, their cytotoxicity
Department of Chemistry and Centre for Scientific Modeling and Computation,
8
–10
precludes clinical use.
The Chinese University of Hong Kong, Shatin, Hong Kong, China
c
Department of Microbiology, The Chinese University of Hong Kong, China
There are not many accounts on the syntheses of bicyclic
carbanucleosides. In 1993, Rodriguez et al. pioneered the con-
struction of dideoxyribonucleoside analogues with a bicyclo
†
Electronic supplementary information (ESI) available: Experimental pro-
cedures, characterisation data, copies of NMR spectra for all compounds,
detailed DFT modelling and CIF files. CCDC 727043 and 1000117. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/
c4ob01763c
[
3.1.0]hexane backbone (Fig. 2) which constrains the confor-
1
1
mation of the analogue back to the C2′-exo (North). From
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9439–9445 | 9439

View Article Online
Paper
Organic & Biomolecular Chemistry
with milder conditions for the construction of chiral hydrinde-
none intermediates.
Results and discussion
Theoretical calculation
1
9
Gaussian 09 program package was used to calculate the
structures of the carbanucleosides with the density functional
theory (DFT) quantum chemical method. Equilibrium geome-
tries of all molecules were fully optimized at the B3LYP/6-31G
2
0
(
d) level of theory. Stationary points were characterized as
either minima (no imaginary frequency) or transition struc-
tures (only one imaginary frequency) by calculation of the
vibrational frequencies using analytical gradient procedures at
the same level. Minimum energy pathways connecting the
reactants and products were confirmed by intrinsic reaction
Fig. 3 Conformation of bicyclic carbanucleosides calculated using the
DFT model: top: C1’-exo conformation; bottom: C3’-endo
conformation.
2
1
coordinate (IRC) calculations.
then on, different kinds of conformationally restricted carbo-
Synthesis
cyclic nucleosides that lock the conformation as either North
1
2
or South were reported.
Retrosynthesis of the target carbanucleosides (3–5) shows that
1
3
Coxsackie B viruses are single strand positive-sense RNA it could be synthesized from cycloadduct 6, which would be
viruses and contain six serotypes of pathogenic enteroviruses prepared from D-(−)-ribose (8) via triene 7, using an IMDA reac-
that trigger illness ranging from gastrointestinal distress to tion as the key step (Scheme 1).
full-fledged pericarditis and myocarditis. They resist a wide
variety of chemical treatment and to the best of our knowl- lidenated to afford acetonide 9, which was then subjected to
Starting from D-(−)-ribose (8), the 2,3-diol in 8 was isopropy-
2
2
2
3
edge, there is still no well accepted treatment for the Coxsackie vinylation to give alkene 10 in good yield. Glycol cleavage oxi-
B viruses and it is interesting to see whether our compounds dation of the vicinal diol in 10 gave lactol 11, which with allyl-
1
4
could be potential drugs to treat the viruses.
magnesium bromide furnished dienes 12 and 13 in a ratio of
However, to date, only carbobicyclic nucleosides with a 6 : 1, respectively, and in excellent overall yields. Aqueous
bicyclo[3.1.0]hexane and a bicyclo[3.3.0]octane system have indium allylation only led to decomposition of the starting
been described. There is still no publication on the syntheses material. Oxidation of 12 with IBX provided a mixture of
of carbanucleosides with a bicyclo[4.3.0]nonane system. There- lactols 14 and 15 in a 3 : 2 ratio, which could not be separated
fore, it is fascinating to see if they exhibit any antiviral activity. by column chromatography (Scheme 2). We hoped to utilize
Furthermore, according to our theoretical calculations, carba- the equilibrated open-chain form of 14 and to eliminate
nucleosides with a cyclohexane ring cis-fused to a cyclopentane the homoallylic alcohol to form IMDA precursor 17
ring should exist in either a C1′-exo or a C3′-endo confor- (Scheme 3). Different elimination conditions were examined,
mation, which is in close agreement with the North conformer but no positive results were observed after exhaustive
of conventional nucleosides (Fig. 3) and may show RNA anti- experimentation.
viral activity.
Since the ring opening–elimination approach failed, we
To synthesize this class of carbobicyclic nucleosides for bio- changed our synthetic avenue to a protection and deprotection
logical studies, we first have to synthesize the bicyclo[4.3.0]- approach (Scheme 4). The allylic hydroxyl group in the diol 12
nonane core. Functionalized hydrindenones (bicyclo[4.3.0]- was protected as silyl ether 17 in good yield. The homoallylic
nonenones) are versatile intermediates in complex natural hydroxyl group was then attempted to be eliminated. However,
product synthesis, e.g., the CD ring in steroids. It is noteworthy after trying different methods, silyl ether 17 only reacted with
that syntheses of chiral functionalized hydrindenones were trifluoroacetic anhydride (TFAA) to give trifluoroacetate 18
1
5
not an easy task via an intramolecular cycloaddition. To which was then subjected to elimination in the presence of
date, only a few publications have reported the intramolecular DBU. However, no triene 19 was detected, beyond recovery of
Diels–Alder (IMDA) reaction on sugars and most of these
1
6
papers documented the syntheses of trans-decaline systems.
Only one report described the synthesis of a bicyclo[4.3.0]-
1
7
nonenone, involving a lengthy synthetic route. Jarosz et al.
reported the syntheses of chiral bicyclo[4.3.0]non-2-enes from
a sugar allyltin. However, the choice for protecting groups was
limited as Lewis acid was employed in the preparation of the
1
8
IMDA precursors. In this paper, we report a shorter route Scheme 1 Retrosynthesis of the target nucleosides.
9440 | Org. Biomol. Chem., 2014, 12, 9439–9445
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 2 Synthesis of lactol 14 and 15.
Fig. 4 X-ray structure of 6.
2
6
Emmons (HWE) olefination was tried but no reaction was
observed with different kinds of bases such as dimsyl sodium
and n-butyllithium. The mixture of trienes 7 was subjected to
oxidation with IBX as an oxidant and furnished trienone 20 in
Scheme 3 Equilibrium between the ring and chain form of 14.
87% yield. Trienone 20 was highly unstable and polymerized
quickly after standing in vacuum. Trienone 20 underwent the
IMDA reaction in toluene in the presence of methylene blue in
a sealed tube to give hydrindenone 6 in good yield. The struc-
ture and absolute configuration of 6 was confirmed by X-ray
crystallography (Fig. 4).
Initially we believed that cis to trans isomerization of the
diene in 20 should precede the cycloaddition. However, theore-
tical calculations disagreed with this idea since the relative
energy of the isomerization is surprisingly high (99 kcal
Scheme 4 Attempted synthesis of triene 19.
−1
−1
mol ), which was 50 kcal mol higher than the published
2
7
value. Therefore, the stereoselectivity of the cycloadduct
could be explained by the proposed transition state models as
shown in Fig. 5. For E-20, two possible transition states, TS-1
and TS-2, in the endo mode of cycloaddition, would lead to the
formation of two thermodynamically favorable cis-fused
cycloadducts 6 and 21. Theoretical calculations also agreed
with these results. However, the steric repulsion between the
isopropylidene and the newly formed 6-membered ring in the
transition state might inhibit the formation of 21. TS-3 might
give 22 but unfortunately 22 was unable to detect. For Z-20,
TS-5 was not favorable due to the steric repulsion between the
isopropylidene and the newly formed 6-membered ring and
in TS-6, the highly twisted conformation might inhibit the
formation of the trans-fused cycloadduct 22.
Scheme 5 Synthesis of 6.
Since there was no trans-fused cycloadduct 22 isolated in
alcohol 17 quantitatively. The use of other bases did not yield the previous step, epimerization was tried but failed to occur
any desired 19.
after many attempts and under a wide variety of conditions
2
8
We therefore revised our synthetic strategy and discovered (Scheme 6).
that lactol 11 was a good candidate for the Wittig reaction
with allyltriphenylphosphonium bromide to introduce the carbanucleosides were carried on with the cis-fused bicyclic
diene functionality (Scheme 5). After screening with different carbocycle 6. The hydrindenone 6 was epoxidized to give
bases, potassium tert-butoxide (KO Bu) in THF gave the best epoxide 23 in good yield (Scheme 7). The β-epoxide was the
2
4
Since epimerization was not successful, syntheses of bicyclic
2
5
t
yield of triene 7 of 88% with a Z to E ratio of 4 : 1. The geo- only isolated product and no α-epoxide was detected.
metric isomers were not separable by column chromatography. The extremely high diastereoselectivity might be explained
2
9
In order to enhance the selectivity, Horner–Wadsworth– by the ketone directing effect proposed by Armstrong.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9439–9445 | 9441

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 8 Synthesis of epoxides 30 and 31.
Scheme 9 Reduction of epoxides 32 and 33.
Reduction of epoxide was attempted with LiBH , NaBH , super
4
4
4
hydride or LiAlH under various conditions, but unfortunately
only alcohols 24 and 25 were isolated. Interestingly, reduction
with DIBAL-H in the presence of AlMe provided the desired
3
diol 26 in moderate yield.
As the reduction was not as easy as expected and the yield
was not satisfactory, an alternative approach was performed
(
Scheme 8). Hydrindenone 6 was firstly reduced with NaBH
the corresponding alcohols 27 and 28 in a 1 : 4 ratio. Alcohol
7 could be oxidized back to 6 by PDC in good yield. Protec-
4
to
Fig. 5 Proposed transition state models of the cyclization of trienone
2
0.
2
tion of 28 as silyl ether 29 proceeded smoothly and this was
followed by epoxidation to give epoxides 30 and 31 in a 1 : 1
ratio. When both epoxides were subjected to reduction, only
the β-epoxide 30 was susceptible to reduction, and alcohol 32
was obtained in good yield. The α-epoxide 31 was reluctant to
be opened with DIBAL-H even under refluxing CH Cl
2
2
(
Scheme 9).
The difficulty of ring-opening in α-epoxide 33 might be
Scheme 6 Attempted epimerization of 6 to 22.
attributable to the steric hindrance of the adjacent cyclo-
pentane ring so that the hydride was unable to approach the
reacting site. To minimize the formation of α-epoxide, we
chose to epoxidize alkene–alcohol 28, and the ratio of epoxide
34 to 24 was slightly increased to 1 : 2 (Scheme 10). The
hydroxyl group in 24 was protected as TBS ether to furnish
silyl ether 30 again.
After the reduction of the epoxide, the newly formed
hydroxyl group in 32 was protected as acetate 35 and as ethoxy
methyl (EOM) ether 36, respectively, in excellent yields
(Scheme 11). Removal of the silyl protecting group in both
Scheme 7 Synthesis of diol 26.
9442 | Org. Biomol. Chem., 2014, 12, 9439–9445
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 10 Epoxidation of 28.
Scheme 13 Synthesis of pyrimidine 44.
Scheme 11 Synthesis of 37 and 38.
Scheme 14 Synthesis of the target adenosine analogue 3.
Scheme 12 Attempted to synthesize pyrimidine 39 and 40.
substrates proceeded smoothly to furnish alcohols 37 and 38
in excellent yields, which were the precursors for coupling
with the nucleobases.
With alcohol 37 and 38 in hand, direct coupling of nucleo-
bases with the free alcohols was examined (Scheme 12). Scheme 15 Syntheses of thymidine analogue 4 and uridine analogue 5.
3
0
Firstly, the Mitsunobu reaction was used but no desired
product was isolated using either PPh or PBu with DIAD in
3
3
refluxing toluene. Alcohols 37 and 38 were then activated by the reactivity, a stronger electrophile, 5-nitro-4,6-dichloropyri-
conversion into sulfonate esters (R = Ms, Tf) which were dis- midine, was used instead. Pyrimidine 45 was then obtained in
placed with purine bases. Unfortunately, both sulfonate esters good yield, and reduction of the nitro group to the amino
failed to form purine 39 and 40, respectively.
group was achieved using iron powder as a reductant in AcOH
As the convergent route failed to give any desired nucleo- (Scheme 13). Using a standard procedure, pyrimidine 44 was
side, the linear approach was adopted to synthesize the target transformed into the target carbanucleoside 3 (Scheme 14).
3
1
molecules. Triflate 41 was successfully converted into azide 42
in good overall yield from alcohol 37 (Scheme 13). Azide 42 logue 5 were similar and the results are shown in Scheme 15.
The syntheses of thymidine analogue 4 and uridine ana-
3
2
was reduced to the corresponding amine 43 by catalytic hydro- Amine 44 was reacted with 3-methoxymethacryloyl isocyanate
3
3
genation in excellent yield. Reaction of amine 43 with and 3-methoxy-2-propenoyl isocyanate to produce urea 47
-amino-4,6-dichloropyrimidine furnished pyrimidine 44 in and 48 in 74% and 99%, respectively. Both ureas 47 and 48
5
miserable yield and could be explained by the low electrophili- were then subjected to acid hydrolysis to remove all the pro-
city of 5-amino-4,6-dichloropyrimidine. In order to improve tecting groups as well as to facilitate the cyclization. Thymi-
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9439–9445 | 9443

View Article Online
Paper
Organic & Biomolecular Chemistry
Conclusion
To conclude, bicyclo[4.3.0]nonane carbanucleosides 3, 4, 5
were synthesized from D-(−)-ribose in 16 steps or 19 steps with
5%, 7% and 12% overall yield, respectively, involving a Wittig
alkenation and an IMDA reaction as the key steps. The cyclo-
adduct 6, a chiral functionalized hydrindenone (bicyclo[4.3.0]-
nonenone), should open an avenue as a versatile intermediate
in complex natural product synthesis. X-ray crystallography
and NOE studies of these carbanucleosides show that they
display the C1′-exo conformation. These compounds demon-
strated weak in vitro antiviral activity against Coxsackie B3
virus. It is worth modifying the compounds and reassess for a
wider range of viruses. Research in this direction is in
progress.
Fig. 6 X-ray structure of adenosine 3.
Fig. 7 NOE results of nucleosides 4 and 5.
Acknowledgements
This work was supported by a Hong Kong RGC General
dine carbanucleoside 4 was obtained in moderate yield (73%) Research Fund (Ref. CUHK 4030/09P).
and uridine carbanucleoside 5 was obtained in excellent yield
(91%).
Notes and references
1
D. K. Tosh, H. O. Kim, S. Pal, J. A. Lee and L. S. Jeong, in
Modified Nucleosides, Wiley-VCH Verlag GmbH & Co. KGaA,
Weinheim, Germany, 2008, pp. 525–560.
Review: (a) E. De Clercq, Nucleosides, Nucleotides Nucleic
Acids, 2009, 28, 586–600; (b) D. M. Huryn and M. OKabe,
Chem. Rev., 1992, 92, 1745–1768; (c) A. D. Borthwick and
Determination of conformation
With all the carbanucleosides in hand, their conformations
were determined by both X-ray crystallography and NOE experi-
ments. Disappointing results were obtained as all the carbanu-
cleosides displayed the undesired C1′-exo conformation. The
crystal structure of adenine 3 is shown in Fig. 6. The NOE
results of both nucleosides 4 and 5 also agreed with the C1′-
exo conformation (Fig. 7).
2
K.
d) L. Agrofoglio, E. Suhas, A. Farese, R. Condom,
S. R. Challand, R. A. Earl and R. Guedj, Tetrahedron, 1994,
0, 10611–10670; (e) V. E. Marquez and M. I. Lim, Med. Res.
Biggadike,
Tetrahedron,
1992,
48,
571–623;
(
5
Rev., 1986, 6, 1–40.
H. Bricaud, P. Herdewijn and E. De Clerq, Biochem. Phar-
macol., 1983, 32, 3583–3590.
4 W. Saenger, in Principle of Nucleic Acid Structure, Springer-
Verlag, New York, 1984, pp. 9–28.
5 H. Mitsuya, in Anti-HIV Nucleosides: Past, Present and
Future, R. G. Landes Co., Texas, 1997.
6 A. Kalman, T. Koritsanszky, J. Beres and G. Sagi, Nucleo-
sides Nucleotides, 1990, 9, 235–243.
7 R. M. Ransohoff, P. Narayan, D. F. Ayers, F. M. Rottman
and T. W. Nilsen, Antiviral Res., 1987, 1, 317–327.
8 D. L. Hill, S. Straight, P. W. Allan and L. L. Bennett Jr., Mol.
Pharmacol., 1971, 7, 375–380.
Biological studies
3
Antiviral activity against enterovirus, Coxsackie B3 virus, was
assessed using the plaque reduction assay. Briefly, African
green monkey cells (Vero) were used to cultivate virus-com-
pound mixtures at various dilutions. After 72 hours of incu-
bation at 37 °C, infected cells were fixed and the number of
plaques was counted. The percentage of reduction in the
number of plaques in wells with compounds was calculated.
Compounds suspected to have antiviral effect were further
examined for cytotoxicity using the 3-(4,5-dimethylthiazol-2-
yl)-2,5-diphenyltetrazolium bromide (MTT) assay.
As a result, 3 nucleosides 3, 4 and 5 showed screening
results suggestive of having weak in vitro antiviral activity
against Coxsackie B3 virus. Adenosine 3 (concentration = 1 µg
9 S. Yaginuma, N. Muto, M. Tsujino, Y. Sudate, M. Hayashi
and M. Otani, J. Antibiot., 1981, 34, 359–366.
−
1
mL ) showed 4.9% of inhibition and thymidine 4 (concen- 10 E. De Clercq, Antimicrob. Agents Chemother., 1985, 28, 84–
−
1
tration = 1 µg mL ) showed 5.5% of inhibition and higher
89.
−
1
concentration (concentration = 100 µg mL ) led to highly 11 J. B. Rodriguez, V. E. Marquez, M. C. Micklaus, H. Mitsuya
−
1
toxic uridine 5 (concentration = 1 µg mL ) which gave the
and J. J. Barchi Jr., Tetrahedron Lett., 1993, 34, 6233–6236.
best result, showing more than 12.3% of inhibition and 12 (a) Q. Chao and V. Nair, Tetrahedron, 1997, 53, 1957–1970;
further structural modification is recommended.
(b) M. Šála, H. Hŕebabecký, M. Dračínský, M. Masojídková,
9444 | Org. Biomol. Chem., 2014, 12, 9439–9445
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
A. M. De Palma, J. Neyts and A. Holý, Collect. Czech. Chem.
Commun., 2010, 75, 1–20.
3 H. A. Rotbart, Antiviral Res., 2002, 53, 83–98.
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
GAUSSIAN 09 (Revision B.01), Gaussian Inc., Pittsburgh, PA,
2010.
1
1
4 B. N. Fields, D. M. Knipe, R. M. Chanock, J. L. Melnick, 20 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
B. Roizman and R. E. Shope, in Fields Virology, Raven Press, 21 C. Gonzalez and H. B. Schlegel, J. Phys. Chem., 1991, 94,
New York, 1985, pp. 739–794.
5 (a) C. C. Browder, F. P. Marmsaeter and F. G. West, Org. 22 P. A. Levene and R. S. Tipson, J. Biol. Chem., 1936, 115,
Lett., 2001, 3, 3033–3035; (b) B. B. Snider and T. C. A. Kirk, 731–747.
J. Am. Chem. Soc., 1983, 105, 2364–2368; (c) E. J. Corey and 23 T. K. M. Shing, D. A. Elsley and J. G. Gillhouley, J. Chem.
T. A. Engler, Tetrahedron Lett., 1984, 25, 149–152; Soc., Chem. Commun., 1989, 1280–1282.
d) D. A. Smith, K. Sakan and K. N. Houk, Tetrahedron Lett., 24 (a) G. Wittig and G. Geissler, Ann., 1953, 580, 44–57;
5523–5527.
1
1
(
1
986, 27, 4877–4880.
(b) G. Wittig and U. Schollkopf, Chem. Ber., 1954, 87, 1318–
6 (a) S. Mondal, R. N. Yadav and S. Ghosh, Org. Lett., 2011,
1330.
1
3, 6078–6081; (b) R. S. Nandurdikar, A. V. Subrahmanyam 25 (a) S. D. Tilley and M. B. Francis, J. Am. Chem. Soc., 2006,
and K. P. Kaliappan, Eur. J. Org. Chem., 2010, 2788–2799;
c) C. Taillefumier and Y. Chapleur, Can. J. Chem., 2000, 78,
08–722.
128, 1080–1081; (b) J. G. Atkinson, M. H. Fisher, D. Horley,
A. T. Morse, R. S. Stuart and E. Synnes, Can. J. Chem., 1965,
43, 1614–1624.
(
7
1
1
1
7 M. D. Clay, D. Riber and A. G. Fallis, Can. J. Chem., 2005, 26 (a) L. Horner, H. Hoffmann and H. G. Wippel, Chem. Ber.,
8
3, 559–568.
8 S. Jarosz, E. Kozlowska and A. Jezewski, Tetrahedron, 1997,
3, 10775–10782.
9 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
1958, 91, 61–63; (b) L. Horner, H. Hoffmann, H. G. Wippel
and G. Klahre, Chem. Ber., 1959, 92, 2499–2505.
27 W. M. Marley and P. M. Jeffers, J. Phys. Chem., 1975, 79,
5
2085–2087.
t
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, 28 Several bases were tried such as NaH, NaOMe, LDA, KO Bu
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
and DBU. We also tried to deprotonate the α-H with NaH
and quenched by D O. The proton signal disappeared
which proved that deprotonation did exist.
2
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, 29 A. Armstrong, P. A. Barsanti, P. A. Clarke and A. Wood,
H. Nakai, T. Vreven, Jr. J. A. Montgomery, J. E. Peralta, J. Chem. Soc., Perkin Trans. 1, 1996, 1373–1380.
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, 30 W.-B. Lu, S. Sengupta, J. L. Petersen, N. Akhmedov and
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, X. Shi, J. Org. Chem., 2007, 72, 5012–5015.
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, 31 (a) Y. Liu, Y. Lu, M. Prashad, O. Repic and T. A. Blacklock,
J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
Adv. Synth. Catal., 2005, 347, 217–219; (b) S. E. Hazlet and
C. A. Dornfeld, J. Am. Chem. Soc., 1944, 66, 1781–1782.
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, 32 G. Shaw and N. R. Warrener, J. Chem. Soc., 1958, 157–161.
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, 33 (a) C. Cesario and M. J. Miller, J. Org. Chem., 2009, 74,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,
5730–5733; (b) L. Santana, M. Teijeria and E. A. Uriarte,
J. Heterocycl. Chem., 1999, 36, 293–295.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 9439–9445 | 9445