High-pressure hetero-diels-alder route to (±)-6,6,6-trifluoro- β-C-naphthyl glycosides

Article abstract of DOI:10.1021/ol900285w

The first de novo synthesis of a β-C-naphthyl glycoside displaying a convenient functionality for subsequent transformations into complex C-aryl glycosides is reported. The synthesis of this (±)-β-C-1,5- dibenzyloxynaphthyl 6,6,6-trifluoro-3-amino glycosi

Full text of DOI:10.1021/ol900285w

ORGANIC
LETTERS
2009
Vol. 11, No. 7
1619-1622
High-Pressure Hetero-Diels-Alder Route
to (()-6,6,6-Trifluoro-ꢀ-C-Naphthyl
Glycosides
Lucie Maingot,† Ste´phane Leconte,† Isabelle Chataigner,‡ Arnaud Martel,† and
Gilles Dujardin*,†
UCO2M UMR 6011 CNRS & FR 2575, UniVersite´ du Maine, 72085 Le Mans, France,
and IRCOF UMR 6014 CNRS & FR 3038, UniVersite´ de Rouen,
76821 Mont-Saint-Aignan, France
gilles.dujardin@uniV-lemans.fr
Received February 10, 2009
ABSTRACT
The first de novo synthesis of a ꢀ-C-naphthyl glycoside displaying a convenient functionality for subsequent transformations into complex
C-aryl glycosides is reported. The synthesis of this (()-ꢀ-C-1,5-dibenzyloxynaphthyl 6,6,6-trifluoro-3-amino glycoside relies on a hyperbaric
HDA reaction involving a new 2-vinylnaphthalenic dienophile.
ꢀ-C-Naphthyl-2-deoxy glycosides 1 are pivotal precursors
of natural C-aryl glycosides such as angucyclines1 and
medermycines2 as a result of their ability to afford the
corresponding bromojuglone 2, which acts as dienophile in
the formation of the key B ring (Scheme 1). The importance
of naphthyl glycosides 1 has stimulated much effort for their
† Universite´ du Maine.
Scheme 1. Proposed Route to Modified ꢀ-C-Aryl Glycosides
‡ Universite´ de Rouen.
(1) (a) Boyd, V. A.; Drake, B. E.; Sulikowski, G. A. J. Org. Chem.
1993, 58, 3191. (b) Boyd, V. A.; Sulikowski, G. A. J. Am. Chem. Soc.
1995, 117, 8472. (c) Krohn, V.; Agocs, A.; Ba¨uerlein, V. J. Carbohydr.
Chem. 2003, 22, 579.
(2) (a) Brimble, V.; Davey, R. M.; McLeod, M. D.; Murphy, M. Aust.
J. Chem. 2003, 787. (b) Brimble, M. A.; Davey, R. M.; McLeod, M. D.;
Murphy, M. Org. Biomol. Chem. 2003, 1690.
(3) Jaramillo, V.; Knapp, S. Synthesis 1994, 9, 1.
(4) (a) Andrews, F. L.; Larsen, D. S. Tetrahedron Lett. 1994, 35, 8693.
(b) Matsuo, V.; Miki, V.; Nakata, V.; Matsumura, V.; Toshima, K. J. Chem.
Soc., Chem. Commun. 1996, 225. (c) Toshima, K.; Matsuo, V.; Ishizuka,
T.; Ushiki, Y.; Nakata, M.; Matsumura, S. J. Org. Chem. 1998, 63, 2007.
(d) Hosoya, T.; Ohashi, Y.; Matsumoto, T.; Suzuki, K. Tetrahedron Lett.
1996, 37, 663.
synthesis3 by classical C-glycosylation,4 organometallic
condensation,1a,b or construction of the naphthalene ring from
a C-glycosylfuran.5
The introduction of fluorine (1-3 F atoms) at the C-6
position of C-naphthyl glycosides is a challenging task, likely
to provide analogues of biological interest. However, altering
the sugar unit after C-glycosylation is quite difficult.6 The
(5) (a) Kaelin, D. E.; Lopez, O. D.; Martin, S. F. J. Am. Chem. Soc.
2001, 123, 6937. (b) Chen, C. L.; Martin, S. F. Org. Lett. 2004, 6, 3581.
(6) (a) Collet, S.; Re´mi, J.-F.; Cariou, C.; La¨ıb, S.; Guingant, A.; Vu,
N. Q.; Dujardin, G. Tetrahedron Lett. 2004, 45, 4911. (b) Vu, N. Q.;
Dujardin, G.; Collet, S.; Raiber, E.-A.; Guingant, A.; Evain, M. Tetrahedron
Lett. 2005, 46, 7669.
(7) Hayman, C. M.; Larsen, D. S.; Brooker, S. Aust. J. Chem. 1998, 51,
545.
10.1021/ol900285w CCC: $40.75
Published on Web 03/12/2009
2009 American Chemical Society

elaboration of 6-fluorinated glycosyl donors for subsequent
C-glycosylation has been investigated in the literature7 and
by our group.8 However, 6,6,6-trifluoro-olivosyl donors are
inefficient for the C-glycosylation of 1,5-dihydroxynaphtha-
lene8 or its derivatives.7
Unfortunately, as in the ester series, we were unable to
transform adduct 5a into the corresponding glycoside.8
In this context, we reinvestigated the more direct [4 + 2]
approach involving less reactive vinylnaphthalene derivatives
as dienophiles. Access to C-naphthyl glycosides could then
result from a simple hydroboration-oxidation applied to the
type-8 adduct (after prior reduction in the ester series). The
desired cis configuration requires an endo-selective cycload-
dition, which could in turn result from appropriate conditions.
In the ester series, we had found previously that an
unactivated dienophile such as 2-vinylnaphthalene was inert
toward “prosugar” heterodienes 3a,b (Table 1, entries 1 and
We therefore investigated an alternative [4 + 2] route
toward C-naphthyl glycosides based on a SnCl4-catalyzed
heterocycloaddition between an R-methoxyvinylnaphthalene
and a “prosugar” heterodiene9 bearing an activating ester
group at the pivotal pro-C-6 position. We also described the
access to ꢀ-C-naphthyl glycosides from the resulting dihy-
dropyran via a new hydroboration/reduction/oxidation tan-
dem reaction (HyBRedOx).10 This method allows the
introduction of one or two fluorine atoms at the C-6 position
prior to the HyBRredOx sequence.10a This strategy efficiently
provided type-5 heteroadducts from ketone enol ether di-
enophiles 4 displaying a 1,5-dihydroxynaphthalene moiety
and heterodienes 3 (Scheme 2) but failed to transform these
Table 1. HDA Reactions of 2-Vinylnaphthalenes: Model Study
Scheme 2. [4 + 2] Route to C-Naphthyl Glycosides
entry
diene
conditionsa
adduct
yield (%)b
drc
1
2
3
4
5
6
7
3a
3b
6a
6b
3c
6c
6c
A
A
A
A
A
A
B
0
0
0
0
70
0
10
11
>97/3
>97/3
76
a Conditions A: CH2Cl2, 42 °C, 5 days. Conditions B: toluene, 110 °C,
c
12 days. b Isolated products. Determined by H NMR analysis.
1
heteroadducts into the desired precursors of C-glycosyl
bromojuglones.11 In this paper, we report the first access to
a (()-6,6,6-trifluoro-ꢀ-C-naphthyl glycoside via high pres-
sure Eu(fod)3-catalyzed hetero-Diels-Alder (HDA) reaction
of heterodiene 6 with a new dienophile, the 2-vinyl-1,5-
dibenzyloxynaphthalene 7.
Our first attempts to access 6,6,6-trifluoro-C-naphthyl
glycoside (Scheme 2, EWG ) RF ) CF3) focused on type-5
adducts (R2 ) OMe). From oxadienes 6 and activated
dienophiles 4, the reaction did not proceed under SnCl4-
catalyzed conditions. In contrast, Eu(fod)3-catalyzed reaction
between 6a and 4c led to the endo adduct 5a in good yield
(Scheme 3).
2).9a Expectedly, a similar behavior was observed toward
trifluoromethylated analogues 6a,b12 (entries 3 and 4). In
contrast, we were pleased to discover that the Eu(fod)3-
catalyzed HDA reaction can occur endo-selectively under
mild thermal conditions when starting from the 4-N-
phthalimido-substituted heterodiene 3c (entry 5). The higher
reactivity of N-Pht heterodiene 3c, when compared to
O-alkylated ones, could result from more favorable geometric
factors in the endo TS, as previously suggested in the case
of arylidene pyruvic acid esters.9a Under the same conditions,
trifluoromethylated N-Pht-heterodiene 6c did not react (entry
6). However, prolonging the reaction time at higher tem-
(8) Leconte, S., PhD Thesis, Universite´ du Maine, 2000.
(9) (a) Martel, A.; Leconte, S.; Dujardin, G.; Brown, E.; Maisonneuve,
V.; Retoux, R. Eur. J. Org. Chem. 2002, 514. (b) For pioneering work on
the [4 + 2] route to C-aryl glycosides, see: Schmidt, R. R.; Frick, W.;
Haag-Zeino, B.; Apparao, S. Tetrahedron Lett. 1987, 28, 4045.
(10) (a) Vu, N. Q.; Brown, E.; Gree, D.; Gree, R.; Dujardin, G.
Tetrahedron Lett. 2003, 44, 6425. (b) Vu, N. Q.; Leconte, S.; Brown, E.;
Gree, D.; Gree, R.; Dujardin, G. J. Org. Chem. 2005, 70, 2641.
(11) Maingot, L.; Nguten, Q. V.; Collet, S.; Guingant, A.; Martel, A.;
Dujardin, G. Eur. J. Org. Chem. 2009, 412.
Scheme 3. Stereocontrolled Access to Adduct 5a
(12) Hojo, M.; Masuda, R.; Kokuryo, Y.; Shiodo, H.; Matsuo, S. Chem.
Lett. 1976, 499.
1620
Org. Lett., Vol. 11, No. 7, 2009

perature (refluxing toluene) led to adduct 11, isolated in 76%
yield in a high endo-selectivity (entry 7).
Encouraged by these results, we investigated the synthesis
of a vinylnaphthalene that could afford juglone precursors:
the 2-vinyl-1,5-dibenzyloxynaphthalene 7 (Scheme 4). This
transition states involving the vinylnaphthalene and its
dibenzyloxy analogue were analyzed by DFT calculations
at the B3LYP/6-31G(d) level. The transition states involving
the monobenzyloxynaphthalene were also determined in
order to compare the steric and the electronic effect of the
benzyloxy group close to the reacting center. The computed
activation parameters in the gas phase at 50 °C are given in
Table 3.
Scheme 4. Synthesis of Dienophile 7
Table 3. Gas-Phase Activation Parameters (kcal/mol) at 323 K
Computed from the endo TS-(10,14,16)-a,b at the B3LYP/
6-31G(d) Level of Theory
new potent dienophile was prepared in three steps from the
monobenzylated 1,5-dihydroxynaphthalene via ortho-bro-
mination, O-protection, and Stille coupling with tributylvi-
nylstannane. The use of NBS in the first step prevented
dibromination in the ortho/para phenolic positions.1a,b
Eu(fod)3-catalyzed HDA reactions between dienes 3c or
6c and dienophile 7 were attempted under the thermal
conditions initially used with 2-vinylnaphthalene (Table 2).
TS-10
TS-16
TS-14
(R1 ) R2 ) H) (R1 ) OBn, R2 ) H) (R1 ) R2 ) OBn)
entry
a
b
a
b
a
b
1
2
3
∆Eq
20.0
21.5
36.9
21.0
25.6
40.8
25.3
20.6
35.7
20.3
27.0
43.6
26.4
21.4
36.9
21.0
∆Gq323 35.6
∆Hq323 19.5
Table 2. HDA Reactions of Dienophile 7
The geometry of the TS proved that the course of the
reaction is significantly asynchronous as illustrated by Table
4. As is generally observed with similar push-pull dienes,
Table 4. Bond Lengths (Å) of the Bond Formed at the TS
entry
TS
CsC
CsO
1
2
3
10b
16b
14b
1958
1966
1968
2414
2338
2333
entry diene
conditions
adduct yield (%)a
1
2
3
4
5
6
7
8
9
3c
6c
3c
6c
6c
3c
6c
3c
6c
toluene, 110 °C, 12 d
toluene, 110 °C, 12 d
0
0
0
0
0
toluene, MW, 170 °C, 0.5 h
toluene, MW, 170 °C, 0.5 h
CH2Cl2, 13 kbar, 50 °C, 4 db
CH2Cl2, 13 kbar, 50 °C, 1 d
CH2Cl2, 13 kbar, 50 °C, 1 d
CH2Cl2, 13 kbar, 50 °C, 4 d
CH2Cl2, 13 kbar, 50 °C, 4 d
the CsC bond is formed slightly prior to the CsO bond.
The reaction is therefore mainly controlled by the ability of
the system to stabilize the partial charges.
14
15
14
15
25
15
44c (74)d
TS a is disfavored in the case of TS-16 and TS-14 as a
result of the steric hindrance caused by R1 substituent (Table
3). This rotation of the vinyl group in the conformation
adopted by the dienophile in the TS could give a first
explanation for the lack of reactivity observed and could be
related to the position of the catalyst in the TS. A constant
growth of the activation energy (∆Eq) is observed from TS-
10a to TS-16b and -14b, whereas the free Gibbs energies
50c (88)d
a Purified product. b No Eu(fod)3 catalyst was used in this case. c dr
>97/3. d Yield based on recovered starting material.
No reaction occurred, even when the reaction time was
prolonged or the temperature increased (entries 1 and 2).
Microwave irradiation was also ineffective (entries 3 and
4). Whatever conditions were used, usually only starting
material was recovered.
The difference in reactivity between 2-vinylnaphthalene
and its dibenzyloxy analogue 7 was quite unexpected. In
order to check the causes of this lack of reactivity, the
(13) The experimental conditions (Lewis acid catalysis and pressure),
which are not taken into account in these calculations, may have a significant
influence on the course of the cycloaddition process.
(14) (a) Van Eldick, R.; Klaerner, F.-G. In High Pressure Chemistry;
Wiley-VCH: Weinheim, 2003. (b) Matsumoto, K.; Hamana, H.; Iida, H.
HelV. Chim. Acta 2005, 88, 2033.
Org. Lett., Vol. 11, No. 7, 2009
1621

of TS-10a and TS-16b at 323 K are similar, within the error
of the calculations. The expected free Gibbs energy gap is
only observed for TS-14b (1.2 kcal/mol).13 This indicates
that the lack of reactivity of dienophile 7 is mainly due to
electronic effects and not to the steric hindrance caused by
the presence of the benzyloxy group.
Scheme 5
.
Synthesis of (()-Trifluoro-ꢀ-C-naphthyl Glycoside
19
Research from past decades has shown that pressure in
the range of 1-20 kbar strongly influences the rate of
processes accompanied by a decrease in volume. The highly
negative activation volume (typically -23 to -51 cm3·mol-1)
characterizing Diels-Alder cycloadditions has generated
many studies that unambiguously demonstrate a powerful
pressure-induced acceleration of these reactions.14 Increase
in endo selectivity is generally observed under these operat-
ing conditions.15 Synergistic effects of Lewis acid catalysis
and high pressures have also been observed.16 Interestingly,
Schmidt’s group demonstrated the efficiency of hyperbaric
conditions (6.2 kbar) on HDA reactions involving styrenes.9b
Scheeren and co-workers evidenced the synergistic effects
of Eu(fod)3-catalysis and high pressure conditions in the
formation of dihydropyrans from vinyl ethers.17
Compressing heterodiene 6c with 7 to 13 kbar in the
absence of any Lewis acid, however, was not sufficient to
promote the cycloaddition in our case (Table 2, entry 5). In
contrast, activation by Lewis acid and 13 kbar pressure at
50 °C showed the positive impact of the multiactivated
process and allowed the formation of the cis-adducts 14 and
15, in a complete endo-selective manner (entries 6 and 7).
However, the isolated yields were low. Increasing reaction
time to 4 days under otherwise identical conditions ensured
good conversions and enhanced yields of 44% and 50%,
respectively (entries 8 and 9). Under these conditions,
residual dienophile 7 could be recovered and yields based
on unrecovered starting material were good (74% and 88%,
respectively).
subsequent treatment of the borane by trimethylamine
N-oxide in diglyme,19 1H NMR of the crude product showed
the total disappearance of the vinylic proton. After purifica-
tion on silicagel, the pure C-naphthyl glycoside (()-19 was
obtained in 55% yield with a complete diastereocontrol.
To conclude, the present work exemplifies the first
successful de novo access to a rac-ꢀ-C-naphthyl-2-deoxy
glycoside that can act as a C-glycosylbromojuglone precur-
sor. The key step of this synthesis is a high pressure/Eu(fod)3-
catalyzed HDA reaction between an activated heterodiene
(that delivers both amino and trifluoromethyl groups in the
final glycoside) and a new dienophile, the 2-vinyl-1,5-
dibenzyloxynaphthalene (7). The ꢀ-C-naphthyl-6,6,6-trif-
luoro-3-amino glycoside described herein provides a fine
illustration of the regiocontrolled introduction of fluorine and
nitrogen atoms that can be carried out on the glycoside
moiety by this strategy. The generalization of this approach
to other modifications of the glycoside unit (e.g., mono- or
bis-fluorination at the C-6 position via the adduct 14), and
its asymmetric extension is under progress in our laboratories.
From these new and promising [4 + 2] adducts obtained
with a high cis-selectivity, we first investigated access to
the corresponding ꢀ-C-naphthyl glycosides in the trifluoro-
methyl series (Scheme 5). N-Deprotection of adduct 15 was
conveniently conducted under mild aminolysis conditions by
treatment with an ethanolic solution of methylamine at
40 °C. Subsequent dibenzylation of the crude primary
amine18 afforded the tetrabenzylated derivative 18 in good
overall yields. Hydroboration of 18 was conducted with
BH3·Me2S complex at 40 °C in THF. After 24 h and
Acknowledgment. We thank the Re´gion Pays de la Loire
for a PhD grant to L.M. and CRIHAN for computating
facilities offered.
Supporting Information Available: Experimental pro-
cedures and spectroscopic data for synthetic compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(15) (a) Boger, D. L.; Robarge, K. D. J. Org. Chem. 1988, 53, 3373.
(b) Le Strat, F.; Vallette, H.; Toupet, L.; Maddaluno, J. Eur. J. Org. Chem.
2005, 5296. (c) Pichon, N.; Harrison-Marchand, A.; Mailliet, P.; Maddaluno,
J. J. Org. Chem. 2004, 69, 7220.
OL900285W
(16) (a) Aben, R. W. M.; Minuti, L.; Scheeren, H. W.; Taticchi, A.
Tetrahedron Lett. 1991, 32, 6445. (b) Kinsman, A. C.; Kerr, M. A. Org.
Lett. 2000, 2, 3517. (c) Chataigner, I.; Hess, E.; Toupet, L.; Piettre, S. R.
Org. Lett. 2001, 3, 515. (d) Minuti, L.; Taticchi, A.; Lanari, D.; Marrocchi,
A.; Gacs-Baitz, E. Tetrahedron: Asymmetry 2003, 14, 2387.
(17) Aben, R. W. M.; de Gelder, R.; Scheeren, H. W. Eur. J. Org. Chem.
2002, 3126.
(18) Amine 17 proved to be hardly separable from dimethylphtalimide,
produced in equimolar quantities from the aminolysis of 15.
(19) Kabalka, G. W.; Hedgecock, H. C. J. J. Org. Chem. 1975, 40, 1776.
1622
Org. Lett., Vol. 11, No. 7, 2009