Intramolecular imino Diels - Alder reaction: Progress toward the synthesis of uncialamycin

Article abstract of DOI:10.1021/jo901291t

(Chemical Equation Presented) We herein described an intramolecular imino Diels-Alder reaction promoted with BF3 3 OEt2/DDQ affording substituted quinolines. Using this procedure, we prepared the chiral quinoline moiety of the uncialamycin, a new enediyne natural product.

Full text of DOI:10.1021/jo901291t

pubs.acs.org/joc
Intramolecular Imino Diels-Alder Reaction: Progress toward the
Synthesis of Uncialamycin
Sandy Desrat and Pierre van de Weghe*
ꢀ
ꢀ
EA Substances Licheniques et Photoprotection, Faculte des Sciences Biologiques et Pharmaceutiques,
Universiteꢀ de Rennes 1, 2 avenue du Prof Leꢀon Bernard, F-35043 Rennes Cedex, France
pierre.van-de-weghe@univ-rennes1.fr
Received June 17, 2009
We herein described an intramolecular imino Diels-Alder reaction promoted with BF3 OEt2/DDQ
3
affording substituted quinolines. Using this procedure, we prepared the chiral quinoline moiety of the
uncialamycin, a new enediyne natural product.
Introduction
The structure of uncialamycin 1, similar to that of dyne-
micin A 2,3 combines a ten-membered enediyne with an
anthraquinone substructure (Figure 1). The strategy for the
total synthesis of the title compound 1 described by Nicolaou
and co-workers was based on the addition of an acetylide 4 to
a quinolinium species, an intramolecular acetylide addition
affording the enediyne system, and then an Hauser annulation
to complete the synthesis. The first key synthetic intermediate,
the chiral quinoline moiety 5, was prepared from the commer-
In 2005, Davies, Andersen, and co-workers disclosed
the uncialamycin 1, a new “enediyne” natural product
isolated from an undescribed streptomycete obtained
from the surface of a lichen Cladonia uncialis.1 The first
biological evaluations showed that 1 possesses potent in
vitro antibacterial activity against Staphylococcus aurens,
Escheridia coli, and Burkholderia apia. Despite the small
amount of product available (∼300 μg were isolated) the
structure of 1 was resolved but assigning the absolute
configuration of C26 was not possible. Nicolaou and co-
workers recently proved without ambiguity the complete
structure of 1 and determined the absolute stereochemistry
at C26 after total synthesis of the racemic mixture and then
reported the first asymmetric synthesis.2 Having ample
quantities of 1 and its C26-epimer in hand, they studied its
biological properties in DNA-cleavage, antibacterial, and
cytotoxic activities. These investigations revealed impress-
ive high potent antitumor activities and broad-spectrum
antibacterial properties.
€
cially available 5-methoxyisatin 6 including a Friedlander
quinoline synthesis and an enantioselective reduction of
ketone to fix the stereogenic center C26.2
These recent findings prompt us to report our progress
toward the synthesis of the title compound 1. Our approach
focused on the preparation of a quinoline that possesses the
well-defined chiral center C26. The quinoline and tetrahydro-
quinoline derivatives still attracted interest due to their im-
portance as synthetic intermediates and as a key struc-
tural core in several natural products which have shown a
wide range of biological activities.4 Hence a variety of syn-
thetic routes have been reported and the development of
new approaches still remains an active field of research.
*To whom correspondence should be addressed. Phone: þ33 2 23 23 38
03. Fax: þ33 2 23 23 44 25.
€
Classical methods, such as Friedlander, Combes, Skraup,
(1) (a) Davies, J.; Wang, H.; Taylor, T.; Warabi, K.; Huang, X.-H.;
Andersen, R. J. Org. Lett. 2005, 7, 5233–5236. (b) Davies, J.; Andersen, R. J.;
Wang, H.; Warabi, K.; Huang, X.-H. Patent WO2007/038868.
(3) Maier, M. E.; Boβe, F.; Niestroj, A. J. Eur. J. Org. Chem. 1999, 1–13.
(4) (a) Katritzky, A. R.; Rachwal, S.; Rachwal, B. Tetrahedron 1996, 52,
ꢀ ꢀ
15031–15070. (b) Kouznetsov, V. V.; Vargas Mendez, L. Y.; Melendez
ꢀ
Gomez, C. M. Curr. Org. Chem. 2005, 9, 141–161. (c) Madapa, S.; Tusi,
Z.; Batra, S. Curr. Org. Chem. 2008, 12, 1116–1183.
(2) (a) Nicolaou, K. C.; Zhang, H.; Chen, J. S.; Crawford, J. J.; Pasunoon,
L. Angew. Chem., Int. Ed. 2007, 46, 4704–4707. (b) Nicolaou, K. C.; Chen, J.
S.; Zhang, H.; Montero, A. Angew. Chem., Int. Ed. 2008, 47, 185–189.
6728 J. Org. Chem. 2009, 74, 6728–6734
Published on Web 07/28/2009
DOI: 10.1021/jo901291t
r
2009 American Chemical Society

Desrat and van de Weghe
JOCArticle
SCHEME 2. Unsuccessful Imino Diels-Alder Reaction
2-pyrroline in the presence of 5 mol % of camphor sulfonic
acid, Batey and co-workers isolated the tetrahydroquinoline
core of martinelline whereas Ma and co-workers combined the
methyl 4-aminobenzoate with ethyl glyoxalate and N-Cbz 2,3-
dihydro-1H-pyrrole in the presence of squaric acid, an unusual
catalyst, to lead to the tetrahydroquinoline nucleus. The
intramolecular Povarov reaction catalyzed by Dy(OTf)3 has
also been used to prepare the alkaloids luotonin A which
possess a quinoline core.10 The intramolecular Povarov reac-
tion was also developed in combinatorial synthesis providing
chemical librairies built around the quinoline scaffold.11
Having these different approaches in mind, we report a
detailed account of the use of intramolecular imino Diels-
Alder reactions to prepare polysubstituted quinolines. Rely-
ing on these results we describe our work for the construction
of the quinoline core of the uncialamycin 1.
FIGURE 1. Structures of uncialamycin (1) and dynemicin A (2)
and Nicolaou’s retrosynthetic analysis of 1.
SCHEME 1. Preparation of Quinolines from the Povarov Re-
action
and Doeber-Miller reactions, are widely recognized and
frequently used for the preparation of quinolines from anilines
but they do not allow the formation of the quinoline nucleus
with diversity. In addition, the harsh reaction conditions can
lead to several byproducts and sometimes poor yields.5,6 The
recent development of alternative methods has been reported
with the use of transition metal as catalysts and some draw-
backs have been overcome.7 Among these methods, the most
direct to build quinoline scaffolds consists of adding electron-
rich alkenes (or alkynes) to electron-deficient aromatic imines
(formed in situ from aniline and aldehyde derivatives) followed
by an oxidation reaction (Scheme 1). Although this imino
Diels-Alder reaction, also called the Povarov reaction, was
first reported about 40 years ago, this reaction only recently
received more attention since it was shown that this cycloaddi-
tion reaction can be promoted or catalyzed by Lewis or protic
acids. However, its use in total synthesis still remains scarce.8
For example, the total synthesis of the alkaloid martinel-
line was simultaneously reported by Batey9a and Ma.9b
Mixing the methyl 4-aminobenzoate with 2 equiv of N-Cbz
Results and Discussion
Intramolecular Imino Diels-Alder Reaction. To build the
quinoline intermediate via the most simple and quickest path-
way, we first investigated the reactivity of 1,3,5-tri(p-methoxy-
phenyl)hexahydro-1,3,5-triazene 712 with the alkene 813 or
alkyne 913 in the presence of Lewis acid (Scheme 2). Although
a few examples of [4þ2] cycloaddition have been described
between aromatic methyleneamines in its trimer form and
dienophiles,14 no cycloadduct was formed after stirring in
various reaction conditions with the alkene 8 or alkyne 9.
The dienophile 8 or 9 was recovered after treatment of the
reaction mixture. This lack of reaction could be due to either
the low reactivity of the triazine 7 or the electronic character of
the dienophile, which is not electron-rich enough.
Because the dienophile 8 or 9 is not electron-rich enough to
react with an aromatic methyleneamine we thought to study
the intermolecular cycloaddition with a more electron defi-
cient N-aryl imine, such as p-anisidine ethyl glyoxylate imine
derivatives.15 Again, although different reaction conditions
ꢀ
(5) (a) Marco-Contelles, J.; Perez-Mayoral, E.; Samadi, A.; Carreiras,
M.; Soriano, E. Chem. Rev. 2009, 109, 2652–2671. (b) Jones, G. In Compre-
hensive Heterocyclic Chemistry II; Katritzky, A. R., Rees, C. W., Eds.; Pergamon
Press: New York, 1996; Vol. 5, pp 167-300.
(10) Twin, H.; Batey, R. A. Org. Lett. 2004, 6, 4913–4916.
(11) For examples, see: (a) Spaller, M. R.; Thielemann, W. T.; Brennan,
P. E.; Bartlett, P. A. J. Comb. Chem. 2002, 4, 516–522. (b) Muhuhi, J.;
Spaller, M. R. J. Org. Chem. 2006, 71, 5515–5526.
ꢀ
(6) For selected recent examples, see: (a) Martinez, R.; Ramon, D. J.;
Yus, M. J. Org. Chem. 2008, 73, 9778–9780. (b) Vander Mierde, H.; van der
Voort, P.; Verpoort, F. Tetrahedron Lett. 2008, 49, 6893–6895. (c) Wu, Y.-C.;
Liu, L.; Wang, D.; Chen, Y.-J. J. Org. Chem. 2006, 71, 6592–9595.
(7) For selected recent examples, see: (a) Horn, J.; Marsden, S.; Nelson, A.;
House, D.; Weingarten, G. G. Org. Lett. 2008, 10, 4117-4120 and references
cited therein. (b) Vander Mierde, H.; van der Voort, P.; De Vos, D.; Verpoort, F.
Eur. J. Org. Chem. 2008, 1625–1631. (c) Lekhok, K. C.; Prajapati, D.; Boruah, R.
C. Synlett 2008, 655–658.
ꢀ
(12) Barluenga, J.; Bayon, A. M.; Campos, P.; Asensio, G.; Gonzales-
~
Nunez, E.; Molina, Y. J. Chem. Soc., Perkin Trans. 1 1988, 1631–1636.
(13) See the Supporting Information.
(14) (a) Campos, P. J.; Lamaza, I.; Rodriguez, M. A. Tetrahedron Lett.
1997, 38, 6741–6744. (b) Malassene, R.; Sanchez-Bajo, L.; Toupet, L.;
Hurvois, J.-P.; Moinet, C. Synlett 2002, 1500–1504.
(8) (a) Povarov, L. S.; Mikhailov, B. M. Izv. Akad. Nauk. SSR, Ser. Khim.
1963, 953–956. (b) Kouznetsov, V. V. Tetrahedron 2009, 65, 2721–2750. (c)
For a review concerning the imino Diels-Alder reaction, see: Buonora, P.;
Olsen, J.-C.; Oh, T. Tetrahedron 2001, 57, 6099–6138.
(9) (a) Powell, D. A.; Batey, R. A. Org. Lett. 2002, 4, 2913–2916. (b) Xia,
C.; Heng, L.; Ma, D. Tetrahedron Lett. 2002, 43, 9405–9409.
(15) For examples of [4þ2] cycloaddition affording quinoline derivatives,
ꢁ
see: (a) Alves, M. J.; Azoia, N. G.; Fortes, A. G. Tetrahedron 2007, 63, 727–
734. (b) Boglio, C.; Lemi, G.; Hasenknopf, B.; Thorimbert, S.; Lacote, E.;
^
Malacria, M. Angew. Chem., Int. Ed. 2006, 45, 3324–3327. (c) Hermitage, S.;
Howard, J. A. K.; Jay, D.; Pritchard, R. G.; Probert, M. R.; Whiting, A. Org.
Biomol. Chem. 2004, 2, 2451–2460.
J. Org. Chem. Vol. 74, No. 17, 2009 6729

Desrat and van de Weghe
JOCArticle
SCHEME 3. First Attempted Intramolecular Cycloaddition
Reaction
Povarov reaction.17 They proposed that Tf2NH catalyzes
two distinct reactions: a cycloaddition between aldimines
and electron-rich olefins to give the corresponding tetrahy-
droquinolines in situ and at the same time a hydrogen-
transfer process from tetrahydroquinolines to aldimines
affording the resulting quinolines and amines.
The imines 18 and 19 were prepared as previously reported
from tartaric acid13,18 and subjected to the same reaction
conditions as above. The imine 18 led to the quinoline 20 and
the amine 21 in a 20/21=0.33:0.66 ratio while 19 gave 20 and
22 in a 20/22=0.5:0.5 ratio and the tetrahydroquinoline 23
and the dihydroquinoline 24 were not detected in the crude
1
reaction mixture after H NMR analysis (Scheme 4). This
means that the oxidation of cycloadduct by the imine
proceeds faster or at the same rate as the intramolecular
cycloaddition. These results confirm the double role played
by the starting material: the imine acts as precursor of the
cycloadduct as well as an oxidant to convert the resulting
cycloadduct into quinoline. Therefore the presence of an
oxidant that could react faster than the imine in the reaction
mixture has been envisaged. Thus, we examined the intra-
molecular cycloaddition reaction of imine 18 with BF3 OEt2
3
(1 equiv) in CH2Cl2 in the presence of various oxidants. As
shown in Scheme 5, the reaction works well in the presence of
2 equiv of DDQ to afford the quinoline 20 in good yield
(72%) without any traces of amine 21. In the presence of
1 equiv of DDQ a complex reaction mixture was obtained.
This confirms the need of using 2 equiv of oxidant for
cycloaddition from alkenes. When the DDQ was replaced
with O2 (1 atm or bubbling O2 in the reaction mixture), a
complex mixture was obtained and the quinoline was iso-
lated in very low yield (up to 23%). Hoping that CAN could
act both as a promoter of the imino Diels-Alder reaction19
and an oxidant, the imine 18 was placed in the presence of
2.1 equiv of CAN in CH2Cl2. Although the disappearance of
18 was complete, only a small amount of quinoline 20 was
isolated after purification on silica gel column chromatog-
raphy from a complex reaction mixture (26% yield). Having
shown the importance of the presence of DDQ in the reaction
mixture to obtain only the quinoline 20, the first reaction
depicted in Scheme 4 was conducted again as follows: After
conversion of the imine 18 in quinoline 20 and amine 21,
2 equiv of DDQ was added to the reaction mixture. After an
additional1 h ofstirring, the amine 21was totally converted to
quinoline 20. This experiment seems to indicate here that the
DDQ can act either as an oxidant to transform the tetra-
hydroquinoline 23 to quinoline 20 without formation of
amine 21 or as an oxidant of amine 21 after its formation.
Having found a satisfactory system, we turned our attention
have been tested (Lewis acid, solvent, and temperature), the
formation of cycloadduct was not observed. Facing this
failure, we planned to see the same reaction in its intramo-
lecular version. We felt that there were two major advantages
to evaluate this intramolecular cycloaddition. As in intra-
molecular cycloaddition, the diene and the dienophile will be
close to each other; this reaction should be allowed and only
one regioisomer would be obtained. As shown in Scheme 3,
the precursor of cycloaddition 14 was synthesized in a few
steps from the propargyl alcohol 11. Reaction of the acryloyl
€
chloride in the presence of Hunig’s base with the racemic
alcohol 12 gave the acryloyl ester 13. After osmylation of 13
the intermediate diol was obtained in 78% yield and then
quantitatively converted into the corresponding glyoxal by
using an excess of sodium periodate. The imine 14 was
isolated in 50% yield after addition of p-anisidine to the
crude glyoxal in toluene in the presence of molecular sieves.16
When compound 14 was treated with 1 equiv of BF3 OEt2 in
3
CH2Cl2 at room temperature, the formation of a mixture of
the quinoline 16 with the amine 17 in a 1:1 ratio (determined
from the crude reaction mixture by 1H NMR) was observed
after complete disappearance of the starting material in place
of the expected dihydroquinoline 15. The derivative 16 was
isolated in 47% yield while the desilylated amine 17 decom-
posed during the purification. This result demonstrates that
the intramolecular imino Diels-Alder favors the cycloaddi-
tion. The presence of the amine 17 suggests that half an
equivalent of imine 14 reacts as an oxidant to convert the
intermediate dihydroquinoline 15 in quinoline 16. During
our work, similar observations were reported by Takasu and
co-workers in the course of their studies of a catalytic
to the amount of BF3 OEt2 required to achieve this intra-
3
molecular imino Diels-Alder reaction in good yields. When
the imine 18 with 2 equiv of DDQ in the presence of 1, 0.5, 0.2,
or 0.1 equiv of BF3 OEt2 was carried out in CH2Cl2 at room
3
temperature after 1 h of stirring, the chemical yield slightly
decreased (72%, 63%, 54%, and 58%, respectively). In the
absence of BF3 OEt2, no quinoline was formed and only
3
decomposition of the starting material was observed.
(17) Shindoh, N.; Tokuyama, H.; Takemoto, Y.; Takasu, K. J. Org.
Chem. 2008, 73, 7451–7456.
(16) Although the imine 10 was quantitatively obtained after treatment of
the reaction mixture, the yield after isolation is low because a part of the imine
was decomposed during the silica gel column chromatography.
(18) Howard, B. E.; Woerpel, K. A. Org. Lett. 2007, 9, 4651–4653.
(19) Han, B.; Jia, X.-D.; Jin, X.-L.; Zhou, Y.-L.; Yang, L.; Liu, Z.-L.; Yu,
W. Tetrahedron Lett. 2006, 47, 3545–3547.
6730 J. Org. Chem. Vol. 74, No. 17, 2009

Desrat and van de Weghe
JOCArticle
SCHEME 4. Cycloaddition Reactions
SCHEME 5. Choice of the Oxidant
TABLE 1. Scope of the Aromatic Group
entry
R
yield, %
product
1
2
3
4
5
6
7
8
OMe
H
Me
NMe2
Cl
CF3
F
NO2
72
24
47
20
36
37
38
39
40
41
42
N.R.a
57
55
Whereas the BF3 OEt2 (1 equiv)/DDQ (2 equiv for
3
alkenes and 1 equiv for alkynes) appears as the best com-
promise to built quinolines using an intramolecular Povarov
reaction, we then examined the reactivity of other substrates
in these reaction conditions. For this purpose we prepared
some precursors by varying the aromatic groups and the
dienophile that is either an olefin or an alkyne. The glyoxylic
acid moiety was synthesized following two different routes,
from the tartaric acid 25 for the alkenes’ derivatives 28 and
from the acryloyl chloride for alkynes’ compounds 33
(Scheme 6). As the glyoxylic and imine’s intermediates are
not easy to isolate pure, we chose to determine the yields of
the synthesis of quinoline 35 from 27 and 32. The 1H NMR
spectral data were consistent with expectations for the
proposed structures 30 and 34 and revealed that the products
have been formed in good to quantitative conversion.
65
N.R.b
aNo cycloaddition. bNo imine formation.
TABLE 2. Alkene vs. Alkyne
We first investigated the influence of the electron density
of the aromatic group on the cycloaddition reaction. Various
imine derivatives were synthesized from the cinnamyl glyox-
ylate 28a as outlined in Scheme 6. The results of this study
are reported in Table 1. As indicated above, the yield of
the formation of quinolines was determined from the dicin-
namyl tartrate 27a obtained after purification in 72% yield.
The intramolecular imino Diels-Alder reaction of the crude
mixture of 30 was carried out in the presence of 1 equiv of
entry
R
R1
yield, % from 27 yield, % from 32 product
1
2
3
4
5
6
H
OMe
H
H
N.R.a
15
32
N.R.a
<5b
44
43
44
45
36
20
46
OMe n-Pr
Ph
OMe Ph
H
24
72
56
59
OMe CH2OBn
64
a48-50% conversion of the imine formation and no cycloaddition.
bDetermined by 1H NMR.
BF3 OEt2 and 2 equiv of DDQ in CH2Cl2 at room tem-
3
perature. The preparation of imine derivatives 30 works
well in most cases (complete conversion) except when using
the 4-nitroaniline, in which case no reaction was observed.
The quinolines 35 were isolated in low to good yields;
however, it appears difficult to rationalize the results based
on the electronic density of the aryl group. The electron-rich
imine 18 gave the best result (72% yield over 3 steps) and
in the same time the electron-deficient imine 25 (R = F)
afforded 65% yield in quinoline 41.
J. Org. Chem. Vol. 74, No. 17, 2009 6731

Desrat and van de Weghe
JOCArticle
SCHEME 6. General Scheme of Quinoline Synthesis
SCHEME 7. Preparation of the Chiral Quinoline 52
We then compared the intramolecular cycloaddition be-
tween alkenes 30 and alkynes 34. Various compounds were
prepared as described in Scheme 6. As shown in Table 2, the
reaction sequence afforded low to good yields. Except with
the terminal olefine and alkyne (entries 1 and 2), we observed
a good conversion of the crude imine 30 or 34 into quinoline
35. The moderate yields are due to an incomplete conversion
of the diol 27 or 32 into glyoxylic derivative 28 or 33. It is
noteworthy that there is no significant difference between the
alkenes’ route and the alkynes’ route. Because only 1 equiv of
DDQ is required, it seems more interesting to use the latter.
Preparation of the Quinoline Core of Uncialamycin. Rely-
ing on these various results, we decided to prepare the
quinoline core of uncialamycin 1 from the alkyne ketone
47.20 Enantioselective reduction of 47 with (R)-Alpine bor-
ane according to Brown21 afforded cleanly the alcohol 48
with 80% ee as determined from the 1H NMR after deriva-
tization into its corresponding O-methylmandelic ester 54.22
The alkynol 48 was converted in acryloyl ester 49 and the diol
50 was isolated as a diastereomeric mixture after osmylation
reaction in 55% yield. The diol 50 was converted into its
glyoxylic acid derivative, which was treated with p-anisidine
in toluene in the presence of molecular sieves to give the
€
(21) Oestreich, M.; Frohlich, R.; Hoppe, D. J. Org. Chem. 1999, 64, 8616–
8626.
(22) (a) A partial epimerization probably occurred in the course of the
formation of the mandelic ester 54. (b) Trost, B. M.; Belletire, J. L.; Godleski,
S.; McDougal, P. G.; Balkovec, J. M.; Baldwin, J. J.; Christy, M. E.; Ponticello, G.
S.; Varga, S. L.; Springer, J. P. J. Org. Chem. 1986, 51, 2370–2374.
(20) (a) Marshall, J. A.; Xie, S. J. Org. Chem. 1995, 60, 7230–7237. (b)
Giner, J.-L. Tetrahedron Lett. 1998, 39, 2479–2482.
6732 J. Org. Chem. Vol. 74, No. 17, 2009

Desrat and van de Weghe
JOCArticle
SCHEME 8. Synthesis of Quinoline 56
acetate 7:3 as eluent and by a followed recrystallization in
ethanol.
Synthesis of Quinoline 52 (Scheme 7). 5-(Benzyloxy)pent-3-
yn-2-one (47):24 To a solution of 1-[(prop-2-ynyloxy)methyl]-
benzene (3.00 g, 20.5 mmol, 1 equiv) in tetrahydrofuran under
argon atmosphere at -78 °C was added n-butyl lithium (2.5 M in
hexanes, 9.03 mL, 22.6 mmol, 1.1 equiv). After 30 min at -78 °C
N-acetylmorpholine (4.78 mL, 41.0 mmol, 2 equiv) was added
and the reaction mixture was allowed to warm to rt overnight.
The reaction was then quenched with a saturated aqueous
solution of ammonium chloride. The product was extracted
with ethyl acetate (3 times) and the combined organic layers
were washed with brine, dried over sodium sulfate, and con-
centrated under reduced pressure. The crude product was
purified by column chromatography on silica gel with cyclohex-
ane/ethyl acetate 95:5 as eluent to obtain 47 as a colorless oil
(1.35 g, 7.18 mmol, 35%). Rf 0.3 (cyclohexane/ethyl acetate 8:2);
1H NMR (270 MHz, CDCl3) δ 7.40-7.32 (m, 5H), 4.61 (s, 2H),
4.32 (s, 2H), 2.36 (s, 3H); 13C NMR (67.5 MHz, CDCl3) δ 183.9,
136.7, 128.5, 128.1, 128.0, 87.3, 85.7, 72.1, 56.9, 32.6; HRMS
(ESI) calcd for C12H12O2Na 211.0735, found 211.0735.
imine 51 (Scheme 7). After imino Diels-Alder cycloaddition
the quinoline 52 was isolated in 65% yield overall from the
diol 50 with 89% ee as determined by HPLC.23 To continue
toward the synthesis of uncialamycin 1, we envisage these
following steps: lactone ring-opening, decarboxylation, and
a diastereoselective intermolecular addition of acetylide.
Finally, to confirm the formation of the quinoline, the
lactone opening of the racemic quinoline 5513 was carried
out with an excess of methyl magnesium chloride to afford
the gem-dimethyl alcohol 56. Crystals suitable for X-ray
diffraction analysis were grown by slow evaporation of
dichloromethane at room temperature from an ethanol/
dichloromethane solution of 56. The X-ray crystal structure
analysis of 56 shows the expected quinoline core
(Scheme 8).13
(R)-5-(Benzyloxy)pent-3-yn-2-ol (48):25 To a commercial sol-
ution of (R)-Alpine borane (0.5 M in THF, 8.50 mL, 4.25 mmol,
2 equiv) under argon atmosphere at 0 °C was added 47 (400 mg,
2.12 mmol, 1 equiv). The tetrahydrofuran was removed under
reduced pressure at 0 °C and the resulting oil was stirred at rt
overnight. The mixture was cooled to 0 °C and 4 mL of diethyl
ether, 0.5 mg of acetaldehyde, and 0.5 mL of diethanolamine
were successively added. The resulting precipitate was elimi-
nated by filtration and the filtrate was washed with a 0.1 M
aqueous solution of HCl, washed with brine, dried over sodium
sulfate, and concentrated under reduced pressure. The crude
product was purified by column chromatography on silica gel
with cyclohexane/ethyl acetate 8:2 as eluent to obtain 48 as a
Conclusions
In summary, we have described that BF3 OEt2 in the
3
presence of DDQ promotes the intramolecular Povarov
reaction and oxidative aromatization affording substituted
quinolines. The dienophile could be either an alkene or an
alkyne without the yield of cyclization being affected. One
equivalent of DDQ is necessary in the case of the cycloaddi-
tion reaction with the alkynes while it takes 2 equiv for the
reaction with alkenes. In the absence of DDQ, the cycloaddi-
tion reaction occurs but the quinoline was obtained as a
mixture with an amine, which results in hydrogen transfer
from the dihydro- or tetrahydroquinoline to the starting
imine.
colorless oil (280 mg, 1.47 mmol, 69%). Rf 0.2 (cyclohexane/
1
ethyl acetate 7:3); ee 80% ; [R]22 þ14.0 (c 1.0, CHCl3); H
D
NMR (270 MHz, CDCl3) δ 7.47-7.23 (m, 5H), 4.63-4.53 (m,
3H), 4.19 (d, J=1.4 Hz, 2H), 2.21-2.05 (br s, 1H), 1.46 (d, J=
6.5 Hz, 3H); 13C NMR (67.5 MHz, CDCl3) δ 137.4, 128.4, 128.1,
127.9, 88.5, 79.9, 71.7, 58.3, 57.4, 24.2; HRMS (EI) calcd for
C12H14O2 190.0994, found 190.1012.
(R)-5-(Benzyloxy)pent-3-yn-2-yl acrylate (49): To a solution
€
of 48 (1.00 g, 5.26 mmol, 1 equiv) and Hunig’s base (2.72 mL,
15.8 mmol, 3 equiv) in 20 mL of dichloromethane at 0 °C under
argon atmosphere was added acryloyl chloride (854 μL, 10.5 mmol,
2 equiv). The mixture was allowed to warm to rt overnight and
quenched with a saturated aqueous solution of ammonium
chloride. The product was extracted with ethyl acetate (3 times)
and the combined organic layers were washed with brine, dried
over Na2SO4, and concentrated under reduced pressure. The
crude product was purified by column chromatography on silica
gel with cyclohexane/ethyl acetate 95:5 as eluent to obtain 49 as
These reaction conditions were used to prepare the chiral
quinoline moiety of the enediyne uncialamycin 1. The C26
chiral center of the uncialamycin was fixed at the beginning
of the synthesis from a well-known enantioselective reduc-
tion of R-alkyne ketone.
Overall, we believe that the described intramolecular
imino Diels-Alder reaction will allow a rapid access to
polysubstituted quinolines in a simple and straightforward
way.
a colorless oil (1.26 g, 5.16 mmol, 98%). Rf 0.4 (cyclohexane/
1
ethyl acetate 8:2); [R]22 þ74.0 (c 1.0, CHCl3); H NMR (270
D
MHz, CDCl3) δ 7.47-7.26 (m, 5H), 6.45 (dd, J = 16.6 and
1.4 Hz, 1H), 6.13 (dd, J=16.6 and 10.5 Hz, 1H), 5.86 (dd, J=
10.5 and 1.4 Hz, 1H), 5.58 (dq, J=6.8 and 1.6 Hz, 1H), 4.58 (s,
2H), 4.19 (d, J=1.6 Hz, 2H), 1.54 (d, J=6.8 Hz, 3H); 13C NMR
(67.5 MHz, CDCl3) δ 164.9, 137.3, 131.3, 128.4, 128.1, 128.0,
127.8, 84.8, 80.8, 71.6, 60.4, 57.2, 21.3; HRMS (EI) calcd for
C12H14O2 190.0994, found 190.0993.
Experimental Section
General Procedure for Intramolecular Imino Diels-Alder Re-
action: Preparation of Quinoline 20 from 18 (Table 1, entry 1).
To a 0.2 M solution of the crude imine 18 in dichloromethane
under argon atmosphere at rt was added 2 equiv of dichlorodi-
cyanoquinone and then 1 equiv of boron trifluoride diethyl
etherate. The mixture was stirred for 30 min and the solvent was
removed under reduced pressure. The dark residue was purified by
column chromatography on silica gel with cyclohexane/ethyl
(R)-5-(Benzyloxy)pent-3-yn-2-yl 2,3-dihydroxypropanoate (50):
To a mixture of 49 (1.00 g, 4.09 mmol, 1 equiv) and N-methyl-
morpholine N-oxide (663 mg, 4.91 mmol, 1.2 equiv) in 30 mL of
(23) Chiralpack AD-H column (5 mm, 4.6 ꢀ 250 mm) with a UV
detection at 259 nm at 20 °C, eluent heptane/isopropanol=8/2, retention
times 13.08 and 15.80 min. See the Supporting Information.
(24) Hauptmann, H.; Mader, M. Tetrahedron Lett. 1977, 36, 3151–3154.
(25) Stipa, P.; Finet, J.-P.; Le Moigne, F.; Tordo, P. J. Org. Chem. 1993,
58, 4465–4468.
J. Org. Chem. Vol. 74, No. 17, 2009 6733

Desrat and van de Weghe
JOCArticle
acetone/water 9:1 was added a solution of osmium tetroxide
(0.5% w/v in tert-butyl alcohol, 2.08 mL, 0.041 mmol, 0.01 equiv).
The mixture was stirred overnight at rt in the dark and was
quenched with a saturated aqueous solution of sodium bisulfite.
The resulting mixture was extracted with ethyl acetate (3 times). The
combined organic layers were washed with brine, dried over sodium
sulfate, and concentrated under reduced pressure. The crude oil
was purified by column chromatography on silica gel with cyclo-
hexane/ethyl acetate 1:1 as eluent to obtain 50, a mixture of
2 diastereoisomers as a pale yellow oil (630 mg, 2.26 mmol,
55%). Rf 0.1 (cyclohexane/ethyl acetate 1:1); [R]22D þ59.9 (c 1.0,
CHCl3); 1H NMR (270 MHz, CDCl3) δ 7.49-7.16 (m, 5H), 5.66-
5.54 (m, 1H), 4.58 (s, 2H), 4.30-4.16 (m, 3H), 3.97-3.79 (m, 2H),
3.36-3.22 (br s, 1H), 2.56-1.82 (br s, 1H), 1.56 and 1.47 (d, J=
6.8 Hz, 3H); 13C NMR (67.5 MHz, CDCl3) δ 172.0, 137.2, 128.7,
128.5, 128.3, and 128.2 (2 dias), 128.1 and 128.0 (2 dias), 84.0, 81.8,
and 81.7 (2 dias), 71.7and 71.6 (2 dias), 64.0, 62.3, 57.2, 21.3;
HRMS (EI) calcd for C12H12O 172.0888, found 172.0900.
(d, J=8.9 Hz, 2H), 5.75 (q, J=6.8 Hz, 1H), 4.60 (s, 2H), 4.21
(s, 2H), 3.84 (s, 3H), 1.65 (d, J=6.8 Hz, 3H).
(R)-9-(Benzyloxymethyl)-7-methoxy-1-methylfuro[3,4-b]quinolin-
3(1H)-one (52): To a solution of crude imine 51 (682 mg, 1.94 mmol,
1 equiv) in 12 mL of dichloromethane under argon atmosphere at rt
were added dichlorodicyanoquinone (440 mg, 1.94 mmol, 1 equiv)
and boron trifluoride diethyl etherate (244 μL, 1.94 mmol, 1 equiv).
The mixture was stirred for 30 min and the solvent was
removed under reduced pressure. The dark residue was pur-
ified by column chromatography on silica gel with cyclohex-
ane/ethyl acetate 7:3 as eluent and by a followed recrystally-
zation in ethanol to obtain 52 as a light yellow solid (440 mg,
1.26 mmol, 65% over 3 steps). Rf 0.3 (cyclohexane/ethyl
acetate 1:1); mp 174-176 °C ; ee 89%; [R]22 -12.8 (c 1.0,
D
1
CHCl3); H NMR (270 MHz, CDCl3) δ 8.28 (d, J=9.5 Hz,
1H), 7.48 (dd, J=9.5 and 2.4 Hz, 1H), 7.43-7.34 (m, 5H), 7.19
(d, J=2.4 Hz, 1H), 5.90 (q, J=6.5 Hz, 1H), 5.06 (ABq, J=
13.0 Hz, 2H), 4.72 (ABq, J=11.6 Hz, 2H), 3.93 (s, 3H), 1.69 (d,
J=6.5 Hz, 3H); 13C NMR (67.5 MHz, CDCl3) δ 168.2, 160.0,
146.1, 142.1, 137.7, 136.7, 136.4, 133.0, 128.9, 128.6, 128.3,
128.1, 123.6, 101.1, 76.8, 73.6, 65.8, 55.6, 21.1; HRMS (EI)
calcd for C21H19NO4 349.1314, found 349.1313.
(R)-5-(Benzyloxy)pent-3-yn-2-yl 2-(4-methoxyphenylimino)-
acetate (51): To a solution of 50 (540 mg, 1.94 mmol, 1 equiv)
in 10 mL of tetrahydrofuran/water 9:1 at rt was added sodium
periodate (830 mg, 3.88 mmol, 2 equiv). The mixture was stirred
overnight at rt and the iodine salts were removed by filtration
through a pad of Celite. A saturated aqueous solution of
Na2S2O5 was added to the filtrate and the resulting mixture
was extracted with diethyl ether (3 times). The combined organic
layers were washed with brine, dried over sodium sulfate, and
concentrated under reduced pressure. To a solution of this
yellow oil in 8 mL of toluene under argon atmosphere at rt
Acknowledgment. We gratefully acknowledge Prof.
ꢀ
Philippe Uriac (LSLP, Universite de Rennes 1, France)
and Prof. Jean-Pierre Hurvois (UMR 6226, Universite de
ꢀ
Rennes 1, France) for fruitfull discussions. We are grateful
ꢀ
for the support provided by Rennes Metropole, the Region
Bretagne, and the Universite de Rennes 1. We also thank the
ꢀ
ꢀ
˚
was added 5 g of molecular sieves 4 A in powder and p-anisidine
ꢁ
Ministere de la Recherche for the fellowship of S.D.
(240 mg, 1.94 mmol, 1 equiv). The resulting mixture was stirred
overnight at rt and the molecular sieves was removed by filtra-
tion. The filtrate was concentrated under reduced pressure
to obtain 51 as a yellow oil (680 mg, quant.), which was not
purified and isolated. Rf 0.5 (cyclohexane/ethyl acetate 1:1); 1H
NMR (270 MHz, CDCl3) δ 7.96 (s, 1H), 7.45-7.22 (m, 7H), 6.94
Supporting Information Available: Full details of the
1
experimental procedures including H, 13C spectra and X-ray
crystallographic data. This material is available free of charge
via the Internet at http://pubs.acs.org.
6734 J. Org. Chem. Vol. 74, No. 17, 2009