A hetero Diels-Alder approach to the synthesis of the first angucyclinone and angucycline 5-aza-analogues

Article abstract of DOI:10.1016/j.tetlet.2004.04.123

The synthesis of two new heterodienes and their regioselective [4+2]cycloaddition reactions with several bromo-naphthoquinones allowed us to prepare the first angucyclinone and angucycline 5-aza-analogues.

Full text of DOI:10.1016/j.tetlet.2004.04.123

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 4911–4915
A hetero Diels–Alder approach to the synthesis of the first
angucyclinone and angucycline 5-aza-analogues
a
Sylvain C. Collet, Jean-Francois Remi, Claire Cariou, Samia Laıb,
a
a
a
ꢀ
Andre Y. Guingant, Nguyen Quang Vu and Gilles Dujardin
€
ß
a,
b
b
*
ꢀ
aLaboratoire de Synthꢁese Organique, UMR CNRS 6513, FR CNRS 2465, Facultꢀe des Sciences et des Techniques,
UFR Sciences, 2 rue de la Houssiniꢁere, BP 92208 – 44322 Nantes Cedex 03, France
bLaboratoire de Synthꢁese Organique, UMR CNRS 6011, Universitꢀe du Maine, F-72085, Le Mans Cedex-9, France
Received 18 March 2004; revised 21 April 2004; accepted 21 April 2004
Available online 8 May 2004
Abstract—The synthesis of two new heterodienes and their regioselective [4+2]cycloaddition reactions with several bromo-naph-
thoquinones allowed us to prepare the first angucyclinone and angucycline 5-aza-analogues.
Ó 2004 Elsevier Ltd. All rights reserved.
Angucyclines are secondary metabolites of microor-
ganisms, which belong to the group of Actinomycetes
spore-forming bacteria. These natural products are
structurally characterised by an angular tetracyclic
benz[a]anthracene framework originating from a de-
caketide chain precursor and by the unusual presence of
a C-glycosidic moiety. Among the angucycline group
some members exhibit biological activities (e.g., anti-
tumoural, antiviral, inhibition of enzymes, etc.) al-
though, until to now, none of them has been recognised
as a clinically useful drug.1 Creating chemical diversity
from structures already validated by Nature is an
important task in the ongoing search for new drug
candidates2 and this problem has been recently
addressed in the field of polyketides by recourse to
combinatorial biosynthesis techniques.3 A chemical
approach to the problem would be to synthesise angu-
cycline or angucyclinone analogues bearing structural
modifications on the aglycone moiety and (or) on the
associated sugar unit, if present. Towards this end we
have explored the abilities of two new dienes, namely
(EE)-1a and (EE)-1b, to react with several naphthoqui-
none-type dienophiles. In the present communication we
wish to present the results of this preliminary study,
which establish that the strategy outlined in Scheme 1 is
effective for the preparation of 5-aza-analogues of an-
gucyclines (angucyclinones) with the B-ring aromatised
(e.g., 2). It is worth mentioning that the nitrogen ele-
ment has not been found in the angucycline class in
contrast to the related class of tetracycline antibiotics.
Nitrogen can be found, however, in the structurally re-
lated angucyclins phenanthroviridine 3 and the more
complex jadomycins.4
The push–pull dienes (EE)-1a and (EE)-1b, required for
our study, were not known in the literature. Diene (EE)-
1a was prepared in three simple steps from the com-
mercially available cyclohexane-1,3-dione 4 as depicted
in Scheme 2.
The 1,3-dione was first transformed (80% yield) to the
enaminone 5 following a reported procedure.5 Treat-
ment of the latter compound with a slight excess of N-
(dimethoxymethyl)-N,N-dimethylamine afforded diene
1a as a 3/1 mixture of E,E:Z,E-isomers. Heating a tol-
uene solution of this mixture at reflux finally delivered
the desired and most stable EE-isomer 1a6 in 92% yield
from 5. Diene (EE)-1b was prepared from 5-methyl-
cyclohexane-1,3-dione 67 in a manner identical to that
described for diene (EE)-1a and in an overall yield of
52%.
Keywords: Angucycline; Angucyclinone; Angucyclin(one) 5-aza-ana-
logues; Azadiene; Bromo-naphthoquinone; Hetero Diels–Alder reac-
tion; C-Glycoside.
* Corresponding author. Tel.: +33-02-51-12-54-92; fax: +33-02-51-12-
54-02; e-mail: guingant@chimie.univ-nantes.fr
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.04.123

4912
S. C. Collet et al. / Tetrahedron Letters 45 (2004) 4911–4915
O
O
R
R
O
1
3
A
4 + 2
B
C
C-glycosidic
residue
D
N
N
O
N
Me
Me
2
(EE) -1a R = H
(EE) -1b R = Me
R
O
Me
Me
O
O
3 R =
HO
MeHN
O
N
OH
Scheme 1.
O
O
HO
O
O
i
ii
+
O
N
O
NH2
N
Me2N
NMe2
(E,E)-1a
4
5
(Z,E)-1a
O
Me
iii
O
Me
N
O
NMe2
(EE) -1b
6
Scheme 2. Synthesis of dienes (EE)-1a and (EE)-1b. Reagents and conditions: (i) NH4OAc, reflux in toluene (Dean–Stark apparatus) for 5 h;
(ii) NMe2–CH(OMe)2, reflux in anhydrous THF for 2 h 30 min; (iii) reflux in dry toluene for 1 h.
The cycloaddition reaction of diene (EE)-1a towards a
set of diversely substituted naphthoquinone-type di-
enophiles was next considered. The first such dienophile
investigated was naphthoquinone itself. Heating both
partners in a toluene solution at reflux for days gave
only faint indication of success, the reaction leading to
the expected angucyclinone 5-aza-analogue 7 in less
than 10% isolated yield. Similar results were obtained
with juglone as the dienophile and no improvement of
the yield could be achieved in the presence of a Lewis
acid. We then reasoned that the presence of an electron-
withdrawing substituent on the naphthoquinone double
bond should enhance [4+2]cycloaddition based on FMO
considerations. In full agreement with this expectation,
we were pleased to observe that the readily prepared 2-
bromo-[1,4]naphthoquinone 88 reacted with diene (EE)-
1a (heating in acetonitrile at 40 °C for 4 h then at 60 °C
overnight) to give 7 as the sole isolated compound in
65% yield. The formation of 7 can be accounted for by
assuming that, due to the acidity of the hydrogen at
C-12b (angucycline numbering), the initial Diels–Alder
adduct, supposed to arise from an endo event, is subject
to an aromatisation process resulting of the facile
elimination of dimethylamine and HBr (Scheme 3). This
cascade sequence thus represents an efficient one-pot
approach towards the 5-aza-angucyclinone skeleton in
which the ABCD framework is assembled in a single
operation and in an acceptable overall yield.
In an effort to investigate the scope of the preceding
[4+2]cycloaddition–aromatisation cascade we next ex-
plored the reactivity of diene (EE)-1a with several 2-
bromo-[1,4]naphthoquinones substituted at carbons C-5
or C-8. The results of these experiments are depicted in
Scheme 4. The 5-acetoxy-2-bromo-[1,4]naphthoquinone
9,9 5-hydroxy-2-bromo-[1,4]naphthoquinone 11,9b 8-
methoxy-2-bromo-[1,4]naphthoquinone 1310 and 8-hy-
droxy-2-bromo-[1,4]naphthoquinone 1511 were all pre-
pared following the reported procedures. Their
cycloaddition with diene (EE)-1a12 closely paralleled the
cycloaddition of 8 with the same diene and gave the
angucyclinone 5-aza analogues 10, 12, 14 and 16,

S. C. Collet et al. / Tetrahedron Letters 45 (2004) 4911–4915
4913
O
O
O
H
O
O
O
O
H
12b
+
(EE)-1a
H
N
N
Br
Br
Br
O
NMe
2
8
O
O
O
"one pot"
reaction
NHMe . HBr
2
65% yield
N
7
Scheme 3. Proposed reaction mechanism for the formation of the angucyclinone 5-aza-analogue 7.
OAc
O
O
O
OH
O
O
OH
O
OAc
O
O
25%
64%
N
N
Br
Br
15%
O
(+ 14% of 12)
O
OH
O
O
12
9
11
10
NMe2
i
40%
O
O
O
O
O
O
O
O
O
O
N
N
52%
Br
75%
Br
OMe
13
OH
OH
OMe
14
16
15
ii
51%
Scheme 4. Synthesis of angucyclinone 5-aza-analogues 10, 12, 14, 16.12 Reagents and conditions: (i) 6 N H2SO4/EtOH (1/3), reflux 1 h 30 min;
(ii) 1.1 equiv Ag2O, 20 equiv MeI, 40 °C, 16 h.
respectively, each of them featuring, like 7, a 3,4-dihy-
dro-2H-benzo[j]phenanthridin-1,7,12-trione framework.
These compounds were obtained in moderate to good
yields except for the 5-aza analogue 12, which was iso-
lated in significantly reduced yields. In that case, non-
negligible amounts (15%) of the dimethylamine addition
product to 11 were also isolated. The chemical structures
of the four angucyclinone 5-aza-analogues were deter-
yield along with small amounts (14%) of its deacetylated
product 18 (Scheme 5).
Having establish the suitability of the sequential
[4+2]cycloaddition–aromatisation protocol to generate
angucyclinone 5-aza-analogues, we turned our attention
to the application of the same protocol to the synthesis
of angucycline 5-aza-analogues. Towards this end, the 2-
bromo-[1,4]naphthoquinone 22, bearing a C-glycoside
residue at carbon C-6 was selected as the test dienophile.
As illustrated in Scheme 6, its synthesis could be effi-
ciently accomplished in three steps. In the first opera-
tion, and following the procedure reported by Toshima
et al.14 for aryl C-glycosidation of unprotected sugars,
the commercially available 1,5-naphthalene diol 18 and
1
mined by H NMR13 and subsequently confirmed by
establishing structural correlations between 10 and 12 as
well as between 16 and 14, as indicated in Scheme 4.
Another significant aspect of these cascade reactions
worth noting is the complete regioselectivity of the ini-
tial [4+2]cycloaddition, which is mainly governed by the
bromine atom.
D
-deoxyglucose 19 were reacted in the presence of
trimethylsilyltriflate to give 20, which was next subjected
to peracetylation conditions to give the pentaacetate 21.
Exposure of 21 to an excess of NBS in aqueous acetic
acid proceeded without events as expected from
literature precedents,9a;15 and provided the targeted
Additionally, we also effected the synthesis of adduct 17
by reacting diene (EE)-1b with the 5-acetoxy-2-bromo-
[1,4]naphthoquinone 9 in a manner similar to the
preparation of 10.12 Compound 17 was isolated in 57%

4914
S. C. Collet et al. / Tetrahedron Letters 45 (2004) 4911–4915
O
Me
O
Me
O
R
O
O
OAc
ref. 12
+
N
N
Br
O
NMe2
17 R = OAc 57%
9
(EE)-1b
14%
18 R = OH
Scheme 5. Synthesis of angucyclinone 5-aza-analogue 17.
OH
OH
OH
O
HO
HO
i
OH
+
67-77%
HO
HO
O
OH
OH
20
18
19
OH
OAc
O
OAc
O
AcO
AcO
AcO
O
ii
90%
OAc
OAc
iii
AcO
70-75%
Br
O
OAc
21
22
OAc
AcO
AcO
O
iv
85-90%
O
O
OR
O
N
23 R = Ac
v
Quant.
24 R = H
Scheme 6. Synthesis of angucycline 5-aza-analogue 24. Reagents and conditions: (i) 18 (2 equiv), TMSOTf (0.33 equiv), dry CH3CN, rt for 15 h, then
Et3N; (ii) pyridine/Ac2O (3/1), 80 °C for 15 h; (iii) NBS (6 equiv), AcOH/H2O (1/2), 65–70 °C for 3 h; (iv) 1a, CH3CN, 40 °C for 3 days; (v) NH4OAc
(8 equiv) in MeOH/H2O (4:1), rt for 7 h.
dienophile 22 in good yield. Admixture of 22 with diene
(EE)-1a followed by a prolonged heating in acetonitrile
afforded the angucycline 5-aza-analogue 23 along with
its monodeacetylated product 24. Treatment of the
mixture with an excess of ammonium acetate16 effected
the total and chemoselective conversion of 23 into the
angucycline 5-aza-analogue 24, which was isolated in
85–90% yield for the two steps.
an aromatised B-ring and bearing various substituents at
carbons C-8, C-10 or C-11 in ring D. The extension of
this protocol to the synthesis of further analogues of
angucyclinones and angucyclines as well as the utilisa-
tion of diene 1a in the synthesis of natural products of
biological significance are currently underway.
Acknowledgements
In summary, the simple and efficient synthesis of two new
‘push–pull’ heterodienes, namely (EE)-1a and (EE)-1b,
were described. Starting from these dienes we established
a domino [4+2]cycloaddition–aromatisation protocol as
a simple and efficient method for the one-pot synthesis of
angucyclinone and angucycline 5-aza-analogues having
ꢀ
The support of the ‘Region des Pays de la Loire’ (CER
2000-2006) is gratefully acknowledged. We are also
grateful to Profs. C. Monneret (Curie Institute, Paris)
and J. P. Gesson (University of Poitiers) for helpful
discussions.

S. C. Collet et al. / Tetrahedron Letters 45 (2004) 4911–4915
4915
L.; Fleitz, F. J.; Daffner, M.; Stevens, J. E. J. Org. Chem.
1991, 56(1), 91–95.
References and notes
10. Hannan, R. L.; Barber, R. B.; Rapoport, H. J. Org. Chem.
1979, 44(13), 2153–2158.
11. Gesson, J.-P.; Mondon, M. J. Chem. Soc., Chem. Com-
mun. 1982, 421–423.
12. A representative experimental procedure is as follows
1. (a) Rohr, J.; Thiericke, R. Nat. Prod. Rep. 1992, 103–137;
(b) Krohn, K.; Rohr, J. Top. Curr. Chem. 1997, 127–195.
2. Breinbauer, R.; Vetter, I. R.; Waldmann, H. Angew.
Chem., Int. Ed. 2002, 41, 2878–2890.
3. (a) Piel, J.; Hertweck, C.; Shipley, P. R.; Hunt, D. M.;
Newman, M. S.; Moore, B. S. Chem. Biol. 2000, 7, 943–
955, and references cited therein; (b) Moore, B. S.; Piel, J.
Antonie van Leeuwenhoek J. Microbiol. 2000, 78, 391–398,
and references cited therein.
4. (a) De Frutos, O.; Echavarren, A. M. Tetrahedron Lett.
1996, 37, 8953–8956; (b) Mohri, S.; Stefinovic, M.;
Snieckus, V. J. Org. Chem. 1997, 62, 7072–7073; (c)
Valderrama, J. A.; Gonzalez, M. F.; Valderrama, C.
Tetrahedron 1999, 55, 6039–6050.
(preparation
of
the
8-hydroxy-3,4-dihydro-2H-
benzo[j]phenanthridine-1,7,12-trione 16). To a solution
of diene 1a (1 mmol) in dry CH3CN (5 mL), 2-bromo-8-
hydroxy-[1,4]naphthoquinone 15 (1 mmol) was added at
room temperature with stirring. After 4 h at this temper-
ature and 18 h at 40 °C, the solvent was removed under
reduced pressure. The residue was purified by silica gel
chromatography (Toluene/AcOEt ¼ 4:1) to give the trione
1
16 (75% yield): H NMR (CDCl3, 300 MHz): d 2.30 (m,
2H), 2.94 (t, J ¼ 7:1 Hz, 2H), 3.23 (t, J ¼ 6:3 Hz, 2H),
7.27–7.35 (m, 1H), 7.65–7.73 (m, 2H), 9.48 (s, 1H), 12.11
(s, 1H). 13C NMR (CDCl3, 75 MHz): d 21.5, 33.4, 39.1,
115.2, 120.1, 124.7, 126.7, 128.8, 134.6, 137.6, 141.1, 151.2,
162.2, 169.6, 182.1, 187.1, 197.5. FT-IR (KBr, cmꢀ1):
3500, 1700, 1680, 1634. MS (EI, 70 eV): m=z ¼ 293 (Mþꢁ,
56), 265 (100), 237 (15), 209 (12), 181 (3), 180 (4), 153 (8),
126 (9). Anal. Calcd C17H11NO4: C, 69.62; H, 3.78; N,
4.78. Found C, 69.39; H, 3.84; N, 4.71.
5. Huang, Y.; Hartmann, R. W. Synth. Commun. 1998, 28,
1197–1200.
6. Spectroscopic data for diene 1a: 1H NMR (CDCl3,
300 MHz): d 2.01 (m, 2H), 2.32 (t, J ¼ 6:5 Hz, 2H), 2.46
(t, J ¼ 6:3 Hz, 2H), 3.03 and 3.06 (2s, 6H), 5.43 (s, 1H),
7.62 (s, 1H). 13C NMR (CDCl3, 75 MHz): d 22.1, 30.5,
34.4, 36.8, 40.4, 110.7, 152.0, 172.5, 199.9. FT-IR (liquid
film, cmꢀ1): 2939, 2815, 1615, 1557. MS (EI, 70 eV):
m=z ¼ 166 (Mþꢁ, 91), 138 (18), 123 (46), 109 (32), 71 (27),
44 (100), 28 (9). HRMS (EI): calcd for C9H14N2O:
166.1106, found 166.110 [Mþꢁ].
13. Regioselectivities of the [4+2]cycloaddition reactions were
determined by recourse to HMQC and HMBC experi-
3
ments (observation of J couplings).
7. Szychowski, J.; MacLean, D. B. Can. J. Chem. 1979, 57,
1631–1637.
8. (a) Heinzman, S. W.; Grunwell, J. R. Tetrahedron Lett.
1980, 21, 4305–4308; (b) Jung, M. E.; Hagenah, J. A.
J. Org. Chem. 1983, 48, 5359–5361.
9. (a) Boyd, V. A.; Sulikowski, G. A. J. Am. Chem. Soc.
1995, 117, 8472; (b) Grunwell, J. R.; Karipides, A.; Wigal,
C. T.; Heinzman, S. W.; Parlow, J.; Surso, J. A.; Clayton,
14. Toshima, K.; Matsuo, G.; Ishizuka, T.; Ushiki, Y.;
Nakata, M.; Matsumura, S. J. Org. Chem. 1998, 63,
2307–2313.
15. Krohn, K.; Agocs, A.; Baeuerlein, C. J. Carbohydr. Chem.
2003, 22, 579–592.
16. Ramesh, C.; Mahender, G.; Ravindranath, N.; Das, B.
Tetrahedron 2003, 59, 1049–1054.