First efficient synthesis of novel oxophenyl-arcyriaflavin analogs

Article abstract of DOI:10.1016/j.tetlet.2005.07.006

New oxophenylarcyriaflavins were synthesized in a few efficient steps. The key steps involved at first a palladium cross-coupling between the 3-bromo-4-(1H-indol-3-yl)1-methylpyrrole-2,5-dione and the 2-formylphenylboronic acid or a methyl 2-trialkylstann

Full text of DOI:10.1016/j.tetlet.2005.07.006

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6071–6074
First efficient synthesis of novel oxophenyl-arcyriaflavin analogs
*
´
´
´ ´
´
Aurelie Bourderioux, Sylvain Routier, Valerie Beneteau and Jean-Yves Merour
Institut de Chimie Organique et Analytique, Universite´ d’Orle´ans, UMR CNRS 6005, rue de Chartres, B.P. 6759,
45067 Orle´ans Cedex 2, France
Received 31 May 2005; revised 30 June 2005; accepted 1 July 2005
Abstract—New oxophenylarcyriaflavins were synthesized in a few efficient steps. The key steps involved at first a palladium cross-
coupling between the 3-bromo-4-(1H-indol-3-yl)1-methylpyrrole-2,5-dione and the 2-formylphenylboronic acid or a methyl 2-tri-
alkylstannylbenzoate, followed by an intramolecular acylation in a C-2 indolic position. All the sequence was carried out without
any indolic protective group.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Indolocarbazole derivatives are a family of natural or
synthetic products commonly exemplified by the Rebec-
camycin I, a weak inhibitor of topoisomerase I.1 In
order to develop selective kinase inhibitors, interacting
with the ATP binding site of the enzymes, modifications
of the indolocarbazole moiety appeared as a valid alter-
native to the synthesis of glycosylated compounds.
Indeed, recent developments in this domain confirmed
the inhibition of Cyclin dependent kinases (CDK) by
indolocarbazole aglycones such as arcyriaflavin A II or
several (het)aryl carbazoles in a nanomolar range.2
H
N
R1
N
O
O
O
O
N
N
H
H
N
H
N
II : Arcyriaflavin A
R
R
H
O
O
N
O
OH
OH
HO
R1
X
I : Rebeccamycin,
In addition, the use of indene versus indole generated
the CEP-7055 III, a strong inhibitor of VEGF-R2
(KDR kinase), which is a candidate for clinical trials
in the antiangiogenic area (Fig. 1).3
X = OMe, R = Cl, R1 = H
N
R2
R1 = CH2Oipr,
R2 = (CH2)3OCOCH2N(CH3)2
III :
R2
N
O
R2
O
N
O
O
For our own part, we demonstrated that one of the two
indoles could be replaced by a naphthyl (compounds IV)
or a phenyl group (compounds V).4,5 In addition, a few
studies reported bis-indolic compounds, which con-
tained a seven-membered ring but, so far, no significant
activities have been reported. For instance, derivatives
such as homoarcyriaflavin,6 arcyriacyanin A,7 and the
marine alkaloid caulersine, were recently described and
synthesized (Fig. 2).8–10
R1
R1
N
H
IV : naphtocarbazoles
N
H
V : phenyl carbazoles
Figure 1. Most representative indolocarbazole derivatives.
Here we report the design and synthesis of new hybrid
derivatives 5 and 11. They contained an indole as I–V,
a seven-membered ring related to VIII and a phenyl
group like V. The central cycloheptatrienone moiety
was coupled with a maleimide or a maleic anhydride.
The resulting original structures were obtained applying
Keywords:
Anticancer
agents;
Indolocarbazoles
synthesis;
Arcyriaflavin.
*
Corresponding author. Tel.: +33 2 38 49 48 53; fax: +33 2 38 41 72
81; e-mail: sylvain.routier@univ-orleans.fr
0040-4039/$ - see front matter ꢀ 2005Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.07.006

6072
A. Bourderioux et al. / Tetrahedron Letters 46 (2005) 6071–6074
H
N
H
N
the retrosynthetic scheme shown in Figure 3, in a few
steps from 3-bromo-4-(1H-indol-3-yl)-1-methylpyrrole-
2,5-dione 110,11 and involved as key steps (i) a cross-cou-
pling palladium catalyzed reaction, (ii) an oxidation,
and (iii) an intramolecular cyclization in the C-2 indolic
position.
O
O
O
O
NH
N
H
N
H
N
H
VIII : Arcyriacyanin A
VI : Homoarcyriaflavin
First reaction (route A) involved the preparation of 1,
which was classically obtained from indole and 2,3-
dibromomaleimide in the presence of LiHMDS in THF
in a fair good yield.11 A palladium cross-coupling Suzuki
type reaction (Table 1) between 1 and the commercially
available 2-formylphenylboronic acid 2 (1.1 equiv) led
to 3. According to our previous work using the 2-meth-
oxynaphthalene boronic acid,5 we carried out the reac-
tion in presence of K2CO3 (1.8 equiv) in a refluxing
mixture of dioxane/water 85/15 using Pd(OAc)2 as a cat-
alyst (entry 1). The reaction afforded the desired com-
pound 3 in 30% yield after 13 h. Increasing the reaction
time was unfruitful and led to the degradation of the mix-
ture proving the low stability of 3, certainly due to the
presence of the free nitrogen atom in aqueous basic con-
ditions. Decreasing the reaction time to 8 h led to 3 in
only 26%. Fortunately, the reaction with 1.5equiv of
boronic acid led after 6 h to the desired compound 3 in
72% yield. This result was higher than the one obtained
in the naphthalenic studies (55%)5 (Scheme 1).
O
MeO
H
N
N
H
O
VIII : Caulersine
Figure 2. Bisindolyl compounds containing a central seven-membered
ring.
O
X
X
O
O
O
COR
1
N
N
H
5
O
H
R = H, X = NCH3
R = OH, X = O
X = NCH3
11 X = O
The next step was aimed at generating a C-2 indolic car-
bonyl function. The literature reported several useful
Figure 3. Retrosynthetic scheme.
Table 1. Conditions of the palladium catalyzed reactions. All the reactions were carried out at reflux
Entry Reagents Catalyst (equiv) Additive (equiv) Solvent
Reaction time (h) Product (yield, %)
1
1 (1 equiv), 2 (1.1 equiv) Pd(OAc)2 (0.1)
K2CO3 (1.8)
Dioxane/water 85/15 (a) 13
3 (30)
(b) 8
6
23
2
8
3 (26)
3 (72)
7 (47)
8 (97)
9 (12)
9 (74)
9 (70)
2
3
4
5
6
7
1 (1 equiv), 2 (1.5equiv) Pd(OAc) 2 (0.1)
K2CO3 (1.8)
Sn2Bu6 (1.2)
Sn2Me6 (1.2)
Dioxane/water 85/15
Toluene
Toluene
Dioxane
Dioxane
6
6
Pd(PPh3)4 (0.1)
Pd (PPh3)4 (0.1)
1 (1 equiv), 7 (1.2 equiv) PdCl2(PPh3)2 (0.2) CuI (0.1)
1 (1 equiv), 7 (1.5equiv) PdCl 2(PPh3)2 (0.2) CuI (0.1)
1 (1 equiv), 8 (1.5equiv) PdCl 2(PPh3)2 (0.2) CuI (0.1)
6
5
Dioxane
O
O
O
O
N
N
N
N
CHO
CO2H
O
O
O
O
O
Br
i
ii
iii
Route A
N
N
N
H
N
H
O
H
H
1
3
9
4
5
O
O
O
O
N
N
O
O
CO2Me
CO2H
O
O
O
iv
v
Br
vi
Route B
N
H
1
N
N
N
H
O
H
H
10
11
Scheme 1. Reagents and conditions: (i) 2-formylphenylboronic acid 2 (1.5equiv), Pd(OAc) 2 (0.1 equiv), K2CO3 (1.8 equiv), dioxane/water 85/15,
rflx., 6 h, 72%; (ii) NH2SO3H (8.1 equiv), NaClO2 (2.2 equiv), dioxane/water, 6 ꢁC, 1 min, 81%; (iii) BF3ÆEt2O (40 equiv), DCE, rflx., 6 h, 89%; (iv)
from 7 (1.5equiv), PdCl 2(PPh3)2 (0.2 equiv), CuI (0.1 equiv), dioxane, rflx., 6 h, 74%, from 8 idem, 5h, 70%; (v) aq KOH (45%, 60 equiv), acetone, rt,
20 h then HCl concn to pH = 1, rt, 3 days, 98%; (vi) see (iii), 20 h, 70%.

A. Bourderioux et al. / Tetrahedron Letters 46 (2005) 6071–6074
6073
methods in the absence of a nitrogen indolic protective
group. These implied the use of carboxylic acids, and
the formation of an acylium ion by Friedel–Crafts
related reactions. So we oxidized the aldehyde function
of 3 into a carboxylic acid with an aqueous solution of
sodium chlorite (2.2 equiv) and sulfamic acid (8.1 equiv)
between 7 and 10 ꢁC.12 After only 1 min the starting
material completely disappeared and the reaction affor-
ded the desired carboxylic acid 4 in 57% yield. If the
reaction was performed at 6 ꢁC by accurate control of
the temperature during the oxidant addition, the acid
4 was obtained in 81% yield. An increase of the time
to 5min led only to degradation. Numerous variations
indicated that these conditions were extremely efficient,
but extremely time and concentration sensitive.
with no significant results. The same reaction was car-
ried out using hexamethylditin (in spite of its greater
toxicity). Indeed, whereas the use of Sn2Bu6 proved to
be quite ineffective, Nicolaou reported the achieving of
71% yield shifting from Bu to Me.20 In our hand, after
a reaction time of 2 h, compound 8 was obtained in
97% yield (entry 4).
The Stille procedure between the bromo compound
1 and the stannylated derivative 7 (1.2 equiv) was
performed with PdCl2(PPh3)2 as a catalyst in refluxing
dioxane in the presence of CuI. After 8 h only 12% of
the desired compound 9 was obtained (entry 5). Increas-
ing the amount of 7 to 1.5equiv reduced the reaction
time to 6 h and the yield increased to 74%. Under similar
conditions, the trimethyltin derivative 8 led to the same
product 9 after 5h in 70% yield (entries 6 and 7). The
ester was saponified using aqueous KOH in acetone at
room temperature during 5h. Then, the acidification
step led, after 12 additional hours, to the maleic com-
pound 10 in 98% yield without any difficulty. The intra-
molecular cyclization corresponding to the final step was
performed with BF3ÆEt2O during 20 h to afford the sec-
ond targeted analog 11 in 70% yield. During the course
of this cyclization, no side reaction was observed in spite
of the presence of the anhydride function.
Following route A, the ring closure leading to the
central seven-membered cycle was the next goal. As
described in the literature, we chose to perform it
through an electrophilic cyclization. Common reagents
for such a reaction were AlCl3 or phosphoric reactants
such as PPSE.13–15 Milder conditions involved the use
of an anhydride in presence of BF3ÆEt2O.16–18 For our
own part, we carried out the reaction in the presence
of a large excess of BF3ÆEt2O as a Lewis acid in refluxing
DCE without any pre-activation of the carboxylic acid
function.16–18 After 12 h, compound 5 was isolated in
66% yield along with numerous nonidentified by-prod-
ucts. The yield could be increased to 89% by lowering
the reaction time to 5h 30 min. After completion of
the reaction, an aqueous treatment and extractions led
directly to pure 5. To our knowledge, it is the first exam-
ple of such an acylation using BF3ÆEt2O without an ind-
olic protection.
In this paper we described two different ways to obtain
the N-methylmaleimide compound 5 and the anhydride
11 in three steps from compound 1. Route A, which
included a Suzuki reaction led to 5 in 52% overall yield.21
Route B involving a Stille reaction afforded the anhydride
11 in 51% global yield. The two routes were quite similar
but way B implied the preparation of a trimethyltin deriv-
ative by an additional step. The biological evaluation of 5
and 11 as kinase inhibitors are being currently performed.
Further investigations to explore the reactivity of the
maleimide and anhydride functions and a generalization
of the synthetic sequence to substituted indoles are
already in progress and will be reported in due course.
Considering that the Suzuki reaction and the highly sen-
sitive oxidation to access compound 4 were limiting
steps to scale up the synthesis, route B was designed.
The 2-tributylstannyl and trimethylstannyl derivatives
7 and 8 were reported as efficient reagents in such a
strategy.19,20 Starting from the methyl-2-bromo benzo-
ate 6, compound 7 could be prepared using Pd(PPh3)4
as a catalyst. Unfortunately, the yield was limited to
41% after 21 h using a low amount of palladium catalyst
(0.008 equiv). Surprisingly, increasing the amount of
catalyst to 0.1 equiv did not decrease the reaction time.
After 23 h, the yield of compound 5 was slightly
increased to 47% (Scheme 2, entry 3).
Acknowledgements
´
ˆ
´
We thank Canceropole Grand Ouest, the Region Centre
and the Association pour la Recherche contre le Cancer
(ARC) for the financial support of this work.
In order to enhance the yield, we decided to explore
some other conditions by slight step by step modifica-
tions. So we used PdCl2(PPh3)2 as a catalyst, in the pres-
ence or absence of LiCl in refluxing THF or dioxane
References and notes
1. (a) Prudhomme, M. Curr. Med. Chem. Anti Cancer Agents
2004, 6, 509–521; (b) Kno¨lker, H.-J.; Reddy, K. R. Chem.
Rev. 2002, 102, 4303–4427.
2. Sanchez-Martinez, C.; Shih, C.; Faul, M. M.; Zhu, G.;
Paal, M.; Somoza, C.; Li, T.; Kumrich, C. A.; Winneroski,
L. L.; Xun, Z.; Brooks, H. B.; Patel, B. K. R.; Schultz, R.
M.; DeHahn, T. B.; Spencer, C. D.; Watkins, E.;
Considine, S. A.; Dempsey, J. A.; Ogg, C. A.; Campbell,
R. M.; Anderson, B. A.; Wagner, J. Bioorg. Med. Chem.
Lett. 2003, 13, 3835–3839.
CO2Me
SnR3
CO2Me
Br
iv
7 R = Bu
8 R = Me
6
Scheme 2. Reagents and conditions: for 7: Pd(PPh3)4 (0.1 equiv),
Sn2Bu6 (1.2 equiv), tol., rflx., 23 h, 47%, for 8 Sn2Me6 (1.2 equiv), 2 h,
97%.
3. Gingrich, D. E.; Reddy, D. R.; Iqbal, M. A.; Singh, J.;
Aimone, L. D.; Angeles, T. S.; Albom, M.; Yang, S.; Ator,
M. A.; Meyer, S. L.; Robinson, C.; Ruggeri, B. A.;

6074
A. Bourderioux et al. / Tetrahedron Letters 46 (2005) 6071–6074
´
Dionne, C. A.; Vaught, J. L.; Mallamo, J. P.; Hudkins, R.
18. Sayah, B.; Pelloux, L.; Vallee, Y. J. Org. Chem. 2000, 65,
L. J. Med. Chem. 2003, 46, 5375–5388.
2824–2826.
´
4. Routier, S.; Peixoto, P.; Merour, J. Y.; Coudert, G.; Dias,
N.; Bailly, C.; Pierre, A.; Leonce, S.; Caignard, D. H. J.
Med. Chem. 2005, 48, 1401–1413.
19. Kalivretenos, A.; Stille, J. K.; Hegedus, L. S. J. Org.
Chem. 1991, 56, 2883–2894.
20. Nicolaou, K. C.; Hepworth, D.; King, N. P.; Raymond,
M.; Finlay, V.; Scarpelli, R.; Manuela, M.; Pereira, A.;
Bollbuck, B.; Bigot, A.; Werschkun, B.; Winssinger, N.
Chem. Eur. J. 2000, 6, 2783–2800.
´
5. Routier, S.; Merour, J. Y.; Dias, N.; Bailly, C.; Lozach,
O.; Meijer, L., submitted for publication.
6. Mahboobi, S.; Burgemeister, T.; Dove, S.; Kuhr, S.; Popp,
A. J. Org. Chem. 1999, 64, 8130–8137.
21. Synthesis of 5: To a solution of acid 4 (745mg, 2.15mmol)
in dichloroethane (120 mL), BF3ÆEt2O (11 mL, 86.6 mmol)
was added dropwise. The mixture was refluxed for 5h
30 min before quenching with iced water (100 mL). CH2Cl2
(100 mL) was added and the aqueous layer was extracted
three times with CH2Cl2 (60 mL). The combined organic
layers were washed with an aqueous saturated solution of
NaHCO3 (120 mL) then with water (3 · 120 mL). After
drying over Na2SO4, the solvents were removed under
reduced pressure. The orange solid was washed with a
slight amount of cold MeOH to give 5 (626 mg, 89%) : mp
230–233 ꢁC (dec). IR (KBr, cmÀ1) m 3272 (NH), 1768, 1703,
1388, 750. 1H NMR (DMSO-d6, 250 MHz) d (ppm) 2.98 (s,
3H), 7.27 (t, 1H, J = 7.4 Hz), 7.47 (t, 1H, J = 7.4 Hz), 7.63
(d, 1H, J = 8.2 Hz), 7.71–7.86 (m, 2H), 8.30 (d, 1H,
J = 8.0 Hz), 8.63 (d, 1H, J = 8.3 Hz), 8.70 (d, 1H,
J = 8.1 Hz), 13.09 (br s, 1H, NH). 13C NMR (DMSO-d6,
62.9 MHz) d (ppm) 24.1 (CH3), 111.3 (Cq), 112.8 (CH),
121.8 (CH), 123.8 (Cq), 125.0 (CH), 126.4 (CH), 126.9
(CH), 128.1 (Cq), 129.3 (CH), 130.2 (CH + Cq), 130.9
(CH), 131.7 (Cq), 137.1 (Cq), 138.0 (Cq), 141.1 (Cq), 168.9
(C@O), 170.1 (C@O), 179.8 (C@O). HRMS-ESI 328.0848
calcd for C20H12N2O3, found 328.0855 (M+Å).
7. Brenner, M.; Mayer, G.; Terpin, A.; Steglich, W. Chem.
Eur. J. 1997, 3, 70–74.
8. Su, J.-Y.; Zhu, Y.; Zeng, L.-M.; Xu, X.-H. J. Nat. Prod.
1997, 60, 1043–1044.
9. Wahlstro¨m, N.; Stensland, B.; Bergman, J. Tetrahedron
2004, 60, 2147–2153.
10. Fresneda, P. M.; Molina, P.; Saez, M. A. Synlett 1999, 10,
1651–1653.
11. Ohkubo, M.; Nishimura, T.; Jona, H.; Honma, T.;
Morishima, H. Tetrahedron 1996, 52, 8099–8112.
12. Lindgren, B. O.; Nilsson, T. Acta Chem. Scand. 1973, 27,
888–889.
13. Bergman, J.; Lidgren, G.; Gogoll, A. Bull. Soc. Chim.
Belg. 1992, 101, 643–660.
´
14. Joseph, B.; Alagille, D.; Merour, J. Y.; Leonce, S. Chem.
Pharm. Bull. 2000, 48, 1872–1876.
15. Bergman, J.; Wahlstrom, N.; Yudina, L. N.; Tholander,
J.; Lidgren, G. Tetrahedron 2002, 58, 1443–1452.
16. Sahar Al Akoum, E.; Legrand, A.; Rigo, B.; Couturier, D.
J. Heterocycl. Chem. 1999, 36, 997–1000.
17. Medio-Simon, M.; Houshang, E.-A.; Pindur, U. Liebigs
Ann. Chem. 1991, 6, 601–602.