New methodology for acridine synthesis using a rhodium-catalyzed benzannulation

Article abstract of DOI:10.1016/j.tetlet.2004.02.022

A new methodology for the synthesis of acridine derivatives is disclosed. The starting materials are commercially available quinolines, which can be converted, via five high efficient steps, in a key quinoline intermediate substituted with a TBS-protected

Full text of DOI:10.1016/j.tetlet.2004.02.022

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 2783–2786
New methodology for acridine synthesis using
a rhodium-catalyzed benzannulation
Philippe Belmont,* Jean-Christophe Andrez and Charlotte S. M. Allan
Universitꢀe Claude Bernard, Lyon I. UMR CNRS 5181, Mꢀethodologies de Synthꢁese et Molꢀecules Bioactives,
B^atiment CPE, 43 boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex, France
Received 21 January2004; revised 5 February2004; accepted 6 February2004
This paper is dedicated to Professor J. Lhomme
Abstract—A new methodologyfor the synthesis of acridine derivatives is disclosed. The starting materials are commercially
available quinolines, which can be converted, via five high efficient steps, in a keyquinoline intermediate substituted with a TBS-
protected-enol–ether and an internal alkyne. The key and last step is a rhodium-catalyzed benzannulation of the quinoline inter-
mediate yielding the desired poly-substituted acridines derivatives.
Ó 2004 Elsevier Ltd. All rights reserved.
Acridine derivatives (Scheme 1) have been known since
the 19th centurywhere theywere first used as pigments
and dyes.1 Their antiseptic activityhas been discovered
in the early1900s and some derivatives were extensively
used during World War I for their antibacterial and
antimalarial properties.1 In the 1920s, their potential in
the fight against cancer was first noted. Since then, a
large number of acridines drugs, natural alkaloids or
synthetic molecules, have been tested as antitumour
agents, a recent target being their telomerase and
topoisomerase inhibition activity.2 Acridines are much
sought after targets since a range of these compounds
continue to be used todayfor the treatment of acute
leukaemia (amsacrine),3 as anticancer agents (ledakrin),4
for their antibacterial properties (acriflavine and ethac-
ridine),2 for action against parasites in the treatment of
malaria, trypanosomiasis and leishmaniasis (quinacrine,
acranil)2 and last but not least for treatment of Alzhei-
merÕs disease (tacrine).2;5 Acridine and acridone (Scheme
1) can be interconverted relativelyeasilyand several
synthetic routes have been designed for this matter.2;6;7
Most routes to acridines and their derivatives proceed
through the corresponding acridone (Scheme 1). This is
largelydue to the ease of formation of diphenylamine-2-
carboxylic acids (via the Ullman reaction), which under
strong acids can cyclize to the corresponding acridones.8
Although discovered in the early1900s, these reaction
are still in use today.9 The transformation from acridone
to acridine requires harsh reductive (sodium amalgam,1
aluminium and mercuryamalgam 10) and oxidative
(chromic acid, iron(III)chloride/HCl,1 nitric acid9) con-
ditions. Among the methods for forming acridines,
which do not proceed through an acridone intermediate,
11
the best known is probablythe Bernthsen reaction,
which involves heating a diphenylamine and an organic
acid with zinc chloride for up to 40 h. Yields are not high
and the temperature required for best results is between
200 and 270 °C.11 The cyclization of diphenylamine-2-
carboxaldehyde is also widely used for acridine synthe-
sis. Again strong acid conditions (trifluoroacetic acid,
sulfuric acid) and/or high temperatures are required.12
The Pfitzinger quinoline synthesis has been adapted
successfullyto access acridines but requires reflux for
several hours in sodium hydroxide solution.13 Due to the
harsh conditions needed in all these routes, theyappear
unsuitable for the synthesis of functionalized acridines
O
OH
8
5
9
1
4
7
6
2
3
N
N
H
N
Acridine
Acridone
Acridone
Scheme 1. Acridine and acridone structures and numbering.
* Corresponding author. Tel.: +33-4-72445868; fax: +33-4-72432963;
e-mail: belmont@cpe.fr
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.02.022

2784
P. Belmont et al. / Tetrahedron Letters 45 (2004) 2783–2786
O
O
O
a, b
c, d
CH3
CH3
H
Cl
Cl
N
N
N
C4H9
3
1
5
e
OH
O
O
CO2Et
C4H9
f
OEt
N
N
C4H9
7
6
Scheme 2. Reagents and conditions: (a) 3 equiv MeMgBr, THF, 40 °C, 98%; (b) 10 equiv MnO2, toluene, 80 °C, 80%; (c) NaI, CH3CN, 0.5 equiv HCl
4 N, reflux, 80%; (d) 1-hexyne, 0.07 equiv PdCl2(PPh3)2, 0.3 equiv CuI, 1.5 equiv Et3N, toluene, rt, 65%; (e) LDA, NC(CO)OEt, 27%; (f) 1 equiv CSA,
CHCl3, reflux, 15%.
incorporating sensitive functional groups. This obser-
vation led us to look for alternative routes.
intermediate, which is unreactive in the desired trans-
formation, but that can give rise, upon quenching in the
presence of an alcohol (methanol or ethanol), to the
corresponding adduct 9 or 10 via an O-cyclization-
alkylation process.19 These structures are veryinterest-
ing due to their structural features, recalling Pentalongin
and Dehydroherbarin derivatives, which possess a broad
range of biological activities.20
Two routes have been investigated. In the first route
(Scheme 2), we attempted to adapt a benzannulation
reaction described for the synthesis of naphthalene
ring.14
Thus, we observed that intermediate 6 undergoes an
acid-catalyzed benzannulation to yield 7. Compound 6
was obtained in five steps from commerciallyavailable
chloro-quinoline 1 beginning with the addition of
MeMgBr (98%) and oxidation to the corresponding
methyl-ketone 3 with MnO2 (80%). Finkelstein reac-
tion15 on compound 3 yielded the iodo derivative 4
(80%)16 which undergoes Sonogashira reaction17 with
1-hexyne (65%) to give 5. Mander reaction18 with ethyl-
cyanoformate (27%) produced the desired b-ketoester
quinoline 6. The low yield in the Mander reaction is
Additionally(Scheme 2), the acid-catalyzed benzannu-
lation (CSA, HCCl3, reflux), of quinoline derivative 6,
led onlyto 15% of the desired acridine 7, which was
accompanied byseveral other products. No character-
ization was possible due to the difficulties of separation.
All attempts to obtain cleanlya predominant C-alkyla-
tive product from 6 failed.
We then chose a different route in which the reac-
tive methyl-ketone derivative 4 was protected (Scheme
4) in order to prevent the O-alkylation reaction to occur.
explained bythe reactivityof precursor
5. This sub-
stance tends to dimerize, in the neat state, leading to
compound 8 (detected bymass spectrometry, Scheme 3).
In solution, it produces a presumed exo-methylene
Based on work published on naphthalene derivatives,21
we transformed 4 (from the previous route) to the
O-TBS protected derivative 11 with TBSOTf and 2,6-
lutidine (80%).
O
OH
CH2
CH3
TBS
O
I
O
I
N
5
N
N
C4H9
C4H9
a
c
CH3
CH2
N
N
4
11
CH2
O
b
neat
C4H9
TBS
CH2
TBS
R
O
O
C4H9
N
R-OH
N
H2C
H3C
N
R
O
H3C O-R
O
13a: R= -C4H9, 13b: R= -P h
12a: R= -C4H9, 12b: R= -Ph
O
N
C4H9
N
C4H9
Scheme 4. Reagents and conditions: (a) 3 equiv TBSOTf, 3 equiv 2,6-
lutidine, CH2Cl2, 80%; (b) 1-hexyne or phenylacetylene, 0.07 equiv
PdCl2(PPh3)2, 0.3 equiv CuI, 1.5 equiv Et3N, toluene, rt, 80–90%; (c)
4 mol % [Rh(CO)2Cl]2, toluene, 120 °C, 60–70%.
Dimer: 8
9: R = -Me or 10: R = -Et
Scheme 3.

P. Belmont et al. / Tetrahedron Letters 45 (2004) 2783–2786
2785
O
I
O
H3CO
H3CO
a-c
d
CH3
H
Cl
N
N
17
14
18
TBS
CH2
TBS
CH2
O
O
O
I
H3CO
H3CO
e
N
19a-d
N
R
f
TBS
R
19a, 20a: R = -C4H9
19b, 20b: R = -P h
19c, 20c: R = -CH2-O-THP
19d, 20d: R = -CH(OEt)2
H3CO
N
20a-d
Scheme 5. Reagents and conditions: (a) 3 equiv MeMgBr, THF, 40 °C, 98%; (b) 10 equiv MnO2, toluene, 80 °C, quant.; (c) NaI, CH3CN, 0.5 equiv
HCl 4 N, reflux, 80–98%; (d) 2.2 equiv TBSOTf, 3 equiv Et3N, CH2Cl2, 77%; (e) 1-alkyne, 0.07 equiv PdCl2(PPh3)2, 0.3 equiv CuI, 1.5 equiv Et3N,
toluene, rt; (f) 10 mol % [Rh(CO)2Cl]2, toluene, 2–4 h, 120 °C.
After an efficient Sonogashira reaction with 1-hexyne
(12a, 90%), metal-catalyzed cylization to the corre-
sponding acridine 13a, was possible with 4 mol %22 of
the Rh(I) complex [Rh(CO)2Cl]2 in 70% yield.23 Sono-
gashira reaction with phenylacetylene on derivative 11
gave the intermediate 12b, which smoothlyunderwent
the rhodium-catalyzed annulation in 60% yield. This
methodologyprovides 1,3-disubstituted acridines.
Acknowledgements
P.B. thanks the ÔAssociation pour la Recherche sur le
CancerÕ (A.R.C) for financial support and C.S.M.A. the
ERASMUS exchange program of the Universityof
Durham (UK). Prof. M. A. Ciufolini is thanked for
helpful discussions and financial support, A. Vialettes
for preliminarywork, Dr. D. Bouchu and Ms. L.
Rousset-Arzel for mass spectrometryspectra.
The same synthetic pathway was conducted also
on 2-chloro-6-methoxy-quinoline-carboxaldehyde 14
(Scheme 5).
References and notes
1. Albert, A. The Acridines, 2nd ed.; Edward Arnold Ltd:
London, 1966.
2. Demeunynck, M.; Charmantray, F.; Martelli, A. Curr.
Pharm. Design 2001, 7, 1703–1724, and references cited
therein.
3. Finlay, G. J.; Atwell, G. J.; Baguley, B. C. Oncol. Res.
1999, 11, 249–254.
4. Lee, H. H.; Wilson, W. R.; Ferry, D. M.; van Zijl, P.;
Pullen, S. M.; Denny, W. A. J. Med. Chem. 1996, 39,
2508–2517.
5. Carlier, P. R.; Chow, E. S.-H.; Han, Y.; Liu, J.; El Yazal,
J.; Pang, Y.-P. J. Med. Chem. 1999, 42, 4225–4231.
6. Acheson, R. M. An Introduction to the Chemistry of
Heterocyclic Compounds, 1st ed.; Wiley-Interscience:
New York, 1956.
7. Katritzky, A. R.; Rees, C. W.; Scriven, E. F. V. Compre-
hensive Heterocyclic Chemistry II. vol. 5.; Pergamon,
Oxford, UK, 1996.
The iodo-methylketone derivative 17 was synthesized
using the previous route. Action of TBSOTf with Et3N
on 17 gave the O-TBS protected derivative 18 (77%),
which underwent efficient Sonogashira reaction with
several alkynes: 1-hexyne (19a, 70%), phenylacetylene
(19b, 60%), tetrahydro-2-(2-propynyloxy)-2H-pyran
(19c, 70%),24 3,3-diethoxy-1-propyne (19d, 72%). The
Rh(I)-catalyzed cyclization produced the corresponding
1,3,7-tri-substituted acridines 20a (60%), 20b (50%), 20c
(60%),25 20d (40%).26 We were able to use alkynes 19c
and 19d, bearing sensitive protecting groups, in this
transformation. In addition, experiments were per-
formed under thermal conditions without catalyst.
Heating derivative 19c for 5 h at 120 °C in toluene, led to
the formation of 10–15% of the desired acridine as
27
shown bythe analysis of the proton NMR spectrum.
8. Acheson, R. M. An Introduction to the Chemistry of
Heterocyclic Compounds, 3rd ed.; Wiley-Interscience:
New York, 1976.
9. Sourdon, V.; Mazoyer, S.; Pique, V.; Galy, J. Molecules
2001, 6, 673–682.
10. Atwell, G. J.; Rewcastle, G. W.; Baguley, B. C.; Denny,
W. A. J. Med. Chem. 1987, 30, 664–669.
11. Bernthsen, A.; Bender, F. Chem. Ber. 1883, 16, 1802–
1819.
12. Gamage, S. A.; Spicer, J. A.; Rewcastle, G. W.; Denny, W.
A. Tetrahedron Lett. 1997, 38, 699–702; Rosevear, J.;
Some work is still performed to improve the cyclization
efficiencyand to analyze more carefullythe free-metal
temperature-based cyclization.
In summary, we have developed a new and efficient
strategyto access 1,3-disubstituted and 1,3,7-tri-substi-
tuted acridines, in a six steps pathwayfrom commer-
ciallyavailable quinolines. Extension of this
methodologyto other heterocyclic structures is under-
wayand will be published in due course.

2786
P. Belmont et al. / Tetrahedron Letters 45 (2004) 2783–2786
Wilshire, J. F. K. Aust. J. Chem. 1981, 34, 839–853;
Albert, A. In Heterocyclic Compounds; Elderfield, R. C.,
1-alkyne tetrahydro-2-(2-propynyloxy)-2H-pyran and
70 lL (0.51 mmol, 1.5 equiv) of Et3N are put in 5 mL of
THF. This mixture is sonicated for 5 min with Ar
bubbling. Then, CuI (20 mg, 0.3 equiv) is added and the
mixture is sonicated for 5 min. Finally, PdCl2(PPh3)2
(17 mg, 0.07 equiv) is added and the mixture is sonicated
for 10 min. The reaction is left overnight at rt, then diluted
with AcOEt and washed three times with NH4Cl and
brine. After evaporation, the residue was submitted to
purification on silica preparative TLC. We obtained
106 mg (70%) of the desired quinoline 19c. Rf : 0.32
(cyclohexane/AcOEt, 8/2). NMR 1H (300 MHz, CDCl3)
d 8.13 (s, 1H), 7.95 (d, J ¼ 9:2 Hz, 1H), 7.32 (dd, J ¼ 2:6,
9.2 Hz, 1H), 7.00 (s, 1H), 5.18 (s, 1H), 4.94 (m, 1H), 4.77
(s, 1H), 4.55 (s, 2H), 3.91 (s, 3H) overlapped with 3.91–
3.83 (m, 1H), 3.56–3.52 (m, 1H), 1.82–1.53 (m, 6H), 0.95
(s, 9H), 0.17 (s, 6H). MS (EI): 453 (MþÅ). HREIMS:
453.2335 calcd for C26H35NO4Si (MþÅ), found 453.2340.
25. A typical experiment for the Rh(I)-catalyzed cyclization
reaction: To a toluene (1.5 mL) solution of the Sonogash-
ira product 19c (0.05 g, 0.110 mmol) was added 10 mol %
[Rh(CO)2Cl]2 (0.004 g, 0.011 mmol) under Ar atmosphere.
The resulting mixture was heated at 120 °C for 3 h. The
mixture was then diluted with AcOEt and washed three
times with NH4Cl and brine. After evaporation of the
volatile, purification on a silica preparative TLC (cyclo-
hexane/AcOEt, 8/2) we obtained 30 mg (60%) of the
acridine derivative 20c as a light yellow oil. Rf : 0.18
Ed.; J. Wileyand Sons: New York, NY, 1952; Vol. 4.
13. Smolders, R. R.; Waefelaer, A.; Coomans, R.; Francart,
D.; Hanuise, J.; Voglet, N. Bull. Soc. Chim. Belg. 1982, 91,
33–42.
14. Ciufolini, M. A.; Weiss, T. J. Tetrahedron Lett. 1994, 35,
1127–1130. Although it requires refluxing chloroform and
the use of camphorsulfonic acid, these conditions are
milder than the previous methods. Moreover, this benz-
annulation can be achieved on structures bearing sensitive
functional groups such as lactones and acetonides (per-
sonal communication of the author on unpublished work).
15. Meth-Cohn, O.; Narine, B.; Tarnowski, B. J. Chem. Soc.
Perkin Trans. 1 1981, 2509–2517.
16. Note that direct Sonogashira reaction on chlorinated
quinolines have alreadybeen reported in the literature:
Ciufolini, M. A.; Mitchell, J. W.; Roschangar, F. Tetra-
hedron Lett. 1996, 37, 8281–8284.
17. Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron
Lett. 1975, 16, 4467–4470.
18. Mander, L. N.; Sethi, S. P. Tetrahedron Lett. 1983, 24,
5425–5428.
19. HMBC and HSQC NMR data (not shown) confirmed the
structure of adduct 9 and 10.
20. Mondal, S.; Nogami, T.; Asao, N.; Yamamoto, Y. J. Org.
Chem. 2003, 68, 9496–9498.
21. Dankwardt, J. W. Tetrahedron Lett. 2001, 42, 5809–5812.
22. Raising the Rh(I) complex equivalent to 0.2 or 1 equiv led
apparentlyto complexation with the quinoline aromatic
nitrogen (since NMR shifting was observed), thus limiting
the yield to 30% and causing deprotection of the starting
material TBS-protected enol–ether group.
23. Several other metallic complexes have been used with
lower or no success: PtCl2 and PdCl2(CH3CN)2 were
unefficient, PdCl2 led to low starting material transforma-
tion.
1
(cyclohexane/AcOEt, 8/2). NMR H (300 MHz, CDCl3) d
8.87 (s, 1H), 8.09 (d, J ¼ 9:6 Hz, 1 H), 7.80 (s, 1H), 7.46
(dd, J ¼ 2:6, 9.4 Hz, 1H), 7.17 (d, J ¼ 2:8 Hz, 1H), 6.87 (s,
1H), 4.95 (d, J ¼ 13:0 Hz, 1H), 4.80 (t, J ¼ 3:4 Hz, 1H),
4.70 (d, J ¼ 13:0 Hz, 1H), 3.99 (s, 3H) overlapped with
3.99–3.93 (m, 1H), 3.60–3.55 (m, 1H), 1.79–1.54 (m, 6H),
1.14 (s, 9H), 0.35 (s, 6H). MS (EI): 453 (MþÅ). HREIMS:
453.2335 calcd for C26H35NO4Si (MþÅ), found 453.2335.
26. In this case, starting material was recovered.
27. Further heating led to decomposition and did not improve
the yield.
24. A typical experiment for the Sonogashira reaction: 0.150 g
(0.34 mmol) of 18, 70 lL (0.51 mmol, 1.5 equiv) of the