Synthesis of RHPS4 via an anionic ring closing cascade

Article abstract of DOI:10.1016/j.tetlet.2008.02.074

A new and convergent synthesis of the potent telomerase inhibitor RHPS4 is presented. The key step is the construction of the pentacyclic framework via an anionic ring closing cascade in which two new rings are formed.

Full text of DOI:10.1016/j.tetlet.2008.02.074

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 2351–2354
Synthesis of RHPS4 via an anionic ring closing cascade
Jesper Langgaard Kristensen *
Department of Medicinal Chemistry, Faculty of Pharmaceutical Sciences, University of Copenhagen, DK-2100 Copenhagen, Denmark
Received 2 January 2008; revised 28 January 2008; accepted 15 February 2008
Available online 19 February 2008
Abstract
A new and convergent synthesis of the potent telomerase inhibitor RHPS4 is presented. The key step is the construction of the
pentacyclic framework via an anionic ring closing cascade in which two new rings are formed.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Antitumour agents; Cyclizations; Fused-ring systems; Nucleophilic aromatic substitutions; Ring closure
1. Introduction
RHPS4 (see Scheme 1) is a potent inhibitor of telome-
rase (IC50 = 0.33 lM).2 It is believed that it mediates its
effect via intercalation and stabilization of G-quadruplex
DNA in the telomere.3 It shows acceptable pharmaceutical
properties and is being evaluated as a candidate for clinical
development.4
RHPS4 was first synthesized via a very intriguing dimer-
ization of a quinaldinium salt.2,5 Although this procedure
gives access to gram quantities of material it lacks flexibi-
lity, that is, it is limited to RHPS4 and very similar ana-
logues. Other synthetic efforts towards RHPS4 and
substituted analogues thereof have focused on constructing
the pentacyclic core of RHPS4 (1 in Scheme 1). Subse-
quently, 1 can be converted to RHPS4 via methylation with
Telomerase inhibitors are an emerging class of com-
pounds that show potential in the fight against cancer.1
Their mode of action is based on the following principle:
telomeres cap the terminal ends of chromosomes with a
repetitive sequence of DNA base pairs (TTAGGG in mam-
mals). During replication 50–200 base pairs are lost in each
cycle, thereby shortening the chromosome, which eventu-
ally leads to cell death. Telomerase restores telomere length
and is up-regulated in most human cancers, thereby allow-
ing the tumour cells to proliferate uncontrolled. Thus, the
inhibition of telomerase prevents the tumour cell from
circumventing this growth-control mechanism.
F
F
F
MeOSO3
N
F
Br
N
N
N
N
F
F
F
F
F
+
+
Ref. 7b
N
N
F
NH2
F
Br
Br
B(OH)2
RHPS4
1
2
3
4
5
Scheme 1. Retrosynthetic analysis of RHPS4 using an anionic cascade ring closure as the key step.
*
Tel.: +45 35 33 64 87; fax: +45 35 33 60 40.
E-mail address: jekr@farma.ku.dk
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.02.074

2352
J. L. Kristensen / Tetrahedron Letters 49 (2008) 2351–2354
F
F
F
5
KF
F
Br
Br
N
1) 3, tBuOK
2) MeI
N
N
F
1) NaNO2, H+
2) CuSO4, KCN
H2N
F
H2N
F
Pd(P(tBu)3)2
Br2
N
95%
51%
97%
75%
F
F
Br
4
6
2
Scheme 2. Synthesis of key intermediate 2 in four steps from 2-fluoro-4-methylaniline.
dimethyl sulfate.7b Previous efforts towards 1 have relied
primarily on two different approaches: thermal extrusion
of nitrogen from triazolyl-substituted acridines at elevated
temperatures6 and palladium-mediated derivatization of
1-substituted acridin-9(10H)-ones followed by a conden-
sation reaction to form the pentacyclic ring system.7
Herein a radically different and flexible synthesis of the
pentacyclic core (1) of RHPS4 is presented. It is based on
an anionic ring closing cascade reaction, previously
reported for the synthesis of the parent quinoline-acridine
ring system of RHPS4.8
F
F
F
N
N
N
F
F
2.2 eq t-BuLi
N
F
Br
THF, - 78 oC
66%
2
1
F
Li
F
N
F
N
Making the two disconnections in 1, as indicated in
Scheme 1, leads to 2 as the key intermediate for the anionic
ring closure. Two further disconnections generate the three
building blocks needed for the assembly of 2: 2-bromo-4-
fluoroaniline (3), 2-bromo-6-fluoro-4-methylbenzonitrile
(4) and 2,5-difluorophenylboronic acid (5). It was envis-
aged that 4 and 5 initially could be joined in a Suzuki–
Miyaura coupling.9 Selective nucleophilic aromatic substi-
tution of 3, using the conditions developed by Gorvin,10
followed by methylation should then lead to 2. Two of
the required building blocks (3 and 5) are commercially
available, whereas 4 can be prepared in two steps. The
entire sequence leading to 2 is outlined in Scheme 2.
Commercially available 2-fluoro-4-methylaniline was
brominated, giving 2-bromo-6-fluoro-4-methylaniline11 in
95% yield. Subsequent diazotization/cyanation gave the
corresponding benzonitrile 4 in a modest 51% yield.
Suzuki–Miyaura coupling of 4 with 5 proceeded very effi-
ciently to give biaryl 6 in 97% yield, using conditions
described earlier for a similar coupling.12 Selective nucleo-
philic aromatic substitution of the most reactive fluorine
situated ortho to the cyano group in 6 by the potassium salt
of 3 followed by in situ quenching of the resulting diaryl-
amine anion with methyl iodide gave 2 in 75% yield.
The stage was now set for the ring closure, see Scheme 3.
Bromine–lithium exchange should generate an aryl lithium
species which would attack the cyano group in a Parham-
type cyclization.13 Subsequent intramolecular nucleophilic
aromatic substitution generates the second ring in the
sequence.
F
F
N
N
Li
Scheme 3. Construction of the pentacyclic core of RHPS4 via an anionic
cascade ring closure.
verted to RHPS4 as described by Stevens,7b thus complet-
ing the synthesis in six steps from 2-fluoro-4-methylaniline.
It is worth noting that no protecting groups are used at
any point in the synthesis and no oxidation/reduction steps
are required. With the right building blocks in hand the
pentacyclic core is assembled in just three steps. This mod-
ular feature makes this synthetic approach to analogues of
RHPS4 very attractive. Substituents on different rings can
be brought into the sequence at a very late stage, and a
large number of analogues of the starting building blocks
3, 4 and 5 with the required functionality to eventually
form the pentacyclic framework are either commercially
available or can be accessed in very few steps. Combining
3, 4, 5 and substituted derivatives thereof in the desired
way gives full control over the substitution pattern of the
resulting pentacyclic core.
In conclusion, a new, flexible and highly convergent
synthesis of RHPS4 has been developed.
2. Experimental
2.1. 2-Bromo-6-fluoro-4-methylaniline
Gratifyingly, the treatment of 2 with 2.2 equiv of tert-
butyllithium at ꢀ78 °C produced the pentacyclic core of
RHSP4 (1) in 66% yield after the recrystallization of the
crude product.14 Pentacycle 1 could subsequently be con-
Bromine (2.26 mL, 44 mmol) in 16 mL of AcOH was
added dropwise to a solution of 2-fluoro-4-methylaniline
(5.00 g, 40.0 mmol) in 6 mL of MeOH and 6 mL of AcOH

J. L. Kristensen / Tetrahedron Letters 49 (2008) 2351–2354
2353
at 0 °C. The reaction mixture was stirred for 1 h, before
1 N NaOH (100 mL) and Na2S2O3 (10 g) were added.
The mixture was extracted with CH2Cl2 (1 ꢁ 100 mL +
2 ꢁ 50 mL) and the combined organic phases were dried
over MgSO4 and concentrated it in vacuo yielding 7.74 g
(95%) of 2-bromo-6-fluoro-4-methylaniline11 as a dark
was added. The flask was sealed and lowered into an oil
bath preheated to 50 °C, and the reaction mixture was stir-
red for 1 h. Afterwards, the mixture was partitioned
between brine (25 mL) and CH2Cl2 (25 mL), and the aque-
ous phase was extracted with CH2Cl2 (2 ꢁ 25 mL). The
combined organic phases were dried over MgSO4 and the
solvent was removed in vacuo. Flash chromatography
(heptane/EtOAc 9:1) yielded 1.075 g (97%) of 6 as an off-
1
solid. H NMR (CDCl3, 300 MHz): 2.21 (s, 3H), 3.94 (br
s, 2H, NH2), 6.72–6.78 (m, 1H), 6.98–7.01 (m, 1H). 13C
1
NMR (CDCl3, 75 MHz): 20.2 (d, JC–F = 2 Hz), 109.3 (d,
white powder, mp 109 °C. H NMR (CDCl3, 300 MHz):
JC–F = 5 Hz), 114.8 (d, JC–F = 19 Hz), 127.6 (d, JC–F
=
2.45 (s, 3H), 7.01–7.21 (m, 5H). 13C NMR (CDCl3,
75 MHz): 21.9 (d, JC–F = 2 Hz), 99.0 (d, JC–F = 16 Hz),
112.8, 116.3 (d, JC–F = 20 Hz), 117.2 (dd, JC–F = 25,
8 Hz), 117.3 (dd, JC–F = 25, 4 Hz), 117.4 (dd, JC–F = 23,
8 Hz), 125.8 (ddd, JC–F = 18, 6, 8 Hz), 127.1 (dd,
JC–F = 3, 1.5 Hz), 139.4, 146.1 (d, JC–F = 7 Hz), 155.1
(dd, JC–F = 243, 3 Hz), 158.1 (dd, JC–F = 242, 3 Hz),
163.4 (d, JC–F = 257 Hz). 19F NMR (CDCl3, 282 MHz):
ꢀ122.6, ꢀ119.9, ꢀ108.1. Anal. Calcd for C14H8F3N: C,
68.02; H, 3.26; N, 5.67. Found: C, 67.89; H, 3.64; N, 5.40.
2 Hz), 128.2 (d, JC–F = 8 Hz), 130.7 (d, JC–F = 15 Hz),
150.8 (d, JC–F = 240 Hz). The crude material (>99% pure
1
as judged from TLC, GC–MS and H NMR) was used in
the following step without further purification.
2.2. 2-Bromo-6-fluoro-4-methylbenzonitrile (4)
The following procedure is a slight modification of that
reported earlier by Paek et al.15 2-Bromo-6-fluoro-4-methyl-
aniline (4.08 g, 20.0 mmol) was suspended in a mixture of
H2O (8.5 mL) and glacial AcOH (14 mL), and concd
H2SO4 (3 mL) was added slowly. The mixture was cooled
to 0 °C before NaNO2 (1.63 g, 23.6 mmol) dissolved in
H2O (3 mL) was added dropwise with vigorous stirring.
In another 500 mL round-bottomed flask, CuSO4 hydrate
(4.06 g, 25 mmol) was dissolved in H2O (16 mL), and
KCN (6.9 g, 0.1 mol) dissolved in H2O (16 mL) was added
slowly. NaHCO3 (14 g) and toluene (20 mL) were added
before the flask was lowered into an oil bath preheated
to 55 °C. To this flask, the above prepared solution of
the diazonium salt was added dropwise with vigorous stir-
ring over 10 min. Subsequently, the mixture was stirred for
15 min before being cooled to rt. CH2Cl2 (50 mL) was
added and the mixture was filtered through Celite into a
separating funnel. The phases were separated and the aque-
ous phase was extracted with CH2Cl2 (3 ꢁ 50 mL). The
combined organic phases were washed with brine
(100 mL) and then dried over MgSO4 and concentrated
in vacuo. Flash chromatography (heptane/EtOAc 9:1)
yielded 2.17 g (51%) of 4 as a yellow powder, which could
be recrystallized from EtOAc/heptane to give 4 as a white
2.4. 3-((2-Bromo-4-fluorophenyl)(methyl)amino)-20,50-
difluoro-5-methylbiphenyl-2-carbonitrile (2)
Potassium tert-butoxide (320 mg, 2.85 mmol) was dis-
solved in dry DMSO (4 mL) at rt. 2-Bromo-4-fluoroaniline
3 (193 lL, 1.69 mmol) was added neat via syringe giving a
dark yellow-brown solution. After 10 min at rt the flask
was cooled to ꢂ10 °C before 6 (348 mg, 1.41 mmol) dis-
solved in DMSO (1 mL) was added dropwise, and the mix-
ture was subsequently stirred for 1 h at rt prior to the
addition of MeI (600 lL, 9.63 mmol). After 30 min, the
mixture was quenched with satd aqueous NH4Cl (50 mL)
and extracted with Et2O (4 ꢁ 50 mL). The combined
organic phase was washed with brine (20 mL), dried
(Na2SO4) and concentrated. Flash chromatography (hep-
tane/EtOAc 9:1) yielded 458 mg (75%) of 2 as a white
powder, mp 121 °C. 1H NMR (CDCl3, 300 MHz): 2.40
(s, 3H), 3.36 (s, 3H), 6.76 (d, 2H, J = 7.8 Hz), 6.99–7.21
(m, 4H), 7.26 (s, 1H) 7.40 (dd, 1H, J = 8.1, 2.9 Hz). 13C
NMR (CDCl3, 75 MHz): 22.1, 41.8 (d, JC–F = 2 Hz),
100.0 (d, JC–F = 1 Hz), 115.6 (d, JC–F = 23 Hz), 116.0,
116.5 (dd, JC–F = 23, 8 Hz), 116.8 (dd, JC–F = 25, 9 Hz),
1
powder, mp 76 °C. H NMR (CDCl3, 300 MHz): 2.44 (s,
3H), 7.01 (d, 1H, J = 9 Hz), 7.33 (s, 1H). 13C NMR
(CDCl3, 75 MHz): 21.8 (d, JC–F = 2 Hz), 102.3 (d,
JC–F = 18 Hz), 112.5, 115.6 (d, JC–F = 20 Hz), 125.2 (d,
117.4 (dd, JC–F = 25, 3 Hz), 118.1, 121.1 (d, JC–F =
25 Hz), 123.1 (d, JC–F = 2 Hz), 123.4 (d, JC–F = 10 Hz),
127.7 (dd, JC–F = 18, 8 Hz), 128.7 (d, JC–F = 9 Hz), 140.6,
JC–F = 2 Hz), 139.4 (d, JC–F = 3 Hz), 147.1 (d, JC–F
=
143.3 (d, JC–F = 4 Hz), 143.8, 152.5, 155.2 (dd, JC–F =
9 Hz), 163.4 (d, JC–F = 260 Hz). 19F NMR (CDCl3,
282 MHz): ꢀ103.6. Anal. Calcd for C8H5BrFN: C, 44.89;
H, 2.35; N, 6.54. Found: C, 44.51; H, 2.39; N, 6.41.
242, 2 Hz), 158.0 (dd, JC–F = 242, 3 Hz), 160.0 (d,
JC–F = 249 Hz). 19F NMR (CDCl3, 282 MHz): ꢀ122.8,
ꢀ120.8, ꢀ115.2. Anal. Calcd for C21H14BrF3N2: C,
58.49; H, 3.27; N, 6.50. Found: C, 58.60; H, 3.40; N, 6.32.
2.3. 20,3,50-Trifluoro-5-methylbiphenyl-2-carbonitrile (6)
2.5. 3,11-Difluoro-6,8-dimethyl-8H-quinolino[4,3,2-kl]
acridine (1)
Benzonitrile 4 (960 mg, 4.49 mmol), 2,5-difluorophenyl-
boronic acid 5 (850 mg, 5.38 mmol) and KF (860 mg,
14.8 mmol) were dissolved in dry THF (15 mL). Nitrogen
was bubbled through the solution for 10 min before bis-
(tri-tert-butylphosphine)palladium(0) (46 mg, 0.09 mmol)
tert-Butyllithium,
1.7 M
in
pentane
(720 lL,
1.22 mmol), was added to THF (3 mL) at ꢀ78 °C. After
5 min, 2 (240 mg, 0.56 mmol) dissolved in THF (1 mL)

2354
J. L. Kristensen / Tetrahedron Letters 49 (2008) 2351–2354
Stevens, M. F. G.; Kelland, L. R. Mol. Pharm. 2001, 60, 981; (c)
Gavathiotis, E.; Heald, R. A.; Stevens, M. F. G.; Searle, M. S. J. Mol.
Biol. 2003, 334, 25.
was added dropwise, causing an instant colouring of the
solution (brown). After the addition, stirring was contin-
ued at ꢀ78 °C for 5 min before glacial AcOH (1 mL) was
added, causing an instant change of colour to orange.
The mixture was warmed to rt and satd aqueous NaHCO3
(25 mL) was added. The mixture was extracted with
CH2Cl2 (3 ꢁ 25 mL). The combined organic phases were
dried (Na2SO4) and the evaporation of the solvent gave a
crude product as a dark orange powder (220 mg). Recrys-
tallization from EtOAc/heptane yielded 123 mg (66%) of
1 as a yellow-orange powder, mp 279 °C (reported:7b
256–258 °C). 1H NMR (DMSO-d6, 500 MHz): 2.60 (s,
3H), 3.67 (s, 3H), 7.24 (s, 1H), 7.48 (ddd, 1H, J = 9.4,
7.6, 3.3 Hz), 7.53 (td, 1H, J = 8.5, 2.8 Hz), 7.60 (dd, 1H,
9.2, 4.4 Hz), 7.92 (dd, 1H, J = 9, 5.8 Hz), 7.96 (s, 1H),
8.31 (dd, 1H, J = 10.6, 2.6 Hz), 8.36 (dd, 1H, J = 9.4,
3.3 Hz). 19F NMR (CDCl3, 282 MHz): ꢀ124.3, ꢀ116.1.
4. Cookson, J. C.; Heald, R. A.; Stevens, M. F. G. J. Med. Chem. 2005,
48, 7198.
5. Oszczapowicz, J.; Jaroszewska-Manaj, J.; Ciszak, E.; Gdaniec, M.
Tetrahedron 1988, 44, 6645.
6. Hagan, D. J.; Gime´nez-Arnau, E.; Schwalbe, C. H.; Stevens, M. F. J.
Chem. Soc., Perkin Trans. 1 1997, 2739.
7. (a) Heald, R. A.; Modi, C.; Cookson, J. C.; Hutchinson, I.; Laughton,
C. A.; Gowan, S. M.; Kelland, L. R.; Stevens, M. F. J. Med. Chem.
2002, 45, 590; (b) Hutchinson, I.; Stevens, M. F. G. Org. Biomol.
Chem. 2007, 5, 114.
8. Kristensen, J. L.; Vedso, P.; Begtrup, M. J. Org. Chem. 2003, 68,
4091.
9. (a) Miyaura, N.; Yanagi, T.; Suzuki, A. Synth. Commun. 1981, 11,
513; (b) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
10. Gorvin, J. H. J. Chem. Soc., Perkin Trans. 1 1988, 1331.
11. Blears, D. J.; Danyluk, S. S.; Schaeffe, T. J. Chem. Phys. 1967, 47,
5037.
12. Goodacre, S. C.; Street, L. J.; Hallett, D. J.; Crawforth, J. M.; Kelly,
S.; Owens, A. P.; Blackaby, W. P.; Lewis, R. T.; Stanley, J.; Smith, A.
J.; Ferris, P.; Sohal, B.; Cook, S. M.; Pike, A.; Brown, N.; Wafford,
K. A.; Marshall, G.; Castro, J. L.; Atack, J. R. J. Med. Chem. 2006,
49, 35.
References and notes
1. (a) Cech, T. R. Angew. Chem., Int. Ed. 2000, 39, 34; (b) Kelland, L. R.
Eur. J. Cancer 2005, 41, 971.
2. Heald, R. A.; Modi, C.; Cookson, J. C.; Hutchinson, I.; Laughton, C.
A.; Gowan, S. M.; Kelland, L. R.; Stevens, M. F. J. Med. Chem. 2002,
45, 590.
13. Parham, W. E.; Bradsher, C. K. Acc. Chem. Res. 1982, 15,
300.
14. Chromatographic purification of 1 leads to a substantial loss of
material due to streaking on the column.
15. Paek, K.; Knobler, C. B.; Maverick, E. F.; Cram, D. J. J. Am. Chem.
Soc. 1989, 111, 8662.
3. (a) Gavathiotis, E.; Heald, R. A.; Stevens, M. F. G.; Searle, M. S.
Angew. Chem., Int. Ed. 2001, 40, 4749; (b) Gowan, S. M.; Heald, R.;