Isoquinoline-based analogs of the cancer drug clinical candidate tipifarnib as anti-Trypanosoma cruzi agents

Article abstract of DOI:10.1016/j.bmcl.2009.10.029

We developed a synthetic route to prepare isoquinoline analogs of the cancer drug clinical candidate tipifarnib. We show that these compounds kill Trypanosoma cruzi amastigotes grown in mammalian host cells at concentrations in the low nanomolar range. Th

Full text of DOI:10.1016/j.bmcl.2009.10.029

Bioorganic & Medicinal Chemistry Letters 19 (2009) 6582–6584
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Isoquinoline-based analogs of the cancer drug clinical candidate tipifarnib
as anti-Trypanosoma cruzi agents
Naveen Kumar Chennamaneni a, Jenifer Arif b, Frederick S. Buckner b, Michael H. Gelb a,c,
*
a Department of Chemistry, Campus Box 351700, University of Washington, Seattle, WA 98195, USA
b Department of Medicine, University of Washington, Seattle, WA 98195, USA
c Department of Biochemistry, University of Washington, Seattle, WA 98195, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We developed a synthetic route to prepare isoquinoline analogs of the cancer drug clinical candidate tipi-
farnib. We show that these compounds kill Trypanosoma cruzi amastigotes grown in mammalian host
cells at concentrations in the low nanomolar range. These isoquinolines represent new leads for the
development of drugs to treat Chagas disease.
Received 28 August 2009
Revised 6 October 2009
Accepted 7 October 2009
Available online 13 October 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Enzyme inhibitors
Anti-parasite therapeutics
Chagas disease
Lanosterol 14a-demethylase
Sterol biosynthesis
Chagas disease is caused by infection with the protozoan para-
site Trypanosoma cruzi (T. cruzi). Approximately 8–11 million peo-
ple in Latin America are infected with this parasite, and 30% of
those can be expected to develop complications ranging from mild
to terminal.1 There is no effective treatment for Chagas disease at
this time.
OMe group in place of this amino group. We are hopeful that these
tipifarnib analogs can be taken into clinical development for Cha-
gas disease because they are highly potent at killing the parasite,
and they are expected to have the desirable oral bioavailability
and pharmacokinetics that tipifarnib has.
Our modeling studies of tipifarnib docked into the active site of
We have previously shown that the protein farnesyltransferase
inhibitor tipifarnib (compound 1, Fig. 1) kills T. cruzi parasites by
T. cruzi lanosterol 14a-demethylase shows that the carbonyl and
N-Methyl portions of the amide of the quinoline ring are in contact
with enzyme surface residues only (Fig. 2). Since such interactions
are usually not strong because of the inherent flexibility of protein
surface residues, we considered making a new series of analogs in
which the quinolone ring is replaced with a quinoline ring. How-
ever, quinolines are well known to be susceptible to oxidative
metabolism in which carbon-2 is hydroxylated leading to the quin-
olone ring after tautomerization of the 2-hydroxyquinoline (Fig. 3).
Thus, we decided to make isoquinoline analogs of the general core
structure 3 (Fig. 3).
binding to lanosterol 14a
-demethylase.2 This enzyme is required
for production of ergosterol, a required component of the parasite’s
membranes, which cannot be replaced by mammalian host cell de-
rived cholesterol.3 Using the X-ray structure of tipifarnib bound to
mammalian protein farnesyltransferase and a homology model of
T. cruzi lanosterol 14a-demethylase with tipifarnib docked into
the active site2, we designed compound 2 (Fig. 1) which no longer
inhibits protein farnesyltransferase (IC50 >5000 nM vs ꢀ1 nM for
tipifarnib) and is about 10-fold more potent at killing T. cruzi
amastigotes (clinically relevant parasite life cycle stage that grows
in mammalian host cells) (EC50 = 0.6 nM) than is our lead com-
pound tipifarnib (EC50 = 4 nM).4 The extra methyl group present
in 2 clashes with the surface of protein farnesyltransferase. In addi-
tion, the amino group of tipifarnib is part of a hydrogen bond net-
work in the active site of protein farnesyltransferase, which does
not exist in the sterol demethylase.4 Compound 2 contains a
Cl
Cl
N
N
N
N
Me
Me
Me
MeO
H2N
Cl
O
Cl
N
O
N
Me
2
tipifarnib, 1
Me
* Corresponding author. Tel.: +1 206 543 7142; fax: +1 206 685 8665.
E-mail address: gelb@chem.washington.edu (M.H. Gelb).
Figure 1. Structure of tipifarnib 1 and analog 2.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.10.029

N. K. Chennamaneni et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6582–6584
6583
Figure 2. Stereo diagram of tipifarnib docked into the active site of the homology model of T. cruzi lanosterol 14
a-demethylase. Tipifarnib is in red, the heme is in yellow, and
the heme-iron is in cyan.
N
N
M
e
MeO
Oxidation
Tautomerization
N
N
HO
O
N
N
H
3
2-Hydroxy
Quinoline
Quinoline
Quinolone
Figure 3. Metabolic oxidation of a quinoline to a quinolone. Compound 3 is the core structure of isoquinoline-based tipifarnib analogs prepared in this study.
The synthesis of the two fragments to make the isoquinoline
tipifarnib analogs is shown in Figure 4. Reaction of substituted
Cl
benzaldehyde 4 with trimethylsilyl cyanide followed by reduction
with LiAlH4 gives hydroxyamine 5. Reductive amination with
p-bromobenzaldehyde gives secondary amine 6. Friedel–Crafts
alkylation leads to ring closure, and the compound is tritylated
on the nitrogen to give 7 in preparation for the next step. The syn-
thesis of the second fragment, methanone 11, is shown in Figure 4
(panel b). These are made by reacting the appropriate Weinreb
amide 9 with a silylated and lithiated N-methylimidazole. Figure
5 shows the joining of the two fragments to give the target
compounds. Compound 7 is converted to the aryl lithium by
lithium–halide exchange, and treatment with methanone 11 gives
tertiary alcohol 12. Alcohol 12 is detritylated and oxidized to iso-
quinoline 13 with MnO2. Conversion of the tertiary alcohol to the
chloride and then to the target compound 14 is carried out with
SOCl2 followed by treatment with methanol. It should also be pos-
sible to heat 13 in the presence of tosic acid and methanol to give
14 in one step5, but this was not attempted. Compound 15 was
made by the same route. To make analog 16, which lacks the phe-
nyl substituent at the 4-position of the quinoline ring, we con-
verted the commercially available 6-bromoisoquinoline to the
aryl lithium and carried through the same sequence shown in
Figure 5.
(a)
Cl
a, b
c
CHO
HO
NH2
4
5
Cl
OH
Br
Cl
d, e
Br
NH
N
Ph3C
6
7
(b)
O
O
OMe
f, g
OH
N
Me
Cl
Cl
9
8
Cl
N
N
h, i
We tested the isoquinolines for inhibition of growth of T. cruzi
amastigotes (Tulahuen strain) inside of mammalian cells (3T3
cells). The parasites stably express the b-galactosidase from Esche-
richia coli, and this enzyme converts yellow colored chlorophenol
N
N
O
11
10
Figure 4. Synthesis of two fragments for isoquinoline tipifarnib analogs. Panel a: (a)
TMSCN, ZnI2, CH2Cl2, 0 °C to rt; (b) LiAlH4, THF, 0 °C to rt, 70% (two steps);
(c) p-bromobenzaldehyde, Et3N, MeOH, then NaBH4, 75%; (d) AlCl3, CH2Cl2, 70 °C,
90%; (e) Ph3CCl, Et3N, DMAP, CH2Cl2, 93%. Panel b: (f) SOCl2, 90 °C; (g) CH3ONHCH3,
pyridine, CH2Cl2, rt, 90% (two steps); (h) n-BuLi, Et3SiCl, THF, À78 °C; (i) 9, n-BuLi,
À78 °C, 75%.
red b-D-galactopyranoside into a red colored product, which is
readily measured by spectrophotometry.5 This allows the number
of parasites in the culture to be readily quantified without resort-
ing to tedious parasite cell counting using microscopy. Data in
Table 1 are provided as values of EC50, the concentration of com-

6584
N. K. Chennamaneni et al. / Bioorg. Med. Chem. Lett. 19 (2009) 6582–6584
Cl
Cl
N
N
N
N
a
OH
Br
+
O
N
N
Ph3C
Cl
Ph3C
Cl
7
11
12
Cl
Cl
N
N
N
N
d, e
b, c
OMe
OH
N
N
Cl
Cl
13
14
Figure 5. Synthesis of isoquinoline tipifarnib analogs. (a) n-BuLi, THF, À78 °C, 56%; (b) TFA, CH2Cl2, 85%; (c) MnO2, dioxane, 90 °C, 64%; (d) SOCl2, rt; (e) MeOH, 90 °C, 80%
(two steps).
pound that reduces parasite growth by 50%. It can be seen that
compound 14 is as potent as the corresponding quinolone 2. Thus,
the carbonyl and N-methyl groups of 2 are not required for efficacy
Table 1
Growth arrest of T. cruzi amastigotes by isoquinoline-based tipifarnib analogsa
in this parasite killing assay. Compound 15 with a 2,6-difluoro-
phenyl replacing the 3-chlorophenyl substituent is also potent in
this assay. Compound 16 lacks this aryl substituent and is not ac-
tive against T. cruzi. This is consistent with our modeling showing
that the phenyl substituent packs well into the active site of T. cruzi
Compd
Structure
EC50 (nM)
Cl
N
N
Me
lanosterol 14a-demethylase. We also studied the lanosterol 14a-
H2N
Tipifarnib
4 (Ref. 2)
demethylase inhibitor posaconazole as a comparator compound.
The data in Table 1 show that the new isoquinolines reported in
this study are nearly as potent as posaconazole. The latter has been
shown to cure mice suffering from a chronic infection with
T. cruzi.3 Posaconaozle is an approved drug for treatment of fungal
infections. There is now discussion of the use of this agent to treat
Chagas disease. The new isoquinolines reported here offer the
advantage that manufacturing cost is expected to be well below
that of posaconazole. Given the outstanding potency of these com-
pounds against T. cruzi growth, detailed pharmacokinetic and Cha-
gas disease efficacy studies in rodents are warranted.
Cl
O
N
Me
Cl
N
N
Me
Me
MeO
2
0.6 (Ref. 4)
Cl
O
N
Me
Cl
N
N
Acknowledgment
Me
MeO
14
15
0.5, 0.9, 1.1b
This work was supported by a grant from the National Institutes
of Health (AI070218).
N
Cl
N
Supplementary data
N
F
MeO
Me
F
0.9, 1.3
Supplementary data (full details for the synthesis and charac-
terization of the isoquinolines) associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.2009.10.029.
N
Cl
N
References and notes
N
Me
MeO
16
120
0.3
1. Division of Parasitic Diseases, Chagas Disease. http://www.cdc.gov/chagas/
2. Hucke, O.; Gelb, M. H.; Verlinde, C. L.; Buckner, F. S. J. Med. Chem. 2005, 48, 5415.
3. Urbina, J. A.; Payares, G.; Moling, J.; Sanoja, C.; Liendo, A.; Piras, M. M.; Piras, R.;
Perez, N.; Wincker, P.; Ryley, J. F. Science 1996, 273, 969.
4. Kraus, J. M.; Verlinde, C. L. M. J.; Karimi, M.; Lepesheva, G. I.; Gelb, M. H.;
Buckner, F. S. J. Med. Chem. 2008, 52, 1639.
N
Cl
Posaconazole
EC50s were determined as described.5
The multiple numbers represent independent determination of EC50
a
5. Buckner, F. S.; Verlinde, C. L. M. J.; La Flamme, A. C.; Van Voorhis, W. C. Anti.
Agents Chemother. 1996, 40, 2592.
b
.