A concise silylamine approach to 2-amino-3-hydroxy-indoles with potent in vivo antimalaria activity

Article abstract of DOI:10.1021/ol101566h

The development of a concise strategy to access 2-amino-3-hydroxy-indoles, which are disclosed as novel antimalarials with potent in vivo activity, is reported. Starting from isatins the target compounds are synthesized in 2 steps and in good yields via oxoindole intermediates by employing tert- butyldimethylsilyl amine (TBDMSNH₂) as previously unexplored ammonia equivalent.

Full text of DOI:10.1021/ol101566h

ORGANIC
LETTERS
2010
Vol. 12, No. 18
3998-4001
A Concise Silylamine Approach to
2-Amino-3-hydroxy-indoles with Potent
in vivo Antimalaria Activity
Sameer Urgaonkar,† Joseph F. Cortese,† Robert H. Barker,‡ Mandy Cromwell,‡
Adelfa E. Serrano,§ Dyann F. Wirth,†,| Jon Clardy,†,⊥ and Ralph Mazitschek*,†,#
Infectious Disease InitiatiVe Program, The Broad Institute of HarVard and MIT,
7 Cambridge Center, Cambridge, Massachusetts 02142, Drug DiscoVery and
DeVelopment, Genzyme Corporation, Waltham, Massachusetts 02451,
UniVersity of Puerto Rico School of Medicine, San Juan, P.R. 00936-5067,
Department of Immunology and Infectious Disease, HarVard School of Public Health,
665 Huntington AVenue, Boston, Massachusetts 02115, Department of Biological
Chemistry and Molecular Pharmacology, HarVard Medical School,
240 Longwood AVenue, Boston, Massachusetts 02115, and Center for Systems Biology,
Massachusetts General Hospital, Boston, Massachusetts 02114
ralph@broad.harVard.edu
Received July 7, 2010
ABSTRACT
The development of a concise strategy to access 2-amino-3-hydroxy-indoles, which are disclosed as novel antimalarials with potent in vivo
activity, is reported. Starting from isatins the target compounds are synthesized in 2 steps and in good yields via oxoindole intermediates by
employing tert-butyldimethylsilyl amine (TBDMSNH2) as previously unexplored ammonia equivalent.
Malaria is the most deadly parasitic infectious disease with an
estimated 300-500 million cases and a death toll of 0.8-1.2
million in 2008 alone.1 Of the four Plasmodium species that
are relevant for humans, P. falciparum accounts for most of
the fatalities.1 These already grim statistics are likely to become
even grimmer as P. falciparum strains that are resistant to
commonly used antimalaria chemotherapeutic agents emerge
and spread throughout Africa and parts of Asia.
The most recent antimalarial drug class was introduced
in 1996.2 In an effort to address the urgent need for new
drugs, especially ones with new cellular targets or chemo-
types that could delay the emergence of resistance, we have
screened small-molecule libraries for novel, drug-like com-
pounds with whole-cell antimalarial activity and limited
susceptibility to established mechanisms of drug resistance.3
† The Broad Institute of Harvard and MIT.
‡ Genzyme Corporation.
§ University of Puerto Rico School of Medicine.
| Harvard School of Public Health.
⊥ Harvard Medical School.
# Massachusetts General Hospital.
(2) Ekland, E. H.; Fidock, D. A. Int. J. Parasitol. 2008, 38, 743.
(3) Baniecki, M. L.; Wirth, D. F.; Clardy, J. Antimicrob. Agents
Chemother. 2007, 51, 716.
(1) Aregawi, M. Cibulskis, R. Otten, M. Williams, R. Dye, C. World
Malaria Report 2008; World Health Organization: Geneva, 2008.
10.1021/ol101566h  2010 American Chemical Society
Published on Web 08/18/2010

Very recently, two similar efforts have also been reported.4,5
Out of ∼79,000 compounds,6 104 inhibitors with nanomolar
activity against drug-sensitive (3D7) and multidrug-resistant
(Dd2, HB3) P. falciparum strains were identified.3 One of
the hits, the 2-amino-5-chloro-3-hydroxy-3-phenylindole 1a
(see Figure 1) was particularly attractive because of its
cylated 2-aminobenzophenones to be the most appropriate
(Scheme 1).9
Scheme 1. Traditional Synthetic Strategy to Access
2-Amino-3-hydroxy-indoles via Cyanohydrins
Unfortunately, this strategy, although readily scalable and
attractive for process scale syntheses, is limited for early stage
exploratory medicinal chemistry due to the lack of commercially
available 2-aminobenzophenones and the inherent lack of
enantioselectivity. As a result, we developed a general, short
and efficient method that would (a) provide analogues in
quantities that satisfy early stage drug discovery requirements,
(b) tolerate a wide variety of functional groups, (c) begin with
commercially available diversely functionalized building blocks,
and (d) avoid the use of cyanide while (e) offering the potential
to enantioselectively access 2-amino-3-hydroxy-indoles.
We envisioned a three-step reaction sequence starting from
isatins, which are widely available and inexpensive starting
materials as shown in Scheme 2.
Figure 1. In ViVo antimalaria activity of 2-aminoindole 1a.
Following inoculation of Swiss Albino mice with P. berghei
parasites, 1a was administered i.p. once daily for 4 days, and
parasitemia was determined on day 5.
unusual and compact 2-amino-3-hydroxy-indole core struc-
ture. In addition, compound 1a achieved excellent exposure
in mouse pharmacokinetic studies, and more importantly,
when tested in a 4-day suppressive P. berghei mouse model
1a demonstrated very good in ViVo efficacy,7 causing a dose-
dependent decrease in parasitemia with undetectable levels
of parasites at the highest dose tested and no signs of adverse
side effects (Figure 1).
Scheme 2. Retrosynthetic Analysis Starting from Isatin (2)
Encouraged by this promising in ViVo activity we began
to explore and optimize this compound class. A thorough
literature survey revealed that 2-amino-3-hydroxy-indoles
are virtually unexplored for biological activity, which we
suspect results at least in part from the limited synthetic
methodology to access this class. To date, four distinct
approaches have been reported.8-11 We found the methods
reported by Bell et al. to prepare the 2-amino-3-hydroxy-
indoles Via reaction of KCN with formylated or dichloroa-
Grignard addition or Rh-catalyzed addition12 of boronic acids
to isatin 2 would give 3-hydroxy-3-aryl-oxindole 3. Oxindole
3 could then react with a protected ammonia equivalent (e.g.,
allylamine) to yield the corresponding protected 2-amino-3-
hydroxy-indole 4, which on deprotection would result in the
desired 2-amino-3-hydroxy-indole 1. Furthermore, boronic
acids13 and electron-rich arenes14 can be added to isatins with
high enantioselectivity. This method would also allow access
to 2-amino-3-hydroxy-indoles enantioselectively.
On the basis of the well-precedented nature of the first
step, we directed our efforts toward the conversion of
3-hydroxyoxindoles 3 to allyl-protected 2-amino-3-hydroxy-
indoles 4. We were somewhat encouraged by variable
(4) Gamo, F.-J.; Sanz, L. M.; Vidal, J.; Cozar, C. d.; Alvarez, E.;
Lavandera, J.-L.; Vanderwall, D. E.; Green, D. V. S.; Kumar, V.; Hasan,
S.; Brown, J. R.; Peishoff, C. E.; Cardon, L. R.; Garcia-Bustos, J. F. Nature
2010, 465, 305
.
(5) Guiguemde, W. A.; Shelat, A. A.; Bouck, D.; Duffy, S.; Crowther,
G. J.; Davis, P. H.; Smithson, D. C.; Connelly, M.; Clark, J.; Zhu, F.;
Jime´nez-D´ıaz, M. B.; Martinez, M. S.; Wilson, E. B.; Tripathi, A. K.; Gut,
J.; Sharlow, E. R.; Bathurst, I.; El Mazouni, F.; Fowble, J. W.; Forquer, I.;
McGinley, P. L.; Castro, S.; Angulo-Barturen, I.; Ferrer, S.; Rosenthal, P. J.;
Derisi, J. L.; Sullivan, D. J.; Lazo, J. S.; Roos, D. S.; Riscoe, M. K.; Phillips,
M. A.; Rathod, P. K.; Van Voorhis, W. C.; M Avery, V.; Guy, R. K. Nature
2010, 465, 311.
(6) Analysis of screening libraries and assay results was supported by
Instant JChem, JKlustor, and Screen; ChemAxon: Budapest, Hungary;
www.chemaxon.com/products.html.
(7) Peters, W. Robinson, B. L. In Parasitic Infection Modelsin; Zak,
O., Sande, M. A., Eds.; Academic Press: London, 1999; pp 757.
(8) Wittig, G.; Kleiner, H.; Conrad, J. Liebigs Ann. Chem. 1929, 469,
(12) Toullec, P.; Jagt, R.; de Vries, J.; Feringa, B.; Minnaard, A. Org.
Lett. 2006, 8, 2715.
1
.
(9) Bell, S. C.; Wei, P. H. L. J. Heterocycl. Chem. 1969, 6, 599
(10) Ishizumi, K.; Inaba, S.; Yamamoto, H. J. Org. Chem. 1974, 39,
2581
.
(13) Shintani, R.; Inoue, M.; Hayashi, T. Angew. Chem., Int. Ed. 2006,
45, 3353.
.
(14) Hanhan, N. V.; Sahin, A. H.; Chang, T. W.; Fettinger, J. C.; Franz,
A. K. Angew. Chem., Int. Ed. 2010, 49, 744.
(11) Benincori, T.; Sannicolo, F. J. Org. Chem. 1988, 53, 1309
.
Org. Lett., Vol. 12, No. 18, 2010
3999

conversions (∼30-50%) obtained upon heating 5-chloro-
3-hydroxyoxindole 3a with an excess of allylamine in the
presence of 5 mol % of PTSA and 4 Å molecular sieves.15
Attempts to increase the yields by employing other drying
agents such as Na2SO4, MgSO4 or the use of Dean-Stark
apparatus were not successful. We anticipated that better
conversions could be achieved with the use of a Lewis acid,
which not only would catalyze the transformation but would
also efficiently entrap the water formed. Exploring a set of
Lewis acids confirmed our hypothesis, with both SnCl4 and
Ti(OiPr)4 efficiently and more importantly, reproducibly
promoting the conversion of 3a to 4a (Scheme 3). Further-
deprotection under condition that would prevent hydrolysis
of the amine. Pleasingly, we were able to directly isolate
the desired 2-aminoindole product 1a in 60% isolated yield.
The structure of 1a was unambiguously established by X-ray
analysis revealing that the newly installed substituent appears
to favor the tautomer with an exoamine rather than an
exoimine geometry (see Supporting Information).
Although TBDMSNH2 has been utilized in the context of
its ligand attributes for metal complexes,19-21 this is, to our
knowledge, the first report on the application of TBDMSNH2
in organic synthesis.
The synthetic utility of this novel SnCl4-promoted ami-
dation reaction with TBDMSNH2 as an ammonia surrogate
was then explored using a series of substituted 3-hydroxy-
oxindoles. A variety of functional groups in the indole ring,
including halogens, nitro, and alkyl ether are well tolerated
in the amidation process (Table 1). The amidation reaction
worked equally well under Ti(OiPr)4 promoted conditions.
The substrate scope with respect to the substituent at the
3-position of oxindoles was also found to be broad. Thus, the
aryl group at the 3-position possessing functional groups such
as ester (entry h) and alkyne (entry d) could be employed in
this process. In the case of an aryl group substituted with an
enolizable ketone, the use of Ti(OiPr)4 in lieu of SnCl4 was
critical for the success of the reaction (entry i, Table 1), as is
the indole with free nitrogen as R2 (entry p).
Scheme 3
.
Amidations with Allylamine and Triphenylsilylamine
Conditionsa
a Conditions: (a) excess allylamine for 4a, (b) 2 equiv of Ph3SiNH2 and
4 equiv NMM for 5.
Interestingly, an aryl group substituted with the sulfon-
amide group yielded the aminoindole with TBDMS group
on sulfonamide nitrogen with this method. However, depro-
tection with HF/pyridine gave the aminoindole 1q (entry q).
It is worth noting that several of these 2-amino-3-hydroxy-
indoles would be difficult to access or would require
functional group manipulations via a cyanohydrin route
significantly lengthening the synthesis. In addition, our
methodology allowed for direct access of N(1)-alkyl imi-
noindole using the corresponding oxindole as starting ma-
terial (entry r).
To ascertain whether an enantioenriched 3-hydroxyoxin-
dole would transform into 2-amino-3-hydroxy-indole under
the reaction conditions without any loss in enantiomeric
excess, we prepared chiral 3-hydroxyoxindole (S)-3a22
(>95% ee) and subjected it to SnCl4 promoted amidation
reaction with TBDMSNH2. Gratifyingly, the corresponding
2-amino-3-hydroxy-indole (S)-1a was obtained in good yield
without any measurable loss of enantiomeric excess (>95%
ee) indicating no racemization occurred during the reaction
(Scheme 4).
more, while microwave conditions reduced the reaction times
from 12-18 h to 40-60 min, the addition of NMM (N-
methyl morpholine) as an acid scavenger allowed the amine
coupling partner to be used in nominal amount.
Disappointingly, the deprotection of the allyl group in 4a
proved challenging despite the plethora of deallylation
protocols for amines and amides.16,17
Next, we turned our attention to identifying an ammonia
equivalent that would allow for more facile deprotection. On
the basis of literature precedents of employing a silylated
ammonia equivalent in unrelated transformations,18 we decided
to explore commercially available Ph3SiNH2. Unexpectedly, as
shown in Scheme 3, the reaction of 3a with Ph3SiNH2 under
the optimized conditions produced O-silylated 2-amino-3-
hydroxy-indole 5, in 60% yield while only trace amounts of
the desired desilylated product 1a were observed. Aminoindole
5 is probably formed by the intramolecular migration of the
silyl group. Attempts to desilylate 5 using TBAF or other
fluoride sources resulted in hydrolysis of 5 and in recovery of
the original hydroxyoxindole starting material 3a.
The initial exploratory SAR against drug-sensitive (3D7)
and drug-resistant (Dd2) parasite strains indicates that
2-amino-3-hydroxy-indoles are in general more active when
R2 is either an ortho-substituted electron-rich aromatic ring
(entries c and m, Table 1) or a naphthyl moiety (entry n,
Encouraged by the finding that a silylamine appears well-
suited to install the desired amine functionality, we suspected
that less sterically demanding analogues of Ph3SiNH2 such
as tert-butyldimethylsilyl amine (TBDMSNH2) might allow
(15) Meyer, R.; Zwiesler, M. J. Org. Chem. 1968, 33, 4274.
(19) West, R.; Boudjouk, P. J. Am. Chem. Soc. 1973, 95, 3983
(20) Bowser, J.; Neilson, R.; Wells, R. Inorg. Chem. 1978, 17, 1882
(21) Quallich, G.; Makowski, T.; Sanders, A.; Urban, F.; Vazquez, E.
J. Org. Chem. 1998, 63, 4116
(22) Hewawasam, P.; Erway, M.; Moon, S.; Knipe, J.; Weiner, H.;
Boissard, C.; Post-Munson, D.; Gao, Q.; Huang, S.; Gribkoff, V.; Meanwell,
N. J. Med. Chem. 2002, 45, 1487.
.
(16) Escoubet, S.; Gastaldi, S.; Bertrand, M. Eur. J. Org. Chem. 2005,
.
2005, 3855
.
(17) Unsatisfactory yields were obtained when a RhCl3-catalyzed allyl
.
deprotection strategy was applied. Zacuto, M.; Xu, F. J. Org. Chem. 2007,
72, 6298
.
(18) Huang, X.; Buchwald, S. Org. Lett. 2001, 3, 3417.
4000
Org. Lett., Vol. 12, No. 18, 2010

The 3-hydroxyl-oxindoles (3) which potentially can be
regenerated by hydrolysis of the 2-amino group did not
exhibit any significant antimalarial activity.
Table 1. Direct Conversion of 3 to 1 with TBDMSNH2
A 2-fold difference in in Vitro activity between the two
enantiomers of 1a was noted and interestingly, racemate was
found to be more active than either of the two enantiomers
(Scheme 4).
Scheme 4. Synthesis of Enantioenriched
2-Amino-3-hydroxy-indole 1a
In summary, we have identified 2-amino-3-hydroxy-
indoles as a novel chemical class with potent in Vitro and in
ViVo antimalaria activity. We have developed a concise
synthetic strategy to efficiently synthesize analogues in
quantities sufficient for medicinal chemistry exploration. This
method establishes the unprecedented use of TBDMSNH2
as an ammonia surrogate and allows for the first enantiose-
lective synthesis of 2-amino-3-hydroxy-indoles. It is likely
that TBDMSNH2 will find use in other transformations
requiring protected ammonia equivalents.
Acknowledgment. We are grateful to Roger Wiegand
(Broad Institute), Ted Sybertz (Genzyme Corporation), Ian
Bathurst (MMV), and all members of the Broad Institute-
Genzyme-MMV Malaria Drug Development Initiative for
thoughtful discussions; Stuart Schreiber (Broad Institute) and
the Broad Chemical Biology Program for access to key
instrumentation and reagents; Erin Tyndall and Justin Dick
for assistance with the P. falciparum viability assay; Chris
Johnson, Galina Beletsky, Kachicholu Agu and Stephen
Jonston (all Broad Institute) for analytical support; Miryam
Garcia Rosa (UPR) for assistance with the animal efficacy
studies; Peter Mu¨ller for generating the X-ray crystallography
data and Li Li for acquisition of HRMS data (both MIT
Chemistry Department). This work was supported by grants
from Medicines for Malaria Venture (MMV) and The Broad
Institute (SPARC).
a Isolated yields. b Ti(OiPr)4 was used instead of SnCl4 and NMM.
c TBS-protected aminoindole was formed, which on treatment with HF/
pyridine gave 1q; yield reported is over two steps. d The indole nitrogen is
methylated (see Supporting Information).
Table 1), which is possibly a result of the increased dihedral
angle of the biaryl system and thereby generating a more
favorable binding conformation for its target(s). 2-Amino-
3-hydroxy-indoles 1 with benzylic and alkyl groups at R2
also displayed potent antimalarial activity for R1 as 5-Cl.
Furthermore, protection of the 3-hydroxyl (Scheme 3 com-
pound 5; EC50 > 5 µM) or 2-amino groups (Scheme 3
compound 4a; EC50 > 5 µM) was found to be detrimental.
Supporting Information Available: Experimental pro-
1
cedures and characterization; H and 13C NMR, chiral SFC
chromatogram, and X-ray structure of 1a. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL101566H
Org. Lett., Vol. 12, No. 18, 2010
4001