NOC chemistry for tuberculosis - Further investigations on the structure-activity relationships of antitubercular isoxazole-3-carboxylic acid ester derivatives

Article abstract of DOI:10.1002/cmdc.201000169

NOC, NOC! Who's there? We have expanded the body of structure-activity relationships (SAR) regarding the 3-isoxazolecarboxylic acid ethyl esters by the design and synthesis of 5-aryl- and 5-[(N-aryl)amino]methyl analogues. Compounds were active against Mycobacterium tuberculosis (Mtb) at sub-micromolar concentrations and, for selected compounds, no toxicity was observed on Vero cells.

Full text of DOI:10.1002/cmdc.201000169

DOI: 10.1002/cmdc.201000169
NOC Chemistry for Tuberculosis—Further Investigations on the Structure–
Activity Relationships of Antitubercular Isoxazole-3-carboxylic Acid Ester
Derivatives
Marco Pieroni,[a] Annamaria Lilienkampf,[a] Yuehong Wang,[b] Baojie Wan,[b] Sanghyun Cho,[b]
Scott G. Franzblau,[b] and Alan P. Kozikowski*[a]
Tuberculosis (TB) is a frequent infectious disease that primarily
involves the lungs and that is spread throughout the world by
Mycobacterium tuberculosis (Mtb). Both morbidity and mortality
related to this disease remain high—each year, approximately
nine million people are diagnosed with active TB and 1.6 mil-
lion die of the disease.[1] The rapid emergence of multidrug-re-
sistant Mtb (MDR-TB) and extensively-resistant Mtb (XDR-TB)
strains, along with the deadly synergism of Mtb with human
immunodeficiency virus (HIV), have further worsened the TB
pandemic, especially in developing countries.[2,3] The ability of
Mtb to persist in a semidormant state is responsible for the 9–
12 month treatment regimen,[4] which leads to poor adherence
to the therapeutic course and frequent relapses. The develop-
ment of an efficacious vaccine is still a challenge,[5] thereby
chemotherapy remains the main means to control the spread
of TB. The current therapeutic arsenal consists of four first-line
TB drugs, isoniazid (INH), pyrazinamide (PZA), ethambutol
(EMB) and rifampin (RMP), and multiple second-line drugs
(e.g., streptomycin, capreomycin, ciprofloxacin, levofloxacin, p-
aminosalicylate, ethionamide, cycloserine, thiacetazone, and ri-
fapentine), which are generally more toxic than the first-line
drugs and have limited availability in the countries where TB is
endemic.[1,6] Although a few novel agents are currently in clini-
cal trials,[7–10] the search for new anti-TB compounds remains
indispensable.
We have previously reported the discovery and optimization
of certain isoxazole-based compounds exhibiting good anti-TB
activity against both the replicating Mtb (R-TB) and nonrepli-
cating persistent Mtb (NRP-TB) (Figure 1).[11,12] These com-
Figure 1. Examples of potent anti-TB isoxazole derivatives, bearing various
linker and aromatic moieties, and details of their biological and chemical
properties (Top); and the overall construction of the anti-TB compound class
(bottom).
pounds consist of a 3-isoxazolecarboxylic acid ester core, pre-
pared by nitrile oxide cycloaddition (NOC) chemistry, which is
connected at the C-5 position to a substituted aromatic or het-
eroaromatic moiety via an alkenyl, alkyloxy, benzyloxy, benzyl-
amino, or phenoxy linker. The overall structure–activity rela-
tionships (SAR) previously obtained for this compound class
have revealed that the linker and the aromatic ring play some
role in modulating both the potency and metabolic stability of
the compound, while the 3-isoxazolecarboxylic acid alkyl ester
is essential for anti-TB activity.[12] However, it should be noticed
that the carboxylic acid derived from the hydrolysis of the
ester, although inactive in vitro, might be the chemical species
responsible for the activity, as supported by results we have re-
cently reported.[12e] Therefore, hydrolysis of the ester at its site
of action may be essential to obtaining the desired anti-TB
effect.
[a] Dr. M. Pieroni, Dr. A. Lilienkampf,+ Prof. Dr. A. P. Kozikowski
Drug Discovery Program, Department of Medicinal Chemistry & Pharma-
cognosy, College of Pharmacy, University of Illinois at Chicago
833 S. Wood Street, Chicago, IL 60612 (USA)
Fax: (+1)312-996-7107
E-mail: kozikowa@uic.edu
[b] Dr. Y. Wang, Dr. B. Wan, Dr. S. Cho, Prof. Dr. S. G. Franzblau
Institute for Tuberculosis Research, College of Pharmacy
University of Illinois at Chicago
833 S. Wood Street, Chicago, IL 60612 (USA)
[+] Current address: School of Chemistry, University of Edinburgh
Our previous studies have established that, with the excep-
tion of the unsubstituted quinoline derivative 2, lipophilic sub-
stituents on the aromatic moiety (e.g., trifluoromethyl, halo-
Joseph Black Building, West Mains Road, Edinburgh EH9 3JJ (UK)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201000169.
ChemMedChem 2010, 5, 1667 – 1672
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1667

MED
gens and alkyl groups) were generally beneficial to the activity,
whereas hydrophilic groups (e.g., hydroxy, amino and N-meth-
ylpiperazine) had a detrimental effect on the anti-TB potency.
However, the better activity is generally coupled with a high
ClogP and low total molecular polar surface area (TPSA) and
predicted water solubility (ALOGpS), features that might repre-
sent, in some cases, a hurdle for oral drug administration. To
overcome this issue, we chose to investigate whether the
linker moiety could be modified so as to improve parameters
such as the ClogP, TPSA, and ALOGpS, without affecting the
anti-TB potency.
A total of 25 compounds were synthesized and tested for
their ability to inhibit the growth of R-TB strain H37Rv in a mi-
croplate Alamar Blue assay (MABA).[15] The compounds were
also tested in a low oxygen recovery assay (LORA),[16] an in
vitro model for the preliminary assessment of the activity
against the persistent Mtb phenotype. Cytotoxicity was as-
sessed using Vero cells. With the exception of the acid 24
(MIC=>128 mm), all of the new compounds synthesized were
found to inhibit the growth of R-TB in the MABA (Tables 1
and 2).
First, we investigated the effect of the aminomethylene
linker (compounds 7–19) on the anti-TB activity. The addition
of a nitrogen atom in the linker led to the desired reduction in
the ClogP, and, in addition, the hydrogen-bond donor/accept-
or ability of the amino moiety may play a role in the interac-
tion with the active site of the target. The quinoline derivative
7 (MIC=6.9 mm) was found to be sevenfold less potent than
the corresponding oxymethylene derivative 1 (MIC=0.9 mm)
previously reported by us.[11] However, replacement of the sub-
stituted quinoline core with an unsubstituted naphthyl ring
led to a twofold improvement in activity (compound 8, MIC=
3.4 mm) as compared to 7.
Next, in order to further simplify the structure, the naphthyl
core was replaced with substituted aromatic rings, i.e., pyridine
and benzene, and the effect of various substitution patterns
on the activity was investigated. In the first round of modifica-
tions, halogens, trifluoromethyl and trifluoromethoxy groups
were chosen as substituents since in our earlier studies these
groups yielded the best overall activity.[12] Indeed, a trifluoro-
methoxy group, either in the para (13, MIC=0.8 mm) or meta
(18, MIC=2.4 mm) position of the phenyl ring, was found to be
the best substituent in the aminomethylene linker series. Re-
placement of the trifluoromethoxy group in the para position
with a trifluoromethyl group (14, MIC=6.5 mm) led to an eight-
fold decrease in anti-TB potency. Difluoro substitution patterns
on the phenyl ring yielded two- to threefold decreased activity
as compared to the naphthyl derivative 8 (9 and 15, MIC=
7.3 mm and 10.9 mm, respectively), whereas an additional fluo-
rine atom (16, MIC=53.8 mm) significantly decreased the anti-
TB potency. Substitution with electron donor alkyl groups, as
in the case of 3,5-dimethyl analogue 10 (MIC=7.0 mm), or with
an ethyl group in the ortho position of the benzene ring (20,
MIC=13.7 mm) yielded an activity comparable to that of the
fluorinated compounds.
Herein, we report the synthesis and anti-TB evaluation of a
series of compounds in which the 3-isoxazolecarboxylic acid
ester core is attached to the aromatic moiety either directly or
via an aminomethylene or an N-methylaminomethylene linker.
The majority of the new compounds synthesized were found
to maintain good anti-TB potency, with the most active com-
pounds having minimum inhibitory concentration (MIC) values
in the sub-micromolar range against R-TB and in the low mi-
cromolar range against NRP-TB.
The target compounds 7 and 25–31, as well as the key inter-
mediate 5-bromomethyl-3-isoxazolecarboxylic acid ethyl ester
(6),[13] were synthesized via a dipolar cycloaddition reaction be-
tween the nitrile oxide derived from 2-chloro-2-(hydroxyimi-
no)acetate and an appropriate dipolarophile (Scheme 1), as de-
Scheme 1. Reagents and conditions: a) ArNH2 (neat), 808C, 3–5 h, (23–75%);
b) ethyl 2-chloro-2-(hydroxyimino)acetate, Et3N (8 h through syringe pump),
anhyd Et2O, RT, 14 h, (42–90%); c) LiOH, THF/MeOH/H2O (3:1:1), RT, 30 min,
(89%); d) HCHO (aq), NaCNBH4, AcOH (cat), CH3CN, RT, 4–5 h, (50–67%); for
complete structures see Tables 1 and 2.
Other modifications aimed at introducing more hydrophilic
groups in the molecule, led to a sharp increase in the MIC
values. In particular, a nitrile group in the meta position (11,
MIC=57.9 mm) yielded the weakest activity in this series. Simi-
larly, replacement of the phenyl ring with a pyridyl ring, as in
compound 19 (MIC=39.6 mm), also seriously undermined the
activity. The unsatisfactory anti-TB activity of 19, as well as of
the nitrile derivative 11 and the methoxy substituted com-
pounds 12 and 17, may partly be due to their very low lipophi-
licity (ClogP=0.9–1.7), a feature likely to hinder compound
penetration through the thick and greasy Mtb cell wall. This
hypothesis is in part supported by the fact that methylation of
the linker nitrogen, which leads to an increase in the ClogP,
scribed previously.[14] Compounds 8–20, bearing an amino-
methylene linker, were synthesized from bromide 6 and the
appropriate aniline under neat conditions at 808C. Similarly,
the reaction of 4-bromo-2,8-bis(trifluoromethyl)quinoline with
an excess of propargylamine afforded the acetylenic intermedi-
ate 5. The N-methylated derivatives 21–23 were obtained via
reductive alkylation of compounds 8, 13, and 14, respectively,
with aqueous formaldehyde and NaBH3CN in acetonitrile with
a catalytic amount of acetic acid. Hydrolysis of ester 7 with
LiOH produced the corresponding acid 24 in excellent yield.
1668
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1667 – 1672

led to derivatives more potent
than the corresponding unme-
thylated precursors: compounds
21, 22 and 23 with MIC values of
1.5 mm, 0.7 mm, and 1.8 mm, re-
spectively. Compound 22, with
an MIC value of 0.7 mm, was the
most active compound in the
series and among the most
active of all the isoxazole-based
anti-TB compounds reported by
our group.[11,12] Moreover, when
compared to some of the mole-
cules previously reported by us
Table 1. Anti-TB activity of compounds 7–23 against Mtb strain H37Rv.
[b]
Compd
R1[a]
R2
H
MIC [mm]
IC50 [mm]
ClogP[c]
ALOGpS[d]
TPSA[e]
77.3
MABA
LORA
7
6.9
>128
>128
3.7
ꢀ4.1
8
9
H
H
3.4
7.3
7.6
nd
nd
2.8
2.0
ꢀ3.8
ꢀ3.2
64.4
64.4
15.2
14.4
with
a similar range of MIC
values (e.g., 1, 3 and 4), deriva-
tive 22 has a better ClogP and
TPSA, and an improved predict-
ed aqueous solubility. To demon-
strate the reliability of the pre-
dictive models used to drive the
design and synthesis of the iso-
xazole ligands, the aqueous solu-
bility of 22 compared to com-
pounds 4 and 30 was also inves-
tigated experimentally (table 2SI
in the Supporting Information).
The lead compound 22 was
indeed found to be more soluble
than 30, which in turn is slightly
more soluble than 4, consistent
with the data derived from the
computational analysis. The syn-
thesis of compounds 25–31, in
which the isoxazole is attached
directly to a substituted aromatic
moiety, was driven by the evi-
dence that the linker had shown
metabolic issues in our earlier
studies.[12d]
10
H
7.0
nd
2.7
ꢀ3.4
64.4
11
12
H
H
57.9
12.0
128
nd
nd
1.7
1.7
ꢀ2.9
ꢀ3.1
88.2
73.6
126
13
14
15
H
H
H
0.8
6.5
9.8
>128
nd
2.6
2.5
2.0
ꢀ3.7
ꢀ3.7
ꢀ3.2
73.6
64.4
64.4
9.0
10.9
25.3
nd
16
17
H
H
53.8
31.4
108.9
28.2
nd
nd
2.2
1.7
ꢀ3.4
ꢀ3.1
64.4
82.8
18
19
20
H
H
H
2.4
39.6
13.7
33.3
127.5
31.0
>128
nd
2.6
0.9
2.6
ꢀ3.7
ꢀ2.9
ꢀ3.6
73.6
86.5
64.4
nd
We have already reported that
a phenyl ring directly attached at
the C-5 position of the isoxazole
ring, when substituted with a
benzyloxy, benzylamine or phen-
oxy side chain, was beneficial for
activity as in compound 4 (Fig-
ure 1).[12c] We chose to further in-
vestigate this finding by intro-
ducing small, mainly lipophilic,
substituents to the phenyl ring.
Indeed, a small set of 5-phenyl-3-
isoxazolecarboxylic acid ethyl
esters (25–31) was found to ex-
hibit excellent anti-TB activity
with MIC values ranging from 0.7
to 2.2 mm. 3,5-Di(trifluoromethyl)
21
22
CH3
1.5
0.7
6.5
7.5
>128
3.3
ꢀ3.7
55.6
CH3
CH3
>128
>128
3.2
3.1
ꢀ3.6
ꢀ3.6
64.8
55.6
23
1.8
0.1
6.4
1.0
RMP
[a] indicates the point of attachment. [b] Cytotoxicity to Vero cells. [c] Calculated using ChemDraw Ultra 7.0
*
(CambridgeSoftꢁ). [d] Predicted water solubility calculated using ALOGPS 2.1 (www.vcclab.org/lab/alogps/).[18]
[e] Total molecular polar surface area (TPSA) calculated used Molinspiration (www.molinspiration.com/services/
psa.html).[19] nd=not determined. MIC values determined by MABA are the mean of replicated experiments
(SD<15%); LORA MIC values represent single measurements.
ChemMedChem 2010, 5, 1667 – 1672
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1669

MED
substitution at the phenyl ring
(30, MIC=0.7 mm) yielded the
best activity within the small set
of compounds synthesized. The
Table 2. Activity of compounds 25–31 against Mtb strain H37Rv.
trifluoromethoxy
substituted
[b]
Compd
R[a]
MIC [mm]
IC50 [mm]
ClogP[c]
ALOGpS[d]
TPSA[e]
69.4
compound 29 (MIC=2.2 mm),
which has the same aromatic
substitution pattern as 13
(MIC=0.8 mm), was the least
active compound. Surprisingly, a
primary amino group at the
meta position (27, MIC=1.6 mm)
was tolerated, whereas in our
earlier studies on isoxazole-
based anti-TB compounds, pri-
mary amino substitution had a
detrimental effect on activity.[12]
All of the compounds were
also tested for their activity in
the LORA. The LORA uses
oxygen-deprived conditions sim-
ulating the NRP-TB phenotype
environment. Among the first-
line TB drugs, only RMP and PZA
have good activity against NRP-
TB. In general, the MIC values
against NRP-TB were found to
be higher (two- to tenfold) than
those against R-TB (Tables 1 and
2). However, several compounds
MABA
LORA
25
2.1
48.7
>128
2.4
ꢀ3.3
26
27
28
1.0
1.6
1.0
25.1
41.5
15.3
nd
nd
3.5
1.9
3.7
ꢀ3.3
ꢀ2.8
ꢀ3.4
52.3
78.4
52.3
>128
29
30
2.2
0.7
17.6
8.1
nd
>128
nd
4.1
4.8
4.1
ꢀ3.8
ꢀ4.1
ꢀ3.9
61.6
52.3
61.6
31
0.9
0.1
13.2
1.0
RMP
[a] indicates the point of attachment. [b] Cytotoxicity to Vero cells. [c] Calculated using ChemDraw Ultra 7.0
*
(CambridgeSoftꢁ). [d] Predicted water solubility calculated using ALOGPS 2.1 (www.vcclab.org/lab/alogps/).[18]
[e] Total molecular polar surface area (TPSA) calculated used Molinspiration (www.molinspiration.com/services/
psa.html).[19] nd=not determined. MIC values determined by MABA are the mean of replicated experiments
(SD<15%); LORA MIC values represent single measurements.
had LORA MIC values in the low micromolar range (<10 mm).
The most active compounds in this assay, namely 21 (LORA
MIC=6.5 mm), 22 (LORA MIC=7.5 mm), 23 (LORA MIC=6.4 mm)
and 30 (LORA MIC=8.1 mm), were also among the most
potent compounds in MABA. Surprisingly, quinoline derivative
7, active against the R-TB strain, failed to exhibit any activity in
the LORA. Notably, the differences between the MIC values ob-
served in LORA and MABA for each compound in the amino-
methylene linker series (8–23) are, on average, lower than
those for compounds where the linker has been removed (25–
31). Therefore, the linker moiety in these isoxazole-based anti-
TB agents may play some role in improving the activity against
the persistent Mtb phenotype.
crease in MIC values of 1.5- to 5.8-fold was observed (data not
shown). Selected compounds did not show any apparent cyto-
toxicity against Vero cells (IC50 =>128 mm).
In conclusion, we have synthesized a number of 3-isoxazole-
carboxylic acid ester derivatives with the aim of optimizing
drug-like properties while maintaining good anti-TB activity.
The new compounds synthesized have an aminomethylene
linker, an N-methylaminomethylene linker, or alternatively, no
linker between the isoxazole C-5 position and the substituted
phenyl ring. All of the new compounds, with the exception of
acid 24, were active against Mtb, with some of them having
sub-micromolar MIC values against R-TB and low micromolar
MIC values against NRP-TB. The most active compounds
against R-TB were 5-[[N-[4-(trifluoromethoxy)phenyl]amino]-
methyl]-3-isoxazolecarboxylic acid ethyl ester (13, MIC=
0.8 mm), along with its N-methyl analogue 22 (MIC=0.7 mm),
The lead compounds also maintained good efficacy against
RMP and kanamycin single-drug-resistant (SDR) strains (ta-
ble 3SI in the Supporting Information). The higher MIC values
noted with both the INH and streptomycin single-drug-resist-
ant strains suggest differences that are not related to the
drug-specific resistance mechanisms of these two strains, but
more likely due to differences in compound metabolism or
efflux. Although this may be considered a weakness for the
further advancement of this compound class, this information
can be used to better understand the molecular target or the
biochemical pathway inhibited by these molecules, which is as
yet undetermined. Finally, when compounds 13, 22 and 30
were tested in the MABA in the presence of 4% bovine serum
albumin (BSA) or 10% fetal bovine serum (FBS), a modest in-
and
5-[2,5-bis(trifluoromethyl)phenyl]-3-isoxazolecarboxylic
acid ethyl ester (30, MIC=0.7 mm). In general, halogenated
compounds exhibited good overall anti-TB potency. The com-
pounds bearing an N-methylaminomethylene linker were more
potent than the corresponding aminomethylene derivatives,
following a trend already noticed according to which lipophi-
licity may, to some extent, affect the activity of these com-
pounds. In addition, the compounds with a substituted phenyl
moiety directly attached to the isoxazole C-5 position exhibited
good activity against R-TB. The compounds bearing an N-
methylamino- or aminomethylene linker maintained better ac-
1670
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1667 – 1672

ethyl ester[13] (6) (1.0 equiv) were heated at 808C until consumption
of the limiting reagent. Upon completion, EtOAc was added and
the mixture was filtered. The filtrate was concentrated in vacuo
and the residue was purified by flash chromatography using gradi-
ent elution (EtOAc/hexane; 0!40%) to give the title compounds.
tivity against NRP-TB than the compounds lacking the linker
moiety, indicating that the linker may play a role in the activity
against the persistent phenotype. Overall, this work highlights
the fact that the derivatives of 3-isoxazolecarboxylic acid alkyl
esters can be subjected to various modifications at the C-5 po-
sition, with the aim of optimizing physicochemical properties,
in this case ClogP, TPSA, and ALOGpS, while maintaining good
anti-TB activity.
General procedure for the synthesis of 21–23: A stirred solution
of amino derivative 8, 13 or 14 (1 equiv) and formaldehyde (37%
aq solution, 0.5 mLmmolꢀ1
)
in CH3CN (10 mLmmolꢀ1
)
at RT
) were
NaBH3CN (3.3 equiv) and glacial AcOH (0.1 mLmmolꢀ1
added. After 2 h, additional glacial AcOH (0.2 mLmmolꢀ1) was
added and the reaction mixture was stirred at RT until consump-
tion of the limiting reagent. The formed precipitate was filtered,
washed with water, and purified by flash chromatography using
gradient elution (EtOAc/hexane; 0!30%) to give the product.
5-[[[N-2,8-bis(trifluoromethyl)-4-quinolinyl]amino]methyl]-3-isox-
azolecarboxylic acid (24): LiOH·H2O (7 mg, 0.166 mmol) was
added to a solution of 7 (18 mg, 0.042 mmol) in THF/MeOH/H2O
(3:1:1; 5 mL), and the reaction mixture was stirred at RT for 0.5 h.
The organic solvents were removed in vacuo, and the resultant
aqueous solution was diluted with water and acidified with 1n HCl
to pH 5–6. The precipitate was filtered off to give 24 as a white
solid (15 mg, 89%); 1H NMR (400 MHz, CD3OD): d=4.91 (s, 2H),
6.75 (s, 1H), 6.97 (s, 1H), 7.67 (t, J=8.0 Hz, 1H), 8.13 (d, J=8.0 Hz,
1H), 8.43 ppm (d, J=8.0 Hz, 1H); 13C NMR (100 MHz, CD3OD): d=
94.3, 102.3, 124.4, 125.0, 127.8, 143.9, 151.3, 160.6, 170.9 ppm;
HRMS-ESI+: m/z [M+H]+ calcd for C16H9F6N3O3: 406.0611, found:
406.0602.
Experimental Section
The general procedures for the preparation of compounds 5–31
1
are reported below. H NMR, 13C NMR, 19F NMR, HRMS, HPLC purity
and a brief description of the biological assays are provided in the
Supporting Information.
Chemistry: 1H NMR and 13C NMR spectra were recorded on a
Bruker spectrometer at 400 MHz and 100 MHz, respectively, with
TMS as an internal standard. 19F NMR spectra were recorded on
Bruker spectrometer at 376 MHz with trifluoroacetic acid (TFA) as
an external standard. Standard abbreviations indicating multiplicity
are used as follows: s=singlet, d=doublet, dd=doublet of dou-
blets, t=triplet, q=quadruplet, m=multiplet and br=broad.
HRMS experiments were performed on Q-TOF-2TM (Micromass).
TLC was performed on Merck 60 F254 silica gel plates. Column chro-
matography was performed using CombiFlashꢁ Rf system with Re-
diSepꢁ columns or alternatively using Merck silica gel (40–60
mesh). Preparative HPLC was carried out on a Shimadzu SCL-10A
VP instrument with an ACE 5-AQ (21.2 mm ꢂ 150 mm) column.
Analytical HPLC was carried out on Agilent 1100 HPLC system with
a Synergi 4 m Hydro-RP 80A column, on a variable wavelength de-
tector G1314A. Method 1: Flow rate=1.4 mLminꢀ1; gradient elu-
tion over 20 min, from 30% CH3CN/H2O to 100% CH3CN with
0.05% TFA. Method 2: Flow rate=1.4 mLminꢀ1; gradient elution
over 20 min, from 10% CH3CN/H2O to 100% CH3CN with 0.05%
TFA.
Acknowledgements
TB Alliance is acknowledged for financial support. A. Lilienkampf
thanks the Academy of Finland (grant 120441) and the Finnish
Cultural Foundation for fellowships. Brian Becker is acknowl-
edged for protein shift assays.
N-2-propynyl-[2,8-bis(trifluoromethyl)]-4-quinolinylamine (5): A
stirred solution of 4-bromo-2,8-bis(trifluoromethyl)quinoline (0.2 g,
0.58 mmol) in dioxane (5 mL) was treated with propargylamine
(0.074 mL, 1.2 mmol) and the reaction mixture was refluxed over-
night. After cooling to RT, the mixture was poured into ice water
(15 mL), extracted with EtOAc (3ꢂ10 mL), and the organic layers
were washed with brine and dried (Na2SO4). After filtration, the sol-
vent was removed in vacuo and the crude material was purified by
flash chromatography (hexane/EtOAc; 7:3) to give compound 5 as
Keywords: drug design
derivatives · tuberculosis
·
drug discovery
·
isoxazole
[1] J. Van den Boogaard, G. S. Kibiki, E. R. Kisanga, M. J. Boeree, R. E. Aar-
noutse, Antimicrob. Agents Chemother. 2009, 53, 849–862.
[2] a) J. T. Brooks, J. E. Kaplan, K. K. Holmes, C. Benson, A. Pau, H. Masur,
Clin. Infect. Dis. 2009, 48, 609–611; b) S. D. Lawn, G. Churchyard, Curr.
Opin. HIV and AIDS 2009, 4, 325–333.
1
a light yellow powder (88 mg, 45%): H NMR (400 MHz, CDCl3): d=
[3] G. Riccardi, M. R. Pasca, S. Buroni, Future Microbiol. 2009, 4, 597–614.
[4] a) J. D. McKinney, W. R. Jacobs, Jr., B. R. Bloom in Emerging infections
(Biomedical Research Reports), (Ed.: R. M. Krause), Academic press, San
Diego, 1998, pp. 51–146; b) K. Hçner zu Bentrup, D. G. Russell, Trends
Microbiol. 2001, 9, 597–605; c) H. I. M. Boshoff, C. E. Barry, Nat. Rev. Mi-
crobiol. 2005, 3, 70–80.
[5] J. Dietrich, T. M. Doherty, APMIS 2009, 117, 440–457.
[6] S. K. Jain, G. Lamichhane, S. Nimmagadda, M. G. Pomper, W. R. Bishai,
Microbe 2008, 3, 285–292.
[7] a) E. C. Rivers, R. L. Mancera, Drug Discovery Today 2008, 13, 1090–1098;
b) M. K. Spigelman, J. Infect. Dis. 2007, 196, S28-S34.
[8] K. Andries, P. Verhasselt, J. Guillemont, H. W. H. Goehlmann, J.-M. Neefs,
H. Winkler, J. Van Gestel, P. Timmerman, M. Zhu, E. Lee, P. Williams, D. de
Chaffoy, E. Huitric, S. Hoffner, E. Cambau, C. Truffot-Pernot, N. Lounis, V.
Jarlier, Science 2005, 307, 223–227.
[9] a) C. K. Stover, P. Warrener, D. R. Van Devanter, D. R. Sherman, T. M.
Arain, M. H. Langhorne, S. W. Anderson, J. A. Towell, Y. Yuan, D. N.
McMurray, B. N. Krelswirth, C. E. Barry, W. R. Baker, Nature 2000, 405,
962–966; b) A. J. Lenaerts, V. Gruppo, K. S. Marietta, C. M. Johnson, D. K.
2.39 (t, J=2.1 Hz, 1H), 4.22 (dd, J1 =2.4 Hz, J2 =5.4 Hz, 2H), 5.29 (br
s, 1H), 6.92 (s, 1H), 7.60 (t, J=7.8 Hz, 1H), 7.98 (d, J=8.4 Hz, 1H),
8.09 ppm (d, J=7.2 Hz, 1H).
General procedure for the synthesis of 6, 7, 25–31: The appropri-
ate acetylene intermediate (1 equiv) and Et3N (3 equiv) were dis-
solved in anhyd THF or ether (15 mLmmolꢀ1). Subsequently, ethyl
2-chloro-2-(hydroxyimino)acetate (3 equiv) in anhyd THF or ether
(2 mLmmolꢀ1) was added to the solution via syringe pump over an
8 h period, and the reaction mixture was stirred overnight at RT.
The reaction mixture was filtered, washed with THF, and the filtrate
was concentrated in vacuo. The crude material was purified by
flash chromatography using gradient elution (EtOAc/hexane; 0!
30–80%). Analytical data for compounds 6[13] and 27[17] matched
the data published previously.
General procedure for the synthesis of 8–20: The appropriate
aniline (1.5 equiv) and 5-bromomethyl-3-isoxazolcarboxylic acid
ChemMedChem 2010, 5, 1667 – 1672
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1671

MED
Driscoll, N. M. Tompkins, J. D. Rose, R. C. Reynolds, I. M. Orme, Antimi-
crob. Agents Chemother. 2005, 49, 2294–2301.
[10] M. Matsumoto, H. Hashizume, T. Tomishige, M. Kawasaki, H. Tsubouchi,
H. Sasaki, Y. Shimokawa, M. Komatsu, PLoS Med. 2006, 3, e466.
[11] J. Mao, B. Wan, Y. Wang, S. G. Franzblau, A. P. Kozikowski, ChemMed-
Chem 2007, 2, 811–813.
[12] a) A. Lilienkampf, J. Mao, B. Wan, Y. Wang, S. G. Franzblau, A. P. Kozikow-
ski, J. Med. Chem. 2009, 52, 2109–2118; b) M. Pieroni, A. Lilienkampf, B.
Wan, Y. Wang, S. G. Franzblau, A. P. Kozikowski, J. Med. Chem. 2009, 52,
6287–6296; c) A. Lilienkampf, M. Pieroni, B. Wan, Y. Wang, S. G. Franz-
blau, A. P. Kozikowski, J. Med. Chem. 2010, 53, 678–688; d) J. Mao, H,
Yuan, Y. Wang, B. Wan, M. Pieroni, Q. Huang, R. B. Van Breemen, A. P. Ko-
zikowski, S. G. Franzblau, J. Med. Chem. 2009, 52, 6966–6978; e) J. Mao,
H. Yuan, Y. Wang, B. Wan, D. Pak, R. He, S. G. Franzblau, Bioorg. Med.
Chem. Lett. 2010, 20, 1263–1268.
[15] S. G. Franzblau, R. S. Witzig, J. C. Mclaughlin, P. Torres, G. Madico, A. Her-
nandez, M. T. Degnan, M. B. Cook, V. K. Quenzer, R. M. Ferguson, R. H.
Gilman, J. Clin. Microbiol. 1998, 36, 362–366.
[16] S. H. Cho, S. Warit, B. Wan, C. H. Hwang, G. F. Pauli, S. G. Franzblau, Anti-
microb. Agents Chemother. 2007, 51, 1380–1385.
[17] Z. Xin, G. Liu, Z. Pei, B. G. Szczepankiewicz, M. D. Serby, H. Zhao,
(Abbott Laboratories, Abbott Park, IL, USA), US patent application:
2004/0167188 A1, US patent number: 7169797 B2, 2004.
[18] I. V. Tetko, J. Gasteiger, R. Todeschini, A. Mauri, D. Livingstone, P. Ertl,
E. V. Radchenko, N. S. Zefirov, A. S. Makarenko, V. Y. Tanchuk, V. V. Proko-
penko, J. Comput. Aid. Mol. Des. 2005, 19, 453–463.
[19] P. Ertl, B. Rohde, P. Selzer, J. Med. Chem. 2000, 43, 3714–3717.
Received: April 20, 2010
[13] P. Bravo, A. Ricca, C. Ticozzi, O. Vajna de Pava, Gazz. Chim. Ital. 1976,
106, 743–759.
Revised: July 15, 2010
[14] A. P. Kozikowski, M. Adamczyk, J. Org. Chem. 1983, 48, 366.
Published online on August 18, 2010
1672
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1667 – 1672