Novel anti-tuberculosis agents from MCR libraries

Article abstract of DOI:10.1016/j.bmcl.2007.04.066

Structure-based design of libraries of multi-component reaction products yields novel potent anti-tuberculosis compounds. Synthesis and preliminary biological results are presented.

Full text of DOI:10.1016/j.bmcl.2007.04.066

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5483–5486
Novel anti-tuberculosis agents from MCR libraries
Alexander Domling,a,* Sepp Achatzb and Barbara Beckb
¨
aDepartment of Pharmaceutical Sciences, University of Pittsburgh, BST3 10019, Pittsburgh, PA 15621, USA
bMorphochem AG, Gmunderstr. 37a, Mu¨nchen, Germany
Received 20 March 2007; revised 13 April 2007; accepted 18 April 2007
Available online 25 April 2007
Abstract—Structure-based design of libraries of multi-component reaction products yields novel potent anti-tuberculosis com-
pounds. Synthesis and preliminary biological results are presented.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Tuberculosis is the world’s number one disease amongst
infections, killing every year 2 million peoples.1 More
than 2 billion humans (>30%) are infected by Mycobac-
terium tuberculosis, the causing organism of tuberculo-
sis. Often tuberculosis is accompanying AIDS.
Whereas tuberculosis is declining in developed coun-
tries, it is strongly augmenting in third world countries.
Despite the obvious urgentness for new and better treat-
ment options, probably due to the recent tractability of
TB in developed countries by current medications and
financial reasons, the development of novel drugs is a
slow process. However a recent outbreak of extreme
resistance tuberculosis (XTR TB) where neither stan-
dard drug nor any of at least three of the six classes of
more toxic and less-effective backup drugs is effective
could potentially prompt more attention to this under-
developed field in drug discovery and development.2
ciated side-effects and a poor efficacy to eradicate dor-
mant pathogens. Moreover the rising incidence of
multi-drug-resistant (MDR) tuberculosis could poten-
tially make tuberculosis quickly incurable and an even
more dangerous worldwide threat.4 Generally recog-
nized mechanisms of resistance include, e.g., drug efflux,
drug metabolism and target adaptation to the drug.
Thus novel approaches in the chemotherapy of tubercu-
losis are highly needed.
A recent proteome-wide screening for isoniacid targets
in M. tuberculosis revealed 16 new isoniacid binding pro-
teins beyond NADH-dependent enoyl-ACP reductase
(InhA) and the NADPH-dependent dihydrofolate
reductase (DfrA).5 Thus, targeting of multiple enzymes
may account for the pleiotropic effects and powerful
mycobactericidal properties of INH.
Recently the complete genome of M. tuberculosis was
published, and novel targets have raised high expecta-
tions for new drugs.3 A host of unusual properties of
the infectious agent account for the difficulties in tuber-
culosis treatment, including a complex cell envelope, col-
onization of macrophages, slow growth cycle and the
ability to remain quiescent, eventually reactivating dec-
ades after infection. Current first-line drugs include eth-
ambutol, isoniazid 1, pyrazinamide 2 and rifampicin and
second-line agents include kanamycin, p-aminosalicylic
acid, ciprofloxacine or cycloserine. However, current
drugs suffer from requiring a long treatment time, asso-
Our approach to the discovery of novel anti-tuberculosis
drugs is therefore based upon the pharmacophor of 1
and 2 (Fig. 1), with the pyridine-4-carboxy and pyrazine
carboxy groups as essential pharmacophores. Thus we
designed two scaffolds based on these scaffolds and pro-
duced libraries with 192 compounds each, based on the
respective carboxylic acids as starting materials using
the Ugi four-component reaction (U-4CR) (Fig. 2).
O
O
N
N
NH2
1
NH2
2
N
H
N
Keywords: Tuberculosis; Multi-component reaction; Isocyanide.
*
Figure 1. Structural formulas of currently used first-line anti-tubercu-
losis drugs isoniazid (INH) and pyrazineamide.
Corresponding author. Tel.: +1 412 383 5300; e-mail:
asd30+@pitt.edu
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.04.066

5484
A. Do¨mling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5483–5486
2
O
O
O
R
R3
O
H
N
O
R4
NH2
NC
R1
R4
OH
OH
N
R1
+
+
+
+
+
+
R2
R3
N
N
2
O
R
R3
O
H
N
O
N
N
N
N
N
R1
R4
NH2
NC
R1
R4
R2
R3
NH2
NH2
NH2
NH2
NH2
NH2
NH2
NH2
N
O
N
NC
N
N
A
B
C
100
100
62
D
100
11
6
E
F
G
H
in all wells
1
2
96
89
50
58
49
68
87
64
52
51
23
63
89
82
43
71
61
71
43
46
54
57
23
69
100
95
77
90
82
83
74
54
83
83
73
82
102
73
76
77
74
67
99
28
62
71
60
72
69
74
33
72
78
72
49
33
64
70
31
64
96
90
57
74
9
CHO
CHO
CHO
3
F
F
F
CHO
4
60
6
Cl
Cl
Cl
CHO
5
74
6
O
6
57
69
80
94
69
76
39
80
72
71
43
67
64
50
69
O
7
76
8
101
95
N
O
O
9
O
10
11
12
92
O
84
O
72
OMe
Figure 2. Anti-mycobacterial libraries based on Ugi’s 4-CR of nicotinic acid and pyrazine carboxylic acid. The screening results of a 96-well plate of
nicotinic acid derived compounds against Mycobacterium tuberculosis are shown (>90% growth inhibition (GI), blue; 80–90% GI, green; 70–80% GI,
orange determined at a single screening concentration of 12.5 lg/ml).
The synthesis was conveniently performed at room
temperature in 96-deep- well polypropylene plates.
Analysis of the crude reaction mixtures after 24 h by
HPLC–MS revealed the formation of virtually all
products, although in different quantities. Moreover it
was found that some wells contained considerable
amounts of the corresponding Passerini products.10
Evaporation of the solvent using a GENEVAC yielded
the crude products.
The biological evaluation of the libraries was performed
in collaboration with the Tuberculosis Antimicrobial

A. Do¨mling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5483–5486
Table 1. Minimum inhibitory concentration (MIC), cytotoxicity and macrophage assay
5485
MICa
IC50
(lg/ml)
SIc
EC90
EC99
EC90/MIC
b
d
d
Compound
Structure
(lg/ml)
O
H
N
N
N
3
3.13
27.9
9.9
n.d.
n.d.
n.d.
N
O
O
4
3.13
31.1
9.9
n.d.
n.d.
n.d.
N
N
O
N
O
N
N
5
3.13
50.3
16.1
>12.5
>12.5
>3.99
N
O
N
N
6
6.25
>62.5
>10
n.d.
n.d.
n.d.
O
N
O
O
N
EC90 and EC99 represent the concentration effecting 90% and 99% reduction in residual mycobacterial growth after 7 days, compared to untreated
controls.
a MIC measured minimum inhibitor concentration in serial dilution beginning at 6.25 lg/ml.
b IC50 cytotoxicity assay on VERO cell line beginning at 10· of MIC.
c SI selectivity index is defined as the ratio of the measured IC50 to the MIC.
d Compounds are screened by serial dilution beginning at 12.5 lg/ml.
Acquisition & Coordinating Facility (TAACF) of the
Southern Research Institute.6
pounds and its results from the biological screening
cascade are shown in Table 1.
The following biological screening cascade was used to
discover active compounds. First the crude compound
libraries were screened at one concentration (12.5 lg/
ml) against M. tuberculosis. Compounds exhibiting
<90% inhibition are not evaluated further. Out of the
best ranking compounds we resynthesized and purified
several examples.7 These compounds were rescreened
at a lower concentration against M. tuberculosis strain
H37Rv to determine the actual minimum inhibitory
concentration in a broth microdilution Alamar blue
assay (MABA). Concurrent with the MIC determination,
selected compounds are tested for cytotoxicity. Com-
pounds with a selectivity index >10 are tested for killing
of M. tuberculosis strain H37Rv in monolayer of mouse
peritoneal macrophages at 4-fold concentrations equiva-
lent to 0.25, 1, 4 and 16 times the MIC. Typical com-
We were delighted to find that a rather high percentage
of this library showed anti-tuberculosis activity. One
current model of isoniazid activity foresees isoniacid to
be an inhibitor of inhA-encoded NADH-dependent en-
oyl-ACP (acyl carrier protein) reductase enzyme.8 This
enzyme is responsible for one step in the biosynthesis
of the unusual fatty acids of the Mycobaterium sp. It
exhibits specificity for long-chain C18 and C16 enoyl thi-
oester substrates. The mode-of-action postulates the
acylation of the C4 of NADH by isoniazid. Thus
according to this model isoniazid is a prodrug which be-
comes activated by the mycobacterial catalase-peroxi-
dase enzyme KatG. The herein described inhibitors are
comprised of a nicotinicacyl moiety as in isoniazid.
However the bulkiness and the leaving group ability dif-
fer very much from the hydrazide isoniazid. Regarding

5486
A. Do¨mling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5483–5486
Pyrazine-2-carboxylic acid (1-benzylcarbamoyl-pentyl)-
cyclohexyl-amide 3, C24H32N4O2 MW: 408.55 g/mol;
the recently observed promiscuity of isoniazid and the
different chemotypes of the herein described inhibitors
one might speculate that other mycobacterial targets
are responsible for the observed activity.
HPLC–MS
(ESI-TOF):
tR = 3.46 min;
m/z = 409
[M+H]+; 431 [M+Na]+; 1H NMR (CDCl3, 250 MHz):
d = 0.89–2.28 (m, 19H), 3.19–3.43 (m, 1H), 3.95–4.17 (m,
1H), 4.49 (m, 2H), 7.32 (m, 5H), 8.21–8.94 (m, 3H); 13C
NMR (CDCl3, 60 MHz): d = 13.8, 22.4, 26 .2, 27.9, 29.1,
29.3, 30.0, 30.9, 31.3, 43.7, 58.4, 61.1, 62.0, 63.0, 127.3,
127.8, 128.6, 138.1, 141.0, 142.8, 144.7, 145.5, 146.1, 150.1,
169.5. Due to the cis/trans isomerism at the tertiary amide
bond rotation isomers are seen in the carbon NMR
spectra.
In summary, we have introduced design, combinatorial
synthesis and first biological evaluation of a novel class
of isoniazid derived anti-tuberculosis compounds.
Libraries of these compounds were synthesized using
isocyanide-based MCRs.9 Several compounds showed
cell-based activities in the same range as the mother
compound isoniazid. The synthetic approach used here
allows for the rapid and efficient optimization of this
compound class. Further studies are needed to fully
delineate the biochemical mode of action of representa-
tives of this lead series. The selection of a wider set of
starting materials should lead to compounds with im-
proved biological and pharmacokinetic activities.
N-Allyl-N-(1-allylcarbamoyl-pentyl)-isonicotinamide
4,
C18H25N3O2 MW 315.42 g/mol; HPLC–MS (ESI-TOF):
tR = 3.36 min; m/z = 316 [M+H]+; 338 [M+Na]+; 1H NMR
(CDCl3, 400 MHz): d = 0.84–0.85 (m, 3H), 1.16–1.29 (m,
4H), 1.77–1.99 (m, 2H), 3.76–4.07 (m, 4H), 4.76–4.86
(m, 1H), 4.99–5.14 (m, 4H), 5.55–5.65 (m, 1H), 5.73–5.79
(m, 1H), 7.19–7.20 (m, 2H), 8.60–8.61 (m, 2H); 13C NMR
(CDCl3, 100 MHz): d = 14.1, 22.6, 28.4, 28.6, 41.8, 49.7,
58.6, 116.3, 118.4, 121.1, 133.6, 134.1, 144.0, 150.2, 170.4,
171.2.
N-(3-Methyl-butyl)-N-{phenyl-[(pyridin-3-ylmethyl)-carbam-
oyl]-methyl}-isonicotinamide 5, C25H28N4O2 MW
416.53 g/mol; HPLC–MS (ESI-TOF): tR = 2.35 min;
m/z = 417 [M+H]+; 439 [M+Na]+; 1H NMR (CDCl3,
250 MHz): d = 0.45–1.32 (m, 11H), 3.17–3.22 (m, 1H),
4.46 (m, 1H), 5.88 (m, 1H), 7.30–7.62 (m, 11H), 8.36–8.61
(m, 3H); 13C NMR (CDCl3, 100 MHz): d = 21.7, 25.6, 38.3,
41.0, 46.7, 63.1, 120.7, 123.5, 126.3, 128.8, 128.9, 129.2,
129.6, 133.8, 134.3, 135.5, 144.1, 148.5, 148.8, 149.9, 169.5,
170.1.
Acknowledgments
We thank Dr. Joseph A. Maddry from the Tuberculosis
Antimicrobial Acquisition and Coordination Facility
(TAACF), National Institute of Allergy and Infectious
Diseases and Southern Research Institute, Birmingham,
AL (USA), for the in vitro evaluation of anti-tuberculo-
sis activity.
Isonicotinic acid 4-allylcarbamoyl-1-benzyl-piperidin-4-yl
ester 6, C22H25N3O3 MW 379.46 g/mol; HPLC–MS(ESI-
TOF): tR = 2.69 min; m/z = 380 [M+H]+; 1H NMR
(CDCl3, 250 MHz): d = 2.32–2.76 (m, 8H), 3.53–3.62 (m,
2H), 3.91 (m, 2H), 5.14 (m, 2H), 5.75–5.82 (m, 1H), 7.30–
7.35 (m, 5H), 7.82–7.84 (m, 2H), 8.78–8.80 (m, 2H); 13C
NMR (CDCl3, 100 MHz): d = 32.3, 41.3, 41.9, 48.7, 52.9,
61.9, 62.8, 81.3, 116.5, 122.9, 127.1, 127.3, 128.2, 128.3,
128.8, 128.9, 133.8, 137.1, 137.9, 138.1, 150.7, 163.4, 171.0,
209.2.
References and notes
1. http://www.who.int/tb/en/.
2. Marris, E. Nature 2006. doi:10.1038/443131b.
3. Cole, S. T. et al. Nature 1998, 398, 537.
4. Somoskovi, A.; Parson, L. M.; Salfinger, M. Respir. Res.
2001, 2, 164.
5. Argyrou, A.; Jin, L.; Siconilfi-Baez, L.; Angeletti, R. H.;
Blanchard, J. S. Biochemistry 2006, 45, 13947.
6. http://www.taacf.org/.
7. General procedure for compound resynthesis: 5 mmol
Amin and 5 mmol aldehyde are placed together without
solvent and stirred on rotary evaporator for 4 h at 50 ꢁC.
Five millimolar acid and 5 mmol isonitrile are added with
20 ml of methanol. The mixture is stirred at room
temperature overnight and the solvent is evaporated in
vacuo. The product is purified by silica gel dish chroma-
tography with ethyl acetate as solvent.
´
8. Banerjee, A.; Dubnau, E.; Quemard, A.; Balasubramanian,
V.; Um, K. S.; Wilson, T.; Collins, D.; de Lisle, G.;
Jacobs, W. R., Jr. Science 1994, 263, 227.
9. Doemling, A. Chem. Rev. 2006, 106, 17.
10. General formula of the Passerini side-products:
R2 R3
R2 R3
O
O
H
N
H
N
N
N
R4
R4
O
O
O
O
N
.