Peptide deformylase inhibitors of Mycobacterium tuberculosis: Synthesis, structural investigations, and biological results

Article abstract of DOI:10.1016/j.bmcl.2008.10.040

Bacterial peptide deformylase (PDF) belongs to a subfamily of metalloproteases catalyzing the removal of the N-terminal formyl group from newly synthesized proteins. We report the synthesis and biological activity of highly potent inhibitors of Mycobacterium tuberculosis (Mtb) PDF enzyme as well as the first X-ray crystal structure of Mtb PDF. Structure-activity relationship and crystallographic data clarified the structural requirements for high enzyme potency and cell based potency. Activities against single and multi-drug-resistant Mtb strains are also reported.

Full text of DOI:10.1016/j.bmcl.2008.10.040

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6568–6572
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Peptide deformylase inhibitors of Mycobacterium tuberculosis: Synthesis,
structural investigations, and biological results
a
a
a
a
a
Arkadius Pichota a, , Jeyaraj Duraiswamy , Zheng Yin , Thomas H. Keller , Jenefer Alam , Sarah Liung ,
Gladys Lee a, Mei Ding a, Gang Wang a, Wai Ling Chan a, Mark Schreiber a, Ida Ma a, David Beer a,
Xinyi Ngew a, Kakoli Mukherjee a, Mahesh Nanjundappa a, Jeanette W. P. Teo a, Pamela Thayalan a,
Amelia Yap a, Thomas Dick a, Wuyi Meng b, Mei Xu b, James Koehn b, Shi-Hao Pan b, Kirk Clark b,
Xiaoling Xie b, Carolyn Shoen c, Michael Cynamon c
*
a Novartis Institute for Tropical Diseases, 10 Biopolis Road, #05-01 Chromos, Singapore 138670, Singapore
b Novartis Institutes for Biomedical Research, 250 Massachusetts Avenue, Cambridge, MA 02139, USA
c Veterans Affairs Medical Center, 800 Irving Avenue, Syracuse, NY 13210, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Bacterial peptide deformylase (PDF) belongs to a subfamily of metalloproteases catalyzing the removal of
the N-terminal formyl group from newly synthesized proteins. We report the synthesis and biological
activity of highly potent inhibitors of Mycobacterium tuberculosis (Mtb) PDF enzyme as well as the first
X-ray crystal structure of Mtb PDF. Structure–activity relationship and crystallographic data clarified
the structural requirements for high enzyme potency and cell based potency. Activities against single
and multi-drug-resistant Mtb strains are also reported.
Received 10 June 2008
Revised 24 September 2008
Accepted 1 October 2008
Available online 14 October 2008
Keywords:
Mycobacterium tuberculosis
Ó 2008 Elsevier Ltd. All rights reserved.
PDF
Antibacterial
SAR
X-Ray
Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb),
is the most common infectious disease caused by a single bacte-
rium, and it is estimated that two billion people or one-third of
the world’s population have been infected by Mtb. More than eight
million new cases of active TB disease each year result in two mil-
lion deaths annually, mostly in developing countries.1 TB is pres-
ently treated with a four-drug combination (isoniazid, rifampin,
pyrazinamide, and ethambutol) that imposes a lengthy 6- to 9-
month treatment course, often under the direct observation of a
healthcare provider.2 The long treatment time makes patient com-
pliance a challenge and the emergence of multi-drug-resistant Mtb
strains (MDR) has limited the available treatment options.3 The sit-
uation has recently become even more dire with reports of so
called extensively drug-resistant TB (XDR), where especially in
HIV/Mtb co-infections all treatment options are failing.4 Novel TB
drugs are urgently needed to shorten treatment time and to treat
drug-resistant TB in a more effective way.
peptides, a key step in protein maturation.5 PDF is essential for
bacterial growth6 making it an appealing target for the design of
novel antibiotics.7 Naturally occurring antibiotics like actinonin in-
hibit the activity of the PDF enzyme, and several synthetic PDF
inhibitors based on the actinonin scaffold have progressed to pre-
clinical and clinical development.8
In a previous study, we have validated Mtb PDF as a drug target
and shown that the PDF inhibitor LBK-611 (1) and analogs have the
potential to serve as a new class of antimycobacterial agents.9 We
report here the structure–activity relationship of analogs of this
R
OH
N
MBG
O
N
N
O
O
NH
Y
Peptide deformylase (PDF) is a bacterial metalloenzyme which
deformylates the N-formylmethionine of newly synthesized poly-
O
N
N
1
2
* Corresponding author.
E-mail address: arkadius.pichota@novartis.com (A. Pichota).
Figure 1. Structures of PDF inhibitor LBK-611 (1) and of corresponding bioisosteric
PDF inhibitors (2).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.040

A. Pichota et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6568–6572
6569
initial lead molecule with the general structure (2). The aim of this
R
R
EtO
OEt
i, ii
iii
study was to reduce the peptidic character of the lead compound
by introducing benzimidazoles and benzoxazoles (Y = O, NH; see
Fig. 1) moieties. The synthesis of these amide bioisosters and the
corresponding SAR around the metal-binding group (MBG; see
Fig. 1) as well as the P10 position (R; see Fig. 1) are reported. Addi-
tionally, we describe the first X-ray crystal structure of Mtb PDF in
complex with inhibitors of general structure 2.
The syntheses of PDF inhibitors used in this study are outlined in
Schemes 1–3.10 (S)-2-Pyrrolidin-2-yl-1H-benzoimidazole (5)11 was
prepared from N-Boc-proline (Boc-3) by coupling with 2-amino ani-
line and followed by imidazole formation in presence of acetic acid12
at 60 °C as shown in Scheme 1. (S)-2-Pyrrolidin-2-yl-benzooxazole
(8) was prepared from N-Cbz-proline (Cbz-3) by treatment with
ammonia and further conversion of the resulting N-Cbz-Pro-NH2
(6) to N-Cbz-Pyrrolidine-2-carbonitrile (7) with POCl3 and pyri-
dine.13 Treatment of 7 with AcCl in ethanol led to reactive interme-
diate (S)-Pyrrolidine-2-carboximidic acid ethyl ester (not shown in
the scheme). Subsequent treatment in situ with 2-hydroxy aniline
produced the N-Cbz-protected benzoxazole which was followed
by hydrogenation to give (S)-2-Pyrrolidin-2-yl-benzoxazole (8).
The P10 moieties (14)14 containing different R side chains were
synthesized as shown in Scheme 2.
HO
OH
OH
O
O
O
O
O
O
9
10
11
iv, v
O
O
O
R
OBn
R
vi, vii, viii
N
HN
N
O
O
Ph
12
Ph
ix, x
13
OH
N
R
OBn
N
R
O
N
xi, xii
O
OH
O
5 or 8
Y
N
O
14
15a-15e (Y = NH)
16a-16e (Y = O)
Commercially available dietheylmalonate (9) was alkylated
with the corresponding alkyl bromides, and hydrolysis of the ester
groups yielded the 2-substituted malonic acids 10. Subsequent
treatment of 10 with formaldehyde in piperidine at 80 °C provided
the acrylic acids (11). The corresponding mixed anhydrides were
obtained by treatment with pivaloyl chloride. Introduction of the
chiral auxiliary (S)-benzyl-2-oxazolidinone provided 12 and conju-
gate addition of benzyloxyamine (BnONH2) formed diasteromeric
adducts (9:1 ratio of R to S for R = Bu as reported by Pratt
et al.14). The desired (R)-diastereoisomer could be isolated as the
p-toluene sulfonic acid salt. Upon removal of the chiral auxiliary,
N-formylation provided key P10 fragments 14. The final com-
pounds, 15a–e (Y = NH) and 16a–e (Y = O), were assembled
through standard peptide coupling methodology with 5 or 8.
Carboxylic acids 19a and b and hydroxamic acids 20a and b
were synthesized as shown in Scheme 3. Coupling of (R)-2-butyl-
succinic acid 4-tert-butyl ester (18)15 with either 5 or 8 provided
acids 19a and 19b. Subsequent treatment with benzyloxyamine
followed by hydrogenation gave the corresponding hydroxamic
acids 20a (Y = NH) and 20b (Y = O).
Scheme 2. Reagents and conditions: (i) EtONa, R-Br, EtOH; (ii) KOH; (iii) CH2O,
piperidine, 80 °C; (iv) pivaloyl chloride, NEt(i-Pr)2, THF; (v) (S)-4-benzyl-2-oxazo-
lidinone, BuLi, ꢀ78 to 25 °C; (vi) BnONH2 (neat); (vii) p-TSA, EtOAc; (viii) Na2CO3;
(ix) LiOH, 30% H2O2, THF–H2O, 0 °C; (x) HCO2H, Ac2O, 0–25 °C; (xi) 5 or 8, HATU,
NEt(i-Pr)2, DMF; (xii) 10% Pd/C, H2, EtOH.
bacilli Calmette Guerin (M. bovis BCG) for compounds with general
structure 2. M. bovis BCG was used as surrogate of Mtb to evaluate
the antibacterial activity of the PDF inhibitors and all biological
evaluations were carried out as previously described.9
From the data in Table 1, the nature of the chelating group has a
marked effect on Mtb PDF activity. Carboxylic acids 19a and b are
significantly less active than their corresponding hydroxamic acids
20a and 20b and reverse hydroxamic acid analogs 15a and 16a.
The loss in activity can be explained by the relatively poor chelat-
ing properties of carboxylic acids with nickel compared to
hydroxamic and reverse hydroxamic acids. This trend in activity
is also observed in the cell-based screen where the carboxylic acids
show no activity against M. bovis BCG (>9 lg/ml). The hydroxamic
acids, despite their high potency against Mtb PDF enzyme, have
significantly lower antibacterial activity than the corresponding re-
verse hydroxamic acid analogs. Additionally, the differences in the
Table 1 summarizes the enzyme inhibition against Mtb Ni–PDF
enzyme and the whole-cell activity against Mycobacterium bovis
HN
N
boc
i
ii, iii
O
NH
N
NH
H2N
N
O
5
PG
4
OH
HN
N
3
iv
v
vi, vii, viii
N
N
cbz
cbz
O
CN
O
NH2
6
8
7
Scheme 1. Reagents and conditions: (i) N-Boc-proline (Boc-3), 2-Amino aniline, HATU, NEt(i-Pr)2, DMF; (ii) HOAc, 60 °C; (iii) 25% TFA, CH2Cl2; (iv) N-Cbz-proline (Cbz-3),
NH3, MeOH; (v) POCl3, pyridine; (vi) AcCl, EtOH; (vii) 2-Hydroxy aniline, EtOH, 80 °C; (viii) 10% Pd/C, H2, EtOH.

6570
A. Pichota et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6568–6572
O
i, ii, iii, iv
OH
OH
tBuO
O
O
17
18
Y
N
N
v, vi
H
5 (Y = NH)
8 (Y = O)
O
O
vii, viii
HO
N
N
N
HO
H
O
O
Y
Y
N
N
20a (Y = NH)
20b (Y = O)
19a (Y = NH)
19b (Y = O)
Scheme 3. Reagents and conditions: (i) SOCl2; (ii) (S)-4-benzyl-2-oxazolidinone, BuLi, ꢀ78 to 25 °C; (iii) NaHMDS, tert-butyl bromoacetate, ꢀ78 to 25 °C; (iv) LiOH, 30% H2O2,
THF–H2O, 0 °C; (v) 5 or 8, HATU, NEt(i-Pr)2, DMF; (vi) TFA; (vii) NH2OBn, EDC, HOBt, DMF; (viii) 10% Pd/C, H2, EtOH.
Table 1
PDF–Ni IC50 for analogs of 2 and antibacterial activities against M. bovis BCGa
R
MBG
N
O
Y
N
.
Compound
MBG
R
Y
PDF–Ni IC50
(
lM)
BCG MIC (lg/ml)
15a
15b
15c
15d
15e
16a
16b
16c
16d
16e
19a
19b
20a
20b
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
N(OH)CHO
COOH
Bu
Pr
Pentyl
Cpm
Bn
Bu
Pr
Pentyl
Cpm
Bn
NH
NH
NH
NH
NH
O
O
O
O
O
0.010
0.024
0.015
0.008
0.068
0.013
0.028
0.013
0.021
0.161
0.202
0.803
0.018
0.049
0.6
9
1.2
1.2
>9
0.6
4.5
0.6
1.2
>9
Bu
Bu
Bu
Bu
NH
O
NH
O
>9
>9
4.5
4.5
COOH
CONHOH
CONHOH
a
For assay conditions see Ref. 9.
benzimidazole and benzoxazoles moieties were found to have little
effect on activity in both Mtb PDF and M. bovis BCG assays.
The observed inhibitory activity of the tested compounds
against Mtb PDF with respect to different R groups is consistent
with the presence of a well defined hydrophobic S10 pocket.16
The S10 pocket plays a critical role in binding the methionine found
in the natural substrate and is as such capable of accommodating
small alkyl and cycloalkyl groups. Consequently, inhibitors incor-
porating hydrophobic R groups like propyl, butyl, pentyl and cyclo-
pentylmethyl displayed potent Mtb PDF inhibition. However the
incorporation of a benzyl group resulted in a significant loss in en-
zyme inhibition.
from a combination of bacterial cell wall penetration properties
and recognition by an efficient efflux system.17
The activity against Mtb was assessed for LBK-611 (1) and two
active PDF inhibitors of both series: 15a and d (benzimidazoles)
and 16a and d (benzoxazoles) (Table 2).18 All PDF inhibitors were
found to be highly active (0.05–0.2
H37Rv19 and comparable to the gatifloxacin control (average MIC
of 0.07 g/ml). As expected for compounds with a new mechanism
lg/ml) against Mtb strain
l
of action, the PDF inhibitors retain their excellent activity against
single and multi-drug-resistant Mtb strains (Table 2, 1, 15a and
16a). These results suggest that PDF inhibitors may have potential
as antimycobacterial agents against MDR TB.
Compounds with similar Mtb PDF activity often display differ-
ent cell-based activity profiles. These differences may be derived
Structures of several complexes of Mtb PDF with our inhibitors
were determined at 2.0 to 1.4 Å resolutions. Each of these struc-

A. Pichota et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6568–6572
6571
Table 2
similar to those from other species. The substrate-binding site
can be divided into three sub main pockets, termed the S10, S20
and S30 pockets, and the metal-binding site. The S10 pocket is a
deep hydrophobic pocket, whereas the S20 and S30 pockets are shal-
low, less well-defined, solvent exposed surface depressions. The
active site of Mtb PDF is highly conserved, particularly in the S10
pocket, which is conserved across species. The nickel ion is coordi-
nated by three protein residues (His148, His152 and Cys106) and a
bidentate reverse hydroxamic acid moiety, resulting in a square
pyramidal coordination. The interactions between Mtb PDF en-
zyme and inhibitor 16a are illustrated in Figure 2; most of the
interactions are with conserved residues, and all H-bond interac-
tions are with protein backbone atoms. The crystal structure of
Mtb Ni–PDF–16a complex (Fig. 2) indicates that substituents at
P30 can be well tolerated without affecting PDF activity as their
binding site is partially solvent exposed. This finding suggests that
this position is ideally suited for the introduction of favorable phar-
macokinetic properties through various substituents. This will be
described in a separate letter.
In vitro antitubercular activity against selected strains of M. tuberculosis (MIC in
ml).
lg/
Compound
Mtb
H37Rv
INH
SM
RIF
PZA
Beijing W
MDRe
SDRa
SDRb
SDRc
SDRd
1
0.2f
0.1
0.05
0.15f
0.2
0.1
0.03
nd
0.06
nd
0.5
0.2
nd
0.5
nd
0.25
0.125
nd
0.6
nd
0.5
0.25
nd
0.25
nd
0.125
0.03
nd
0.06
nd
15a
15d
16a
16d
Gatifloxacin was used as positive control (MIC H37Rv = 0.07
l
g/ml).
a
Isoniazid single-drug-resistant (SDR) strain.
Streptomycin-resistant strain.
Rifampicin-resistant strain.
Pyrazinamide-resistant strain.
Beijing W multi-drug-resistant strain.
Average value of eight independent experiments.
b
c
d
e
f
tures contains a complex of Mtb PDF, a bound metal ion (Ni2+) and
an inhibitor. The X-ray crystal structure of 16a with Mtb PDF is
shown in Figure 2.20
In conclusion, we have identified highly potent PDF inhibitors of
Mtb and established an initial SAR. Several potent inhibitors of Mtb
PDF were identified which were also highly active against single
and multi-drug-resistant strains of Mtb, indicating their potential
use as antibacterial agents against MDR TB. The X-ray crystal struc-
ture of an Mtb PDF–inhibitor complex was solved and used to iden-
tify the structural requirements for high enzyme potency.
Based on sequence homology, PDFs are typically classified as
type I (generally from Gram-negative species) or II (generally from
Gram-positive species). Type II PDF has a larger molecular weight
than type I PDF, primarily due to sequence insertions; however,
the type I PDFs have a longer C-terminal extension than type II.
These sequence differences result in structural differences between
type I and type II PDFs.16 Amino acid sequence analysis of Mtb PDF
indicates that it has some features similar to type I PDF, such as the
extended C-terminal portion, while it also has features similar to
type II PDF, such as several insertions.21 However, the core struc-
ture of Mtb-PDF is shown to be more similar to that of a typical
type I PDF (such as Escherichia coli PDF). Similar to E. coli PDF,
Acknowledgments
The authors thank Jeasie Tan for analytical support, and Kathryn
Bracken, Neil Ryder and Beat Weidmann for helpful discussions.
Supplementary data
the structure of Mtb PDF contains three a-helices, seven b-sheets,
and three 310 helices, and all the residues from three highly con-
served motifs are structurally conserved forming the active site.
The unique feature of Mtb PDF is its unusually long C-terminal
extension, part of which forms a b-strand pairing with the b-strand
from N-terminal region. This extension has been shown to be crit-
ical for Mtb PDF activity.22 The substrate-binding site of PDF from
different bacteria species have been described previously.16
The X-ray crystal structure of Mtb PDF and bound inhibitors
were used to identify and explain the structural requirements for
the enzyme potency observed. Overall, the structures of Mtb PDF-
inhibitor complexes reveal that the active-site in Mtb PDF is very
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2008.10.040.
References and notes
1. (a) Dye, C. Lancet 2006, 367, 938; (b) World Health Organization (WHO), Fact
Sheet No. 104: Tuberculosis, 2007.
2. Davies, P. D. O.; Yew, W. W. Expert Opin. Invest. Drugs 2003, 12, 1297.
3. (a) Farmer, P.; Kim, J. Y. Br. Med. J. 1998, 317, 671; (b) Treatment of Tuberculosis:
Guidelines for National Programmes, 3rd ed.; World Health Organization,
Geneva; 2004.
4. (a) Marris, E. Nature 2006, 443, 131; (b) Centers for Disease Control and
Prevention (CDC), Morbidity and Mortality Weekly Report, 2006, 55, 301.
5. (a) Adams, J. M. J. Mol. Biol. 1968, 33, 571; (b) Adams, J. M.; Capecchi, M. Proc.
Natl. Acad. Sci. U.S.A. 1966, 55, 147.
6. (a) Chan, P. F.; O’Dwyer, K. M.; Palmer, L. M.; Ambrad, K. A.; Ingraham, K. A.; So,
C.; Lonetto, M. A.; Biswas, S.; Rosenberg, M.; Holmes, D. J.; Zalacain, M. J.
Bacteriol. 2003, 185, 2051; (b) Mazel, D.; Pochet, S.; Marliere, P. EMBO J. 1994,
13, 914; (c) Yuan, Z.; Trias, J.; White, R. J. Drug Discov. Today 2001, 6, 954; (d)
Clements, J. M.; Ayscough, A. P.; Keavey, K.; East, S. P. Curr. Med. Chem. Anti-
Infect. Agents 2002, 1, 239; (e) Leeds, J. A.; Dean, C. R. Curr. Opin. Pharmacol.
2006, 6, 445.
7. (a) Pei, D. Emerg. Ther. Targets 2001, 5, 23; (b) Giglione, C.; Pierre, M.; Meinnel,
T. Mol. Microbiol. 2000, 36, 1197.
8. (a) Chen, D.; Yuan, Z. Expert Opin. Invest. Drugs 2005, 14, 1107; (b) Johnson, K.
W.; Lofland, D.; Moser, H. E. Curr. Drug Targets Infect. Disord. 2005, 5, 39; (c)
Ramanathan-Girish, S.; McColm, J.; Clements, J. M.; Taupin, P.; Barrowcliffe, S.;
Hevizi, J.; Safrin, S.; Moore, C.; Patou, G.; Moser, H. Antimicrob. Agents
Chemother. 2004, 48, 4835.
9. Teo, J. W. P.; Thayalan, P.; Beer, D.; Yap, A. S. L.; Nanjundappa, M.; Ngew, X.;
Duraiswamy, J.; Liung, S.; Dartois, V.; Schreiber, M.; Hasan, S.; Cynamon, M.;
Ryder, N. S.; Yang, X.; Weidmann, B.; Bracken, K.; Dick, T.; Mukherjee, K.
Antimicrob. Agents Chemother. 2006, 50, 3665.
10. Pichota, A.; Duraiswamy, J.; Yin, Z.; Keller, T. H.; Schreiber, M. WO 2007077186
A1 20070712; Chem. Abstr. 2007, 147, 166191.
11. Patel, D. V.; Yuan, Z.; Jain, R. K.; Lewis, J. G.; Jacobs, J.; WO 2002102791 A1
20021227; Chem. Abstr. 2002, 138, 55864.
12. Balboni, G.; Salvadori, S.; Guerrini, R.; Negri, L.; Giannini, E.; Jinsmaa, Y.; Bryant,
S. D.; Lazarus, L. H. J. Med. Chem. 2002, 45, 5556.
Figure 2. X-Ray crystal structure of 16a bound to Ni–PDF of M. tuberculosis. The
Ni2+ atom is colored in magenta, nitrogen atoms in blue, and oxygen atoms in red.
The compound is shown as yellow stick model in the active site. X-ray analysis
revealed that the benzoxazole part is pointing out of the active site pocket. The
critical interactions are between the chelating unit and the metal atom, the butyl
side chain (methionine mimic), and the central carbonyl atom (H-bond to V5).
Atomic coordinates have been deposited in RCSB protein data bank, with accession
code 3E3U.
13. Almquist, R. G.; Chao, W. R.; Jennings-White, C. J. Med. Chem. 1985, 28, 1067.

6572
A. Pichota et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6568–6572
14. Pratt, L. M.; Beckett, R. P.; Davies, S. J.; Launchbury, S. B.; Miller, A.; Spavold, Z.
M.; Todd, R. S.; Whittaker, M. Bioorg. Med. Chem. Lett. 2001, 11, 2585.
15. Jain, R.; Sundram, A.; Lopez, S.; Neckermann, G.; Wu, C.; Hackbarth, C.; Chen,
D.; Wang, W.; Ryder, N. S.; Weidmann, B.; Patel, D.; Trias, J.; White, R.; Yuan, Z.
Bioorg. Med. Chem. Lett. 2003, 13, 2373.
16. (a) Guilloteau, J. P.; Mathieu, M.; Giglione, C.; Blanc, V.; Dupuy, A.; Chevrier, M.;
Gil, P.; Famechon, A.; Meinnel, T.; Mikol, V. J. Mol. Biol. 2002, 320, 951; (b)
Kreusch, A.; Spraggon, G.; Lee, C. C.; Klock, H.; McMullan, D.; Ng, K.; Shin, T.;
Vincent, J.; Warner, I.; Ericson, C.; Lesley, S. A. J. Mol. Biol. 2003, 330, 309; (c)
Smith, K. J.; Petit, C. M.; Aubart, K.; Smyth, M.; McManus, E.; Jones, J.; Fosberry,
A.; Lewis, C.; Lonetto, M.; Christensen, S. B. Protein Sci. 2003, 12, 349.
17. (a) Nikaido, H. Cell Dev. Biol. 2001, 12, 215; (b) Silva, P. E. A.; Bigi, F.; De La Paz
Santangelo, M.; Romano, M. I.; Martin, C.; Cataldi, A.; Ainsa, J. A. Antimicrob.
Agents Chemther. 2001, 45, 800; (c) Danilchanka, O.; Mailaender, C.;
Niederweis, M. Antimicrob. Agents Chemother. 2008, 52, 2503.
18. Cynamon, M. H.; Alvirez-Freites, E.; Yeo, A. E. T. J. Antimicrob. Chemother. 2004,
53, 403.
19. A lower culture inoculum was used in the TB assay compared to the BCG assay
(105 vs 106 CFU/ml) causing the systematically lower MICs.
20. The details of structure refinement are summarized in the
supplementary material. Coordinates of the final structure have been
deposited in the RCSB Protein Data Bank under the accession number
3E3U.
21. Saxena, R.; Chakraborti, P. K. J. Bacteriology 2005, 187, 8216.
22. Saxena, R.; Chakraborti, P. K. Biochem. Biophys. Res. Commun. 2005, 332,
418.