Novel anti-infection agents: Small-molecule inhibitors of bacterial transcription factors

Article abstract of DOI:10.1016/j.bmcl.2007.07.072

Structure-based drug design was utilized to identify potent small-molecule inhibitors of proteins within the AraC family of bacterial transcription factors, which control virulence in medically important microbes. These agents represent a novel approach to fight infectious disease and may be less likely to promote resistance development. These compounds lack intrinsic antibacterial activity in vitro and were able to limit a bacterial infection in a mouse model of urinary tract infection.

Full text of DOI:10.1016/j.bmcl.2007.07.072

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5652–5655
Novel anti-infection agents: Small-molecule inhibitors
of bacterial transcription factors
Todd E. Bowser,a,* Victoria J. Bartlett,a Mark C. Grier,a Atul K. Verma,a
Taduesz Warchol,a Stuart B. Levya,b and Michael N. Alekshuna,
aParatek Pharmaceuticals, Inc., 75 Kneeland Street, Boston, MA 02111, USA
bCenter for Adaptation Genetics and Drug Resistance, Tufts University School of Medicine, 136 Harrison Avenue,
Boston, MA 02111, USA
Received 27 June 2007; accepted 19 July 2007
Available online 21 August 2007
Abstract—Structure-based drug design was utilized to identify potent small-molecule inhibitors of proteins within the AraC family
of bacterial transcription factors, which control virulence in medically important microbes. These agents represent a novel approach
to fight infectious disease and may be less likely to promote resistance development. These compounds lack intrinsic antibacterial
activity in vitro and were able to limit a bacterial infection in a mouse model of urinary tract infection.
Ó 2007 Elsevier Ltd. All rights reserved.
The emergence of multidrug-resistant bacteria has chal-
lenged researchers to develop novel therapies for the
prevention and treatment of infectious diseases. Tradi-
tional antibiotics exert strong selective pressure for resis-
tance development, since they target bacterial processes
essential for the growth of the organism. Alternative ap-
proaches have involved the modulation of virulence fac-
tors, elements that contribute to infection in vivo but are
not essential for growth of the organism in vitro. Most
attempts along this path have targeted individual viru-
lence factors, such as pili or toxins, but these approaches
have suffered from a limited spectrum of activity. The
targeting of regulatory mechanisms that could affect
multiple virulence factors simultaneously may offer an
alternative approach. Toward this end, the AraC family
of bacterial transcription factors represents a novel tar-
get for this unique therapeutic approach.
otic and oxidative stress resistance, expression of type
III secretion, toxin production, and other processes that
are important during infection.2–6 Previous studies have
demonstrated that the inactivation of genes encoding
AraC proteins attenuates virulence in human7 and ani-
mal models of infection.3,4,6,8–10,12 Since all of the AraC
proteins contain a highly conserved DNA-binding do-
main, it is envisioned that a small-molecule inhibitor di-
rected at or otherwise affecting this domain could
interfere with protein function and act as an effective
agent for preventing infection.
Using a structure-based drug design strategy, we chose
to focus on a subset of AraC proteins including those re-
lated to MarA and other multiple antibiotic resistance
(Mar) proteins.11 Previous work in our laboratory dem-
onstrated that Mar proteins (MarA, SoxS, and Rob)
regulate virulence of Escherichia coli in a murine model
of ascending pyelonephritis.12 Structural information
from crystallographic studies13,14 of Mar proteins and
supportive data from NMR13–15 and mutational experi-
ments16 were used to construct a computer model of the
conserved DNA-binding domain. A set of combinato-
rial chemistry scaffolds was docked to this active-site
template to identify potential small-molecule inhibitors.
High scoring hits were used to search the ACX-SC data-
base (CambridgeSoft Corp., Cambridge, MA) for com-
mercially available compounds with structural
similarity. From this database, approximately 2000
The AraC family of proteins is a large group of tran-
scription factors found in a variety of medically impor-
tant bacteria, particularly the Enterobacteriaceae.1
These proteins regulate virulence by controlling antibi-
Keywords: Antibacterials; Virulence; Hydroxybenzamidazoles;
Transcription.
*
Corresponding author. Tel.: +1 617 275 0040; fax: +1 617 275
0039; e-mail: tbowser@paratekpharm.com
Present address: Schering-Plough Research Institute, 2015 Galloping
Hill Road, Kenilworth, NJ 07033, USA.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.072

T. E. Bowser et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5652–5655
5653
An IC50 of 17 lM was determined for compound 34 in
the SoxS DNA-binding inhibition assay (Table 2). This
value is at least 15-fold lower than the IC50 of >250 lM
estimated for compound 9 (exact IC50 determination of
9 was limited by solubility of the compound). A number
of exploratory analogs of 34 substituted on the benzam-
ide side chain were prepared as before and tested in the
SoxS assay (36–40, Table 2). All of the analogs demon-
strated 100% inhibition at 25 lg/mL in the screening
assay (data not shown) and low micromolar IC50 values
against SoxS. In particular, an IC50 of 820 nM was
observed for the p-dimethylamino derivative 40. This
IC50 represents a more than 300-fold increase in potency
over 9.
R2
R6
H2N
R6
NO2
NO2
F
N
H
a
R2
OH
N
R6
R2
N
b
c
9-40
R1
O
N
R6
R2
N
1-8
The above 1-hydroxybenzamidazoles were designed to
specifically target proteins controlling the expression of
virulence factors which are not essential for growth of
the organism in vitro. Therefore, it was necessary to
confirm that these compounds (e.g., 34, 36–40 for illus-
trative purposes) were devoid of intrinsic antibacterial
activity. No inhibition of bacterial growth in vitro was
observed when these derivatives were screened against
antibiotic susceptible strains of Staphylococcus aureus
and E. coli (Table 2).
Scheme 1. Reagents and conditions: (a) Na2CO3, DMF, rt; (b) NaOH
(aq), rt or NaOMe, MeOH, 25–60 °C or NaH, THF, 40–60 °C; (c) R1-X,
(X = Cl or Br), NaHCO3, DMF, rt.
compounds were purchased and screened for inhibitory
activity with an in vitro protein–DNA-binding assay
using MarA, SoxS, and Rob.17 Five unique structural
classes of inhibitors were identified that negatively af-
fected the binding of Mar proteins to their cognate
DNA. Based on the potential for chemical diversity,
we focused our initial drug discovery effort on the
1-hydroxybenzimidazole class of Mar inhibitors. Ana-
logs of the hit compound 1 were prepared as previously
described18,19 to systematically explore the effects of the
1, 2, and 6-position on activity (Scheme 1). The new
derivatives were screened for their ability to inhibit the
binding of SoxS, a representative Mar protein, to its
DNA target. Screening of the 1-position derivatives
2–9 identified the 1-hydroxy analog 9 as the most active
inhibitor of SoxS (Table 1). This compound demon-
strated improved activity over the original hit
compound 1 (38% versus 14% inhibition, respectively)
and was used as the core structure for further explora-
tion of the 6 and 2-positions.
Since AraC proteins have a highly conserved DNA-
binding domain, representative compounds 38 and 40
were screened for activity against other AraC family
members from E. coli (MarA, Rob), Pseudomonas
aeruginosa (ExsA), Salmonella typhimurium (Rma),
and Proteus mirabilis (PqrA). As shown in Table 3, com-
pounds 38 and 40 demonstrated comparable activity
in vitro for all AraC family members tested.
Using an animal model of ascending pyelonephritis
caused by E. coli,20,12 compounds 37 and 40 were judged
for the ability to affect kidney infection. Previous studies
using this urinary tract infection model have shown that
E. coli mutants with a soxS gene deletion colonize the
mouse kidney in numbers approximately 1-log fewer
than the wild type strain. When administered at
10 mg/kg by way of an intraperitoneal route, compound
37 showed a statistically significant 1-log decrease in the
number of E. coli colony forming units relative to the
untreated control (Table 4). Compound 40 also reduced
the colony number relative to the untreated control, but
the effect did not reach statistical significance.
Several analogs of 9 modified at the 6-position (10–21)
were prepared to explore the replacement of the nitro
group. The removal of the nitro group (10) eliminated
activity against SoxS. However, comparable activity to
9 was observed for the 6-acetyl (17), 6-ethanoneoxime
(18), and 6-methanesulfonyl (19) analogs. Less, but
detectable, inhibition was observed for electron-with-
drawing substitutions (11–13) and the dimethylamino
analog 14.
The results of this preliminary study identify the
1-hydroxybenzimidazoles as effective ‘‘broad spec-
trum’’ small-molecule inhibitors of proteins within
the AraC family. In addition, select derivatives were
able to attenuate an E. coli infection in a murine mod-
el of urinary tract infection. Since these compounds
are devoid of intrinsic antibacterial activity in vitro,
the risk of rapid resistance development is expected
to be less than that observed with traditional antibiot-
ics. The potent activity in vitro and efficacy in vivo
highlight the potential for these derivatives as a novel
treatment for infectious diseases. Efforts are ongoing
to establish a detailed SAR of the 1-hydroxybenzimi-
dazoles against several AraC proteins with the hope
The most dramatic improvement in activity was ob-
tained through substitution of the 2-position phenyl
group of 9 (22–35). Screening of these derivatives iden-
tified the p-amino-substituted analogs (25, 29, 31, and
34) as the most potent inhibitors of SoxS. In particular,
the p-aminoacyl analogs (31, 34) demonstrated complete
inhibition of SoxS in the screening assay. Compound 34
also demonstrated complete inhibition of SoxS when di-
luted to 25 lg/mL (67 lM). Based on these results, 34
was chosen for further testing.

5654
T. E. Bowser et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5652–5655
Table 1. Inhibition of SoxS binding to its cognate DNA by 1-hydroxybenzimidazoles
R1
O
N
R6
R2
N
Compound
R6
R1
R2
% Inhibitiona
1
2
NO2
NO2
CH2CO2H
CH2CO2Et
H
14
0
H
3
NO2
NO2
CH2CH2CO2H
CH2CH2OH
H
0
4
H
0
12
5
NO2
NO2
CH2CN
CH2CH2CH3
H
6
H
0
17
7
NO2
NO2
CH3
CH2CH2NH2
H
8
H
0
38
9
NO2
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
H
0
26
F
CN
H
H
30
26
F3C
H
(H3C)2N
tert-Butyl
H3CO
H3CC(O)
H3CC(NOH)
H3CSO2
HOC(O)
H2NC(O)
NO2
H
28
0
H
H
0
37
H
H
38
35
H
H
12
14
H
p-Br
26
35
NO2
NO2
p-OH
p-OCH3
p-NH2
m-NH2
12
80
NO2
NO2
60
NO2
NO2
o-NH2
2
14
p-OC6H5
p-N(CH3)2
m-N(CH3)2
p-NHC(O)CH3
m-NHC(O)CH3
o-NHC(O)CH3
p-NHC(O)C6H5
p-CH2NHC(O)C6H5
NO2
NO2
75
30
NO2
NO2
100, 58b
32
31
NO2
NO2
100, 100b
44
NO2
a Values are means of three experiments, standard deviation is less than 15%. Compounds screened at 50 lg/mL.
b % Inhibition at 25 lg/mL.
Table 2. In vitro activity of SoxS inhibitors
Table 3. In vitro IC50 (lM) values for 37 and 39 against several AraC
family members: MarA, SoxS, and Rob (E. coli), ExsA (P. aeruginosa),
Rma (S. typhimurium), and PqrA (P. mirabilis)
a
R
OH
N
O
O2N
N
H
Compound
IC50 (lM)
N
MarA
SoxS
Rob
ExsA
Rma
PqrA
34, 36-40
38
40
ndb
1.2
2.8
0.82
4.7
1.3
4.0
1.9
4.9
1.8
ndb
1.4
Compound
R
SoxS IC50 (lM)
Median MICb
a
(lg/mL)
S. aureusc E. colid
a Standard error <15% for all values.
b Not determined.
34
36
37
38
39
40
H
17
10
>64
>64
>64
>64
>64
>64
>64
>64
>64
>64
>64
>64
p-CH3
p-OCH3
p-F
2.5
2.8
3.4
Table 4. Efficacy in vivo for compounds 36 and 39 in a mouse model
of urinary tract infection
m-Cl
Compound
Log decrease CFU/g kidneya
p-N(CH3)2 0.82
37
40
1.0 (p < 0.05b)
0.5
a Standard error <15% for all values.
b Minimum inhibitory concentration (MIC) for bacterial growth.
c Antibiotic susceptible S. aureus RN450.
a CFU, colony forming units.
b Statistically significant by ANOVA with normal distribution of data.
d Antibiotic susceptible E. coli ML308-225.

T. E. Bowser et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5652–5655
5655
11. (a) George, A. M.; Levy, S. B. J. Bacteriol. 1983, 155, 531;
(b) George, A. M.; Levy, S. B. J. Bacteriol. 1983, 155, 541.
12. Casaz, P.; Garrity-Ryan, L. K.; McKenney, D.; Jackson,
C.; Levy, S. B.; Tanaka, S. K.; Alekshun, M. N.
Microbiology 2006, 152, 3643.
of identifying potent compounds with useful activity
against a range of pathogens.
Acknowledgments
13. Kwon, H. J.; Bennik, M. H.; Demple, B.; Ellenberger, T.
Nat. Struct. Biol. 2000, 7, 424.
We thank Pat Cannon, David McKenney, Jaime Quinn,
Caroline Jackson, and Janine Landry for expertise in
performing the in vivo experiments.
14. Rhee, S.; Martin, R. G.; Rosner, J. L.; Davies, D. R. Proc.
Natl. Acad. Sci. U.S.A. 1998, 95, 10413.
15. Dangi, B. P.; Pelupessey, R. G.; Martin, R. G.; Rosner, J.
L.; Louis, J. M.; Gronenborn, A. M. J. Mol. Biol. 2001,
314, 113.
16. Gilette, W. K.; Martin, R. G.; Rosner, J. L. J. Mol. Biol.
2000, 299, 1245.
References and notes
17. A quantitative chemiluminescence-based assay was used
to assay compounds for their ability to inhibit the binding
activity of Mar proteins to their cognate DNA. A
biotinylated double-stranded DNA target (2 nM) was
incubated with test compound and the Mar protein
(20 nM) fused to 6-histidine residues in a streptavidin-
coated microtiter plate. Compounds that inhibit the
binding of the Mar protein to DNA result in a loss of
protein from the plate after the first wash step and a
reduced luminescence signal. As further validation of the
specificity of the compounds for the Mar protein, select
test compounds were incubated with DNA and subjected
to agarose gel electrophoresis. None of the compounds
tested altered the structure of the test DNA in the assay.
This indicates that the test compounds are not non-specific
intercalators of DNA.
1. Alekshun, M. N.; Levy, S. B. Trends Microbiol. 1999, 7,
410.
2. Finlay, B. B.; Falkow, S. Microbiol. Mol. Biol. Rev. 1997,
61, 136.
3. Hauser, A. R.; Kang, P. J.; Engel, J. N. Mol. Microbiol.
1998, 27, 807.
4. Flashner, Y.; Mamroud, E.; Tidhar, A.; Ber, R.; Aftalion,
M.; Gur, D.; Lazar, S.; Zvi, A.; Bino, T.; Ariel, N.; Velan,
B.; Shafferman, A.; Cohen, S. Infect. Immun. 2004, 72,
908.
5. Featherston, J. D.; Bearden, S. W.; Perry, R. D. Mol.
Microbiol. 1996, 22, 315.
6. Champion, G. A.; Neely, M. N.; Brennan, M. A.; Dirita,
V. J. Mol. Microbiol. 1997, 23, 323.
7. Bieber, J.; Zhu, J.; Dziejman, M.; Faruque, S.; Calder-
wood, S.; Mekalanos, J. Science 1998, 280, 2114.
8. Frota, C. C.; Papavinasasundaram, K. G.; Davis, E. O.;
Colston, M. J. Infect. Immun. 2004, 72, 5483.
9. Hava, D.; Camilli, A. Mol. Microbiol. 2002, 45, 1389.
10. Randall, L. P.; Woodward, M. J. Med. Microbiol. 2001,
50, 770.
18. De Stevens, G.; Brown, A. B.; Rose, D.; Chernov, H. I.;
Plummer, A. J. J. Med. Chem. 1967, 10, 211.
19. Gardiner, J. M.; Procter, J. Tetrahedron Lett. 2001, 42,
5109.
20. Keane, W. F.; Freedman, L. R. Yale J. Biol. Med. 1967,
40, 231.