Small molecule inhibitors of bacterial transcription complex formation

Article abstract of DOI:10.1016/j.bmcl.2017.08.036

Knoevenagel condensation was employed to generate a set of molecules potentially capable of inhibiting the RNA polymerase-σ⁷⁰/σ^A interaction in bacteria. Synthesis was achieved via reactions between a variety of indole-7-carbaldehydes and rhodanine, N-allylrhodanine, barbituric acid or thiobarbituric acid. A library of structurally diverse compounds was examined by enzyme-linked immunosorbent assay (ELISA) to assess the inhibition of the targeted protein–protein interaction. Inhibition of bacterial growth was also evaluated using Bacillus subtilis and Escherichia coli cultures. The structure–activity relationship studies demonstrated the significance of particular structural features of the synthesized molecules for RNA polymerase-σ⁷⁰/σ^A interaction inhibition and antibacterial activity. Docking was investigated as an in silico method for the further development of the compounds.

Full text of DOI:10.1016/j.bmcl.2017.08.036

Accepted Manuscript
Small Molecule Inhibitors of Bacterial Transcription Complex Formation
Daniel S. Wenholz, Ming Zeng, Cong Ma, Marcin Mielczarek, Xiao Yang,
Mohan Bhadbhade, David St.C. Black, Peter J. Lewis, Renate Griffith, Naresh
Kumar
PII:
S0960-894X(17)30835-1
DOI:
http://dx.doi.org/10.1016/j.bmcl.2017.08.036
BMCL 25230
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
19 July 2017
16 August 2017
16 August 2017
Please cite this article as: Wenholz, D.S., Zeng, M., Ma, C., Mielczarek, M., Yang, X., Bhadbhade, M., Black, D.S.,
Lewis, P.J., Griffith, R., Kumar, N., Small Molecule Inhibitors of Bacterial Transcription Complex Formation,
Bioorganic & Medicinal Chemistry Letters (2017), doi: http://dx.doi.org/10.1016/j.bmcl.2017.08.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Small Molecule Inhibitors of Bacterial
Transcription Complex Formation
Daniel S. Wenholz a, Ming Zeng a, Cong Ma b,c, Marcin Mielczarek a, Xiao Yang b,c, Mohan
Bhadbhade a, David St. C. Black a, Peter J. Lewis b, Renate Griffith d, and Naresh Kumar a,b*
a School of Chemistry, UNSW Sydney, Kensington, NSW 2052, Australia
b School of Environmental and Life Sciences, University of Newcastle, Callaghan, NSW
2308, Australia
c Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic
University, Hong Kong
d School of Medical Sciences, UNSW Sydney, Kensington, NSW 2052, Australia
e Australian Centre for Nanomedicine, UNSW Sydney, Kensington, NSW 2052, Australia
These authors contributed equally.
ABSTRACT
Knoevenagel condensation was employed to generate a set of molecules potentially capable
of inhibiting the RNA polymerase-σ70/σA interaction in bacteria. Synthesis was achieved via
reactions between a variety of indole-7-carbaldehydes and rhodanine, N-allylrhodanine,
barbituric acid or thiobarbituric acid. A library of structurally diverse compounds was
examined by enzyme-linked immunosorbent assay (ELISA) to assess the inhibition of the
1

targeted protein-protein interaction. Inhibition of bacterial growth was also evaluated using
Bacillus subtilis and Escherichia coli cultures. The structure-activity relationship studies
demonstrated the significance of particular structural features of the synthesized molecules
for RNA polymerase-σ70/σA interaction inhibition and antibacterial activity. Docking was
investigated as an in silico method for the further development of the compounds.
Keywords: Antibacterial activity; Bacterial Transcription; RNA Polymerase; Structure-
activity relationship (SAR); Indole
The introduction of antibiotics to medicine has allowed the efficient treatment of bacterial
infections posing a substantial threat to human health and life.1 This revolutionary progress is
now under threat by the emergence of bacterial resistance to antibiotics due to their excessive
and often unjustified use.1-8 The rapidly developing antibiotic resistance problem among
highly pathogenic species is significantly escalated by continuously decreasing numbers of
new antibacterial drug candidates reaching the market.1, 3, 7-9 Novel antibacterial agents that can
undergo successful development towards efficient antibiotics are urgently needed in order to
counteract the dramatically intensifying bacterial resistance to currently used drugs.3, 9
A key priority of antibacterial drug research is the identification of molecules exhibiting
novel mechanisms of activity rather than the development of existing classes of antibiotics to
which resistance has already occurred.3, 9 Bacterial metabolic pathways, especially the gene
expression processes, have been exploited as attractive drug targets.10 One of the most
promising directions to explore is targeting bacterial transcription,10-11 the process of
transcribing genetic information encoded in DNA into RNA via the enzyme RNA polymerase
(RNAP).10 Two forms of DNA-dependent RNAP are present in bacterial cells: the core
enzyme and the holoenzyme (HE).12-16 The HE is formed between the RNAP core enzyme,
which consists of five subunits (α2ββ′ω), and a transcription factor, such as a σ factor, giving
the enzyme transcription specificity by allowing it to recognize promoter sites on the DNA.12-
2

17
Whilst a wide range of σ factors exist both within and between bacterial species, an
essential σ factor, σA in Gram-positive B. subtilis and σ70 in Gram-negative E. coli, is involved
in the transcription of ‘housekeeping’ genes.12, 17-18
A novel antibacterial strategy seeks to inhibit HE formation via disruption of the protein-
protein interaction (PPI) that occurs at the binding ‘hotspot’ between the highly conserved
solvent-exposed clamp helix (CH) region of the β′ subunit of the RNAP core and region 2.2
19
of the σ70/σA factor,12-13, and thus inhibit bacterial transcription. As the σ70/σA factors
demonstrate high structural similarity across a variety of bacteria species,10, 17 compounds that
inhibit the β′-CH-σ70/σA interaction are expected to show broad spectrum activity.
2.2
Additionally, since σ70/σA factors regulate transcription initiation only in bacterial cells,13
drugs targeting this PPI would be less likely to produce adverse effects.
Recently a series of benzoic acid-based molecules has been synthesized and evaluated for
transcription inhibitory activity, with compounds 1, 2, 3 and 4 (Figure 1) efficiently inhibiting
the interaction between core RNAP and σ70, exhibiting 85-100% transcription inhibition at 10
µM.20-21 Additionally, molecule 1 proved to be a good broad spectrum inhibitor of bacterial
growth showing MIC values of 12.5 and 6.25 µg/mLL againnst Staphylococcus epidermidis and
E. coli, respectively.20
3

Figure 1. Highly active synthetic compounds inhibiting transcription initiation in bacteria.20,
22-24
The pharmacophore model previously developed by our group based on a B. subtilis RNAP
homology model and the amino acids in σA found in mutagenesis studies to be responsible for
the RNAP HE formation19, 24 was used for the design of several classes of structurally related
indole-based inhibitors of the β′-CH-σA interaction such as bis-indole GKL003 5 (Figure
2.2
1).22-24 In our preliminary studies involving isothermal titration calorimetry experiments, the
bis-indoles were found to inhibit transcription by competitively binding to the β′-CH region
of the RNAP core enzyme.24 Through the use of ELISA GKL003 5 was determined to inhibit
the PPI between β′-CH and B. subtilis σA by 63% at 15 µM.25 Further development of this
series led to molecules 6 and 7 (Figure 1) as the most potent mono- and bis-indole-based
inhibitors, with of 86% and 60% inhibition of the β′-CH-σA PPI at 15 µM as measured by
ELISA, respectively, and they inhibited E. coli growth at 200 µM (16% and 21% inhibition,
respectively).22-23
4

The structure-activity relationship (SAR) studies performed on a large family of
compounds related to 1-4 showed that incorporation of a carboxylic acid group significantly
increased their inhibitory activity against the β′-CH-σ70 interaction.20 Moreover, the
allylrhodanine moiety was found to be essential for antibacterial activity.20 These
observations were used to design a series of mono-indole compounds based upon the
structures of 6 and 7 that incorporated features of 1-4. A focus was placed on including
compounds with lower molecular weight and increased polarity in order to overcome the
25
solubility issues encountered by the previous compounds.22-23, These compounds were
synthesized and evaluated for their ability to inhibit the interaction between the β′-CH region
of RNAP core and σA factor by ELISA, and for antibacterial activity against B. subtilis and E.
coli. Biological evaluation was followed by molecular docking and SAR studies.
The indoles 8a and 8b were obtained following the well-established Bischler indole
synthesis modified by our group.26-27 The methyl 4,6-dimethoxyindole-2-carboxylate 9a was
afforded via the Hemetsberger indole synthesis.28 The indoles 8a, 8b and 9a were formylated
under Vilsmeier-Haack conditions at position 7 (Scheme 1) to give the corresponding indole-
7-carbaldehydes 12a-c in yields of 85-95%.29 The methyl ester 12c was also converted into
the carboxylic acid 13 via alkaline hydrolysis in 95% yield (Scheme 1).
In order to synthesize the benzyl ester 15, the methyl indole-2-carboxylate 9a was first
hydrolyzed to the indole-2-carboxylic acid 9b30 using a solution of potassium hydroxide in an
ethanol/water mixture, followed by DCC-mediated coupling reaction between the acid 9b and
benzyl alcohol in the presence of DMAP as a catalyst to give the benzyl indole-2-carboxylate
14 in 35% yield. Subsequent Vilsmeier-Haack formylation of the ester 14 furnished indole-7-
carbaldehyde 15 in 90% yield (Scheme 1).
Scheme 1. Synthesis of compounds 9b, 12-15.
5

(a) POCl3 (1.0 equiv), DMF, 0 °C • room temp., 2 h, 85-95%; (b) KOH (excess), EtOH/H2O
3:1 (v/v), reflux, 2 h, 95-99%; (c) BnOH (1.2 equiv), DCC (3.0 equiv), DMAP (0.2 equiv),
DCM, 0 °C • room temp., 3 h, 35%; (d) POCl3 (1.1 equiv), DMF, 0 °C • room temp., 12 h,
99%.
The reaction between indole-7-carbaldehyde 12a and rhodanine 16 was carried out in a 1:1
molar ratio in the presence of piperidine as a catalyst by heating in anhydrous ethanol under
reflux. However, as indicated by TLC analysis, even after 72 h the reaction did not proceed to
completion. In order to shift the reaction equilibrium towards the formation of the product
10a, the amount of rhodanine was increased to 1.2 equivalents along with the addition of β-
alanine (2 equiv) and glacial acetic acid as a solvent.31 TLC analysis showed that the starting
indole-7-carbaldehyde 12a had been fully consumed after 3 h of heating under reflux. As the
product of the reaction was found to be very insoluble in the majority of commonly used
solvents such as EtOAc, EtOH or DCM, column chromatography was excluded as a potential
method for its purification. Therefore, a number of solvent mixtures were tested to
recrystallize the crude product. A DMF/MeOH mixture was finally used for recrystallization
to give the compound 10a as a dark red powder in 60% yield (Scheme 2). This method was
applied to the reactions of the indole-7-carbaldehydes 12b, 12c, 13 and 15 with rhodanine 16
or allylrhodanine 17 to afford the desired products 10a-i in 30-68% yield (Scheme 2). The 1H
6

NMR spectra of the products display a characteristic singlet at around 7.80-8.50 ppm which
indicates the successful formation of the double bond linker.
The single crystal X-ray crystallographic analyses confirmed (Z)-stereochemistry of the
molecules 10c and 10d (Figure 2). This observation is in a good agreement with literature
data.32-33
The Knoevenagel condensations between indole-7-carbaldehydes 12a-c or 13 and
barbituric acid 18 or thiobarbituric acid 19 resulted in the formation of the compounds 11a-h
as dark red powders in 27-100% yield after recrystallization from a DMF/MeOH mixture
(Scheme 2).34 The reactions were found to proceed efficiently by heating the substrates in
either an acetic anhydride/acetic acid 1:10 mixture or in acetic anhydride under reflux for 2 h.
Furthermore, the use of a catalyst was not crucial for these reactions.
7

Scheme 2. Synthesis of compounds 10a-i and 11a-h.
(a) β-Alanine (2 equiv), CH3CO2H, 100 °C, 3 h (30-68%); (b) acetic anhydride or acetic
anhydride and CH3CO2H, 90 °C, 2 h (27-100%).
8

A
B
Figure 2. ORTEP diagrams of the crystal structures of the compounds 10c (A) and 10d (B)
(front view, 50% probability thermal ellipsoids at 100 K shown).
The library of the seventeen synthesized molecules was evaluated for biological activity.
Firstly, the inhibition of the interaction between B. subtilis σA and the β′-CH region of the
core RNAP was examined by ELISA at 15 µM compound concentration and is expressed as a
percentage of the positive control (interaction between σA and β′-CH in the absence of the
inhibitor) where the positive control interaction is 100%. Secondly, bacterial growth
9

inhibition was evaluated at 0.2 mM compound concentration using two representative
species, B. subtilis (Gram-positive) and E. coli (Gram-negative), and is expressed as a
percentage of the positive control (bacterial growth in the absence of the inhibitor) where the
positive control bacterial growth is 100% (Table 1).22-25
Table 1. Evaluation of antibacterial activity of the synthesized molecules.
β′-CH-σA
interaction
inhibition at
15 µM by
ELISA [%]
Growth
inhibition of
E. coli at
Growth
inhibition of B.
subtilis at 0.2
mM [%]
Entry clog
P
MW
0.2 mM [%]
10a
10b
10c
10d
10e
10f
10g
10h
10i
11a
11b
11c
11d
11e
11f
5.3
6.1
4.6
5.4
3.3
4.1
3.0
3.9
5.7
3.3
4.2
2.6
3.5
1.3
2.2
1.1
2.0
5.2
430.9
471.0
396.5
436.5
378.4
418.5
364.4
404.5
494.6
425.8
441.9
391.4
407.4
373.3
389.4
359.3
375.4
447.5
21
34
17
0
15
14
45a
66
28
79b
26
30
28
30a
24
0
0
8
44a
44
45
22
0
3
46
36
74
63
15
14
45
14
28
17
28
71
21
21
6
41a
0
83
0
0
20
57
0
0
53
89b
91b
0
11g
11h
12
0
0
a precipitation at 0.2 mM, b affects the exponential phase of bacterial growth.
The top three hits (10h, 10i and 11h) exhibited greater than 60% inhibition of the β′-CH-σA
interaction in the ELISA assay at 15 µM (74%%, 63% andd 71%, resspectiveely). Howwever, 11h
was found to be inactive against both E. coli and B. subtilis, whilst 10h and 10i showed
moderate and low antibacterial activity.
Although compound 10f possessed only moderate β′-CH-σA inhibition of 46 % at 15 µM, it
showed good antibacterial activity against E. coli, with growth inhibition of 79% and
moderate activity against B. subtilis, showing potential for further development as an
antibacterial agent, particularly against Gram-negative bacteria. Similarly, compound 11c
displayed only moderate inhibition of the PPI but inhibited the growth of B. subtilis (83% at
10

0.2 mM), but showed no effect on E. coli growth. The molecules with high antibacterial
activity, but very low RNAP-σ inhibitory activity, such as 11e and 11f (89% and 91%
inhibition of B. subtilis growth at 0.2 mM, respectively) might exhibit antibacterial activity
via additional mechanisms different from inhibition of the β′-CH-σA interaction. However,
compounds 10f, 11e and 11f were observed to affect the exponential phase of bacterial,
growth and this suggests that they inhibit transcription.24
To investigate potential problems of the compounds in reaching their targets, calculated
logP values and molecular weights are provided in Table 1. Both values are high for 10i and
10b, possibly explaining the low growth inhibition observed, however, this is not the case for
10g, 10h, 11g and 11h, all of which however, contain ionizable COOH groups that may
prevent efficient uptake by the bacteria. Compound 10d is very lipophilic, and this combined
with the fact that it does not inhibit the PPI, suggests a different mechanism of growth
inhibition. The compounds all incorporate substructures similar to those of the literature
compounds as exemplified by 1-4. These exhibited antibacterial activity against a mutated E.
coli strain deficient in multidrug efflux systems, but not against wild-type E. coli.20, 35 They
have also been shown to have off-target effects.35 These issues could be related to the
presence of common interference substructures (PAINS)36 and will need to be investigated
further in the development of the compounds presented here. However, we have previously
not observed any interference in several detailed assays of compound 5 at different
concentrations.24
The synthesized compounds, 10a-i and 11a-k, were docked using Genetic Optimized
Ligand Docking (GOLD) through the Discovery Studio (DS) interface onto the β′-CH region
of a new RNAP homology model. This was repeated with the most active compound from the
SB series, 1, as well as the three active compounds from the GKL series, 5, 6 and 7.20, 22-23, 25
Compounds with carboxylic acid groups were also docked in their negatively ionized
11

deprotonated state. As the solvent exposed residue side chains are not static in the core
enzyme, it was deemed to be important to allow flexibility in the amino acid side chains in
and around the binding hotspot area.37 The side chains of the residues in the PPI hotspot on β′-
CH24 were set to flexible in the docking protocol, allowing transient pockets or grooves to
appear in the relatively flat β′-CH surface to facilitate greater interaction with small
molecules.
12

A
B
C
D
Figure 3. Positions and intermolecular interactions of σA and docked compounds at the β′-
2.2
CH binding hotspot24 of a B. subtilis RNAP holoenzyme homology model. Hotspot residues
coloured in cyan, a critical Arg267 residue in green, and hydrogen bond interactions are
shown as magenta dashed lines. (A) σA 2.2 region in homology model of HE. (B) Docked
1(1). (C) Docked 10h(1). (D) Docked 10i. (1) indicates deprotonated carboxylic acid group.
13

Table 3. Docking results for synthesized compounds and known inhibitors.
Carboxylic
Acid Group
Entry
Docking GOLD
Score
State
-
-
-
-
10a
10b
10c
10d
10e
10f
10g
10g(1)
10h
10h(1)
10i
49.2
51.8
52.1
45.3
50.7
54.2
55.5
66.2
54.8
71.2
63.6
50.4
42.7
45.7
47.3
51.0
51.4
50.8
47.8
62.0
51.2
70.7
59.6
67.6
74.1
56.3
68.0
-
-
COOH
COO-
COOH
COO-
-
-
-
-
-
-
-
-
10j
11a
11b
11c
11d
11e
11f
11g
11g(1)
11h
11h(1)
1
COOH
COO-
COOH
COO-
COOH
COO-
-
1(1)
5
6
-
-
7
It was observed that the majority of the docked compounds were placed by GOLD into a
groove running between Arg264 and Arg267 on the β′-CH region, as shown in Figure 3.
Although this position does not cover the entire hotspot area, it does allow strong interactions
with the three arginine residue that are most important in the PPI, namely Arg264, Arg267
and Arg270 as identified from mutagenesis data.24 This position also has great potential to
interrupt interactions with the σA-Glu277 residue, which has been identified as playing an
important role in the PPI as part of the σA2.2 hotspot.19
14

Figure 4. Relationship between docking GOLD scores and β′-CH-σA interaction inhibition of
synthesized compounds 10a-i and 11a-h along with known active inhibitors 5-7.22-24
The compounds were mapped to the published pharmacophore from our group24 (data not
shown), however, no level of correlation with the biological data was observed. In
comparison to the pharmacophore mapping results, there was good correlation between the
β′-CH-σA inhibition activities and the highest GOLD score of each compound (R2 = 0.64),
taking into account the structural diversity of the compounds (Figure 4). Furthermore, the two
compounds with the highest β′-CH-σA inhibitory activities, 10h and 11h, achieved the two
highest GOLD scores of 71.2 and 70.7, respectively. The high GOLD scores observed also
suggest that the ELISA results were likely due to strong interaction with the β′-CH region,
rather than being artifacts of interference compounds. Although the high GOLD scores for
10h and 11h were only obtained by the ionised forms of both compounds, the non-ionisable
compound with the highest β′-CH-σA inhibition, 10i, also achieved the best docking result of
all neutral synthesized compounds with a GOLD score of 63.6. This trend was also seen with
the known inhibitors, 1, 5 and 7, the docking results of which all showed GOLD scores close
to 70. With a GOLD score of 56.3, mono-indole 6 scored the lowest out of the known
15

inhibitors, consistent with it exhibiting the lowest β′-CH-σA inhibition of the four.23 These
results show that docking shows potential as a tool in the future development of RNAP HE
formation inhibitors.
The β′-CH-σA inhibitory activities and antibacterial activities along with the docking results
of the synthesized molecules were also analyzed for possible structure-activity relationships.
It was observed that the replacement of the 3-phenyl and 3-(4′-chlorophenyl) group with a
carboxylic acid or ester at the 2-position of the indole backbone led to greatly increased
ability to inhibit the β′-CH-σA interaction, as can be seen in a comparison between the β′-CH-
σA inhibitory activities of 11b and 11d with 11h. This was further supported by the predicted
binding modes from the docking analysis, wherein the carboxylic acid and ester groups at the
2-position were predicted to form hydrogen bonds with Arg267 and Arg270 (Figure 3),
whilst no interaction was predicted for the phenyl groups at the 3-position with the protein.
This was consistent with the fact that both the 3-phenyl group and the 3-(4′-chlorophenyl)
group were unable to map to any feature on the pharmacophore. These observations show
that he 3-phenyl group in not essential for the antibacterial activity of the compounds.
It was also observed that a carboxylic acid group at the 2-position of the indole system
produced much higher β′-CH-σA inhibitory activities than those observed for the esters at this
position. This is apparent when comparing the β′-CH-σA inhibitory activities of 10f with 10j
(46% and 74%, respectively) or 11f with 11h (17% and 71%, respectively). This may be due
to the carboxylic acid group’s ability to ionize and form charge-charge interactions with the
arginine groups of the binding hotspot, typically Arg270, as suggested by the docking results.
However, the incorporation of a CO2Me group at the 2-position improved antibacterial
activity. This is shown in the comparison of compounds 10h and 10f, where inhibition of the
E. coli growth increased from 30% to 79% or the compounds 10i and 10e, where inhibition of
the B. subtilis growth increased from 6% to 45% when the CO2H group was replaced by the
16

CO2Me group. Moreover, such a replacement resulted in a dramatic increase in antibacterial
activity against both E. coli and B. subtilis for the molecules 11g and 11e, or 11h and 11f.
Notably, the CO2Me group was present in all the inhibitors that affected the exponential
phase of bacterial growth (10f, 11e and 11f). This may be due to the increase in clogP of the
esters, allowing greater passive diffusion through the bacterial cell membrane.
It was observed that the RNAP-σ inhibitory activity of the esters could be improved by
replacing the CO2Me group with a larger CO2Bn group, as shown via comparison of 10f and
10i. This change increased the β′-CH-σA inhibitory activity from 46% to 63%, which may be
the result of an increase in hydrophobic interactions. This is in agreement with the docking
results, which show the benzyl ring interacting with Met287 with a different predicted mode
of binding to the other esters (Figure 3). However, this modification resulted in a decrease in
antibacterial activity, possibly due to the increase in molecular weight resulting in a reduction
in passive transport through bacterial membranes. This result highlights a potential future
direction in modification of the ester group with other less bulky, less hydrophobic groups.
Comparison of the activities of the molecules 10g, 10h, 11g and 11h indicates that the
allylrhodanine (present in the molecule 10h) is most beneficial for RNAP-σ inhibitory
activity among the 5- and 6-membered heterocyclic rings. The N-allyl group of
allylrhodanine 17 is important not only for RNAP-σ inhibitory activity, but also for
antibacterial activity as highlighted in the case of the compounds 10e and 10f.
In conclusion, 17 novel compounds were successfully synthesized utilizing Knoevenagel
condensation. The ability of the compounds to inhibit the interaction between σA2.2 and the β′-
CH region of the core RNAP was evaluated by ELISA at 15 µM. AAntibaccterial aactivitty of the
molecules was determined based on the inhibition of E. coli and B. subtilis growth in culture
at 0.2 mM. The compounds were then docked onto the solvent exposed face of the β′-CH
region of a B. subtilis σA homology model using GOLD. The SAR studies supported by the
17

molecular docking resulted in the identification of molecular features potentially beneficial
for inhibition of the β′-CH-σA interaction and bacterial growth. Modifications of the ester
group at the C2 position of the indole are a promising direction for the design of compounds
that exhibit high inhibition of both the RNAP-σ assembly formation and bacterial growth. In
particular, a methyl group could be replaced with ethyl, n-propyl or iso-propyl groups as the
larger ester groups could enhance RNAP-σ inhibitory activity as observed for the benzyl
ester. As confirmed by calculation of molecular properties and much fewer solubility-related
issues encountered during the examination of antibacterial activity, this research has
delivered compounds with significantly improved aqueous solubility compared to the
previously synthesized bis-indoles22, 25 and mono-indoles23. This is mainly due to reduction of
molecular weight and size, and increase in hydrophilicity by the incorporation of polar groups
such as the carboxylic group. Furthermore, this study has demonstrated that in silico docking
using GOLD shows potential as a tool for the further development of RNAP-σ inhibitors and
future studies may benefit immensely through implementation of this technique during lead
development.
Acknowledgments
This work was supported by a Project Grant from the National Health and Medical
Research Council (NHMRC) Australia to N.K., P.L. and R.G. [NHMRC Grant
APP1008014]. M.M. is thankful to UNSW Australia for a Tuition Fee Scholarship (TFS) and
to N.K. for a Living Allowance Scholarship. We would like to give a special thank you to
Thanh Le, Oscar Thach, Samuel K. Kutty and Murat Bingul for their aid in this study. We are
also thankful to the Nuclear Magnetic Resonance Facility and the Bioanalytical Mass
Spectrometry Facility (BMSF) in the Mark Wainwright Analytical Centre at UNSW
Australia for the opportunity of using their instruments.
18

Abbreviations Used
CH, clamp helix; DOPE, Discrete Optimized Protein Energy; DS, Discovery Studio;
GOLD, Genetic Optimized Ligand Docking; HE, holoenzyme; PPI, protein-protein
interaction; PDF, probability density function; RMSD, root mean-square-deviation; RNAP,
RNA polymerase.
19

REFERENCES
1. Hart, C., Antibiotic resistance: an increasing problem? Br. Med. J. 1998, 316 (7140),
1255-1257.
2. Yoshikawa, T. T., Antimicrobial resistance and aging: beginning of the end of the
antibiotic era? J. Am. Geriatr. Soc. 2002, 50 (s7), 226-229.
3. Shekhar, C., Bacteria: Drug Resistance Spreads, but Few New Drugs Emerge. Chem.
Biol. 2010, 17 (5), 413-414.
4. Neu, H. C., The crisis in antibiotic resistance. Science 1992, 257 (5073), 1064-1074.
5. Mazel, D.; Davies, J., Antibiotic resistance in microbes. Cell. Mol. Life Sci. 1999, 56
(9), 742-754.
6. McAdam, A. J.; Hooper, D. C.; DeMaria, A.; Limbago, B. M.; O'Brien, T. F.;
McCaughey, B., Antibiotic resistance: how serious is the problem, and what can be done?
Clin. Chem. 2012, 58 (8), 1182-1186.
7. Peleg, A. Y.; Hooper, D. C., Hospital-acquired infections due to gram-negative
bacteria. N. Engl. J. Med. 2010, 362 (19), 1804-1813.
8. Boucher, H. W.; Talbot, G. H.; Bradley, J. S.; Edwards, J. E.; Gilbert, D.; Rice, L. B.;
Scheld, M.; Spellberg, B.; Bartlett, J., Bad bugs, no drugs: no ESKAPE! An update from the
Infectious Diseases Society of America. Clin. Infect. Dis. 2009, 48 (1), 1-12.
9. Chopra, I., The 2012 Garrod Lecture: Discovery of antibacterial drugs in the 21st
century. J. Antimicrob. Chemother. 2012, 68, 496-505.
10. Villain-Guillot, P.; Bastide, L.; Gualtieri, M.; Leonetti, J.-P., Progress in targeting
bacterial transcription. Drug Discov. Today 2007, 12 (5), 200-208.
20

11. Ma, C.; Yang, X.; Lewis, P. J., Bacterial Transcription as a Target for Antibacterial
Drug Development. Microbiol. Mol. Biol. Rev. 2016, 80 (1), 139-160.
12. Vassylyev, D. G.; Sekine, S.-i.; Laptenko, O.; Lee, J.; Vassylyeva, M. N.; Borukhov,
S.; Yokoyama, S., Crystal structure of a bacterial RNA polymerase holoenzyme at 2.6 Å
resolution. Nature 2002, 417 (6890), 712-719.
13. Burgess, R. R.; Anthony, L., How sigma docks to RNA polymerase and what sigma
does. Curr. Opin. Microbiol. 2001, 4 (2), 126-131.
14. Finn, R. D.; Orlova, E. V.; Gowen, B.; Buck, M.; van Heel, M., Escherichia coli RNA
polymerase core and holoenzyme structures. EMBO J. 2000, 19 (24), 6833-6844.
15. Patikoglou, G. A.; Westblade, L. F.; Campbell, E. A.; Lamour, V.; Lane, W. J.; Darst,
S. A., Crystal structure of the Escherichia coli regulator of 70, Rsd, in complex with 70
domain 4. J. Mol. Biol. 2007, 372 (3), 649-659.
16. Mekler, V.; Pavlova, O.; Severinov, K., Interaction of Escherichia coli RNA
polymerase 70 subunit with promoter elements in the context of free 70, RNA polymerase
holoenzyme, and the •- 70 complex. J. Biol. Chem. 2011, 286 (1), 270-279.
17. Malhotra, A.; Severinova, E.; Darst, S. A., Crystal structure of a 70 subunit fragment
from E. coli RNA polymerase. Cell 1996, 87 (1), 127-136.
18. Blanco, A. G.; Canals, A.; Bernués, J.; Solà, M.; Coll, M., The structure of a
transcription activation subcomplex reveals how 70 is recruited to PhoB promoters. EMBO
J. 2011, 30 (18), 3776-3785.
19. Johnston, E. B.; Lewis, P. J.; Griffith, R., The interaction of Bacillus subtilis A with
RNA polymerase. Protein Sci. 2009, 18 (11), 2287-2297.
21

20. Villain-Guillot, P.; Gualtieri, M.; Bastide, L.; Roquet, F.; Martinez, J.; Amblard, M.;
Pugniere, M.; Leonetti, J.-P., Structure-activity relationships of phenyl-furanyl-rhodanines as
inhibitors of RNA polymerase with antibacterial activity on biofilms. J. Med. Chem. 2007, 50
(17), 4195-4204.
21. André, E.; Bastide, L.; Michaux-Charachon, S.; Gouby, A.; Villain-Guillot, P.;
Latouche, J.; Bouchet, A.; Gualtiéri, M.; Leonetti, J.-P., Novel synthetic molecules targeting
the bacterial RNA polymerase assembly. J. Antimicrob. Chemother. 2006, 57 (2), 245-251.
22. Mielczarek, M.; Devakaram, R. V.; Ma, C.; Yang, X.; Kandemir, H.; Purwono, B.;
Black, D. StC.; Griffith, R.; Lewis, P. J.; Kumar, N., Synthesis and biological activity of
novel bis-indole inhibitors of bacterial transcription initiation complex formation. Org.
Biomol. Chem. 2014, 12 (18), 2882-2894.
23. Mielczarek, M.; Thomas, R. V.; Ma, C.; Kandemir, H.; Yang, X.; Bhadbhade, M.;
Black, D. StC.; Griffith, R.; Lewis, P. J.; Kumar, N., Synthesis and biological activity of
novel mono-indole and mono-benzofuran inhibitors of bacterial transcription initiation
complex formation. Bioorg. Med. Chem. 2015, 23 (8), 1763-1775.
24. Ma, C.; Yang, X.; Kandemir, H.; Mielczarek, M.; Johnston, E. B.; Griffith, R.;
Kumar, N.; Lewis, P. J., Inhibitors of Bacterial Transcription Initiation Complex Formation.
ACS Chem. Biol. 2013, 8 (9), 1972-1980.
25. Kandemir, H.; Ma, C.; Kutty, S. K.; Black, D. StC.; Griffith, R.; Lewis, P. J.; Kumar,
N., Synthesis and biological evaluation of 2, 5-di (7-indolyl)-1, 3, 4-oxadiazoles, and 2-and
7-indolyl 2-(1, 3, 4-thiadiazolyl) ketones as antimicrobials. Bioorg. Med. Chem. 2014, 22 (5),
1672-1679.
22

26. Pchalek, K.; Jones, A. W.; Wekking, M. M.; Black, D. StC., Synthesis of activated 3-
substituted indoles: an optimised one-pot procedure. Tetrahedron 2005, 61 (1), 77-82.
27. Mitchell, P. S.; Sengul, I. F.; Kandemir, H.; Nugent, S. J.; Chen, R.; Bowyer, P. K.;
Kumar, N.; Black, D. StC., Bromination of 4, 6-dimethoxyindoles. Tetrahedron 2012, 68
(39), 8163-8171.
28. Bingul, M.; Cheung, B. B.; Kumar, N.; Black, D. StC., Synthesis of symmetrical and
unsymmetrical diindolylmethanes via acid-catalysed electrophilic substitution reactions.
Tetrahedron 2014, 70 (40), 7363-7369.
29. Black, D. StC.; Bowyer, M.; KUMAR, N.; Mitchell, P., Calix [3] indoles, new
macrocyclic tris (indolylmethylene) compounds with 2, 7-linkages. J. Chem. Soc. Chem.
Commun. 1993, (10), 819-821.
30. Black, D. StC.; Kumar, N.; Wong, L., Synthesis of 4, 6-dimethoxyindoles. Aust. J.
Chem. 1986, 39 (1), 15-20.
31. Mori, M.; Moraca, F.; Deodato, D.; Ferraris, D. M.; Selchow, P.; Sander, P.; Rizzi,
M.; Botta, M., Discovery of the first potent and selective Mycobacterium tuberculosis Zmp1
inhibitor. Bioorg. Med. Chem. Lett. 2014, 24 (11), 2508-2511.
32. Zvezdanovic, J.; Daskalova, L.; Yancheva, D.; Cvetkovic, D.; Markovic, D.;
Anderluh, M.; Smelcerovic, A., 2-Amino-5-alkyllidenethiazol-4-ones as promising lipid
peroxidation inhibitors. Monatsh. Chem. 2014, 145, 945-952.
33. Mukhopadhyay, C.; Ray, S., Rapid and straightforward one-pot expeditious synthesis
of 2-amino-5-alkylidene-thiazol-4-ones at room temperature. Tetrahedron Lett. 2011, 52
(48), 6431-6438.
23

34. Klikar, M.; Bureš, F.; Pytela, O.; Mikysek, T.; Pad lková, Z.; Barsella, A.; Dorkenoo,
K.; Achelle, S., N, N•-Dibutylbarbituric acid as an acceptor moiety in push–pull
chromophores. New J. Chem. 2013, 37 (12), 4230-4240.
35. Mariner, K. R.; Trowbridge, R.; Agarwal, A. K.; Miller, K.; O'Neill, A. J.; Fishwick,
C. W.; Chopra, I., Furanyl-rhodanines are unattractive drug candidates for development as
inhibitors of bacterial RNA polymerase. Antimicrob. Agents Chemother. 2010, 54 (10), 4506-
4509.
36. Baell, J. B.; Holloway, G. A., New substructure filters for removal of pan assay
interference compounds (PAINS) from screening libraries and for their exclusion in
bioassays. J. Med. Chem. 2010, 53 (7), 2719-2740.
37. Wells, J. A.; McClendon, C. L., Reaching for high-hanging fruit in drug discovery at
protein–protein interfaces. Nature 2007, 450 (7172), 1001-1009.
38. Bruker SADABS, Bruker AXS Inc.: Madison, Wisconsin, USA, 2001.
39. Bruker APEX2 and SAINT, Bruker AXS Inc.: Madison, Wisconsin, USA, 2007.
40. Sheldrick, G. M., Acta Crystallogr. 2008, (A64), 112.
41. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.; Puschmann, H.,
OLEX2: a complete structure solution, refinement and analysis program. J. Appl. Crystallogr.
2009, 42 (2), 339-341.
42. Kabsch, W., Automatic processing of rotation diffraction data from crystals of
initially unknown symmetry and cell constants. J. Appl. Crystallogr. 1993, 26 (6), 795-800.
43. Sali, A.; Blundell, T., Comparative protein modelling by satisfaction of spatial
restraints. Protein structure by distance analysis 1994, 64, C86.
24

44. Chen, J.; Campbell, A. P.; Urmi, K. F.; Wakelin, L. P.; Denny, W. A.; Griffith, R.;
Finch, A. M., Human 1-adrenoceptor subtype selectivity of substituted homobivalent 4-
aminoquinolines. Bioorg. Med. Chem. 2014, 22 (21), 5910-5916.
45. MacDougall, I. J.; Griffith, R., Selective pharmacophore design for 1-adrenoceptor
subtypes. J. Mol. Graph. Model. 2006, 25 (1), 146-157.
GRAPHIC ABSTRACT
74%
inhibition
A
of `-CH-
interaction at
15 µM
25