Identification of the anti-mycobacterial functional properties of piperidinol derivatives

Article abstract of DOI:10.1111/bph.13744

Background and Purpose: Tuberculosis (TB) remains a major global health threat and is now the leading cause of death from a single infectious agent worldwide. The current TB drug regimen is inadequate, and new anti-tubercular agents are urgently required to be able to successfully combat the increasing prevalence of drug-resistant TB. The purpose of this study was to investigate a piperidinol compound derivative that is highly active against the Mycobacterium tuberculosis bacillus. Experimental Approach: The antibacterial properties of the piperidinol compound and its corresponding bis-Mannich base analogue were evaluated against M. smegmatis and Gram-negative organisms. Cytotoxicity studies were undertaken in order to determine the selectivity index for these compounds. Spontaneous resistant mutants of M.?smegmatis were generated against the piperidinol and corresponding bis-Mannich base lead derivatives and whole genome sequencing employed to determine the genetic modifications that lead to selection pressure in the presence of these compounds. Key Results: The piperidinol and the bis-Mannich base analogue were found to be selective for mycobacteria and rapidly kill this organism with a cytotoxicity selectivity index for mycobacteria of >30-fold. Whole genome sequencing of M.?smegmatis strains resistant to the lead compounds led to the identification of a number of single nucleotide polymorphisms indicating multiple targets. Conclusion and Implications: Our results indicate that the piperidinol moiety represents an attractive compound class in the pursuit of novel anti-tubercular agents. Linked Articles: This article is part of a themed section on Drug Metabolism and Antibiotic Resistance in Micro-organisms. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v174.14/issuetoc.

Full text of DOI:10.1111/bph.13744

Identification of the anti-mycobacterial functional properties of piperidinol
derivatives
1
3
2
2
Collette S. Guy1,2, Esther Tichauer , Gemma L. Kay , Daniel J. Phillips , Trisha L. Bailey ,
James Harrison1, Christopher M. Furze1, Andrew D. Millard3, Matthew I. Gibson2,3, Mark J.
Pallen3, Elizabeth Fullam1*

1) School of Life Sciences, University of Warwick, Coventry, CV4 7AL, United Kingdom
2) Department of Chemistry, University of Warwick, Coventry, CV4 7AL, United Kingdom
3) Medical School, University of Warwick, Coventry, CV4 7AL, United Kingdom
These authors contributed equally
* Correspondence: Elizabeth Fullam, School of Life Sciences, University of Warwick,
Coventry, CV4 7AL, UK. E-mail: e.fullam@warwick.ac.uk
Running title: piperidinol mycobacterial investigation
Key words: drug discovery, antimicrobial, target identification, chemical bioinformatics,
tuberculosis, Mycobacterium tuberculosis
Supporting information is included.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/bph.13744
This article is protected by copyright. All rights reserved.

Abstract
Background and purpose: Tuberculosis (TB) remains a major global health threat and is
now the leading cause of death from a single infectious agent worldwide. The current TB
drug regimen is inadequate and new anti-tubercular agents are urgently required to be able to
successfully combat the increasing prevalence of drug-resistant TB. The purpose of this study
was to investigate a piperidinol compound derivative that is highly active against the
Mycobacterium tuberculosis bacillus.
Experimental approach: The antibacterial properties of the piperidinol compound and its
corresponding bis-Mannich base analogue were evaluated against Mycobacterium smegmatis
and Gram-negative organisms. Cytotoxicity studies were undertaken in order to determine the
selectivity index for these compounds. Spontaneous resistant mutants of Mycobacterium
smegmatis were generated against the piperidinol and corresponding bis-Mannich base lead
derivatives and whole genome sequencing employed to determine genetic modifications as
result to selection pressure in the presence of these compounds.
Key results: The piperidinol and the bis-Mannich base analogue were found to be selective
for mycobacteria and rapidly kill this organism with a cytotoxicity selectivity index for
mycobacteria of >30 fold. Whole genome sequencing of M. smegmatis strains resistant to the
lead compounds led to the identification of a number of single nucleotide polymorphisms
(SNPs) indicating multiple-targets.
Conclusion and implications: Our results indicate that the piperidinol moiety represents an
attractive compound class in the pursuit of novel anti-tubercular agents.
This article is protected by copyright. All rights reserved.

Abbreviations:
CFU: colony forming units; Clint: intrinsic clearance; E. coli: Eschericia coli; Fractional
inhibitory concentration (FIC); HIV: human immunodeficiency virus; INH: isoniazid;
INDEL: insertion/deletion of bases; Luria-Bertani (LB); MIC: minimum inhibitory
concentration; MBC: minimum bactericidal concentration; M. bovis BCG: Mycobacterium
bovis bacillus Calmette-Guérin; M. smegmatis: Mycobacterium smegmatis; M. tuberculosis:
Mycobacterium tuberculosis; NAT: Arylamine N-acetyltransferase; P. putida: Pseudomonas
putida; REMA: resazurin microtiter assay; RIF: rifampicin; SET solution: sucrose-EDTA-
Tris-HCl; SNP: single nucleotide polymorphism; TB: Tuberculosis; TE buffer: Tris-EDTA;
WT: wild-type
This article is protected by copyright. All rights reserved.

Tables of Links
TARGETS
Enzymesa
Arylamine N-acetyltransferase
LIGANDS
Piperidinol (compound 1)
SMILE:
CN1CCC(C2=CC=CC=C2)(O)C(C(C3=CC=CC=C3)=O)C1
PUBCHEM CID:20784
Bis-Mannich base (compound 2)
SMILE:
O=C(CCN(C)CCC(C1=CC=CC=C1)=O)C2=CC=CC=C2
PUBCHEM CID: 408365
These Tables list key protein targets and ligands in this article which are hyperlinked to
corresponding entries in http://www.guidetopharmacology.org, the common portal for data
from the IUPHAR/BPS Guide to PHARMACOLOGY (Southan et al., 2016) and are
permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 (Alexander et
al., 2015)
This article is protected by copyright. All rights reserved.

Introduction
Tuberculosis (TB), caused by the bacillus Mycobacterium tuberculosis, is a major global
pathogen and has now surpassed human immunodeficiency virus (HIV) to be the leading
cause of death from a single infectious agent (WHO, 2015). In 2014, 9 million new cases of
active TB were reported and 1.5 million deaths resulted from TB infection, with a significant
number of deaths occurring in individuals who are co-infected with HIV and TB (WHO,
2015, Dye and Williams, 2010). The current drug-regimen to treat TB is over 40 years old
and is hindered by its long duration, which is typically 6-9 months in the case of drug-
sensitive strains and frequently accompanied by toxic side-effects (Zumla et al., 2015). A
variety of reasons, including patient non-compliance, have led to the emergence of drug
resistant TB strains that are highly resistant to most, if not, all – in the case of totally drug
resistant strains of TB – of the current antibiotics that are available (Udwadia et al., 2012,
Klopper et al., 2013). Together this highlights the urgent need for the discovery of new
antibiotics that can be used to shorten treatment time, tackle the increasing problem of
clinical drug resistance and be used in combination with HIV treatments. However, there has
been relatively little interest in the pursuit of new therapeutic agents for TB by the
pharmaceutical industry, which may be due in part to the greatest burden of disease being
present in developing countries (Cole, 2014). Despite this, over the last 10 years there has
been an increase in the discovery and development of new TB drugs either through high-
throughput phenotypic screens or target-based approaches (Reynolds et al., 2012, Stanley et
al., 2012, Lechartier et al., 2014, Christophe et al., 2009). This has resulted in a promising
TB drug pipeline (Wallis et al., 2016, Zumla et al., 2013), comprising new antibiotic agents
including the accelerated approval of bedaquiline (Andries et al., 2005), which is the first
FDA-approved TB drug in 40 years, and delamanid for use in multi-drug resistant TB (Gler
et al., 2012), along with new therapeutic regimens (Zumla et al., 2015, Zumla et al., 2013).
The mycobacterial N-acetyltransferase (NAT) enzyme has been identified as a potential
target for new TB agents (Bhakta et al., 2004). Deletion of the nat gene in Mycobacterium
bovis bacillus Calmette-Guérin (BCG) results in a significant decrease in the mycolates found
in the mycobacterial cell wall, with increased antibiotic susceptibility and killing within
macrophages (Bhakta et al., 2004, Anderton et al., 2006). In previous work, a high-
throughput target-based screen of 5,000 drug-like compounds was undertaken to find specific
inhibitors of NAT enzymes (Westwood et al., 2010, Abuhammad et al., 2012, Fullam et al.,
This article is protected by copyright. All rights reserved.

2011, Fullam et al., 2013). From this screen the piperidinol derivative: 3-benzoyl-4-phenyl-1-
methylpiperidinol (Figure 1), compound 1 was identified as an inhibitor of mycobacterial
NAT enzymes and a potent inhibitor of M. tuberculosis (Abuhammad et al., 2014,
Abuhammad et al., 2012, Westwood et al., 2010, Fullam et al., 2009, Fullam et al., 2008).
The unique in vitro mechanism of inhibition of compound 1 with the NAT enzyme from
Mycobacterium marinum has been elucidated and involves a specific covalent modification
of a cysteine residue located within the active site of the NAT enzyme (Abuhammad et al.,
2012, Abuhammad et al., 2010, Fullam et al., 2009, Fullam et al., 2008).
The aim of this study was to further evaluate the antimicrobial properties of the lead cyclic
piperidinol derivative 1 along with its analogous bis-Mannich base derivative 2 (Figure 1)
which depending on the in vivo physiological conditions encountered may be produced via a
retro-Aldol mechanism from compound 1. In order to achieve this, a combination of
approaches involving anti-microbial susceptibility testing, cytotoxicity studies, resistance
mapping by whole genome sequencing and genetic validation using the non-pathogenic
organism M. smegmatis have been undertaken. The combined results demonstrate that the
piperidinol scaffold 1 and its derivative 2 are specific for the Mycobacterium genus, resulting
in rapid killing, have an encouraging cytotoxicity selectivity index with a multiple and
diverse mechanism of action.
This article is protected by copyright. All rights reserved.

Materials and Methods
Chemical synthesis of 4-hydroxy-1-methyl-4-phenylpiperidin-3-yl)(phenyl)methanone:
compound 1 and 3,3’-(methylazanediyl)bis(1-phenylpropan-1-one): compound 2
The synthesis of 4-hydroxy-1-methyl-4-phenylpiperidin-3-yl)(phenyl)methanone (compound
1) and 3,3’-(methylazanediyl)bis(1-phenylpropan-1-one) (compound 2) were carried out as
described previously (Abuhammad et al., 2012, Wang et al., 2000, Gul et al., 2002, Blicke
and Burckhalter, 1942). Full experimental details and characterisation data of the compounds
used in this study are in Supplementary Information (Supplementary Figures 1 and 2).
Bacterial strains, cell lines, culture conditions and chemicals
M. smegmatis MC2155 was routinely grown at 37 oC in 7H9 broth (BD Difco) supplemented
with 0.2 % glycerol and 0.05 % Tween 80 or Luria-Bertani (LB) medium supplemented with
0.05 % Tween 80. Escherichia coli (E. coli) and Pseudomonas putida (P. putida) were
o
o
routinely cultured in LB-medium at 37 C and 30 C respectively. Human alveolar basal
epithelial A549 cells (Public Health England, ECACC 86012804) were cultured at 37 oC with
5 % CO2 atmosphere in Ham’s F-12K (Kaighn’s) Medium (Gibco, UK) supplemented with
10 % fetal-bovine serum,
100 Units/mL penicillin, 100 µg/mL streptomycin, and
250 ng/mL amphotericin B (HyClone). Ovine red blood cells were purchased from TCS
Biosciences. All chemicals were purchased from Sigma-Aldrich unless otherwise stated.
Determination of minimum inhibitory concentration
The measurement of the minimum inhibitory concentration (MIC) of compounds 1 and 2
were determined using the resazurin reduction microplate assay (REMA) as described
previously (Palomino et al., 2002). M. smegmatis, E. coli and P. putida were grown to mid-
log phase and the inoculum standardised to 1 x 106 colony forming units (CFU)/mL before
addition to the prepared 96-well flat-bottom microtiter plate with 2-fold serial dilutions of
each drug in media. An antibiotic control was also added to each plate (rifampicin for M.
smegmatis, ampicillin for E. coli and tetracycline for P. putida) and the last column of the
plate was used as a control without the addition of compound. In the case of M. smegmatis
o
the plates were incubated for 24 h at 37 C without shaking before addition of 25 L
resazurin (one tablet of resazurin (VWR) dissolved in 30 mL of sterile PBS). Following a
further incubation at 37 oC the plates were assessed for colour development. The MIC values
were determined as the lowest concentration of drug that prevented the colour change of
This article is protected by copyright. All rights reserved.

resazurin (blue – no bacterial growth) to resofurin (pink – bacterial growth). The MIC values
were determined with 5 independent experimental repeats and data are presented as mean ±
SEM.
Determination of the minimum bactericidal activity (MBC)
The MBC was determined by setting up a microtiter plate as performed for the MIC
determination. Instead of adding resazurin at 24 h, each well from the microtiter plate was
plated on to solid-LB medium and the CFUs determined after a 3-4 day incubation time at 37
oC. The lowest concentration at which no CFUs were counted was determined to be the
MBC. The MBCs were carried out with 5 independent experimental repeats.
Time kill studies
Cultures of M. smegmatis (105 CFU) were incubated with 2x MBC of compound 1 (625
g/mL), 2x MBC compound 2 (312.5 g/mL) and 2x MBC rifampicin (25 g/mL), along
with a no-drug control. Cultures were collected at defined time intervals (0, 30, 60, 90, 120,
180, 240, 300, 360 min and 24 h) and plated on solid LB-media and incubated at 37 oC for 3
days. Cell viability was assessed by determining the CFUs. The time kill studies were carried
out with 5 independent experimental repeats and data are presented as mean ± SEM.
Determination of compound interactions using a REMA checkerboard assay
In order to determine whether the compounds interacted with isoniazid a checkerboard assay
was employed. Fractional inhibitory concentrations (FICs) were calculated by use of the
following formula: FIC (X + Y) = [MIC of compound X in combination with Y]/[MIC of X
alone]. To evaluate interaction profiles the fractional inhibitory index (ΣFIC) was calculated
as FIC of compound X + FIC of compound Y where ΣFICs of ≤ 0.5 designate synergistic
activity, ΣFICs of ≥ 4.0 indicate antagonism, and values in between correspond to additivity
(Rand et al., 1993, Lechartier et al., 2012).
Cytotoxicity assay
The cytotoxicity of the compounds was measured against a human lung epithelial cell line
(A549). Briefly, cells were incubated (106 cells/well) with 2-fold serial dilutions of
compounds (compound 1: 31.25-500 g/mL; compound 2: 62.5-1000g/mL) in a 12 well
plate (Thermo Fisher #150628), including a cell only control. Following incubation for 24 h
This article is protected by copyright. All rights reserved.

o
at 37 C, in the presence of 5 % CO2 cell viability was determined by adding 100 L
resazurin for 1 h at 37 oC and measuring the absorbance of the resofurin metabolite at 570 nm
using a BioTek Synergy HT microplate reader. All assays were undertaken with 5
independent experimental repeats and data are presented as mean ± SEM.
Haemolysis assay
The haemolysis activity of compounds 1 and 2 were tested against ovine red blood cells.
Stock concentrations of compounds 1 (500 g/mL) and 2 (1 mg/mL) were prepared in PBS
containing 5 % DMSO. The compounds were serially diluted giving final concentrations of
compound 1 of 15.6-250 g/mL and compound 2 of 31.25-500 g/mL. A positive control
comprised lysis buffer (10 mM Tris pH 7.8, 0.32 M sucrose, 5 mM MgCl2, 10 % Triton X-
100) and the negative control comprised PBS containing 5 % DMSO. 150 L of each
o
compound was added to L of ovine blood and incubated at 22 C for 1 h, after which
o
the samples were centrifuged for 5 min
(2,000 rpm, 22 C). 10 L of the supernatant
was added to 90 L of PBS in a 96 well flat bottom microtiter plate (Thermo Fisher,
#167008) and the absorbance read at 450 nm (BioTek HT Synergy). All assays were
undertaken with 5 independent experimental repeats and data are presented as mean ± SEM.
Agglutination assay
The agglutination activity of compounds 1 and 2 were tested against ovine red blood cells.
Briefly, 150 L of ovine blood was incubated with 150 L compound 1 or 2 (2-fold
dilutions: final concentration of compound 1 15.6-250 g/mL, final concentration of
compound 2
31.25-500 g/mL). Separately, 150 L of 25 % polyethylenimine
was added as positive control or 150 L PBS as a negative control. Following addition of the
compounds the ovine blood was incubated at room temperature for 1 h, after which 50 L of
the sample was added to a round bottom 96 well microtiter plate (Corning, #3790), which
was then incubated at room temperature for a further 30 min. The plate was then assessed for
signs of agglutination. All assays were undertaken with 5 independent experimental repeats
and data are presented as mean ± SEM.
Metabolic stability in vitro
The intrinsic clearance (CLint) of compounds 1 and 2 were determined using mouse (CD-1)
liver microsomes (Gibco). Briefly, microsomes (100 g final protein concentration) and test
This article is protected by copyright. All rights reserved.

compound in 0.1 M phosphate buffer pH 7.4 were prepared. In parallel an NADPH-
regenerating system (Promega) was prepared in 0.1 M phosphate buffer (pH 7.4). The
solutions were pre-incubated at 37 °C for 10 min before assessment of the intrinsic clearance
was initiated by mixing the two solutions (50 μL of each; final compound concentration 1
μg/ml) at 37 °C. After 0, 5, 10, 15, 20 and 30 min the reactions were terminated by the
addition of 100 μL of acetonitrile containing 1 g/mL verapamil and placed on ice for 30
min. The samples were then centrifuged at 12,000 × g for
10 min and analysed by LC-
MS to determine the quantity of the parent compound remaining over time. Carbamazepine
(1 g/mL) was used as a control for low intrinsic clearance.
Generation of spontaneous resistant mutants of M. smegmatis to compounds 1 and 2.
M. smegmatis resistant mutants were generated by plating 108 mid-log cells (OD600 0.6) on
LB-agar containing 2.5x MIC of compound 1 and 5x MIC of compound 2. In the case of
compound 1, following the initial identification of mutants from the initial 2.5x MIC plate the
selected colonies were inoculated into liquid LB media containing 2x MIC compound 1 twice
before selecting on solid medium containing 2.5x MIC compound 1 and then proceeding as
for compound 2. In the case of compound 2, following identification, resistant mutants were
subsequently inoculated into liquid media in the absence of the compound to mid-log phase
and selected on solid LB-agar containing 5x MIC of compound 2. The colonies were
subsequently inoculated into LB-medium in the absence of compound before selecting on
LB-agar plate containing either 2.5x MIC of compound 1 or 5x MIC of compound 2 to
confirm resistance to the compound. Following this validation step for the generation and
identification of spontaneous mutants, the mutant strains and wild-type M. smegmatis were
inoculated in parallel into 50 mL liquid media containing 2x MIC of the respective compound
(no compound was added in the case of the wild-type M. smegmatis strain) and grown to
mid-log phase. Genomic DNA was prepared (total of 9 samples (7 resistant mutant strains
and 2 wild-type strains)) by centrifugation of the culture and resuspension in 25 % sucrose,
50 mM EDTA, 50 mM Tris HCl, pH 8.0 (SET solution) and 50 L of 20 mg/mL lysozyme
and incubated at 37 oC for 16 h. After this incubation step, 5 L of 10 mg/mL RNAse A was
o
added and the resuspension incubated at 37 C for 30 min, after which 250 L of Proteinase
K solution (400 mg/mL Proteinase K, 100 mM Tris HCl pH 8.0, 0.5 % SDS) was added and
o
incubated at 55 C for 2 h. DNA was extracted using phenol-chloroform-isoamyl alcohol
(24:24:1), chloroform-isoamyl alcohol (24:1), precipitated with ethanol, dried by air and
This article is protected by copyright. All rights reserved.

resuspended in TE buffer (10 mM Tris, 1 mM EDTA pH 8.0). The amount and purity of the
DNA was checked with a NanoDrop (Thermo Scientific).
Genomic DNA from each M. smegmatis mutant and wild-type strain was converted into
Nextera XT libraries for sequencing according to the manufacturer’s instructions with a few
modifications. Libraries were prepared using the dual indexing principal and amplified
libraries were purified using 25 μL AMPure XP beads and 45 μL of purified library, retained
and stored at -20 °C until denaturing for sequencing. Libraries were pooled in equimolar
concentrations (determined by analysis on Agilent Bioanalyser 2100 and HS dsDNA qubit
assay) and 12 pM sequenced on an Illumina MiSeq platform V2 2x 250 bp paired end
protocol. Reads were aligned to the
M. smegmatis str. MC2155 (accession number
NC_008596.1) using the Burrows-Wheeler Aligner (BWA) MEM algorithm v0.7.5a-r405 (Li
and Durbin, 2009). Resulting BAM files were manipulated with SAMtools (Li et al., 2009) to
produce mpileup files, with the following parameters ‘mpileup -B -f ’ .VarScan v2.3
(Koboldt et al., 2012) was used for single nucleotide polymorphism (SNP) and
insertion/deletion (INDEL) calling, with the following parameters ‘ --min-var-freq .80 --p-
value 0.05 --min-avg-qual 30’. SNPs that were common to both the wild-type control and the
mutant strain were not considered in further analysis. Read data has been submitted to the
EBI under the accession number PRJEB15140.
Genetic validation
Genes identified as putative targets from the whole-genome sequencing data were cloned into
the replicative vector pMV261 for over-expression in mycobacteria under control of the
hsp60 promoter. The gene of interest was amplified by PCR (Q5 High-Fidelity Polymerase
(New England Biolabs) from M. smegmatis genomic DNA. The primers used are as shown in
Supplementary Table 1. The resulting PCR product was digested with the appropriate
restriction enzymes and ligated (T4 DNA ligase, NEB) into the pMV261 vector. The
resulting construct was transformed into E. coli TOP10 (Invitrogen) and verified by
sequencing (GATC Biotech). The resulting constructs were electroporated into M. smegmatis
along with an empty pMV261 vector control.
Data and analysis
Blinding and randomisation were not applied in these studies. All data are represented by
mean ± SEM of at least five individual experiments and analysed using GraphPad Prism
This article is protected by copyright. All rights reserved.

version 7.0a software program (GraphPad Software Inc., San Diego, CA, USA). Statistical
significance between different groups was determined by unpaired t-test for two groups or
one-way ANOVA with Bonferroni's post hoc test to compare all pairs of columns for more
than two groups. A value of P < 0.01 was considered statistically significant. The data and
statistical analysis comply with the recommendations on experimental design and analysis in
pharmacology (Curtis et al., 2015).
This article is protected by copyright. All rights reserved.

Results
Antimicrobial properties of compound 1 and 2
Compound 1 has previously been shown to be highly active against M. tuberculosis with a
MIC of < 5g/mL (Abuhammad et al., 2012, Jeney and Zsolnai, 1956). We used the
resazurin microplate assay (REMA) to determine the MIC for compound 1 and its bis-
Mannich base derivative 2 (Figure 1) against M. smegmatis. In addition, in order to determine
the specificity of compounds 1 and 2 for the Mycobacterium genus these compounds were
also tested for their potential to inhibit the growth of the Gram-negative organisms E. coli and
P. putida. The MIC values for compounds 1 and 2 are shown in Table 1 and indicate that the
piperidinol class of compounds and its corresponding bis-Mannich base are able to inhibit the
growth of M. smegmatis with MIC values of 62.5 g/mL and 125 g/mL respectively. Whilst
the MIC for compounds 1 and 2 are higher than that for
M. tuberculosis it is clear
that these compounds do not have any effect on the growth of E. coli or P. putida at
concentrations as high as 250 g/mL.
To further understand the inhibitory properties of compounds 1 and 2 the MBCs were
determined against M. smegmatis. The MBCs were found to be 312.5 g/mL for compound 1
and ranged between 78-156 g/mL for compound 2 indicating that both compounds are
bactericidal against actively growing M. smegmatis. There is good correlation between the
MIC and MBC values for compound 2, whereas compound 1 displayed an MBC 5x greater
than its MIC value.
To determine whether compounds 1 and 2 are able to kill M. smegmatis completely and the
rate at which the antimicrobial activity is exerted, time-kill experiments were conducted in
which the reduction of CFUs was quantified as a function of exposure time of M. smegmatis
to 2x MBC of compounds 1 and 2 over a 24-hour time period. The kill curve is shown in
Figure 2. A monophasic kill curve was observed for both compounds 1 and 2 with a 2-log
reduction in CFUs after 90 minutes for compound 1 and 3 hours for compound 2 with
apparent complete eradication of viable M. smegmatis within a 3 hour time period for
compound 1 and a 5 hour time period for compound 2. In comparison, rifampicin was added
at 2x MBC and no reduction in CFUs were observed during this time period (Figure 2) (One-
way ANOVA: P < 0.001).
This article is protected by copyright. All rights reserved.

Compound interaction assessed by a checkerboard assay
To evaluate the impact of potential compound combinations and interactions on M.
smegmatis a checkerboard assay was used. Isoniazid was serially-diluted two-fold
horizontally across the plate (0 g/mL - 10g/mL) and either compound 1 or 2 (0 g/mL -
1000 g/mL) was serially diluted two-fold vertically down the plate. In addition, control
wells in which either isoniazid, compound 1 or compound 2 alone were tested
(Supplementary Figure 3). The sum of the fractional inhibitory index (FIC) for each
combination was calculated whereby FIC ≤ 0.5 denotes synergy and FIC ≥ 4 denotes
antagonism. Compound 1 has a FIC of 1.5 and compound 2 has a FIC of 3. Importantly,
no antagonistic interaction is observed between either compound 1 or compound 2 and the
front-line anti-tubercular agent isoniazid. 
Cytotoxicity of compounds 1 and 2
To determine the selectivity index of compounds 1 and 2 we measured the cytotoxicity of
both compounds against the A549 human lung epithelium cell line using the resazurin
reduction viability assay and the haemolytic activity of compounds 1 and 2 against ovine red
blood cells using both a haemoglobin release assay and an agglutination assay (Figure 3 and
Supplementary Figure 4). The maximum tolerated dose of compounds 1 and 2 against A549
cells was found to be between
125 g/mL-250 g/mL (t-test compound 1: P < 0.001; t-
test compound 2: P < 0.001) (Figure 3A). No lysis or agglutination of ovine blood red cells
was observed at any of the concentrations tested up to 250 g/mL for compound 1 (t-test: P <
0.001) and up to 500 g/mL for compound 2 (t-test: P < 0.001) (Figure 3B and
Supplementary Figure 4). These results therefore indicate that compounds 1 and 2 have a
higher selectivity for the clinically relevant pathogenic organism M. tuberculosis compared to
the human and ovine cells tested with a selectivity index of > 30-fold. These results are in
agreement with previous studies that also found low toxicity to compound 1 for the mouse
macrophage cell line RAW 264.7 and human cell line U937 (Abuhammad et al., 2012).
Metabolic stability of compounds 1 and 2
We subsequently evaluated the in vitro metabolic stabilities of compound 1 and 2 using
mouse liver microsomes and LC-MS and found that both compounds had high intrinsic
clearance rates
[CLint] > 48 L/min/mg protein. A slow [CLint] control using
carbamazepine was used for comparison (Supplementary Figure 5).
This article is protected by copyright. All rights reserved.

Target identification of compounds 1 and 2
Compound 1 has shown encouraging inhibitory potency against M. tuberculosis and
biochemical
studies have shown that compound 1 inhibits mycobacterial NAT enzymes (Abuhammad et
al., 2014, Abuhammad et al., 2012, Jeney and Zsolnai, 1956). In order to identify any
additional intracellular targets for the piperidinol class of compounds spontaneous resistant
mutants in M. smegmatis were generated. Using the determined MIC values of compounds 1
and 2 (Table 1) M. smegmatis resistant mutants were generated on solid LB-agar containing
2.5x MIC of compound 1 (due to solubility issues) and 5x MIC compound 2. Three colonies
for compound 1 and four colonies for compound 2 were selected for subsequent inoculation
in liquid medium in the absence of any compound and re-selection on solid medium in the
presence of either compound 1 at 2.5 x MIC and compound 2 at 5x MIC. All colonies
selected from the original plate maintained their ability to grow on solid medium with the
addition of compound 1 and 2 following this procedure. Two of the resistant mutants to
compound 1 and two of the resistant mutants to compound 2 were re-tested in the REMA
assay whereby it was shown that the MIC of the mutants to compound 1 increased 5 fold (t-
test: P < 0.001) and the mutants to compound 2 increased by 2-3-fold (t-test: P < 0.01) when
compared to the wild-type M. smegmatis strain (Table 2). Interestingly isolate 3 to compound
1 also had increased resistance to compound 2 by 2-3 fold (t-test: P < 0.01) and similarly
isolate 2 to compound 2 also increased its resistance to compound 1 between 2-3 fold (t-test:
P < 0.01) indicating cross-resistance and a potential common resistance mechanism. Each
mutant remained fully susceptible to isoniazid and rifampicin (Table 2).
Given that resistance to compounds 1 and 2 was maintained in the absence of any selection
pressure on the mutant strains generated, it was postulated that a genetic resistance
mechanism is involved in the observed resistant phenotype. To identify the genetic mutation
causing the observed phenotype the genomes of the mutant strains were subjected to whole-
genome sequencing using Illumina technology (Table 3).
This article is protected by copyright. All rights reserved.

Bioinformatics analysis identified nineteen putative SNPs compared to the laboratory M.
smegmatis wild-type control strains. Three SNPs were predicted to occur in pseudogenes, six
SNPs were predicted to occur in intergenic regions and two SNPS were predicted as
synonymous mutations. Alleles with non-synonymous SNPs were found to occur in six M.
smegmatis genes (Table 3) and included two genes with unknown function (MSMEG_2746
and MSMEG_3244), a gene predicted to be involved with homing (HNH) endonuclease
function (MSMEG_2148), a putative 3-dehydroquinate synthase gene (MSMEG_3033,
aroB), a putative monooxygenase gene (MSMEG_4083) and a putative gene involved in
amino acid transport (MSMEG_5807). It is interesting to note that more alleles with non-
synonymous SNPs were identified for the resistant mutants isolated from compound 2 which
were generated at 5x MIC compound 2 compared to 2.5x MIC compound 1 due to solubility
issues encountered. It was particularly interesting to find that four of the predicted SNPs
occurred in the aroB gene resulting in an amino acid change of G278R and C352G for all
four resistant isolates to compound 2 and two additional SNPs in aroB resulting in the
additional amino acid change of G280D and L359P in three out of four of these resistant
isolates (Table 3). These results led us to investigate the role of this potential aroB target
further.
The aroB gene from M. smegmatis was expressed using the hsp60 promoter from the
replicative pMV261 vector and the susceptibility of this overexpressing strain to compounds
1 and 2 was undertaken using the REMA assay and the MICs determined. The results
demonstrate that over-expression of aroB leads to partial resistance to both compound 1 and
2 when compared to the wild-type strain (t-test compound 1: P < 0.01; t-test compound 2: P <
0.01) although there is little difference between overexpression of aroB compared to empty
vector alone (Table 4).
This article is protected by copyright. All rights reserved.

Discussion and Conclusions
Piperidinol compound 1 and its corresponding bis-Mannich base 2 have previously been
demonstrated to have potent anti-tubercular activity against both M. tuberculosis and M.
bovis BCG (Abuhammad et al., 2014, Abuhammad et al., 2012, Jeney and Zsolnai, 1956) and
we have shown these compounds also have specificity for the Mycobacterium genus and
good cytotoxicity selectivity. Despite no evident signs of cytotoxicity, the compounds appear
to have a high intrinsic clearance rate in mouse microsomes and hence further chemical
optimisation of these compounds will be required in order to improve their metabolic
degradation profile whilst retaining anti-tubercular activity.
Treatment of TB is increasingly hindered by the emergence of drug-resistant strains and this
has led to increased efforts in the identification of new potent inhibitors of M. tuberculosis
and pharmacologically validated targets. In order to tackle this problem, there has been a
recent increase in the utilisation of chemical genomics approaches to determine the cellular
targets of these newly identified anti-tubercular agents. The generation of spontaneous
resistant mutants followed by whole genome sequencing has been successfully used to
determine a number of intracellular targets including, for example, SQ109, an FDA approved
drug to be used only for drug-resistant strains of TB (Tahlan et al., 2012), dinitrobenzamide
and benzothiazole derivatives which are in the late stages of pre-clinical development
(Christophe et al., 2009, Neres et al., 2012) and the natural product pyridomycin (Hartkoorn
et al., 2012). From this approach a range of new and previously pharmacologically validated
targets have been identified, including the MmpL3 membrane transporter (Tahlan et al.,
2012) that is involved in the export of trehalose monomycolate and cell wall biosynthesis,
and InhA, an enoyl (ACP)-reductase that is a key component in the type II fatty acid synthase
system (FasII) and is the cellular target for isoniazid (Banerjee et al., 1994).
The piperidinol compound 1 has been shown to inhibit the NAT enzyme from M. marinum
through the generation of a polyvinyl moiety that forms a covalent adduct with the cysteine
residue located in the NAT active-site (Abuhammad et al., 2014, Abuhammad et al., 2012,
Abuhammad et al., 2010). However, the potential link between the anti-tubercular activity
observed of 1 and the potential endogenous role of NAT in the bactericidal activity of this
piperidinol derivative has yet to be evaluated and other potential targets have not been
precluded (Abuhammad et al., 2014, Abuhammad et al., 2012). Therefore, in this study, we
This article is protected by copyright. All rights reserved.

utilised a chemical genetic approach as a route to determine other potential cellular targets of
compounds 1 and 2. Given that M. smegmatis has been successfully used in high-throughput
screening campaigns to identify new TB drugs, including the clinically approved anti-
tubercular agent bedaquiline (Andries et al., 2005), we chose to use this organism to generate
spontaneous resistant mutants due to the ease and speed of manipulation despite the slight
increase in the MIC of compounds 1 and 2 compared to M. tuberculosis. Interrogation of the
resultant resistant isolates’ genomes has led to the identification of a multiple number of high
quality SNPs for each mutant. The high number of SNPs observed for this piperidinol
scaffold is higher than that for similar chemical bioinformatics studies (Hartkoorn et al.,
2012, Hartkoorn et al., 2014, Tahlan et al., 2012) suggesting a potential diverse activity
mechanism profile.
Mycobacteria have a highly unique cell wall structure, comprising a mycolic acid-
arabinogalactan-peptidoglycan (mAGP) core (Brennan and Nikaido, 1995). Investigation into
the endogenous role of the NAT enzyme in mycobacteria by deletion of the nat gene in M.
bovis BCG has determined that NAT has a key role in the biosynthesis of mycobacterial
mycolic acids and has been postulated to have a potential role in maintaining the homeostasis
of the acetyl CoA pool (Bhakta et al., 2004). NAT enzymes utilize the cofactor acetyl CoA as
an acetyl donor for arylamine substrates (Sim et al., 2007). Whilst, under these experimental
conditions, our studies did not identify any SNPs in the nat gene, it was of particular interest
to note the identification of a number of SNPs in the aroB enzyme in all resistant isolates
generated with compound 2. The aroB gene putatively encodes for a 3-dehydroquinate
synthase enzyme involved in the second step of the shikimate pathway with AroB catalysing
the cyclisation of 3-deoxy-D-arabino-heptulosonate-7-phosphate in 3-dehydroquinate. The 3-
dehydroquinate is subsequently metabolised in the shikimate pathway to chorismate which is
required for the biosynthesis of all aromatic amino acids and other key metabolites. In M.
tuberculosis the aroB gene (Rv2538) is predicted to be essential for the survival of M.
tuberculosis (Sassetti et al., 2003, Parish and Stoker, 2002). Enzymes within the shikimate
pathway represent excellent targets for the development of new antibiotics as they are
essential for bacterial survival with no human counterpart (Bentley, 1990, Ducati et al.,
2007). Given the role of the AroB enzyme in the essential biosynthesis of aromatic amino
acids, naphthoquinones, menaquinones, and mycobactin in M. tuberculosis we selected this
prospective genetic target for further validation. Overexpression of aroB using the replicative
pMV261 vector, with a relative copy number of 2-3 fold (Stover et al., 1991), resulted in
This article is protected by copyright. All rights reserved.

partial resistance of these strains to compounds 1 and 2, which could in part be explained by
the presence of the chromosomal wild-type aroB gene still being expressed. Partial resistance
has been found to occur in a similar chemical bioinformatics study to investigate the
mechanism of resistance to clofazimine, a drug that is used in the multidrug therapy of
leprosy (Hartkoorn et al., 2014). Although, it is possible that these studies have also
identified off-target compensatory pathways in response to these compounds. However, it is
probable that given the high number of SNPs identified for this piperidinol derivative for M.
smegmatis that these compounds are active against a diverse set of intracellular targets which
subsequently results in difficulties in restoring the observed chemical-genetic phenotype
through overexpression of a single gene.
In summary, these studies indicate that the cyclic piperidinol class of compound along with
its bis-Mannich base derivate are specific for the Mycobacterium genus and rapidly kill M.
smegmatis with encouraging cytotoxicity profiles. Genetic interrogation of spontaneous
resistant mutants of M. smegmatis generated under the selective pressure of compounds 1 and
2 has led to the identification of potential genetic targets and pathways that are involved in
the multiple and diverse mode of action for these piperidinol chemical moieties. Previous in
vitro biochemical studies demonstrated the formation of a polyvinyl ketone moiety which is
likely to be highly reactive at the proteomic level. Given the increase in TB drug-resistant
strains that often results from drugs acting on a single biological target, chemical moieties
with anti-tubercular inhibitory activities that act upon multiple cellular targets are highly
attractive for the development of new TB agents from a clinical standpoint and are worthy of
further investigation.
This article is protected by copyright. All rights reserved.

Acknowledgements
We would like to thank the Media Preparation facility in the School of Life Sciences. EF
acknowledges support in the form of a Sir Henry Dale Fellowship jointly funded by the
Wellcome Trust and the Royal Society (Grant Number 104193/Z/14/Z) and the Antimicrobial
Resistance Cross Council Initiative supported by the seven research councils’ (grant number
MR/N006917/1). MIG acknowledges support from the ERC (grant number 638661) and the
Antimicrobial Resistance Cross Council Initiative supported by the seven research councils’
(grant number MR/N006917/1). Equipment used in these studies was supported by the Royal
Society award (Grant number RG120405). DP thanks the Institute of Advanced Studies,
University of Warwick for a IAS Early Career Fellowship, JH is an INTEGRATE Early
Career Fellow funded by the EPSRC (Grant number EP/M027503/1).
Author contributions
Conceived and designed the experiments: CG ET GK MIG MJP EF. Performed the
experiments: CG ET GK DP CF TB JH EF. Analysed the data: CG ET GK CG DP AM MIG
EF. Analysed bioinformatics data: ET, GK, ADM. Contributed reagents/materials/analysis
tools: CG GK AM DP MIG MJP EF. Wrote the paper: CG ET ADM EF
Conflicts of interest:
The authors declare no conflicts of interest.
Supporting Information
Supporting information is provided for the synthetic protocols for compounds 1 and 2 along
with 5 Supplementary Figures and 1 Supplementary table.
This article is protected by copyright. All rights reserved.

References
WHO,
2015.
Global
Tuberculosis
Report
[Online].
Available:
http://apps.who.int/iris/bitstream/10665/191102/1/9789241565059_eng.pdf [Accessed 2016].
Abuhammad A, Fullam E, Bhakta S, Russell AJ, Morris GM, Finn PW, et al. (2014).
Exploration of piperidinols as potential antitubercular agents. Molecules 19: 16274-90.
Abuhammad A, Fullam E, Lowe ED, Staunton D, Kawamura A, Westwood IM, et al. (2012).
Piperidinols that show anti-tubercular activity as inhibitors of arylamine N-acetyltransferase:
an essential enzyme for mycobacterial survival inside macrophages. PLoS One 7: e52790.
Abuhammad AM, Lowe ED, Fullam E, Noble M, Garman EF, Sim E (2010). Probing the
architecture of the Mycobacterium marinum arylamine N-acetyltransferase active site. Protein
Cell 1: 384-92.
Alexander SP, Fabbro D, Kelly E, Marrion N, Peters JA, Benson HE, et al. (2015). The
Concise Guide to PHARMACOLOGY 2015/16: Enzymes. Br J Pharmacol 172: 6024-109.
Anderton MC, Bhakta S, Besra GS, Jeavons P, Eltis LD, Sim E (2006). Characterization of
the putative operon containing arylamine N-acetyltransferase (nat) in Mycobacterium bovis
BCG. Mol Microbiol 59: 181-92.
Andries K, Verhasselt P, Guillemont J, Gohlmann HW, Neefs JM, Winkler H, et al. (2005).
A diarylquinoline drug active on the ATP synthase of Mycobacterium tuberculosis. Science
307: 223-7.
Banerjee A, Dubnau E, Quemard A, Balasubramanian V, Um KS, Wilson T, et al. (1994).
inhA, a gene encoding a target for isoniazid and ethionamide in Mycobacterium tuberculosis.
Science 263: 227-30.
Bentley R (1990). The shikimate pathway--a metabolic tree with many branches. Crit Rev
Biochem Mol Biol 25: 307-84.
Bhakta S, Besra GS, Upton AM, Parish T, Sholto-Douglas-Vernon C, Gibson KJ, et al.
(2004). Arylamine N-acetyltransferase is required for synthesis of mycolic acids and complex
lipids in Mycobacterium bovis BCG and represents a novel drug target. J Exp Med 199:
1191-9.
Blicke FF, Burckhalter JH (1942). The Preparation of beta-Keto Amines by the Mannich
Reaction. J. Am. Chem. Soc. 64: 451-454.
Brennan PJ, Nikaido H (1995). The envelope of mycobacteria. Annu Rev Biochem 64: 29-
63.
Christophe T, Jackson M, Jeon HK, Fenistein D, Contreras-Dominguez M, Kim J, et al.
(2009). High content screening identifies decaprenyl-phosphoribose 2' epimerase as a target
for intracellular antimycobacterial inhibitors. PLoS Pathog 5: e1000645.
Cole ST (2014). Who will develop new antibacterial agents? Philos Trans R Soc Lond B Biol
Sci 369: 20130430.
This article is protected by copyright. All rights reserved.

Curtis MJ, Bond RA, Spina D, Ahluwalia A, Alexander SP, Giembycz MA, et al. (2015).
Experimental design and analysis and their reporting: new guidance for publication in BJP.
Br J Pharmacol 172: 3461-71.
Ducati RG, Basso LA, Santos DS (2007). Mycobacterial shikimate pathway enzymes as
targets for drug design. Curr Drug Targets 8: 423-35.
Dye C, Williams BG (2010). The population dynamics and control of tuberculosis. Science
328: 856-61.
Fullam E, Abuhammad A, Wilson DL, Anderton MC, Davies SG, Russell AJ, et al. (2011).
Analysis of beta-amino alcohols as inhibitors of the potential anti-tubercular target N-
acetyltransferase. Bioorg Med Chem Lett 21: 1185-90.
Fullam E, Kawamura A, Wilkinson H, Abuhammad A, Westwood I, Sim E (2009).
Comparison of the Arylamine N-acetyltransferase from Mycobacterium marinum and
Mycobacterium tuberculosis. Protein J 28: 281-93.
Fullam E, Talbot J, Abuhammed A, Westwood I, Davies SG, Russell AJ, et al. (2013).
Design, synthesis and structure-activity relationships of 3,5-diaryl-1H-pyrazoles as inhibitors
of arylamine N-acetyltransferase. Bioorg Med Chem Lett 23: 2759-64.
Fullam E, Westwood IM, Anderton MC, Lowe ED, Sim E, Noble ME (2008). Divergence of
cofactor recognition across evolution: coenzyme A binding in a prokaryotic arylamine N-
acetyltransferase. J Mol Biol 375: 178-91.
Gler MT, Skripconoka V, Sanchez-Garavito E, Xiao H, Cabrera-Rivero JL, Vargas-Vasquez
DE, et al. (2012). Delamanid for multidrug-resistant pulmonary tuberculosis. N Engl J Med
366: 2151-60.
Gul HI, Calis U, Vepsalainen J (2002). Synthesis and evaluation of anticonvulsant activities
of some bis Mannich bases and corresponding piperidinols. Arzneimittelforschung 52: 863-9.
Hartkoorn RC, Sala C, Neres J, Pojer F, Magnet S, Mukherjee R, et al. (2012). Towards a
new tuberculosis drug: pyridomycin - nature's isoniazid. EMBO Mol Med 4: 1032-42.
Hartkoorn RC, Uplekar S, Cole ST (2014). Cross-resistance between clofazimine and
bedaquiline through upregulation of MmpL5 in Mycobacterium tuberculosis. Antimicrob
Agents Chemother 58: 2979-81.
Jeney E, Zsolnai T (1956). [Studies in search of new tuberculostatic drugs. I. Hydrazine
derivatives, carbolic acid, phenols, quaternary ammonium compounds and their
intermediaries]. Zentralbl Bakteriol Orig 167: 55-64.
Klopper M, Warren RM, Hayes C, Gey Van Pittius NC, Streicher EM, Muller B, et al.
(2013). Emergence and spread of extensively and totally drug-resistant tuberculosis, South
Africa. Emerg Infect Dis 19: 449-55.
This article is protected by copyright. All rights reserved.

Koboldt DC, Zhang Q, Larson DE, Shen D, Mclellan MD, Lin L, et al. (2012). VarScan 2:
somatic mutation and copy number alteration discovery in cancer by exome sequencing.
Genome Res 22: 568-76.
Lechartier B, Hartkoorn RC, Cole ST (2012). In vitro combination studies of benzothiazinone
lead compound BTZ043 against Mycobacterium tuberculosis. Antimicrob Agents Chemother
56: 5790-3.
Lechartier B, Rybniker J, Zumla A, Cole ST (2014). Tuberculosis drug discovery in the post-
post-genomic era. EMBO Mol Med 6: 158-68.
Li H, Durbin R (2009). Fast and accurate short read alignment with Burrows-Wheeler
transform. Bioinformatics 25: 1754-60.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. (2009). The Sequence
Alignment/Map format and SAMtools. Bioinformatics 25: 2078-9.
Neres J, Pojer F, Molteni E, Chiarelli LR, Dhar N, Boy-Rottger S, et al. (2012). Structural
basis for benzothiazinone-mediated killing of Mycobacterium tuberculosis. Sci Transl Med 4:
150ra121.
Palomino JC, Martin A, Camacho M, Guerra H, Swings J, Portaels F (2002). Resazurin
microtiter assay plate: simple and inexpensive method for detection of drug resistance in
Mycobacterium tuberculosis. Antimicrob Agents Chemother 46: 2720-2.
Parish T, Stoker NG (2002). The common aromatic amino acid biosynthesis pathway is
essential in Mycobacterium tuberculosis. Microbiology 148: 3069-77.
Rand KH, Houck HJ, Brown P, Bennett D (1993). Reproducibility of the microdilution
checkerboard method for antibiotic synergy. Antimicrob Agents Chemother 37: 613-5.
Reynolds RC, Ananthan S, Faaleolea E, Hobrath JV, Kwong CD, Maddox C, et al. (2012).
High throughput screening of a library based on kinase inhibitor scaffolds against
Mycobacterium tuberculosis H37Rv. Tuberculosis (Edinb) 92: 72-83.
Sassetti CM, Boyd DH, Rubin EJ (2003). Genes required for mycobacterial growth defined
by high density mutagenesis. Mol Microbiol 48: 77-84.
Sim E, Westwood I, Fullam E (2007). Arylamine N-acetyltransferases. Expert Opin Drug
Metab Toxicol 3: 169-84.
Southan C, Sharman JL, Benson HE, Faccenda E, Pawson AJ, Alexander SP, et al. (2016).
The IUPHAR/BPS Guide to PHARMACOLOGY in 2016: towards curated quantitative
interactions between 1300 protein targets and 6000 ligands. Nucleic Acids Res 44: D1054-
68.
Stanley SA, Grant SS, Kawate T, Iwase N, Shimizu M, Wivagg C, et al. (2012).
Identification of novel inhibitors of M. tuberculosis growth using whole cell based high-
throughput screening. ACS Chem Biol 7: 1377-84.
This article is protected by copyright. All rights reserved.

Stover CK, De La Cruz VF, Fuerst TR, Burlein JE, Benson LA, Bennett LT, et al. (1991).
New use of BCG for recombinant vaccines. Nature 351: 456-60.
Tahlan K, Wilson R, Kastrinsky DB, Arora K, Nair V, Fischer E, et al. (2012). SQ109 targets
MmpL3, a membrane transporter of trehalose monomycolate involved in mycolic acid
donation to the cell wall core of Mycobacterium tuberculosis. Antimicrob Agents Chemother
56: 1797-809.
Udwadia ZF, Amale RA, Ajbani KK, Rodrigues C (2012). Totally drug-resistant tuberculosis
in India. Clin Infect Dis 54: 579-81.
Wallis RS, Maeurer M, Mwaba P, Chakaya J, Rustomjee R, Migliori GB, et al. (2016).
Tuberculosis-advances in development of new drugs, treatment regimens, host-directed
therapies, and biomarkers. Lancet Infect Dis 16: e34-46.
Wang S, Sakamuri S, Enyedy IJ, Kozikowski AP, Deschaux O, Bandyopadhyay BC, et al.
(2000). Discovery of a novel dopamine transporter inhibitor, 4-hydroxy-1-methyl-4-(4-
methylphenyl)-3-piperidyl 4-methylphenyl ketone, as a potential cocaine antagonist through
3D-database pharmacophore searching. Molecular modeling, structure-activity relationships,
and behavioral pharmacological studies. J Med Chem 43: 351-60.
Westwood IM, Bhakta S, Russell AJ, Fullam E, Anderton MC, Kawamura A, et al. (2010).
Identification of arylamine N-acetyltransferase inhibitors as an approach towards novel anti-
tuberculars. Protein Cell 1: 82-95.
Zumla A, Chakaya J, Centis R, D'ambrosio L, Mwaba P, Bates M, et al. (2015). Tuberculosis
treatment and management--an update on treatment regimens, trials, new drugs, and adjunct
therapies. Lancet Respir Med 3: 220-34.
Zumla A, Nahid P, Cole ST (2013). Advances in the development of new tuberculosis drugs
and treatment regimens. Nat Rev Drug Discov 12: 388-404.
This article is protected by copyright. All rights reserved.

Table 1. Minimum inhibitory concentrations of compounds 1 and 2
Bacterium
Compound 1
MIC (g/mL)
5
Compound 2 MIC
Reference
(g/mL)
M. tuberculosis
nd
(Abuhammad et al., 2012,
Jeney and Zsolnai, 1956)
(Abuhammad et al., 2014)
This study
M. bovis BCG
M. smegmatis
E. coli
6.25-12.5
62.5±0
250±0
nd
125±0
250±0
>500±0
This study
P. putida
250±0
This study
nd – not determined. All assays in this study were undertaken with n=5 and data are presented as
mean ± SEM.
This article is protected by copyright. All rights reserved.

Table 2. Activity of compounds 1 and 2 against selected resistant mutants
Isolatea
MIC compound MIC compound MIC INH MIC RIF
1
2
(g/mL)
(g/mL)
(g/mL)
(g/mL)
Compound 1
Compound 1
Compound 2
Compound 2
n/a
1
375.0±62.5
375.0±62.5
171.6±38.2
265.6±46.9
56.2±6.3
156±0
25±0
5±0
10±0
3
375.0±62.5
375.0±62.5
281.3±31.3
112.5±12.5
12.5±0
12.5±0
12.5±0
12.5±0
1
10±0
2
3.2±0.5
3.2±0.5
M.
WT
smegmatis
aIsolate refers to mutant numbers listed in Table 3
INH: isoniazid; RIF: rifampicin; WT: wild-type strain: n/a: not applicable
The averages of five independent experiments are shown and the data are presented as mean ± SEM.
This article is protected by copyright. All rights reserved.

Table 3: Single nucleotide polymorphisms detected in Mycobacterium smegmatis mutants resistant to compounds 1 and 2
SNP locus
M. smegmatis
chromosome
position
Amino
Codon
acid
Gene
Annotation
Change
change
Compound 1
Compound 2
2223880
2223882
cCccGc P380R MSMEG_2148 HNH
–
–
–
–
–
–
–
–
R
K
R
K
R
K
AaaCaa K381Q
endonuclease
2816638
gGagAa G332E MSMEG_2746 Unknown
E
E
–
–
E
–
–
3105561
3105568
3105783
3105805
3328246*
GggAgg G278R MSMEG_3033 aroB
gGcgAc G280D
TgcGgc C352G
cTacCa L359P
ccTccC P384P MSMEG_3244 Hypothetical
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
R
–
G
–
R
D
G
P
R
D
G
P
R
D
G
P
–
P*
P*
P*
4158542*
5874050
ttTttC
F227F MSMEG_4083 Monooxygenase
–
–
–
–
–
F*
F*
F*
F*
gCcgGc A306G MSMEG_5807 Amino acid
G
–
–
–
–
transport
Genomic positions are relevant to M. smegmatis str. MC2155 (accession number NC_008596.1)
WT alleles are represented by ‘– ‘; * represents non-synonymous SNP
This article is protected by copyright. All rights reserved.

Table 4: Susceptibility profiles of M. smegmatis strains to compounds 1 and 2
MIC (g/mL)
Strain
Compound 1
56.2±6.3
Compound 2
112.5±12.5
Rifampicin
7.5±1.1
M. smegmatis mc2155 – Wild type
M. smegmatis:aroB
112.5±12.5
112.5±12.5
300.0±49.9
275.0±61.2
7.5±1.1
7.5±1.1
M. smegmatis:pMV261
The averages of five independent experiments are shown and the data are presented as mean ± SEM.
This article is protected by copyright. All rights reserved.

Figure 1. Compounds identified as inhibitors of the NAT enzyme. Structures of piperidinol
compound 1 and the bis-Mannich base compound 2
This article is protected by copyright. All rights reserved.

Figure 2. Time kill curve of compounds 1 and 2. Exponentially growing cultures of M. smegmatis
were treated with 2x MBC of compound 1 (), 2x MBC compound 2 (), 2x MBC rifampicin ()
and a no drug control (). CFUs were counted at the time points indicated. The averages of five
independent experiments are shown and the data are presented as mean ± SEM.
This article is protected by copyright. All rights reserved.