Analogs of nitrofuran antibiotics are potent GroEL/ES inhibitor pro-drugs

Article abstract of DOI:10.1016/j.bmc.2020.115710

In two previous studies, we identified compound 1 as a moderate GroEL/ES inhibitor with weak to moderate antibacterial activity against Gram-positive and Gram-negative bacteria including Bacillus subtilis, methicillin-resistant Staphylococcus aureus, Kleb

Full text of DOI:10.1016/j.bmc.2020.115710

Bioorganic&MedicinalChemistry28(2020)115710
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Analogs of nitrofuran antibiotics are potent GroEL/ES inhibitor pro-drugs
Mckayla Stevensa, Chris Howea, Anne-Marie Raya, Alex Washburna, Siddhi Chitrea,
⁎
Jared Sivinskib, Yangshin Parka,c,d, Quyen Q. Hoanga,c,d, Eli Chapmanb, Steven M. Johnsona,
a Indiana University School of Medicine, Department of Biochemistry and Molecular Biology, 635 Barnhill Dr., Indianapolis, IN 46202, United States
b The University of Arizona, College of Pharmacy, Department of Pharmacology and Toxicology, 1703 E. Mabel St., PO Box 210207, Tucson, AZ 85721, United States
c Stark Neurosciences Research Institute, Indiana University School of Medicine. 320 W. 15th Street, Suite 414, Indianapolis, IN 46202, United States
d Department of Neurology, Indiana University School of Medicine. 635 Barnhill Drive, Indianapolis, IN 46202, United States
A R T I C L E I N F O
A B S T R A C T
Keywords:
In two previous studies, we identified compound 1 as a moderate GroEL/ES inhibitor with weak to moderate
antibacterial activity against Gram-positive and Gram-negative bacteria including Bacillus subtilis, methicillin-
resistant Staphylococcus aureus, Klebsiella pneumonia, Acinetobacter baumannii, and SM101 Escherichia coli (which
has a compromised lipopolysaccharide biosynthetic pathway making bacteria more permeable to drugs).
Extending from those studies, we developed two series of analogs with key substructures resembling those of
known antibacterials, nitroxoline (hydroxyquinoline moiety) and nifuroxazide/nitrofurantoin (bis-cyclic-N-
acylhydrazone scaffolds). Through biochemical and cell-based assays, we identified potent GroEL/ES inhibitors
that selectively blocked E. faecium, S. aureus, and E. coli proliferation with low cytotoxicity to human colon and
intestine cells in vitro. Initially, only the hydroxyquinoline-bearing analogs were found to be potent inhibitors in
our GroEL/ES-mediated substrate refolding assays; however, subsequent testing in the presence of an E. coli
nitroreductase (NfsB) in situ indicated that metabolites of the nitrofuran-bearing analogs were potent GroEL/ES
inhibitor pro-drugs. Consequently, this study has identified a new target of nitrofuran-containing drugs, and is
the first reported instance of such a unique class of GroEL/ES chaperonin inhibitors. The intriguing results
presented herein provide impetus for expanded studies to validate inhibitor mechanisms and optimize this
antibacterial class using the respective GroEL/ES chaperonin systems and nitroreductases from E. coli and the
ESKAPE bacteria.
GroEL
GroES
Molecular chaperone
Chaperonin
Proteostasis
Small molecule inhibitors
ESKAPE pathogens
Antibiotics
Antibacterials
Nitroreductase
Pro-drugs
1. Introduction
baumannii, Pseudomonas aeruginosa, and Enterobacter species.3–8 While
several classes of antibiotics that target diverse biological pathways
The rise of antibiotic-resistant infections has become a significant
threat to human health worldwide. This issue has been recently re-
emphasized by the World Health Organization’s 2019 report to
Secretary-General of the United Nations “No time to wait: Securing the
future from drug-resistant infections” and the Centers for Disease
Control and Prevention (CDC) in their 2019 report titled “Antibiotic
Resistance Threats in the United States”.1,2 World-wide, over 700,000
deaths from drug-resistant infections occur annually.1 In the US
alone, > 2.8 million infections by antibiotic resistant bacteria occur
annually, with 35,000 deaths. Of particular prominence is a subset of
bacteria referred to as the ESKAPE pathogens – an acronym that re-
presents Gram-positive Enterococcus faecium and Staphylococcus aureus
bacteria, and Gram-negative Klebsiella pneumonia, Acinetobacter
have been successfully used to treat these bacteria for decades, we are
in an era where derivatizing analogs to circumvent bacterial resistance
has led to diminishing returns for developing effective new clinical
candidates.8–14 In some instances, bacterial strains have emerged that
are resistant to all contemporary antibacterials, as well as older, more
toxic drugs that clinicians are reverting to in more desperate situations
(e.g. polymyxins).15,16 To counter the diminishing antibacterial pipe-
line, it is crucial that new antibacterial candidates are developed that
function through unexploited biological pathways. Furthermore, of
particular urgency is to identify new antibiotic candidates that are ef-
fective against Gram-negative bacteria, since their lipopolysaccharide
(LPS) outer membranes and efficient efflux pumps make them highly
impermeable and intrinsically resistant to many antibacterial agents.
Abbreviations: HSP, Heat shock protein; BSP, bis-sulfonamido-2-phenylbenzoxazole; SCA, salicylanilide; HQ, hydroxyquinoline series; NF, nitrofuran series; MDH,
malate dehydrogenase; Rho, rhodanese; Nfz, nifuroxazide; Nft, nitrofurantoin; Nox, nitroxoline; NR, nitroreductase
⁎ Corresponding author.
E-mail address: johnstm@iu.edu (S.M. Johnson).
https://doi.org/10.1016/j.bmc.2020.115710
Received 2 July 2020; Received in revised form 7 August 2020; Accepted 10 August 2020
Availableonline30August2020
0968-0896/©2020ElsevierLtd.Allrightsreserved.

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
Towards identifying antibacterial candidates that function through
fundamentally new mechanisms of action, researchers have recently
begun exploring the possibility of targeting molecular chaperones as
antibacterial strategies.17–25 A broad class of molecular chaperones
consist of several families of heat shock proteins (HSPs) that are cate-
gorized by the relative molecular weights of their subunits: HSP100,
HSP90, HSP70, HSP60, HSP40, and small molecular weight HSPs.26–31
Together, these HSPs form an intricate intracellular network that
maintains protein homeostasis by helping unfolded polypeptides to fold
to their native states, or targeting them for degradation when mis-
folding occurs. Our lab is focused on targeting the HSP60 class of mo-
lecular chaperones, known as GroEL in bacteria, as a mechanistically
unique antibiotic strategy. As all bacteria contain at least one GroEL
homolog that is essential under all conditions, this class of molecular
chaperones represents a potential antibacterial target with broad
spectrum applicability.29,32–42
GroEL is comprised of two homo-heptameric rings that stack back-
to-back with one another.43–54 The 800 kDa bis-toroidal GroEL oligomer
functions in conjunction with a 70 kDa homo-heptameric co-chaperone,
called GroES, to fold substrate polypeptides in an ATP-dependent
manner. The GroEL/ES-mediated folding cycle is initiated when ATP
and an unfolded polypeptide bind to a GroEL cis-ring. The positively
cooperative binding of ATP to each of the seven GroEL subunits powers
a conformational shift of the GroEL apical domains upwards, allowing
them to engage with the GroES co-chaperone lid. Binding of GroES
causes a greater conformational movement of the GroEL apical domains
that releases the bound polypeptide substrate into the now enclosed
GroEL/ES cis-chamber. The polypeptide is then allowed to try and re-
fold on its own, sequestered from the outside cytoplasmic milieu, with
ATP hydrolysis acting as the timing mechanism (~5–10 s). ATP hy-
drolysis in the cis-ring sends an allosteric signal for ATP and another
unfolded polypeptide substrate to then bind to the GroEL trans-ring.
This initiates the release of the GroES lid from the cis-ring, along with
the substrate protein. If the substrate protein has not achieved its native
state, it can re-attempt the folding process or be targeted for degrada-
tion.
Figure 1. Previous investigations identified bis-sulfonamido-2-phenylbenzox-
azole (BSP) and salicylanilide (SCA) analogs that were potent GroEL/ES in-
hibitors with antibacterial activity against Gram-positive bacteria.56–59 Com-
pound 1 was previously found to be a moderate inhibitor of GroEL/ES-mediated
refolding of denatured rhodanese and malate dehydrogenase substrates
(IC50 = 18 & 31 μM, respectively) and a weak to moderate inhibitor of the
proliferation of B. subtilis (EC50 = 83 μM), methicillin-resistant S. aureus
(EC50 = 56 μM), K. pneumoniae (EC50 = 95 μM), A. baumannii (EC50 = 32 μM),
and SM101 E. coli (EC50 = 19 μM).55,57 Owing to its ability to inhibit both
Gram-positive and Gram-negative bacteria, and its structural similarity to
known antibacterials nitroxoline, nifuroxazide, and nitrofurantoin, in this
study, we sought to develop new compound 1 analogs (Scheme 1) that were
more potent and selective inhibitors of GroEL/ES and bacterial proliferation.
In 2014, we reported a study that identified 235 GroEL/ES in-
hibitors from a high-throughput screen of 700,000 compounds.55 Sub-
sequent screening and analog derivatization of two hit-to-lead series –
based on bis-sulfonamido-2-phenylbenzoxazole (BSP) and salicylanilide
(SCA) scaffolds (Figure 1) – identified lead compounds that potently
inhibited the proliferation of E. faecium and S. aureus bacteria (both
Gram-positive), but that were largely ineffective against E. coli and the
KAPE Gram-negative bacteria.56–59 However, from our initial anti-
bacterial testing, compound 1 stood out as another hit inhibitor of in-
terest.57 Despite being only a moderate GroEL/ES inhibitor, compound
1 exhibited weak to moderate antibacterial efficacy against Bacillus
subtilis, methicillin-resistant S. aureus, K. pneumonia, A. baumannii, and
SM101 E. coli (which have a temperature sensitive protein in the LPS
biosynthetic pathway making the bacteria more permeable to drugs at
side groups were assessed that contained a diverse range of sub-
structures, including some that we have found effective with other
GroEL/ES inhibitor scaffolds that have shown antibacterial properties
(e.g. thiophenes, 2-chlorothiophenes, and aryl-sulfonamides).56–58,71
These analogs were tested in a panel of in vitro assays for bacterial
proliferation inhibition, GroEL/ES inhibition (both substrate folding
and ATPase functions), human cell cytotoxicity, and bacterial resistance
towards lead analogs. The results from testing analogs in this panel of
biochemical and cell-based experiments are presented herein, with a
discussion on SAR and the potential of these series as GroEL/ES-tar-
geting antibacterial candidates.
non-permissive temperatures).60,61 Furthermore, the compound
1
scaffold resembles other known antibiotics that are active against
Gram-negative bacteria. For instance, compound 1 shares its hydro-
xyquinoline substructure with nitroxoline, and its bis-cyclic-N-acylhy-
drazone core with nitrofuran-based antibiotics such as nifuroxazide and
nitrofurantoin (Figure 1). While these drugs are primarily used to treat
urinary tract infections, they also exhibit antibiotic effects against a
range of pathogens including the ESKAPE bacteria.62–68
Based on the similarities of compound 1 with nitroxoline, nifurox-
azide, and nitrofurantoin, we developed a library of analogs that probed
the structure–activity relationships (SAR) of the cyclic/aryl sub-
structures on both the right and left-hand sides of the N-acylhydrazone
linker (Scheme 1). The right-hand substructure of these analogs con-
tained either a hydroxyquinoline group (mimicking 1 and nitroxoline)
or a nitrofuran group (mimicking nifuroxazide and nitrofurantoin),
generating two distinct series of compounds. For each series, 15 left-
2. Results and discussion
2.1. Conceptualizing and developing the hydroxyquinoline and nitrofuran-
containing series of compound 1 analogs.
Owing to their similarity to the known antibacterials nitroxoline,
nifuroxazide, and nitrofurantoin, we investigated two primary series of
2

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
Table 1
Compilation of EC50 values for the hydroxyquinoline and nitrofuran analogs
tested in the ESKAPE and E. coli bacterial proliferation assays. Cells that are
shaded darker blue are most potent, while those that are shaded lighter blue to
white are less potent to inactive (> 100 μM).
#
2
3
4
5
6
7
1
8
9
10
11
12
13
14
15
>100 37 >100 >100 >100 >100 80
>100 63 99 74 >100 92 77
70 84 >100 >100 >100 >100 >100
>100 69 82 56 >100 80 63
>100 >100 >100 >100 >100 >100 79
Scheme 1. a Structures of the hydroxyquinoline (HQ – 1–15) and nitrofuran
(NF – 16–28, including nitrofurantoin and nifuroxazide) series of analogs with
the syntheses of compounds 1 and 20 shown as a representative examples. To
develop SAR in each series, the methoxy rings of 1 and 20 were substituted with
a variety of other ring substructures that we have found effective with other
GroEL/ES inhibitor scaffolds that have shown antibacterial properties (e.g.
thiophenes, 2-chlorothiophenes, and aryl-sulfonamides) – please refer to the
>100 48
>100 87
87
71 >100 97
46
85 >100 95
100 >100
>100 99 >100 >100 >100 >100 93
>100 56 >100 >100 >100 >100 >100
>100 >100 >100 >100 81 >100 >100
>100 >100 >100 >100 >100 >100 >100
>100 >100 >100 >100 85 >100 99
>100 >100 >100 >100 >100 >100 >100
>100 >100 >100 84 >100 >100 100
>100 >100 >100 >100 >100 >100 >100
data tables in the Supporting Information for specific compound struc-
a
tures.69,70
.
Reagents and conditions: (a) The respective hydrazides and alde-
hydes were stirred with cat. HCl in DMSO for 18 h, then precipitated in water,
filtered, and dried in vacuo (46–99% yields).
Nitroxoline
Nitrofurantoin 38
18
9.5
27
7.1
3.0
10
2.8
2.5
99
6.6
4.0
compound 1 analogs that also bore hydroxyquinoline and nitrofuran
substructures (Scheme 1). For each series, 15 left-side groups were
assessed that contain a diverse range of substructures, including some
that we have found effective with other GroEL/ES inhibitor scaffolds
that have shown antibacterial properties (e.g. thiophenes, 2-chlor-
othiophenes, and aryl-sulfonamides).56–58,71 A third group of analogs
(29–42) was also investigated to determine which parts of the hydro-
xyquinoline and nitrofuran aryls were required for inhibitor potency in
the respective assays (refer to Tables S1C and S2C in the Supporting
Information for the structures of compounds 29–42).
40 >100 >100 36
16
19
36 >100 >100 >100 2.1
42
0.69
0.42
0.45
16
17
18
19
24
12
31
13
82 >100 40
72 >100 49
9.5
16
54 >100 73
2.4
0.87
Nifuroxazide 8.1
37 >100 >100 54
20
21
22
23
24
25
26
27
28
22
>100 34 >100 >100 >100 >100 10.0
16
11
11
>100 9.6 >100 >100 >100 >100 >100
11 11 >100 >100 >100 >100 17
>100 >100 >100 >100 >100 >100 >100
>100 >100 >100 >100 >100 >100 >100
8.1 >100 >100 >100 >100 1.9
6.0 >100 >100 >100 >100 41
5.2 >100 >100 >100 >100 >100
3.5 >100 >100 >100 >100 >100
Analogs were all synthesized through a one-step coupling reaction
between the respective aryl-aldehydes and N-acylhydrazides in DMSO,
with HCl as a catalyst.69,70 After stirring overnight at room tempera-
ture, the final N-acylhydrazone products were precipitated through the
addition of water, and the solids were filtered, rinsed, and dried in
vacuo (synthetic yields ranged from 46 to 99%). Where necessary,
compounds were further purified via normal and/or reverse-phase
chromatography. Synthesized analogs were analyzed by RP-HPLC for
purity and LC-MS and 1H NMR for structural confirmation (complete
characterization data can be found in the Supporting Information).
While all compounds were found to be > 95% pure using two distinct
sets of RP-HPLC conditions, for some analogs (3, 4, 6, 9, 10, 17, 23),
we noticed a splitting of peaks in the 1H NMR spectra. This phenom-
enon has previously been studied by others and reported as resulting
from hindered rotation around the amide bond, providing rotational
isomers (rotamers).72,73 Thus, we believe that the purity of these
compounds is consistent with HPLC results showing > 95% purity.
log phase (OD600 ~ 0.4–0.6) – to examine inhibition against actively-
replicating bacteria – whereupon final OD600 readings were taken to
assess for bacterial growth and inhibition. An in-depth protocol for this
common set of proliferation assays is provided in the Supporting
Information, and calculated EC50 values are reported in Tables 1 and
S2. For easier visualization of results, tables are heat-mapped according
to inhibitor potencies, with darker cells representing the most potent
analogs, and lighter cells the least potent inhibitors.
Results from the bacterial proliferation assays indicated that the
hydroxyquinoline analogs were largely ineffective against E. coli and
the ESKAPE pathogens, although several weak inhibitors were identi-
fied. Prior to initiating this study, we had hypothesized that the weak
metal-chelating properties of the hydroxyquinoline substructure might
allow these inhibitors to act as siderophores that could be actively taken
up into bacteria; however, the lack of efficacy in this series was perhaps
not surprising as a previous study by Pelletier et al. had indicated that
the antibacterial activity of nitroxoline was reduced upon cation sup-
plementation.65,75–78 Despite this setback, we were excited to see that
many of the nitrofuran-based analogs were much more effective at in-
hibiting bacterial growth. In particular, analogs 16–20, 22–24, and 26
were moderate to strong inhibitors of the Gram-positive E. faecium and
S. aureus bacteria. For reasons that are not clear, and contrary to the
GroEL/ES inhibition results (discussed below), the dimethylaniline (21)
and bulkier sulfonamide-containing analogs (25–28) were less effective
(or inactive) compared to the analogs with the smaller N-acylhydrazide
substructures. Presumably these compounds suffered from efflux and/
2.2. Evaluating analogs for inhibiting the growth of E. Coli and the ESKAPE
bacteria.
Analogs were initially tested for antibacterial efficacy against re-
presentative strains of antibiotic-sensitive E. coli and the ESKAPE bac-
teria (refer to the Materials & Methods section for strain and vendor
information for the respective bacteria). To determine compound effi-
cacy, bacterial proliferation assays were carried out in liquid media
culture supplemented with physiological concentrations of free calcium
and magnesium cations.74 For these assays, bacterial cultures
(OD600 = 0.01) were exposed to test compounds in 8-point, 3-fold di-
lution series (100 μM to 46 nM concentration range), in 384-well plates.
After addition of compounds to cultures, the plates were sealed with
Breathe-Easy gas permeable membranes and allowed to grow to mid-
3

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
or poor permeability through the bacterial cell walls, since they were
the most potent at inhibiting GroEL/ES and would thus be expected the
most potent at killing bacteria if they achieved appreciably high in-
tracellular concentrations, and presuming they were exhibiting on-
target effects against GroEL/ES.
the previous section, the dimethylaniline-nitrofuran (21) and bulkier
sulfonamide-containing analogs (12–15 and 25–28) were generally the
most potent GroEL/ES inhibitors, yet the least effective or inactive
against bacteria, suggesting they were unable to penetrate bacteria or
were quickly effluxed out – otherwise they would have been expected to
exhibit stronger antibacterial effects. Although, this is presuming that
the antibacterial effects of these analogs was from on-target inhibition
of GroEL/ES, which remains to be proven.
While antibacterial effects were limited against the Gram-negative
KAPE bacteria, a few analogs (16–21) were potent against E. coli, with
EC50 values ≤ 11 μM. Compounds 16 and 17 were more potent than
nitroxoline, nifuroxazide, and nitrofurantoin against E. coli (EC50
values < 1 μM), and were even moderate inhibitors of K. pneumoniae.
These were significant findings as our previous studies failed to identify
lead analogs with such high efficacies against E. coli or any of the Gram-
negative KAPE bacteria.56–58 As evidenced by the results of compound
39 (Table S2C), which has an unsubstituted furan ring, the nitro group
was essential for antibacterial effects. This is putatively because ni-
trofuran antibiotics are activated to reactive metabolites by nitror-
eductases in bacteria, which we discuss further below.79–83 Ad-
ditionally, the hydroxyquinoline and nitrofuran aldehyde starting
materials (40 and 41, respectively) were potent inhibitors of nearly all
the bacteria (excluding P. aeruginosa), yet the N-acylhydrazone analogs
were largely ineffective against the KAPE bacteria, supporting our belief
that the linkers were not hydrolyzing and releasing the starting mate-
rials. Therefore, inhibitor potencies likely owed to the final products
themselves, or their metabolites in the case of the nitrofuran series.
2.4. Evaluating analogs for inhibiting GroEL/ES-mediated substrate
refolding functions in the presence of the E. Coli NfsB type-1 nitroreductase.
That none of the nitrofuran analogs were found to be potent GroEL/
ES inhibitors is complicated by the fact that nitrofuran-based antibiotics
are known to act as prodrugs in vivo – they require metabolism by
bacterial nitroreductases to generate reactive metabolites that are as-
sociated with their antibacterial effects.79–83 However, our standard
GroEL/ES-mediated refolding assays and native substrate reporter
counter-screens were conducted without nitroreductases present. This
emphasized the need to re-examine analogs in modified refolding and
native reporter activity assays that included a nitroreductase enzyme to
activate the nitrofuran analogs in situ, which would be more re-
presentative of the bacterial intracellular environment. As an initial test
to see whether or not the activated nitrofuran metabolites would be
more potent GroEL/ES inhibitors, we purchased the E. coli NfsB type 1
nitroreductase and modified our standard GroEL/ES-dMDH refolding
and native MDH counter-screens to generate the reactive metabolites in
situ (detailed protocols for these assays are presented in the Materials &
Methods section). We have reported IC50 results from these assays in
Table S1A – S1C in the Supporting Information, and graphically in the
correlation plot of Figure 2C, where IC50 values are compared between
compounds tested in the GroEL/ES-dMDH refolding assay with and
without the E. coli NfsB nitroreductase.
2.3. Evaluating analogs for inhibiting GroEL/ES-mediated substrate
refolding functions.
Since we identified several analogs that potently inhibited the
growth of both Gram-positive and Gram-negative bacteria, we next
evaluated their abilities to inhibit E. coli GroEL/ES-mediated substrate
folding functions in vitro. We first evaluated all test compounds in our
standard GroEL/ES-mediated refolding assays that employ either ma-
late dehydrogenase (MDH) or rhodanese (Rho) as the denatured sub-
strate reporter enzymes (detailed descriptions of the assay protocols are
presented in the Supporting Information).56–59,71,84 When denatured,
these enzymes are efficiently folded by GroEL/ES in the absence of
inhibitors, and thus act as reporters to determine the degree of inhibi-
tion against the bacterial chaperonin system. Inhibition was examined
in the presence of these two orthogonal substrates in order to support
that inhibitors were acting against the chaperonin system. We further
counter-screened for inhibition of native MDH and Rho – where test
compounds were added after the denatured MDH and Rho substrate
enzymes were completely refolded by GroEL/ES – to identify false po-
sitives that may simply act by inhibiting the enzymatic reporter reac-
tions of the coupled refolding assays. We have found that these series of
four biochemical assays are highly effective at eliminating false-posi-
tives as compounds rarely inhibit both reporter enzymes since their
enzymatic read-outs are so different from one another. IC50 results for
all compounds tested in these four assays are presented in Tables S1A –
S1C in the Supporting Information.
We were excited to see that in the presence of E. coli NfsB, the ni-
trofuran analogs exhibited dramatically increased inhibition of GroEL/
ES refolding functions. The IC50 values for the hydroxyquinoline (1–15)
and other analogs without nitro groups (e.g. 29–42) were nearly
identical in the presence and absence of NfsB, supporting that increased
inhibition was dependent on modification of the nitrofuran moiety and
not other effects on the overall compound scaffold or from NfsB itself.
Furthermore, the nitrofuran metabolites were inactive in the native
MDH reporter counter-screen, indicating that increased inhibition was
obtained through selectively targeting the GroEL/ES-mediated re-
folding cycle. While the degree of potency shift varied between analogs,
ten shifted to IC50 values ≤ 11 μM. As points of comparison, the most
potent nitrofuran analog in the absence of NfsB had an IC50 = 26 μM
(analog 21), with six being completely inactive (IC50 > 100 μM).
Significant potency shifts were observed for the most effective anti-
bacterials (16–20), showing that some of the strongest antibacterial
compounds were also strong GroEL/ES inhibitors when activated by a
nitroreductase enzyme. While activated nifuroxazide and nitrofur-
antoin were only weak to moderate GroEL/ES inhibitors in this new
assay, this may not be surprising as they are reported in the literature to
be preferentially activated by the other E. coli type 1 nitroreductase,
NfsA.80,82,85
Unfortunately, we found that the nitrofuran analogs were only weak
to moderate GroEL/ES inhibitors despite being the most potent at in-
hibiting bacterial growth (see the white diamond symbols in the cor-
relation plots of Figure 2A). Conversely, the hydroxyquinoline analogs
(black circles in Figure 2A) were much stronger GroEL/ES inhibitors,
although they were largely inactive against bacteria. While the hydro-
xyquinolines were slightly more potent in the GroEL/ES-mediated dRho
compared to dMDH refolding assays (Figure 2A), this was likely be-
cause several of them had the coupled effects of also being weak in-
hibitors of the native Rho reporter reaction (Figure 2B); however, no
compounds inhibited native MDH, supporting that the GroEL/ES in-
hibitors were not false-positives. Analogs 29–38, where the various
parts of the hydroxyquinoline substructure were pared away, were
largely inactive in all biochemical assays, indicating the necessity for
the complete hydroxyquinoline moiety for inhibition. As mentioned in
While we observed a general trend whereby the more potent that
nitrofuran inhibitors were in the in situ NfsB-GroEL/ES-dMDH refolding
assay, the more potent they were at inhibiting S. aureus proliferation
(Figure 3B), we are guarded as to whether or not inhibitors were po-
tentially functioning on-target in bacteria. We note the limitations in
such an overly simplistic comparison since we had tested such a small
set of analogs, and only tested them in the presence of E. coli NfsB. As
the different nitrofuran analogs would be expected to exhibit varying
SAR for activation by NfsA and NfsB, it will be important to test in-
hibitors in the presence of both nitroreductases to gain a more complete
picture of how they could be functioning in bacteria. We are in the
4

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
A.
C.
B.
Nft
Inactive
Weak
Nft
100
10
1
Nfz
Nfz
100
10
1
100
Nft
Nfz
Nox
Moderate
10
1
Nox
Strong
Nox
Very Strong
VS
S
M
W
I
0.1
0.1
0.1
0.1
1
10
100
0.1
1
10
100
0.1
1
10
100
GroEL/ES-dMDH Refolding IC50 (μM)
GroEL/ES-dMDH Refolding IC50
(μM)
Native MDH Activity IC50 (μM)
without E. coli NfsB nitroreductase
Hydroxyquinolines
Nitrofurans
Figure 2. Correlation plots of IC50 values for compounds evaluated in the respective biochemical assays. A. Compounds inhibited nearly equipotently in the GroEL/
ES-dMDH and the GroEL/ES-dRho refolding assays (Spearman correlation coefficient comparing log(IC50) values in each assay is 0.8877 (p < 0.0001)). For the
purposes of categorizing inhibitor potencies, we consider compounds with IC50 values plotted in the grey zones to be inactive (i.e. greater than the maximum
concentrations tested), > 33 μM to be weak inhibitors, 11–33 μM moderate inhibitors, 1–11 μM potent inhibitors, and < 1 μM very potent and acting near
stoichiometrically since the concentration of GroEL tetradecamer is 50 nM during the refolding cycle (i.e. 700 nM GroEL monomeric subunits). B. While some
compounds inhibited in the native Rho enzymatic reporter counter-screen, none inhibited native MDH enzymatic activity, supporting that inhibitors were not false-
positives that simply target the enzymatic reporter reactions of the coupled GroEL/ES-mediated refolding assays. C. Nitrofuran analogs exhibited increased inhibition
in the GroEL/ES-dMDH refolding assay in the presence of E. coli NfsB, while hydroxyquinoline analogs did not (Spearman correlation coefficient comparing hy-
droxyquinoline log(IC50) values in each assay is 0.9527 (p < 0.0001)), supporting a pro-drug mechanism of action through metabolism of the nitro group. No
compounds inhibited native MDH enzymatic activity in either the absence or presence of E. coli NfsB (refer to Table S1 in the Supporting Information). Results
plotted in the grey zones represent IC50 values higher than the maximum concentrations tested. Data points for nifuroxazide (Nfz), nitrofurantoin (Nft), and
nitroxoline (Nox) are labelled for comparison.
process of cloning and expressing E. coli nfsA and nfsB and developing
an expanded panel of nitrofuran-based analogs to study inhibitor me-
chanisms in greater detail. Furthermore, investigating the nitror-
eductases of the various ESKAPE bacteria could provide a stronger ra-
tionale for why these compounds were largely inactive against the
KAPE Gram-negative strains. Another limitation was that we were using
the E. coli GroEL/ES chaperonin system as a surrogate in our assays.
While we anticipate inhibition results will translate to the chaperonin
systems of the other ESKAPE bacteria owing to their high sequence si-
milarities (between 55 and 95% amino acid identity between the cha-
peronins from E. coli and the ESKAPE bacteria), this still remains to be
demonstrated. Thus, we are also in the process of generating
recombinant versions of the respective ESKAPE bacteria GroEL and
GroES proteins and will report on such inhibition results in the future.
2.5. Evaluating analogs for inhibiting GroEL-mediated ATPase activity.
Since many proteins use ATP for their biological functions, in-
hibiting GroEL/ES by competitively binding to the ATP sites could
prove problematic for being able to selectively target the chaperonin
system. Thus, we further tested analogs in a well-established GroEL
ATPase assay that employed malachite green to monitor inorganic
phosphate liberated as GroEL hydrolyzed ATP.55,59,84 Briefly, a solution
of GroEL was incubated with test compounds (8-point, 3-fold dilution
A.
B.
C.
100
10
1
100
100
10
1
Nft
Nft
Nfz
Nox
10
Nox
Nfz
Nox
1
Nfz
Nft
0.1
0.1
0.1
0.1
1
10
100
0.1
1
10
100
0.1
1
10
100
GroEL/ES-dMDH Refolding IC50 (μM)
with E. coli NfsB nitroreductase
GroEL/ES-dMDH Refolding IC50 (μM)
with E. coli NfsB nitroreductase
GroEL/ES-dMDH Refolding IC50 (μM)
with E. coli NfsB nitroreductase
Hydroxyquinolines
Nitrofurans
Figure 3. Correlation plots comparing IC50 values for compounds tested in the in situ NfsB-GroEL/ES-dMDH refolding assay with EC50 values for inhibiting E. faecium
(A), S. aureus (B), and E. coli (C) proliferation. While increasing inhibition by the nitrofurans in the in situ NfsB-GroEL/ES-dMDH refolding assay in general provided
more effective inhibition of S. aureus growth (panel B), more thorough studies will need to be conducted – e.g. testing a larger number of analogs in the presence of S.
aureus nitroreductases and GroEL/ES chaperonin system – to gain a clearer picture of whether or not compounds may be functioning on-target against GroEL/ES in
bacteria. Results plotted in the grey zones represent IC50 and EC50 values higher than the maximum concentrations tested. Data points for nifuroxazide (Nfz),
nitrofurantoin (Nft), and nitroxoline (Nox) are labelled for comparison.
5

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
series) and the assay was initiated by addition of ATP. After incubating
for 45 min, the ATPase reaction was quenched by the addition of EDTA.
Malachite green was then added to the assay to bind and detect free
phosphates in solution (absorbance detection at λ = 600 nm). If ana-
logs inhibited ATPase activity, then there would be no free phosphates
for malachite green to bind, leading to minimal absorbance at 600 nm.
The detailed procedure for this assay is presented in the Supporting
Information. As indicated in Tables S1A to S1C in the Supporting
Information, none of the analogs from either series inhibited GroEL by
blocking ATP hydrolysis. Thus, we believe that these inhibitors bind to
sites outside of the ATP pockets. We are pursuing and will report on
studies to identify these unknown binding sites in the future. While
these results alleviate concerns about non-selectively targeting other
ATP-dependent proteins, we further assessed off-target effects through
a more definitive approach by evaluating analog cytotoxicity in two
human cell lines, discussed below.
bacteria to generate resistance to a lead candidate. We examined the
ability of E. coli to generate resistance to 17 (with nifuroxazide and
nitrofurantoin as controls), since resistance to nitrofuran-based anti-
biotics has been well-characterized in this bacterium. While 17, nifur-
oxazide, and nitrofurantoin were all potent inhibitors of E. coli pro-
liferation, 17 was the most potent at inhibiting GroEL/ES in the
presence of NfsB, and thus may exhibit greater on-target effects in
bacteria. However, as discussed above, we do appreciate the limitations
of not employing NfsA. To identify differences in the ability of E. coli to
generate resistance to these three compounds with such distinct
bioactivity profiles, we employed a 12 day resistance assay as we pre-
viously reported for our salicylanilide lead candidate S. aureus in-
hibitors (a detailed protocol is included in the Supporting
Information).58 Briefly, a dilute culture of E. coli (OD600 = 0.01) was
grown in the presence of inhibitors for 24 h (tested in dose–response in
duplicate), then EC50 results were determined from the OD600 readings
of the wells. Over the course of 12 days, we sequentially sub-cultured
bacteria from the respective wells with the highest concentration of
inhibitors where bacteria grew to an OD600 > 0.2, monitoring for
increases in EC50 values over time.
2.6. Evaluating the cytotoxicity of analogs to human colon and small
intestinal cells.
While in previous studies we have employed biochemical counter-
screening with the human mitochondrial HSP60/10 chaperonin system,
our accumulating results indicate that inhibiting HSP60/10 in vitro is a
poor indicator of potential off-target toxicity to human cells.57–59,84
This is highlighted by the fact that we have identified many known
drugs and natural products that are potent inhibitors of HSP60/10
biochemical function in vitro, yet exhibit little to no adverse
effects in cells or animals. For instance, we found that suramin
is a potent HSP60/10 inhibitor, yet it has been used safely for over
100 years as a first-line treatment for Trypanosoma brucei infec-
tions.59,84 In addition, as now identified in this study, bioactivation of
nitrofuran antibiotics by nitroreductase enzymes greatly increases the
extent of inhibition against GroEL/ES refolding activity, and potentially
human HSP60/10; however, this further complicates testing against
HSP60/10 since human cells do not contain nitroreductases. Therefore,
we feel the most appropriate initial assessment of potential in vivo
toxicity is to test compounds for cytotoxicity to human cells in culture.
To assess for potential cytotoxic effects, compounds were tested in
two Alamar Blue-based cell viability assays using human FHC colon and
FHs 74 Int small intestinal cells. Briefly, we grew cells to ~ 80–90%
confluency, then sub-cultured 1,500 cells per well (in 384-well plates)
for 24 h in the absence of test compounds. Compounds were then added
and the cultures were incubated for an additional 48 h, whereupon the
Alamar Blue reporter reagents were added and well fluorescence was
monitored over time. Alamar Blue contains resazurin (non-fluorescent),
which is reduced to resorufin (highly fluorescent) in the presence of
viable cells. A detailed protocol is presented in the Supporting
Information, along with cell cytotoxicity CC50 values in Tables S2A –
S2C. As graphically presented in the Figure 4 correlation plots com-
paring bacterial proliferation EC50 to human cell cytotoxicity CC50 re-
sults, lead nitrofuran inhibitors (16, 17, 20, nitrofurantoin, and nifur-
oxazide) selectively inhibited E. faecium, S. aureus, and E. coli
proliferation with low to no cytotoxicity to human cells (representative
results are shown for human FHs 74 Int small intestine cells, but results
are similar for FHC colon cells and are presented in Figure S1 in the
Supporting Information). Intriguingly, the nitrofuran analogs were ty-
pically less toxic than their hydroxyquinoline counterparts, putatively
because they would need to be metabolized to their active inter-
mediates, yet human cells do not harbor nitroreductases.
While we found that nifuroxazide and nitrofurantoin were initially
more potent than 17 (Figure 5), E. coli quickly developed intermediate
resistance (within 3–5 days) such that all three inhibitors were nearly
equipotent. This initial resistance was putatively through mutations
affecting NfsA function, as has been previously reported by others.80
That EC50 values then somewhat plateaued in the 20–40 μM range is
consistent with NfsB still being able to metabolize the nitrofuran moi-
eties and maintain efficacy, albeit at a reduced capacity. EC50 values
continued to slowly increase over time for all three compounds, with a
particular jump in resistance seen for the second set of replicates for
nitrofurantoin and nifuroxazide to a lesser extent, but not for 17. Thus,
inhibitors that are preferentially activated by NfsB, as may be the case
for 17, might be more effective drug candidates with respect to com-
batting the emergence of drug resistant strains. We are cautious in over-
interpreting these results since the experiment was only conducted in
duplicate for three analogs, and further studies are warranted.
We next confirmed that the resistance generated by the replicate E.
coli strains was irreversible (i.e. putatively through permanent muta-
tions of NfsA and NfsB as previously reported by others) as opposed to
transient means (i.e. by up-regulating efflux pumps). To accomplish
this, we sub-cultured single colonies obtained from the replicate sam-
ples where the bacteria exhibited the greatest degree of resistance to
test compounds (day 12 samples for all compound replicates, except
replicate 2 for nitrofurantoin, which was taken at day 10), for 4 × 12 h
serial passages in fresh media without any test compounds present. We
then performed another 24 h follow-up proliferation assay to determine
EC50 values (dose–response curves are presented for nifuroxazide, ni-
trofurantoin, and 17 in Figures
6 and S2 in the Supporting
Information). Results indicated that these subsequently-cultured bac-
terial strains were still resistant to each of the respective inhibitors they
were generated from, supporting that resistance mechanisms were
permanent. As previous studies by others have extensively character-
ized mutations affecting NfsA and NfsB that E. coli acquire to generate
resistance against nitrofuran antibiotics, we did not perform genotyping
to further characterize the specific resistance mechanisms for these
strains, as they were likely the same.80
Since nifuroxazide, nitrofurantoin, and 17 displayed different in-
hibition and resistance profiles, we next examined if the respective
resistant strains were cross-resistant with each of the other inhibitors.
Intriguingly, while the replicate strains that were initially resistant to
nifuroxazide were cross-resistant to both nitrofurantoin and 17
(Figures 7A and S3A in the Supporting Information), the nitrofur-
antoin-resistant strains were still sensitive to nifuroxazide and 17
(Figures 7B and S3B), and the 17-resistant strains were susceptible to
nifuroxazide, but not nitrofurantoin (Figures 7C and S3C). Thus, it is
evident that strains that have generated resistance to one analog are not
2.7. Investigating the ability of E. Coli to gain resistance to 17,
nifuroxazide, and nitrofurantoin.
As we discovered several nitrofuran-based analogs that selectively
inhibited the growth of E. faecium, S. aureus, and E. coli with minimal
toxicity to human cells, we next investigated how easy it would be for
6

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
A.
B.
C.
Nft
Nft
Nft
Nfz
100
100
10
1
100
10
1
Nfz
Nfz
Nox
Nox
Nox
10
1
0.1
0.1
0.1
0.1
1
10
100
0.1
1
10
100
0.1
1
10
100
E. faecium Proiliferation EC50 (μM)
E. coli Proiliferation EC50 (μM)
S. aureus Proiliferation EC50 (μM)
Hydroxyquinolines
Nitrofurans
Figure 4. Correlation plots examining the selectivity of compounds inhibiting the proliferation of E. faecium (A), S. aureus (B), and E. coli (C) over cytotoxicity to
human FHs 74 Int small intestine cells (results for cytotoxicity to human FHC colon cells are similar, with CC50 values reported in Table S2 and Figure S1 in the
Supporting Information). Results plotted in the grey zones represent EC50, and CC50 values higher than the maximum concentrations listed. Data points for nifur-
oxazide (Nfz), nitrofurantoin (Nft), and nitroxoline (Nox) are labelled for comparison.
necessarily cross-resistant to other analogs. Further studies are war-
ranted to see if combination therapy with two or more inhibitor analogs
might be synergistic and prevent or prolong the emergence of resistant
strains.
indicated that the nitrofurans act as pro-drugs and their reactive me-
tabolites can be potent GroEL/ES inhibitors. Lead nitrofuran analogs
were potent inhibitors of E. faecium, S. aureus, and E. coli proliferation
and exhibited minimal cytotoxicity to human FHC colon and FHs 74 Int
small intestinal cells. While E. coli were able to generate varying de-
grees of irreversible resistance to nifuroxazide, nitrofurantoin, and lead
analog 17 (putatively through mutations in NfsA and NfsB), clones were
not necessarily cross-resistant to the other inhibitors. This finding may
indicate diverging mechanisms of activation and/or targets for struc-
turally distinct inhibitors, suggesting that combination therapy could
prove synergistic and delay the onset of bacterial resistance.
3. Conclusions
In a previous high-throughput screen, we identified compound 1 as
a moderate GroEL/ES inhibitor with weak to moderate antibacterial
efficacy against B. subtilis, MRSA, K. pneumonia, A. baumannii, and
SM101 E. coli (which has a temperature sensitive LPS biosynthetic
pathway). Intriguingly, key substructures of compound 1 resembled
those of nitroxoline (i.e. shared hydroxyquinoline moiety) and ni-
trofuran-based antibacterials such as nifuroxazide and nitrofurantoin
(i.e. shared bis-cyclic-N-acylhydrazone cores). Thus, we developed two
parallel series of hydroxyquinoline and nitrofuran-bearing compound 1
analogs in an effort to increase inhibitor potency against GroEL/ES and
E. coli and the ESKAPE bacteria, while reducing cytotoxicity to human
cells (a compilation of assay results for nitrofurantoin, nifuroxazide,
and lead analogs 16–21 is presented in Table 2). Initially, only the
hydroxyquinoline series was found to contain potent GroEL/ES in-
hibitors in our traditional GroEL/ES-mediated substrate refolding as-
says; however, subsequent testing in the presence of E. coli NfsB
Perhaps most importantly, the present study has identified a new
target for a large class of clinical drugs whose mode of action is am-
biguous. While we had previously identified that GroEL/ES affects si-
milar proteins and pathways as nitrofuran-containing drugs – e.g. per-
turbing protein synthesis, aerobic energy metabolism, DNA synthesis,
RNA synthesis, and cell wall synthesis – further studies are warranted to
elucidate whether targeting GroEL/ES may be a driver of their bioac-
tivities.33 While we note the limitations of having evaluated compounds
in biochemical assays using only the E. coli GroEL/ES chaperonin
system and NfsB as surrogates, a more thorough analysis against the
various chaperonin systems and nitroreductases from E. coli and the
ESKAPE bacteria is a monumental undertaking beyond the scope of the
Nifuroxazide
Nitrofurantoin
Compound 17
A.
B.
C.
100
10
1
100
10
1
100
10
1
1
3
5
7
9
11
1
3
5
7
9
11
1
3
5
7
9
11
Passage Day
Passage Day
Passage Day
Replicate 1
Replicate 2
Figure 5. Evaluating the ability of E. coli to generate resistance to nifuroxazide (A), nitrofurantoin (B), and compound 17 (C) over time. Time-course plots show the
change in EC50 values for each compound over the 12-day serial passage resistance assay (compounds tested in duplicate, as indicated by the black and white
triangles).
7

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
Nifuroxazide
Nitrofurantoin
Compound 17
A.
B.
C.
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
0.1
1
10
100
0.1
1
10
100
0.1
1
10
100
[Inhibitor] (μM)
[Inhibitor] (
μ
M)
[Inhibitor] (μM)
Sensitive E. coli
Resistant E. coli
Resistance Reconfirmation
Figure 6. Dose-response curves for nifuroxazide (A), nitrofurantoin (B), and compound 17 (C) tested against the susceptible parent E. coli (white triangle), the
maximally-resistant strains developed to the respective test compounds (black triangle), and follow-up proliferation assays for resistant strains tested after serial
passaging in the absence of test compounds to account for possible reversible inhibition mechanisms (grey triangles). Results are presented for the replicate 1 samples
from the resistance assay, with results for replicate 2 samples similar and presented in Figure S2 in the Supporting Information. In all instances, the resistant strains
were nearly equally resistant even after culturing in the absence of inhibitors for the 4x12 h passages.
present study. However, having identified this new chaperonin-tar-
geting pro-drug inhibition mechanism for nitrofuran-containing drugs,
this study provides the impetus for us to undertake such an arduous
task. We will report on our ongoing findings in future studies.
preparatory HPLC purification, samples were chromatographically se-
parated using a Waters XSelect CSH C18 OBD prep column (part
number 186005422, 130 Å pore size, 5 µm particle size, 19x150 mm),
eluting with a H2O:CH3CN gradient solvent system. Linear gradients
were run from either 100:0, 80:20, or 60:40 A:B to 0:100 A:B (A = 95:5
H2O:CH3CN, 0.05% TFA; B = 5:95 H2O:CH3CN, 0.05% TFA. Products
from normal-phase separations were concentrated directly, and reverse-
phase separations were concentrated, diluted with H2O, frozen, and
lyophilized. For primary compound purity analyses (HPLC-1), samples
were chromatographically separated using a Waters XSelect CSH C18
column (part number 186005282, 130 Å pore size, 5 µm particle size,
3.0x150 mm), eluting with the above H2O:CH3CN gradient solvent
systems. For secondary purity analyses (HPLC-2) of final test com-
pounds, samples were chromatographically separated using a Waters
XBridge C18 column (part number 186003132, 130 Å pore size, 5.0 µm
particle size, 3.0x100 mm), eluting with a H2O:MeOH gradient solvent
system. Linear gradients were run from either 100:0, 80:20, 60:40, or
20:80 A:B to 0:100 A:B (A = 95:5 H2O:MeOH, 0.05% TFA; B = 5:95
H2O:MeOH, 0.05% TFA). Test compounds were found to be > 95% in
purity from both RP-HPLC analyses. Mass spectrometry data were
4. Materials & methods
4.1. General synthetic Methods
Unless otherwise stated, all chemicals were purchased from com-
mercial suppliers and used without further purification. Reaction pro-
gress was monitored by thin-layer chromatography on silica gel 60
F254 coated glass plates (EM Sciences). Flash chromatography was
performed using a Biotage Isolera One flash chromatography system
and eluting through Biotage KP-Sil Zip or Snap silica gel columns for
normal-phase separations (hexanes:EtOAc gradients), or Snap KP-C18-
HS columns for reverse-phase separations (H2O:MeOH gradients).
Reverse-phase high-performance liquid chromatography (RP-HPLC)
was performed using a Waters 1525 binary pump, 2489 tunable UV/Vis
detector (254 and 280 nm detection), and 2707 autosampler. For
Nifuroxazide resistant E. coli
Nitrofurantoin resistant E. coli
Compound 17 resistant E. coli
A.
B.
C.
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
0.1
1
10
100
0.1
1
10
100
0.1
1
10
100
[Inhibitor] (μM)
[Inhibitor] (μM)
[Inhibitor] (μM)
Nifuroxazide
Nitrofurantoin
Compound 17
Figure 7. Evaluation of cross-resistance between the respective resistant E. coli strains with nifuroxazide, nitrofurantoin, and compound 17. The three panels show
dose–response curves for the three inhibitors tested against strains where resistance was initially generated to nifuroxazide (A), nitrofurantoin (B), and compound 17
(C). Results are presented for the replicate 1 samples from the resistance assay, with results for replicate 2 samples similar and presented in Figure S3 in the
Supporting Information. Results indicate that resistance generated to one inhibitor is not necessarily cross-resistant to the other inhibitors, potentially indicating
different mechanisms of activation and/or targets.
8

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
Table 2
Cell-based and biochemical EC50, CC50, and IC50 results for the top eight lead inhibitors based on average Selectivity Indices (SI) for inhibiting E. coli proliferation
over cytotoxicity to human colon and intestine cells. For the GroEL/ES-dMDH refolding assay and native MDH counter-screens, IC50 results are shown for compounds
tested in the absence and presence of NfsB nitroreductase (w/ and w/o NR, respectively).
Cell-Based Assay EC50 & CC50 (μM)
Biochemical Assay IC50 (μM)
Native MDH GroEL/ES-dMDH refolding:
w/o NR w/ NR
SI (CC50 / EC50
)
Bacterial Proliferation Human Cell Viability
Compound Structure & Name / Number
16
Colon
Intestine
E
S
K
A
P
E
E. coli
Colon
Intestine
w/o NR
w/ NR
24
7.1
16
82
>100
40
0.42
83
71
198
168
>63
>63
>56
>56
>100
7.7
17
12
38
3.0
27
19
40
72
>100
49
36
0.45
0.69
53
85
119
190
88
3.2
84
Nitrofurantoin
>100 >100
>100 >100
>100
>100
>146
>146
>63
>56
>100
Nifuroxazide
8.1
22
16
37
54
0.87
1.9
61
79
75
70
91
40
>63
>63
>56
>56
>100
69
19
10
20
8.1
>100 >100 >100 >100
>100
>53
18
19
31
13
10
36
42
>100 >100 >100
2.1
2.4
>100
61
94
90
>48
26
45
38
>63
>63
>56
>56
>100
>100
12
23
9.5
54
>100
73
21
>100
34
>100 >100 >100 >100
10.0
>100
93
>10
9.3
>63
>56
26
12
collected using either Agilent LC 1200-MS 6130 or Agilent LC 1290-MS
6545 Q-TOF analytical LC-MS instruments at the IU Chemical Genomics
Core Facility (CGCF). 1H NMR spectra were recorded on a Bruker
300 MHz spectrometer at the IU CGCF. Chemical shifts are reported in
parts per million and calibrated to the d6-DMSO solvent peaks at
2.50 ppm. A representative synthesis of all analogs is presented below
in the context of compound 1. 1H NMR, MS, and HPLC characterization
data are presented in the Supporting Information for the remaining
analogs.
cloacae (Jordan) Hormaeche and Edwards strain CDC 442–68 (ATCC
#13047). For protein expression and purification, NEB 5-alpha and
BL21 (DE3) E. coli cells were purchased from New England Biolabs, and
Rosetta™ 2 (DE3) E. coli cells were purchased from EMD Millipore. The
human cell viability assays used FHC colon cells (CRL-1831) and FHs 74
Int small intestine cells (CCL-241) obtained from the ATCC. Ampicillin
was used at a concentration of 50 μg/mL, when appropriate.
1: N'-((8-hydroxyquinolin-5-yl)methylene)-4-methoxybenzohy-
drazide. To a stirring mixture of 4-methoxybenzoic acid hydrazide
(668 mg, 4.02 mmol) and 8-hydroxyquinoline-5-carbaldehyde (583 mg,
3.37 mmol) was added a catalytic amount of HCl (0.09 mL of a 4 N
solution in 1,4-dioxane, 0.36 mmol) in 10 mL of DMSO, then the re-
action was left to stir at rt overnight. The following day, the reaction
was diluted with distilled water and the precipitate was filtered, rinsed
with distilled water, collected, and dried to afford 1 as a yellow solid
(1.07 g, 99% yield). 1H NMR (300 MHz, d6-DMSO) δ 11.70 (s, 1H),
10.44 (br s, 1H), 9.61 (d, J = 8.6 Hz, 1H), 8.94 (d, J = 2.8 Hz, 1H),
8.81 (s, 1H), 7.90–8.01 (m, 2H), 7.68–7.82 (m, 2H), 7.17 (d,
J = 8.1 Hz, 1H), 7.04–7.13 (m, 2H), 3.85 (s, 3H); MS (ESI) C18H16N3O3
[MH]+ m/z expected = 322.1, observed = 322.1; HPLC-1 = 99%;
HPLC-2 = 99%.
6. Expression and purification of E. Coli GroEL and GroES
proteins.
E. coli GroEL and GroES were expressed and purified as previously
reported.55–59,71,84,86,87 Detailed protocols for these protein purifica-
tions are presented in the Supporting Information.
7. Evaluating compounds for inhibition in the GroEL/ES-mediated
dMDH refolding assay.
All compounds were evaluated for inhibiting E. coli GroEL/ES-
mediated refolding of the denatured MDH reporter enzyme as per our
previously reported procedure.55–59,71,84 A detailed protocol for this
assay is presented in the Supporting Information. Results presented
represent the averages of IC50 values obtained from at least four re-
plicates.
5. General materials and methods for biochemical & cell-based
experiments.
The bacterial proliferation assays employed the following bacterial
8. Counter-screening compounds for inhibition of native MDH
enzymatic activity.
strains: NEB 5-alpha Escherichia coli (a derivative of DH5α E. coli, New
England Biolabs #C2987H); Enterococcus faecium
- (Orla-Jensen)
Schleifer and Kilpper-Balz strain NCTC 7171 (ATCC #19434);
All compounds were counter-screened for inhibiting the enzymatic
activity of the native MDH reporter enzyme as per our previously re-
ported procedure55–59,71,84 A detailed protocol for this assay is pre-
sented in the Supporting Information. Results presented represent the
averages of IC50 values obtained from at least five replicates.
Staphylococcus aureus
- Rosenbranch strain Seattle 1945 (ATCC
#25923); Klebsiella pneumoniae - (Schroeter) Trevisan strain NCTC 9633
(ATCC #13883); Acinetobacter baumannii - Bouvet and Grimont strain
2208 (ATCC 19606); Pseudomonas aeruginosa - (Schroeter) Migula strain
NCTC 10,332 (ATCC #10145); Enterobacter cloacae - E. cloacae, subsp.
9

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
9. Evaluating compounds for inhibition in the GroEL/ES-mediated
dMDH refolding assay and native MDH activity counter-screen in
the presence of E. coli NfsB nitroreductase.
Results presented represent the averages of EC50 values obtained from
at least four replicates.
14. Evaluating compound effects on the viability of human colon
and small intestine cells.
These assays were performed nearly identically to those mentioned
above, but with a couple distinct modifications (detailed procedures are
presented in the Supporting Information). For both assays, 10 μg/mL of
the E. coli NfsB nitroreductase enzyme (Sigma product #N9284) was
added to the initial GroEL/ES-dMDH solution. For the GroEL/ES-dMDH
refolding assay, after stamping with compounds, 10 μL of a 2.4 mM
NADH solution was added, followed by a 10 min incubation period at
37 °C to bioactivate the nitrofuran analogs. The remainder of the assay
was conducted as in our standard protocol where no NfsB nitror-
eductase was present. With the extra 10 μL volume from the NADH
addition, this made the compound concentrations during the refolding
cycle part of the assay range from 83 µM to 38 nM (3-fold dilution
series). For the native MDH activity counter-screen in the presence of E.
coli NfsB nitroreductase, the assay was performed as described above
for the GroEL/ES-dMDH refolding assay with NfsB; however, compound
addition to the assay plates was conducted after the EDTA quench step,
followed by a 10 min incubation period at 37 °C to bioactivate the ni-
trofuran analogs. With the extra 10 μL volume from the NADH addition,
this made the compound concentrations during the enzymatic reporter
part of the assay range from 56 µM to 25 nM (3-fold dilution series).
IC50 values for the test compounds were obtained by plotting the %
inhibition results in GraphPad Prism and analyzing by non-linear re-
gression using the log(inhibitor) vs. response (variable slope) equation.
Results presented represent the averages of IC50 values obtained from at
least four replicates in each assay.
All compounds were evaluated for cytotoxicity to human colon
(FHC) and small intestine (FHs 74 Int) cells using an Alamar Blue-based
viability assay analogous to previously reported procedures.56–59,71
A
detailed protocol for the general cell viability assay used for both cell
lines is presented in the Supporting Information. Results presented re-
present the averages of CC50 values obtained from at least four re-
plicates for the FHC and three replicates for the FHs 74 Int cell lines.
15. Evaluating the ability of E. Coli to generate resistance to lead
inhibitors.
To identify potential resistance toward nifuroxazide, nitrofurantoin,
and lead inhibitor 17, a liquid culture, 12-day serial passage assay was
employed as per previously reported procedures, and using the NEB 5-
alpha E. coli.4,56,58,88 A detailed protocol for this assay is presented in
the Supporting Information.
16. Control compounds, calculation of IC50 / EC50 / CC50 values,
and statistical considerations.
For all assays, DMSO was used as negative control and a panel of our
previously discovered and reported chaperonin inhibitors were used as
positive controls: e.g. compounds 8, 9, and 18 from Johnson et. al 2014
and Abdeen et. al 2016;55,57 suramin and compound 2 h-p from Abdeen
et. al 2016;59 compounds 20R, 20L, and 28R from Abdeen et. al 2018;56
and closantel and rafoxanide from Kunkle et. al 2018.58 Bacterial pro-
liferation assays also included antibiotic controls such as vancomycin,
daptomycin, and rifampicin.56,58 All IC50 / EC50 / CC50 results reported
are averages of values determined from individual dose–response
curves in assay replicates as follows: 1) individual I/E/CC50 values from
assay replicates were first log-transformed and the average log(I/E/
CC50) values and standard deviations (SD) calculated; 2) replicate log(I/
E/CC50) values were evaluated for outliers using the ROUT method in
GraphPad Prism (Q of 10%); and 3) average I/E/CC50 values were then
back-calculated from the average log(I/E/CC50) values. To compare log
(IC50) values between different assays, two-tailed Spearman correlation
analyses were performed using GraphPad Prism (95% confidence level).
For compounds where log(I/E/CC50) values were greater than the
maximum compound concentrations tested (i.e. > 1.75, > 1.80, >
1.92, > 2.00, and > 2.40 – or > 56, > 63, > 83, > 100, and >
250 µM, respectively), results were represented as 0.1 log units higher
than the maximum concentrations tested (i.e. 1.85, 1.90, 2.02, 2.10,
and 2.50 – or 71, 79, 105, 126, and 316 µM, respectively) so as not to
overly bias results because of the unavailability of definitive values for
inactive compounds.
10. Evaluating compounds for inhibition in the GroEL/ES-
mediated dRho refolding assay.
All compounds were evaluated for inhibiting E. coli GroEL/ES-
mediated refolding of the denatured Rho reporter enzyme as per our
previously reported procedure.56–59,71,84 A detailed protocol for this
assay is presented in the Supporting Information. Results presented
represent the averages of IC50 values obtained from at least four re-
plicates.
11. Counter-screening compounds for inhibition of native
rhodanese enzymatic activity.
All compounds were counter-screened for inhibiting the enzymatic
activity of the native Rho reporter enzyme as per our previously re-
ported procedure.56–59,71,84 A detailed protocol for this assay is pre-
sented in the Supporting Information. Results presented represent the
averages of IC50 values obtained from at least four replicates.
12. Evaluating compounds for inhibition in the GroEL-mediated
ATPase assay.
All compounds were evaluated for inhibiting E. coli GroEL-mediated
ATPase activity as per our previously reported procedure although
employing only GroEL in solution and not containing GroES or dena-
tured MDH.55,59,84 A detailed protocol for this assay is presented in the
Supporting Information. Results presented represent the averages of
IC50 values obtained from at least four replicates.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
13. Evaluating compounds for inhibition of bacterial cell
proliferation.
Acknowledgments
Research reported in this publication was supported by the National
Institute of General Medical Sciences (NIGMS) of the National Institutes
of Health (NIH) under Award Number R01GM120350. QQH and YP
additionally acknowledge support by NIH grants 5R01GM111639 and
5R01GM115844. The content is solely the responsibility of the authors
All compounds were evaluated for inhibiting the proliferation of E.
coli and each of the ESKAPE bacteria as per previously reported pro-
cedures.56,58 A detailed protocol for the general bacterial growth assay
used for each bacterium is presented in the Supporting Information.
10

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
and does not necessarily represent the official views of the NIH. This
work was also supported in part by startup funds from the IU School of
Medicine (SMJ) and the University of Arizona (EC).
32. Fayet O, Ziegelhoffer T, Georgopoulos C. The groES and groEL heat shock gene
products of Escherichia coli are essential for bacterial growth at all temperatures. J.
Bacteriol. 1989;171:1379–1385.
33. Chapman E, Farr GW, Usaite R, et al. Global aggregation of newly translated proteins
in an Escherichia coli strain deficient of the chaperonin GroEL. Proc. Natl. Acad. Sci.
U.S.A. 2006;103:15800–15805.
Appendix A. Supplementary data
34. Bittner AN, Foltz A, Oke V. Only one of five groEL genes is required for viability and
successful symbiosis in Sinorhizobium meliloti. J. Bacteriol. 2007;189:1884–1889.
35. Kong TH, Coates ARM, Butcher PD, Hickman CJ, Shinnick TM. Mycobacterium-
Tuberculosis Expresses 2 Chaperonin-60 Homologs. Proc. Natl. Acad. Sci. U.S.A.
1993;90:2608–2612.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmc.2020.115710.
36. Hu YM, Henderson B, Lund PA, et al. A Mycobacterium tuberculosis mutant lacking
the groEL homologue cpn60.1 is viable but fails to induce an inflammatory response
in animal models of infection. Infect. Immun. 2008;76:1535–1546.
37. Qamra R, Mande SC, Coates AR, Henderson B. The unusual chaperonins of
Mycobacterium tuberculosis. Tuberculosis (Edinb). 2005;85:385–394.
38. Qamra R, Srinivas V, Mande SC. Mycobacterium tuberculosis GroEL homologues
unusually exist as lower oligomers and retain the ability to suppress aggregation of
substrate proteins. J. Mol. Biol. 2004;342:605–617.
References
1. WHO. No time to wait: Securing the future from drug-resistant infections; World
Health Organization: 2019; p 28.
2. Centers for Disease Control and Prevention. Antibiotic resistance threats in the
United States. Centers for Disease Control and Prevention: Atlanta. Georgia, USA.
2019;2019:148.
39. Hu YM, Coates ARM, Liu A, Lund PA, Henderson B. Identification of the monocyte
activating motif in Mycobacterium tuberculosis chaperonin 60.1. Tuberculosis.
2013;93:442–447.
3. Pendleton JN, Gorman SP, Gilmore BF. Clinical relevance of the ESKAPE pathogens.
Expert Rev Anti-Infe. 2013;11:297–308.
4. Fleeman R, LaVoi TM, Santos RG, et al. Combinatorial libraries as a tool for the
discovery of novel, broad-spectrum antibacterial agents targeting the ESKAPE pa-
thogens. J. Med. Chem. 2015;58:3340–3355.
40. Fan M, Rao T, Zacco E, et al. The unusual mycobacterial chaperonins: evidence for in
vivo oligomerization and specialization of function. Mol. Microbiol.
2012;85:934–944.
5. Karlowsky JA, Hoban DJ, Hackel MA, Lob SH, Sahm DF. Antimicrobial susceptibility
of Gram-negative ESKAPE pathogens isolated from hospitalized patients with intra-
abdominal and urinary tract infections in Asia-Pacific countries: SMART 2013–2015.
J. Med. Microbiol. 2017;66:61–69.
41. Henderson B, Lund PA, Coates AR. Multiple moonlighting functions of mycobacterial
molecular chaperones. Tuberculosis (Edinb). 2010;90:119–124.
42. Lund PA. Multiple chaperonins in bacteria – why so many? FEMS Microbiol. Rev.
2009;33:785–800.
6. Rice LB. Federal funding for the study of antimicrobial resistance in nosocomial
pathogens: No ESKAPE. J. Infect. Dis. 2008;197:1079–1081.
43. Horwich AL, Farr GW, Fenton WA. GroEL-GroES-mediated protein folding. Chem.
Rev. 2006;106:1917–1930.
7. Santajit S, Indrawattana N. Mechanisms of antimicrobial resistance in ESKAPE pa-
thogens. Biomed. Res. Int. 2016;2016:2475067.
44. Horwich AL, Fenton WA. Chaperonin-assisted protein folding: a chronologue. Q Rev
Biophys. 2020;53:e4.
8. Boucher HW, Talbot GH, Bradley JS, et al. Bad bugs, no drugs: No ESKAPE! An
update from the Infectious Diseases Society of America. Clin. Infect. Dis.
2009;48:1–12.
45. Horwich AL, Fenton WA, Chapman E, Farr GW. Two families of chaperonin: phy-
siology and mechanism. Annu. Rev. Cell. Dev. Biol. 2007;23:115–145.
46. Horwich AL, Low KB, Fenton WA, Hirshfield IN, Furtak K. Folding in-Vivo of
Bacterial Cytoplasmic Proteins - Role of Groel. Cell. 1993;74:909–917.
47. Braig K, Otwinowski Z, Hegde R, et al. The crystal structure of the bacterial cha-
peronin GroEL at 2.8 A. Nature. 1994;371:578–586.
9. Boucher H, Miller LG, Razonable RR. Serious infections caused by methicillin-re-
sistant Staphylococcus aureus. Clin. Infect. Dis. 2010;51(Suppl 2):S183–S197.
10. Boucher HW, Talbot GH, Benjamin DK, et al. 10 x '20 Progress-Development of New
Drugs Active Against Gram-Negative Bacilli: An Update From the Infectious Diseases
Society of America. Clin Infect Dis. 2013;56:1685–1694.
48. Sigler PB, Xu Z, Rye HS, Burston SG, Fenton WA, Horwich AL. Structure and function
in GroEL-mediated protein folding. Annu. Rev. Biochem. 1998;67:581–608.
49. Fenton WA, Kashi Y, Furtak K, Horwich AL. Residues in chaperonin GroEL required
for polypeptide binding and release. Nature. 1994;371:614–619.
50. Fenton WA, Horwich AL. GroEL-mediated protein folding. Protein Sci.
1997;6:743–760.
11. Lewis K. Platforms for antibiotic discovery. Nat. Rev. Drug Discov. 2013;12:371–387.
12. World Health Organization. Treatment of Tuberculosis: Guidelines. 4th ed. ed.;
World Health Organization: Geneva, Switzerland, 2010; p 160.
13. World Health Organization. Antibacterial agents in preclinical development: an open
access database. World Health Organization: Geneva, Switzerland, 2019; p 20.
14. Theuretzbacher U, Outterson K, Engel A, Karlen A. The global preclinical anti-
bacterial pipeline. Nat. Rev. Microbiol. 2020;18:275–285.
51. Saibil HR, Fenton WA, Clare DK, Horwich AL. Structure and allostery of the cha-
peronin GroEL. J. Mol. Biol. 2013;425:1476–1487.
52. Weissman JS, Hohl CM, Kovalenko O, et al. Mechanism of GroEL action: productive
release of polypeptide from a sequestered position under GroES. Cell.
1995;83:577–587.
15. Landman D, Georgescu C, Martin DA, Quale J. Polymyxins revisited. Clin. Microbiol.
Rev. 2008;21:449–465.
16. Poirel L, Jayol A, Nordmann P. Polymyxins: Antibacterial Activity, Susceptibility
Testing, and Resistance Mechanisms Encoded by Plasmids or Chromosomes. Clin.
Microbiol. Rev. 2017;30:557–596.
53. Weissman JS, Rye HS, Fenton WA, Beechem JM, Horwich AL. Characterization of the
active intermediate of a GroEL-GroES-mediated protein folding reaction. Cell.
1996;84:481–490.
17. Chiappori F, Fumian M, Milanesi L, Merelli I. DnaK as antibiotic target: Hot spot
residues analysis for differential inhibition of the bacterial protein in comparison
with the human HSP70. PLoS ONE. 2015;10:e0124563.
54. Xu Z, Horwich AL, Sigler PB. The crystal structure of the asymmetric GroEL-GroES-
(ADP)7 chaperonin complex. Nature. 1997;388:741–750.
55. Johnson SM, Sharif O, Mak PA, et al. A biochemical screen for GroEL/GroES in-
hibitors. Bioorg. Med. Chem. Lett. 2014;24:786–789.
18. Arita-Morioka K, Yamanaka K, Mizunoe Y, Ogura T, Sugimoto S. Novel strategy for
biofilm inhibition by using small molecules targeting molecular chaperone DnaK.
Antimicrob. Agents Chemother. 2015;59:633–641.
56. Abdeen S, Kunkle T, Salim N, et al. Sulfonamido-2-arylbenzoxazole GroEL/ES
Inhibitors as Potent Antibacterials against Methicillin-Resistant Staphylococcus
aureus (MRSA). J. Med. Chem. 2018;61:7345–7357.
19. Sass P, Josten M, Famulla K, et al. Antibiotic acyldepsipeptides activate ClpP pep-
tidase to degrade the cell division protein FtsZ. Proc. Natl. Acad. Sci. U.S.A.
2011;108:17474–17479.
57. Abdeen S, Salim N, Mammadova N, et al. GroEL/ES inhibitors as potential anti-
biotics. Bioorg. Med. Chem. Lett. 2016;26:3127–3134.
20. Evans CG, Chang L, Gestwicki JE. Heat shock protein 70 (hsp70) as an emerging drug
target. J. Med. Chem. 2010;53:4585–4602.
58. Kunkle T, Abdeen S, Salim N, et al. Hydroxybiphenylamide GroEL/ES Inhibitors Are
Potent Antibacterials against Planktonic and Biofilm Forms of Staphylococcus
aureus. J. Med. Chem. 2018;61:10651–10664.
21. Piper PW, Millson SH. Spotlight on the microbes that produce heat shock protein 90-
targeting antibiotics. Open Biol. 2012;2:120138.
59. Abdeen S, Salim N, Mammadova N, et al. Targeting the HSP60/10 chaperonin sys-
tems of Trypanosoma brucei as a strategy for treating African sleeping sickness. Bioorg.
Med. Chem. Lett. 2016;26:5247–5253.
22. Neckers L, Tatu U. Molecular Chaperones in Pathogen Virulence: Emerging New
Targets for Therapy. Cell Host Microbe. 2008;4:519–527.
23. Kragol G, Lovas S, Varadi G, et al. The antibacterial peptide pyrrhocoricin inhibits
the ATPase actions of DnaK and prevents chaperone-assisted protein folding.
Biochemistry. 2001;40:3016–3026.
60. Galloway SM, Raetz CR. A mutant of Escherichia coli defective in the first step of
endotoxin biosynthesis. The Journal of biological chemistry. 1990;265:6394–6402.
61. Vuorio R, Vaara M. The lipid A biosynthesis mutation lpxA2 of Escherichia coli re-
sults in drastic antibiotic supersusceptibility. Antimicrob Agents Chemother.
1992;36:826–829.
24. Kumar A, Balbach J. Targeting the molecular chaperone SlyD to inhibit bacterial
growth with a small molecule. Sci. Rep. 2017;7:42141.
25. Lupoli TJ, Vaubourgeix J, Burns-Huang K, Gold B. Targeting the Proteostasis
Network for Mycobacterial Drug Discovery. ACS Infect. Dis. 2018;4:478–498.
26. Hartl FU, Bracher A, Hayer-Hartl M. Molecular chaperones in protein folding and
proteostasis. Nature. 2011;475:324–332.
62. Abouelhassan Y, Yang Q, Yousaf H, et al. Nitroxoline: a broad-spectrum biofilm-
eradicating agent against pathogenic bacteria. Int J Antimicrob Agents.
2017;49:247–251.
63. Thota S, Rodrigues DA, Pinheiro PSM, Lima LM, Fraga CAM, Barreiro EJ. N-
Acylhydrazones as drugs. Bioorg. Med. Chem. Lett. 2018;28:2797–2806.
64. Wagenlehner FM, Munch F, Pilatz A, et al. Urinary concentrations and antibacterial
activities of nitroxoline at 250 milligrams versus trimethoprim at 200 milligrams
against uropathogens in healthy volunteers. Antimicrob. Agents Chemother.
2014;58:713–721.
27. Horwich AL. Molecular chaperones in cellular protein folding: the birth of a field.
Cell. 2014;157:285–288.
28. Gestwicki JE, Shao H. Inhibitors and chemical probes for molecular chaperone net-
works. J. Biol. Chem. 2019;294:2151–2161.
29. Kumar CMS, Mande SC, Mahajan G. Multiple chaperonins in bacteria-novel functions
and non-canonical behaviors. Cell stress & chaperones. 2015;20:555–574.
30. Lund PA. Microbial molecular chaperones. Adv Microb Physiol. 2001;44:93–140.
31. Buchner J. Molecular chaperones and protein quality control: an introduction to the
JBC Reviews thematic series. J. Biol. Chem. 2019;294:2074–2075.
65. Sobke A, Klinger M, Hermann B, et al. The urinary antibiotic 5-nitro-8-hydro-
xyquinoline (Nitroxoline) reduces the formation and induces the dispersal of
Pseudomonas aeruginosa biofilms by chelation of iron and zinc. Antimicrob. Agents
Chemother. 2012;56:6021–6025.
11

M. Stevens, et al.
Bioorganic&MedicinalChemistry28(2020)115710
66. Kresken M, Korber-Irrgang B. In vitro activity of nitroxoline against Escherichia coli
urine isolates from outpatient departments in Germany. Antimicrob. Agents
Chemother. 2014;58:7019–7020.
77. Prachayasittikul V, Prachayasittikul S, Ruchirawat S, Prachayasittikul V. 8-
Hydroxyquinolines: a review of their metal chelating properties and medicinal ap-
plications. Drug Des Devel Ther. 2013;7:1157–1178.
67. Fuchs F, Wille J, Hamprecht A, et al. In vitro activity of mecillinam and nitroxoline
against Neisseria gonorrhoeae - re-purposing old antibiotics in the multi-drug re-
sistance era. J. Med. Microbiol. 2019;68:991–995.
78. Pelletier C, Prognon P, Bourlioux P. Roles of divalent cations and pH in mechanism of
action of nitroxoline against Escherichia coli strains. Antimicrob. Agents Chemother.
1995;39:707–713.
68. Muller AE, Verhaegh EM, Harbarth S, Mouton JW, Huttner A. Nitrofurantoin's effi-
cacy and safety as prophylaxis for urinary tract infections: a systematic review of the
literature and meta-analysis of controlled trials. Clin. Microbiol. Infect.
2017;23:355–362.
79. Nepali K, Lee HY, Liou JP. Nitro-Group-Containing Drugs. J Med Chem.
2019;62:2851–2893.
80. Whiteway J, Koziarz P, Veall J, et al. Oxygen-insensitive nitroreductases: analysis of
the roles of nfsA and nfsB in development of resistance to 5-nitrofuran derivatives in
Escherichia coli. J. Bacteriol. 1998;180:5529–5539.
69. Johnson SM, Petrassi HM, Palaninathan SK, et al. Bisaryloxime ethers as potent in-
hibitors of transthyretin amyloid fibril formation. J Med Chem. 2005;48:1576–1587.
70. Rando DG, Avery MA, Tekwani BL, Khan SI, Ferreira EI. Antileishmanial activity
screening of 5-nitro-2-heterocyclic benzylidene hydrazides. Bioorg Med Chem.
2008;16:6724–6731.
81. Kitamura S, Narai N, Tatsumi K. Studies on bacterial nitroreductases. Enzymes in-
volved in reduction of aromatic nitro compounds in Escherichia coli. J.
Pharmacobiodyn. 1983;6:18–24.
82. Bryant DW, McCalla DR, Leeksma M, Laneuville P. Type I nitroreductases of
Escherichia coli. Can. J. Microbiol. 1981;27:81–86.
71. Washburn A, Abdeen S, Ovechkina Y, et al. Dual-targeting GroEL/ES chaperonin and
protein tyrosine phosphatase B (PtpB) inhibitors: A polypharmacology strategy for
treating Mycobacterium tuberculosis infections. Bioorg Med Chem Lett.
2019;29:1665–1672.
83. Roldan MD, Perez-Reinado E, Castillo F, Moreno-Vivian C. Reduction of poly-
nitroaromatic compounds: the bacterial nitroreductases. FEMS Microbiol. Rev.
2008;32:474–500.
72. Lopes AB, Miguez E, Kummerle AE, Rumjanek VM, Fraga CA, Barreiro EJ.
Characterization of amide bond conformers for a novel heterocyclic template of N-
acylhydrazone derivatives. Molecules. 2013;18:11683–11704.
73. Dinh NH, Tuyet Lan HT, Thu Trang TT, Van Hoan P. Synthesis and NMR
Spectroscopic Characteristics of a Series of Hydrazide-Hydrazones Containing
Furoxan Ring Derived from Isoeugenoxyacetic Acid. J Heterocycl Chem.
2012;49:814–822.
84. Stevens M, Abdeen S, Salim N, et al. HSP60/10 chaperonin systems are inhibited by a
variety of approved drugs, natural products, and known bioactive molecules. Bioorg
Med Chem Lett. 2019;29:1106–1112.
85. Race PR, Lovering AL, Green RM, et al. Structural and mechanistic studies of
Escherichia coli nitroreductase with the antibiotic nitrofurazone. Reversed binding
orientations in different redox states of the enzyme. J. Biol. Chem.
2005;280:13256–13264.
74. Wiegand I, Hilpert K, Hancock RE. Agar and broth dilution methods to determine the
minimal inhibitory concentration (MIC) of antimicrobial substances. Nat. Protoc.
2008;3:163–175.
86. Chapman E, Farr GW, Fenton WA, Johnson SM, Horwich AL. Requirement for
binding multiple ATPs to convert a GroEL ring to the folding-active state. Proc Natl
Acad Sci USA. 2008;105:19205–19210.
75. Miethke M, Marahiel MA. Siderophore-based iron acquisition and pathogen control.
Microbiol Mol Biol Rev. 2007;71:413–451.
87. Chapman E, Farr GW, Furtak K, Horwich AL. A small molecule inhibitor selective for
a variant ATP-binding site of the chaperonin GroEL. Bioorg Med Chem Lett.
2009;19:811–813.
76. du Moulinet d'Hardemare A, Gellon G, Philouze C, Serratrice G. Oxinobactin and
sulfoxinobactin, abiotic siderophore analogues to enterobactin involving 8-hydro-
xyquinoline subunits: thermodynamic and structural studies. Inorg Chem.
2012;51:12142–12151.
88. Kim S, Lieberman TD, Kishony R. Alternating antibiotic treatments constrain evo-
lutionary paths to multidrug resistance. Proc. Natl. Acad. Sci. U.S.A.
2014;111:14494–14499.
12