Optimization of LpxC Inhibitor Lead Compounds Focusing on Efficacy and Formulation for High Dose Intravenous Administration

Article abstract of DOI:10.1021/acs.jmedchem.9b01605

LpxC inhibitors were optimized starting from lead compounds with limited efficacy and solubility and with the goal to provide new options for the treatment of serious infections caused by Gram-negative pathogens in hospital settings. To enable the development of an aqueous formulation for intravenous administration of the drug at high dose, improvements in both solubility and antibacterial activity in vivo were prioritized early on. This lead optimization program resulted in the discovery of compounds such as 13 and 30, which exhibited high solubility and potent efficacy against Gram-negative pathogens in animal infection models.

Full text of DOI:10.1021/acs.jmedchem.9b01605

Subscriber access provided by EDINBURGH UNIVERSITY LIBRARY | @ http://www.lib.ed.ac.uk
Article
Optimization of LpxC inhibitor lead compounds focusing on
efficacy and formulation for high dose intravenous administration
Philippe Panchaud, Jean-Philippe Surivet, Stefan Diethelm, Anne-Catherine Blumstein,
Jean-Christophe Gauvin, Loic Jacob, Florence Masse, Gaelle Mathieu, Azely Mirre,
Christine Schmitt, Michel Enderlin-Paput, Roland Lange, Carmela Gnerre, Swen
Seeland, Charlyse Herrmann, Hans Locher, Peter Seiler, Daniel Ritz, and Georg Rueedi
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01605 • Publication Date (Web): 05 Dec 2019
Downloaded from pubs.acs.org on December 8, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Optimization of LpxC inhibitor lead compounds
focusing on efficacy and formulation for high dose
intravenous administration
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Authors: Philippe Panchaud,* Jean-Philippe Surivet, Stefan Diethelm, Anne-Catherine
Blumstein, Jean-Christophe Gauvin, Loïc Jacob, Florence Masse, Gaëlle Mathieu,
Azely Mirre, Christine Schmitt, Michel Enderlin-Paput, Roland Lange, Carmela Gnerre,
†
Swen Seeland, Charlyse Herrmann, Hans H. Locher, Peter Seiler, Daniel Ritz and
Georg Rueedi
Idorsia Pharmaceuticals Ltd, Hegenheimermattweg 91, CH-4123 Allschwil, Switzerland
KEYWORDS: Antibacterial, formulability, Gram-negative pathogens, hydroxamic acid, iv
administration, LpxC inhibitors, solubility.
ABSTRACT: LpxC inhibitors were optimized starting from lead compounds with limited
efficacy and solubility, and with the goal to provide new options for the treatment of
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
serious infections caused by Gram-negative pathogens in hospital settings. In order to
enable the development of an aqueous formulation for intravenous administration of the
drug at high dose, improvements in both solubility and antibacterial activity in vivo were
prioritized early on. This lead optimization program resulted in the discovery of
compounds such as 13 and 30, which exhibited high solubility and potent efficacy
against Gram-negative pathogens in animal infection models.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Infections caused by multidrug-resistant (MDR) Gram-negative bacteria are nowadays
considered as one of the greatest threats to global health.1, 2 In particular, the incidence
of infections caused by resistant pathogens such as Acinetobacter, Pseudomonas
aeruginosa (Pae) and Enterobacteriaceae including Klebsiella pneumoniae (Kpn) and
Escherichia coli (Eco), is increasing at a steady and alarming pace.3, 4 Development of
novel antibiotics active against these MDR Gram-negative bacteria (GNB) is urgently
needed to provide new therapeutic options for difficult-to-treat infections, and to avoid
ACS Paragon Plus Environment
2

Page 3 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
antimicrobial resistance (AMR) jeopardizing modern medical practice such as general
intensive care, joint replacement, transplant surgery, and cancer therapy.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
LpxC has drawn attention as a promising target for new antibiotic discovery ever since
specific inhibitors exhibiting potent growth inhibition of Pae and Enterobacteriaceae
were discovered.5-7 In addition, such antibiotics with novel mechanisms of action may
overcome cross-resistance with marketed antibiotics since no LpxC inhibitors are
currently used in clinical settings. LpxC (UDP-3-O-((R)-3-hydroxymyristoyl)-N-
acetylglucosamine deacetylase; EC 3.5.1.108) is a zinc-dependent enzyme localized in
8
the cytosol of GNB. It catalyzes the first committed step in the biosynthesis of lipid A, a
major component of lipopolysaccharides (LPS), and is essential for Gram-negative
9
outer membrane integrity and cell envelope biogenesis. The gene encoding LpxC is
broadly distributed across the genomes of GNB while no homologues have been found
in genomes of Gram-positive bacteria and mammals. This offers the possibility of
developing a selective treatment, which would avoid promoting resistance development
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in Gram-positive organisms and be devoid of side effects linked to the inhibition of a
homologous human target.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
In the accompanying manuscript, we reported the discovery of new LpxC inhibitors 1-3
(Figure 1), which exhibited dose-dependent efficacy in a mouse thigh infection model.
Even though the results of these proof-of-concept studies were encouraging, daily
doses of 150 mg/kg did not lead to a reduction of bacterial load below that at treatment
start. Similarly, methylsulfone hydroxamate derivatives reported by Pfizer, such as 4,
suffered from limited free exposures in rodents, mainly due to their high plasma
11
clearance. The optimization strategy reported in this latter case focused on lowering
the human predicted dose by reducing plasma protein binding, which successfully
yielded compounds such as 5 with improved free exposures and efficacy whereas
clearance remained moderate to high.12, 13
ACS Paragon Plus Environment
4

Page 5 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
MeO S
O
MeO S
O
MeO S
2
2
2
H
H
H
N
N
N
N
OH
N
OH
N
OH
N
N
N
O
O
O
1
2
3
H N
2
(S)(S)
HO
HO
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MeO S
O
MeO S
2
2
H
H
N
N
OH
N
OH
O
O
4
5
MeO
F
Figure 1. Methylsulfone hydroxamate LpxC inhibitors reported by Idorsia (1-3) and
Pfizer (4, 5).
Compounds with low clearance in humans are usually preferred since they minimize the
therapeutic doses needed, apart from the case described above that represents an
exception. In addition, low clearance in animals is also advantageous for preclinical
development since toxicology studies are conducted at exposure levels that are
multiples of those estimated for human efficacy. The identification of compounds that
can be formulated as concentrated solutions for conducting such toxicology studies
represents an additional challenge in the context of development of antibiotics for Gram-
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
negative pathogens, where patients are typically treated in hospitals with high daily
doses intravenously.14
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Starting from the reported lead compounds 1-3, we aimed at improving antibacterial
activity in vivo while concomitantly ensuring high solubility in a vehicle adequate for
future safety studies. We took advantage of the various leads’ structural features to
follow a two-pronged approach. Starting from lead compound 1, we improved the
intrinsic solubility and formulability by using the potential of its amino moiety for
ionization. In parallel, we targeted improving efficacy and metabolic stability for lead
compounds 2 and 3, while their hydroxy group represented a potential handle to attach
a solubilizing moiety that would enable the development of a prodrug approach later on
if needed.
RESULTS AND DISCUSSION
Antibacterial activity, solubility and pKa. Compound 1 was regarded as a particularly
appealing starting point for investigating the solubility and formulability of this series of
LpxC compounds. As previously seen in X-ray structures of compounds from this
ACS Paragon Plus Environment
6

Page 7 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chemical series, the terminal substituent on the distal alkyne lies on the outside of the
LpxC enzyme, which allowed for structural modification without detrimental effects on
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
10
the enzyme binding affinity. The compound showed moderate solubility at pH 4, 7 and
9
and potent antibacterial activity with MICs of < 1 µg/mL against the three strains
tested (Table 1). With a measured logD of 0.7 at pH 7.4, 1 was relatively polar among
the compounds evaluated. The various pKa’s of the compound were determined to
investigate the impact of ionization on solubility. The primary amine turned out to be
weakly basic (pKa 6.5) and the hydroxamic acid weakly acidic (pKa 7.5). Therefore,
several ionized species exist under equilibrium at physiological pH, which may limit the
potential to formulate the compound at high concentrations. The target concentration for
formulation was set at 25 mg/mL at a pH ranging from 4.5 to 8.5 to avoid local tolerance
15
issues at the injection site. Given the pKa values of 1, the pH of a suitable formulation
would have to be <4.5 in order to allow for sufficient solubility with >99% of the
compound in the protonated state. Therefore, we decided to look for compounds which
possess a more basic amine since a higher pKa would allow complete protonation of
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the amine at a pH closer to physiological conditions, thus facilitating the preparation of
concentrated formulations.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
8

Page 9 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Exploration of side chains containing an amine function.
O
MeO2S
H
N
N
(R)
OH
N
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R
MIC (µg/mL)a
ΔTm (K)b
pKac
Solubility (µg/mL)
Cmpd
R
logD7.4
R
Eco
Kpn
Pae
Eco
Pae
pH 4 pH 7 pH 9
H N
2
1
6
0.25 0.25
0.5
0.5
0.5
16
1
10.0 11.6
10.6 12.2
11.6 13.4
0.7
1.4
2.1
-0.4
0.4
0.2
1.0
6.5
6.5
5.5
9.2
6.5
9.0
8.0
571
551
412
418
182
15
536
486
95
H
N
0.12 0.25
N
7
0.25
8
0.5
16
2
8
H2N
9.6
9.5
10.7
11.5
>730 552 >730
>810 >810 >810
>780 >780 574
HO
9
1
N
F
10
11
4
4
1
10.6 11.3
N
N
0.5
1
1
9.6
12.0
>810 384
203
12
13
N
0.5
0.5
1
2
11.1 13.2
1.1
0.3
6.2
8.4
>850 329 >850
>730 >730 >730
O
0.5
0.5
8.9
9.9
N
aE. coli ATCC 25922; K. pneumoniae T 6474, QR and ESBL; P. aeruginosa ATCC
b
c
27853. DSF on isolated enzyme, determined at 25 µM compound concentration. pKa
of the hydroxamic acid varied between 7.3 and 7.7.
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Addition of methyl groups to 1 resulted in derivatives 6 and 7, which displayed
comparable antibacterial activity despite having slightly increased binding affinity for the
enzymes (ΔTm values, determined by DSF). However, their solubility decreased in line
with the logD increase.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The introduction of an extra carbon atom in 8 resulted in a significant pKa shift with a
more basic nitrogen (pKa 9.2), which would be almost fully protonated under
physiological conditions. This was reflected in the solubility and logD measurements.
Unfortunately, compound 8 had almost completely lost its antibacterial activity even
though it had retained significant binding to the enzymes (ΔTm). It was hypothesized
that the basic primary amine may lead to a reduced uptake into the GNBs under
investigation since the MIC shifts between the WT Pae strain and the isogenic efflux
mutant were comparable for all compounds shown in Table 1 (4- to 8-fold shift, Table S-
1).
Various nitrogen-containing rings were also considered as side chains.
Hydroxypiperidine derivative 9 displayed moderate antibacterial activity against the
ACS Paragon Plus Environment
10

Page 11 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
three strains as well as very good solubility at all pHs but its pKa was identical to that of
1, thus preventing us from fully exploiting the nitrogen atom’s potential for ionization. An
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
inversion of the piperidine ring as in 10 led to a pKa increase but still resulted in a
weaker antibacterial activity. Although critical for achieving potent activity against these
three bacteria, neither the basicity nor the location of the nitrogen influenced the binding
16
17
to the enzyme. Modification with a fluorine or oxetane substituent in 11 and 12
resulted in a step-wise decrease in the pKa along with variable effects on lipophilicity
and solubility, but this did not result in recovering the targeted level of antibacterial
activity. Only 13, with its smaller azetidine ring, displayed antibacterial activity below
1 µg/mL against the three strains, was very soluble at all pHs tested and possessed a
basic pKa of 8.4, which was deemed ideal to allow sufficient solubility for developing a
concentrated formulation suitable for iv dosing. Surprisingly, it had an MIC of 0.5 µg/mL
against Pae despite having the lowest ΔTm among the compounds tested, suggesting
again that the antibacterial potency is driven not only by target potency but also by drug
accumulation in the cytoplasm.
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The potential of the azetidine side chains was further explored (Table 2). Since
amides are usual metabolites resulting from aldehyde oxidase-mediated reduction of
hydroxamic acids, we first checked whether the presence of the hydroxamic acid moiety
was indeed required for activity. As expected, the corresponding amide derivative 14 did
not show any inhibitory activity and did not bind to the enzymes. Then, the effect of the
azetidine substitution on antibacterial activity and solubility was investigated.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Exploration of the azetidine side chain substitution.
O
MeO2S
H
N
N
(R)
OH
N
O
R
ACS Paragon Plus Environment
12

Page 13 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MIC (µg/mL)a
ΔTm (K)b
Eco Pae
8.9 9.9
Solubility (µg/mL)
pKac
Cmpd
R
logD7.4
R
Eco
Kpn
Pae
pH 4 pH 7 pH 9
13
0.5
0.5
0.5
0.3
0.5
-1.0
0.5
1.1
0.3
1.0
1.4
0.6
8.4
8.4
9.4
5.5
6.7
8.0
6.7
5.2
8.5
>730 >730 >730
N
N
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4d
>32
8
>32
4
>32
1
0
0.3
9.3
>700 706
92
15
8.0
>710 >710 >710
HN
16
17
N
0.25
0.12
1
0.5
0.5
2
1
10.5 12.6
>800 128
681
O
2
8.1
7.1
11.6
10.0
>760 219 >760
N
F
F
18
19
20
21
2
>740 175
279
HN
0.06 0.25
0.06 0.25
0.5
1
10.6 12.9
10.6 13.2
>780 247 >780
>820 209 >820
>750 >750 >750
N
F
F
N
F
0.25
0.5
0.5
9.3
9.6
9.7
10.4
11.2
11.2
N
2
2
3
N
0.25 0.25
0.5
1.0
8.3
>780 398
420
2
0.5
0.5
1
1
0.5
0.5
0.1
0.7
7.9
7.8
>780 >780 >780
>830 >830 >830
N
HO
HO
24
10.3 12.1
N
N
25
1
1
1
9.7
12.4
0.3
-
>780 >780 >780
O
aE. coli ATCC 25922; K. pneumoniae T 6474, QR and ESBL; P. aeruginosa ATCC
b
c
2
7853. DSF on isolated enzyme, determined at 25 µM compound concentration. pKa
d
of the hydroxamic acid varied between 7.2 and 7.7. Amide instead of hydroxamic acid.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Unsubstituted azetidine 15 retained some level of antibacterial activity, notably against
Pae, despite the basic pKa and the high polarity as indicated by the negative logD.
Oxetane derivative 16 displayed MICs comparable to those of 13 but was significantly
less soluble at pH 7 likely due to the reduced nitrogen basicity. In comparison, the pKa
increased with the two fluorinated azetidines 17 and 18. While the former had overall
properties comparable to 16 despite its lower affinity to the enzymes, the latter had
antibacterial activity and solubility that were not satisfactory. When the fluorine was
present on the substituent of the azetidine nitrogen, as in 19, MICs improved but the
solubility at pH 7 remained limited, in line with the pKa and logD. An attempt to further
modify the antibacterial activity by modulating logD and pKa with an additional fluorine
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
atom in 20 proved unsuccessful. As expected, it led to increased logD and
concomitantly decreased pKa, but the antibacterial activity did not improve. Ethyl-
substituted analogue 21 displayed overall properties comparable to those of 13, while
isopropyl derivative 22 exhibited higher logD and lower MICs against
Enterobacteriaceae. Surprisingly, the MIC against Pae was not negatively affected by
the increased lipophilicity. Even though 22 had a pKa in the desired range, it suffered
ACS Paragon Plus Environment
14

Page 15 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
from limited solubility at pH 7 and 9. Hydroxy-containing side chains were also tolerated,
as highlighted by 23 and 24. These derivatives displayed overall properties comparable
to those of 13, though a somewhat slightly lower pKa. One advantage of such
derivatives may be the possible use of the hydroxy function as a handle to attach a
promoiety in using a prodrug approach, which could be an option to reach higher
concentrations in formulations if needed. Interestingly, the slightly less active amide 25
was highly soluble over the entire pH range tested even though it did not contain a
protonable nitrogen. In summary, the azetidine ring has been shown to be a privileged
motif that allows combining potent antibacterial activity and high solubility. Several
compounds of interest were therefore progressed into in vivo evaluation (vide infra).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Metabolic stability. The other chemical series reported in the accompanying
manuscript featured a hydroxymethyl substituent on the cyclopropyl moiety in place of
the amino group found in 1. These early leads 2 and 3 also exhibited dose-dependent
efficacy in the murine thigh infection model, but net static efficacy over 24 h was not
achieved even at relatively high doses. To improve the efficacy in this model, one could
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
focus on improving the intrinsic antibacterial activity (MIC) and/or the exposure (PK). It
had already been reported that most of the metabolism of methylsulfone derivatives
such as 4 occur on the hydroxamic acid moiety via glucuronide formation and aldehyde
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
11
oxidase-mediated reduction to the corresponding amide. In order to gain a more
detailed understanding of our compounds’ metabolism, and the impact of the solvent-
exposed groups in particular, a metabolite identification study was performed on a
compound related to 2 and possessing the same cyclopropyl substitution. Upon
incubation in suspended fresh rat hepatocytes, a glucuronide and the corresponding
amide were indeed observed as expected. The hydroxymethyl side chain also appeared
to be a metabolically labile spot, which, upon oxidation, gave the corresponding
carboxylic acid in both rat microsomes and hepatocytes. We therefore decided to
increase the sterics around the cyclopropyl moiety to block this metabolically labile site.
The effects of these modifications on antibacterial activity and metabolic stability in vitro
are reported in Table 3.
ACS Paragon Plus Environment
16

Page 17 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. Exploration of side chains containing a hydroxy group.
O
MeO2S
H
N
N
(R)
OH
N
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R
MIC (µg/mL)a
Clint,tot
Solubility (µg/mL)
pH 4 pH 7 pH 9
Cmpd
2
R
logD7.4
Eco
Kpn
Pae
RHep b
RLM c
(S)
HO
HO
0.12 0.25 0.25
19
<10
1.3
1.1
0.9
1.6
0.9
0.9
0.5
0.5
600
600
149
344
354
220
588
602
384
401
613
248
628
552
(
S)
(R)
26
0.12 0.25
0.5
0.5
13.4
8.5
55
<10
24
(R)
HO
3
0.25
1
553
HO
HO
(S)
(R)
(S)
2
7
8
0.25 0.25 0.25
<10
16
506
(
S)
2
F
0.25
0.25
0.5
0.5
0.5
1
0.5
0.5
1
13
>760
>760
>780
>780
(
R)
HO
29
30
F
(S)
13
12
HO
HO
(
R)
(
S)
10.2
<10
>780 >780
(
S)
HO
HO
(
S)
(
R)
31
0.25
1
0.5
10
<10
618
>780
(
R)
aE. coli ATCC 25922; K. pneumoniae T 6474, QR and ESBL; P. aeruginosa ATCC
b
6
2
7853. Intrinsic clearance in µL/(min*10 cells) measured in fresh rat hepatocytes in
suspension. The fraction unbound of compounds in media used for all Clint
c
investigations were all in a similar range. Intrinsic clearance in µL/(min*mg) measured
in rat liver microsomes.
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
As reported in the accompanying manuscript, the difference in the stereochemistry of
the cyclopropyl ring in 2 and 26 impacted neither the MIC nor the Cl measured in vitro
int
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
in rat hepatocytes and microsomes. However, when the hydroxymethyl group was
appended to a different position on the cyclopropane ring as in 3, the antibacterial
activity was slightly reduced and hepatocytes Cl slightly decreased together with a
int
concomitant increase in metabolic liability in microsomes. The presence of a methyl
group in 27, which displayed MICs comparable to those of 2, resulted in an increase of
about 3-fold in Cl in hepatocytes but not in microsomes, potentially indicating an
int
enhanced glucuronidation. The opposite trend was observed with an additional fluorine
atom at the same position, and both diastereomers 28 and 29 displayed a slightly lower
Cl . This additional atom had no impact on the antibacterial activity when compared to
int
2 and 26. Incorporation of an additional hydroxymethyl unit on this substituent led to 30
and 31, which, in addition to improved metabolic stability, also displayed excellent
solubility.
ACS Paragon Plus Environment
18

Page 19 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pharmacology. Hydroxy-containing derivatives with acceptable Cl values as well as
int
azetidine analogs showing the targeted antibacterial activity in vitro were assessed for
efficacy against E. coli in the neutropenic murine thigh infection model after
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
subcutaneous administration (Table 4). The fluorinated derivative 29 reduced CFU
below net stasis, i.e. below the level measured at treatment start 2 h after mice were
infected, and exhibited significantly higher efficacy in this model as compared to the
previous lead compounds 2 and 3. While the 3-fold increase in free AUC could explain
the higher efficacy compared to 2, 29 also displayed a stronger CFU reduction than 3
despite a 2-fold lower free AUC and comparable MIC. Diol 30 also displayed an
improved efficacy and a larger free AUC. With its roughly 5-fold higher free AUC, the
soluble azetidine 13 proved more efficacious than its previous amine lead compound 1.
Fluoroethyl derivative 19 displayed efficacy below net stasis, comparable to that of 13,
with the 8-fold improved MIC of 19 over 13 compensating for its 6-fold reduced free
AUC. Finally, the two hydroxy derivatives 23 and 24 revealed different results. While the
efficacy and MIC of 23 was comparable to that of 13, the free exposure of 24 was low
and did only result in marginal reduction of the bacterial burden in vivo.
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 4. Efficacy and exposure of selected derivatives in mice.
O
MeO2S
H
N
N
(R)
OH
N
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R
MICa
Efficacyb
mPPBc
(%)
AUC24hd
free AUC24h
(µg*h/mL)
Cmpd
R
(
µg/mL) (Δlog10CFU/thigh)
(µg*h/mL)
(S)
HO
2
3
0.12
0.25
0.25
0.5
+0.8
+0.8
-1.2
-0.6
+1.8
-0.8
-0.6
-0.9
+2.2
99.1
91.5
98.6
97.3
92.2
63.3
93.5
66.1
90.0
124
1.11
7.64
3.30
9.54
3.49
19.4
3.19
21.4
4.73
(S)
HO
89.5
236
HO
(S)
2
9
0
F
(S)
HO
HO
(
R)
(S)
3
347
(S)
H N
2
1
0.25
0.5
44.7
52.9
49.1
63.2
47.3
13
N
1
9
3
4
0.06
0.5
N
F
2
N
HO
2
0.5
N
HO
a
b
E. coli ATCC 25922. Mice were treated sc at 150 mg/kg starting at 2 h post infection
and sacrificed at 24 h post infection. Median Δlog CFU/thigh of animal treated with
10
c
compounds compared to CFU at treatment start is reported. Determined at 1 µM
d
compound concentration. AUC values originated from PK studies performed in mouse
dosed sc at 150 mg/kg, except for 2 for which the AUC was extrapolated from a sc
dosing at 50 mg/kg.
ACS Paragon Plus Environment
20

Page 21 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ADME properties. The compounds that exhibited interesting efficacy in the murine
infection model were further characterized in rats for their PK properties after
intravenous dosing, with the goal to investigate the correlation between in vitro and in
vivo clearances as a first step towards human PK modeling. Fluorinated cyclopropyl 29
displayed a 7-fold lower unbound clearance compared to 2 (Table 5). It showed a
proportionally higher free exposure, highlighting the positive effect of the fluorine. Diol
30 displayed the lowest clearance among all hydroxy derivatives. Among the
compounds with amine side chains, 13 had lower unbound clearance than 1 in rats.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
This was reflected in the in vitro Cl data from hepatocytes but not liver microsomes.
int
Data from hepatocytes are therefore better suited to predict in vivo clearance,
consistent with the important role played by phase II metabolism for this series of
compounds. The more lipophilic fluoroethyl derivative 19 displayed higher clearance
while the hydroxy-containing azetidine derivative 23 was comparable to 13 overall. In
general, the more metabolically stable the compound was in rat hepatocytes in vitro, the
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
lower its unbound clearance and the higher its free AUC in vivo. Given the intrinsic
clearance shown in human in vitro systems, a lower metabolic clearance is expected in
human as compared to that measured in animals.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Additional in vitro investigations on reactive metabolite formation and drug drug
interactions were performed. Upon incubation of the compounds at 10 µM with human
liver microsomes, the rates of trapping of reactive entities by dansyl glutathione were in
.
all instances lower than 100 pmol/mg h. The compounds did not inhibit any of the major
cytochromes P450 enzymes CYP3A4/2C9/2D6 (IC > 50 µM), and neither did 29, 30,
50
1, 13, and 23 activate the human pregnane X receptor (PXR) at concentration up to
30 µM (2 and 19 not tested in this latter assay).
ACS Paragon Plus Environment
22

Page 23 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 5. In vitro and rat PK parameters of selected derivatives.
O
MeO2S
H
N
N
(R)
OH
N
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R
Clint,tot
rPPB e
(Cl)f
Cl
u
free AUC
Cmpd
2
R
logD7.4
RHep a
RLMb
<10
12
HHepc
3.2
<2
HLMd
<10
<10
<10
<10
48
(%)
(mL/min*kg) (ng*h/mL)
(S)
HO
HO
1.3
0.9
0.5
0.7
0.3
1.0
0.1
19
99.2
98.3
98.2
88.9
81.0
93.3
69.9
2125 (17)
306 (5.2)
172 (3.1)
261 (29)
95 (18)
80.8
550
(
S)
(S)
2
9
0
F
(S)
13
HO
HO
(R)
S)
(
3
10.2
17
<10
14
<2
968
(
S)
H N
2
1
<2
637
13
8.6
665
>1250
<2
1777
291
N
19
15.4
7.3
<2
<10
14
567 (38)
93 (28)
N
F
23
198
<2
1774g
N
HO
a
6
Intrinsic clearance in µL/(min*10 cells) measured in fresh rat hepatocytes in
suspension. All Cl were determined at 1 µM compound concentration, and the fraction
int
b
unbound of compounds in media were all in a similar range. Intrinsic clearance in
c
6
µL/(min*mg) measured in rat liver microsomes. Intrinsic clearance in µL/(min*10 cells)
d
measured in cryopreserved human hepatocytes in suspension. Intrinsic clearance in
e
µL/(min*mg) measured in human liver microsomes. Determined at 1 µM compound
f
concentration. Rats were dosed iv at 10 mg/kg; Cl : unbound clearance calculated by
u
g
correcting total Cl (in parenthesis) with the free fraction in plasma. Calculated from iv
PK at 50 mg/kg.
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Antibacterial activity on an extended panel of strains. Five compounds were further
evaluated against a panel of recent clinical isolates, including drug-resistant pathogens,
such as fluoroquinolone- or carbapenem-resistant strains (Table 6). All compounds
showed potent activity against Enterobacteriaceae with MIC values of ≤4 µg/mL and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
90
against Pae isolates with MIC of ≤ 2µg/mL. While excellent activity was observed
90
against the respiratory-tract pathogen Haemophilus influenzae, there was no useful
activity against Acinetobacter baumannii and Moraxella catarrhalis. This is in agreement
with the published findings that in these GNB the activity of LpxC is dispensable for
18
growth. In addition, all compounds exhibited only weak binding affinity in DSF against
an isolated A. baumannii enzyme (data not shown). Similarly, it is reasonable to assume
that it is also the case for Stenotrophomonas maltophila. In summary, these compounds
have an antibacterial spectrum that would be well suited for the treatment of serious
infections caused by GNB, especially Enterobacteriaceae and P. aeruginosa.
ACS Paragon Plus Environment
24

Page 25 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 6. MIC (µg/mL) of selected derivatives and reference compounds.
90
Strains (n)
C. freundii (10)
E. aerogenes (11)
E. cloacae (10)
E. coli (21)
13
1
19
0.5
1
23
1
29
1
30
1
Ciproa
0.125
0.03
0.06
>8
Tigeb
1
Imic
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
2
2
1
2
4
1
4
4
4
4
1
1
0.25
0.125
2
1
2
2
1
0.25
0.5
>16
16
K. oxytoca (9)
4
4
4
4
≤0.016
>8
0.5
4
K. pneumoniae (21)
M. morganii (10)
P. mirabilis (10)
S. marcescens (11)
A. baumanii (15)
B. cepacia (11)
H. influenzae (9)
M. catarrhalis (10)
P. aeruginosa (16)
4
2
4
4
0.5
1
0.25
0.5
0.25
>16
4
1
0.5
2
1
≤0.016
0.125
0.25
>8
8
1
1
4
4
0.5
>16
16
1
1
0.5
>16
8
2
1
>16
16
>16
2
16
>16
0.25
0.5
>16
4
>16
>16
0.5
0.06
>16
>16
8
≤0.03 ≤0.03 ≤0.03 ≤0.03 ≤0.03
≤0.016
1
>16
2
>16
1
>16
2
>16
2
>16
2
>8
S. maltophilia (10)
>16
>16
>16
>16
>16
8
a
b
c
Ciprofloxacin. Tigecycline. Imipenem.
Formulation investigations. The identification of a formulation suitable for toxicity
studies often represents a challenge in the field of antibacterial compounds targeting
GNB since single doses of ≥1 g are required for human use, and multiples of the
predicted human exposure are needed in animals. Highly concentrated formulations are
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
therefore required since the volume that can be injected is limited and only vehicles that
do not lead to any tolerability or toxicity issues on their own can be used.14
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We decided to test our compounds at various concentrations in an aqueous solution
containing 5% mannitol as the only excipient for osmolarity adjustment and modify the
pH to maximize solubility. The lowest acceptable pH was set at pH 4.5 to avoid local
tolerance issues at the injection site. As the ionization of compounds comprising a
hydroxy-containing side chain devoid of any basic amine takes place at basic pH by
deprotonation of the hydroxamic acid, the upper pH threshold was set at pH 8.5.
Formulations trials were conducted at concentrations of 1, 5, 10 and 25-30 mg/mL, and
the highest concentration for each derivative where a clear solution was obtained is
reported in Table 7. Azetidines 13, 19, and 25 could be successfully formulated at 25-
30 mg/mL under acidic conditions. For the latter two, slightly turbid formulations were
obtained at the highest concentrations. Filtration through a 0.22 µm PVDF filter yielded
clear solutions, which showed the expected concentration after HPLC analysis. Diol 30
also turned out to be soluble at high concentration in a slightly basic formulation,
ACS Paragon Plus Environment
26

Page 27 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
whereas the limited solubility of 29 prevented clear solutions at concentration >1
mg/mL.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 7. Formulations resulting in a clear solution in 5% mannitol/water.
Cmpd
Concentration [mg/mL]
pH
4.5
4.5
4.5
8.4
8.2
13
19
23
29
30
30
25
30
1
25
CONCLUSION
We successfully completed a lead optimization program focused on improving solubility
and formulability parameters as well as antibacterial efficacy rather than simply in vitro
potency. Focusing on solubility in addition to activity early in the lead optimization phase
led to the discovery of a set of compounds fulfilling the targeted attributes, i.e., improved
antibacterial activity against GNB causing complicated urinary tract, intraabdominal
space, or lower respiratory tract infections, and improved properties for intravenous
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
administration. In addition to displaying suitable physicochemical properties enabling
the development of a formulation for intravenous administration of high doses, the
compounds exhibited improved metabolic stability resulting in lower clearance leading
to promising efficacy in an animal infection model. A few of these compounds will be
assessed further and results will be reported in due course.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CHEMISTRY
Compounds 6, 8, 9, 11, 16, 18, 24 and 27-31 were prepared by reacting alkyne 32
and halo-alkynes 33a-m in a Cadiot-Chodkiewicz cross-coupling reaction 19 followed by
cleavage of the THP protecting group under suitable acidic conditions, using either
hydrochloric acid in a 9:1 mixture of methanol and water, or PPTS in ethanol (Scheme
1
).20, 21 Alternatively, coupling alkyne 32 with iodoalkynes 33n-o in a Sonogashira
reaction gave 10, respectively 12, after deprotection.
Scheme 1. Synthesis of derivatives using Cadiot-Chodkiewicz and Sonogashira
coupling a
ACS Paragon Plus Environment
28

Page 29 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
MeO S
2
H
Hal = Br or I
Hal
O
MeO S
N
2
H
N
OH
N
N
N
OTHP
a,b/c
N
+
O
O
R
32
33a-m
6, 8, 9, 11, 16, 18, 24, 27-31
R
F
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H
HO
N
R =
H N
2
N
N
HN
F
N
O
6
8
9
11
16
18
HO
HO
HO
HO
HO
HO
(
S)
(R)
(S)
(R)
(S)
(S)
(R)
F
F
N
(S)
(R)
(S)
HO
(
R)
HO
(S)
24
27
28
29
30
31
O
MeO S
2
H
O
MeO S
N
2
H
N
N
OH
N
I
N
OTHP
d,b
N
+
O
O
R
32
33n,o
10, 12
R
R =
N
N
O
10
12
a
.
Reagents and conditions: (a) CuCl, NH OH HCl, n-BuNH , 0 °C-rt, 2 h. (b) HCl, 9:1
2
2
MeOH/H O, rt, 1 h. (c) PPTS, EtOH, 80 °C, 2 h. (d) CuI, PdCl (PPh ) , Et N, THF,
2
2
3 2
3
40 °C, 1 h.
The side chain of 7, 17 and 23 was functionalized by undergoing a reductive
amination or a deprotection reaction under standard conditions before the final
hydroxamic acid deprotection (Scheme 2).
Scheme 2. Synthesis of derivatives 7, 17 and 23 a
ACS Paragon Plus Environment
29