Diazabicyclooctane Functionalization for Inhibition of β-Lactamases from Enterobacteria

Article abstract of DOI:10.1021/acs.jmedchem.9b02125

Second-generation β-lactamase inhibitors containing a diazabicyclooctane (DBO) scaffold restore the activity of β-lactams against pathogenic bacteria, including those producing class A, C, and D enzymes that are not susceptible to first-generation inhibitors containing a β-lactam ring. Here, we report optimization of a synthetic route to access triazole-containing DBOs and biological evaluation of a series of 17 compounds for inhibition of five β-lactamases representative of enzymes found in pathogenic Gram-negative bacteria. A strong correlation (Spearman coefficient of 0.87; p = 4.7 × 10^-21) was observed between the inhibition efficacy of purified β-lactamases and the potentiation of β-lactam antibacterial activity, indicating that DBO functionalization did not impair penetration. In comparison to reference DBOs, avibactam and relebactam, our compounds displayed reduced efficacy, likely due to the absence of hydrogen bonding with a conserved asparagine residue at position 132. This was partially compensated for by additional interactions involving certain triazole substituents.

Full text of DOI:10.1021/acs.jmedchem.9b02125

Subscriber access provided by Hong Kong University of Science and Technology Library
Article
DIAZABICYCLOOCTANE FUNCTIONALIZATION FOR
INHIBITION OF #-LACTAMASES FROM ENTEROBACTERIA
Flavie Bouchet, Heiner Atze, Matthieu Fonvielle, Zainab Edoo,
Michel Arthur, Mélanie Etheve-Quelquejeu, and Laura Iannazzo
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b02125 • Publication Date (Web): 03 Apr 2020
Downloaded from pubs.acs.org on April 9, 2020
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 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DIAZABICYCLOOCTANE FUNCTIONALIZATION FOR
INHIBITION
OF
β-LACTAMASES
FROM
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
ENTEROBACTERIA
†
1
‡1
‡
‡
‡
Flavie Bouchet, Heiner Atze, Matthieu Fonvielle, Zainab Edoo, Michel Arthur, *
†
†
Mélanie Ethève-Quelquejeu, * Laura Iannazzo *
†
Laboratoire de Chimie et de Biochimie Pharmacologiques et Toxicologiques, Université de
Paris, UMR 8601, Paris, F-75006 France; CNRS UMR 8601, Paris, F-75006, France.
‡
INSERM, Sorbonne Université, Université de Paris, Centre de Recherche des Cordeliers, F-
7
5006, Paris, France.
ABSTRACT: Second-generation β-lactamase inhibitors containing a diazabicyclooctane
(DBO) scaffold restore the activity of β-lactams against pathogenic bacteria, including those
producing class A, C, and D enzymes that are not susceptible to first-generation inhibitors
containing a β-lactam ring. Here, we report optimization of a synthetic route to access triazole-
containing DBOs and biological evaluation of a series of 17 compounds for inhibition of five
β-lactamases representative of enzymes found in pathogenic Gram-negative bacteria. A strong
-21
correlation (Spearman coefficient of 0.87; p = 4.7 10 ) was observed between the inhibition
efficacy of purified β-lactamases and the potentiation of β-lactam antibacterial activity
indicating that DBO functionalization did not impair penetration. In comparison to reference
DBOs, avibactam and relebactam, our compounds displayed reduced efficacy likely due to the
absence of hydrogen bonding with a conserved asparagine residue at position 132. This was
partially compensated by additional interactions involving certain triazole substituents.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 46
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
INTRODUCTION
Penicillin is the first broadly used antibiotic introduced in medicine and the β-lactams remain
the most widely used drug family in spite of the emergence of various resistance mechanisms.1
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 targets of β-lactams are the D,D-transpeptidases, commonly referred to as penicillin-
binding proteins (PBPs), that catalyze the last cross-linking step of peptidoglycan synthesis.2
The most common β-lactam resistance mechanism in Gram-negative bacteria is the production
of one or more β-lactamases, which catalyze the hydrolysis of the amide bond of the four-
3
membered β-lactam ring (Figure 1A). The hydrolysis product is not active accounting for the
absence of inhibition of the PBP targets. This resistance mechanism has been successfully
defeated by β-lactamase inhibitors, clavulanate, sulbactam, and tazobactam (Figure 1B).3,4 In
the presence of these inhibitors, the drug (imipenem in Figure 1C) remains active for acylation
of the PBPs. First-generation inhibitors contain a β-lactam ring and act as suicide substrates
3
forming stable adducts with β-lactamases (Figure 1C). Avibactam (Figure 1B), clinically
approved in combination with ceftazidime in 2015, is the first representative of second-
generation inhibitors. Its structure is based on a distinct scaffold, diazabicyclooctane (DBO),
5
,6
which contains a five-membered cyclic urea rather than a β-lactam ring. The mechanism of
action of avibactam relies on nucleophilic attack of the DBO carbonyl by the active serine of
class A, C and D β-lactamase leading to regioselective opening of the cyclic urea and nitrogen
6
protonation (Figure 1D). This reaction affords a stable carbamoyl-enzyme adduct. In contrast
to β-lactam-containing inhibitors, formation of the carbamoyl-enzyme is fully reversible
regenerating native β-lactamase and avibactam.6-8 The efficacy of avibactam relies on the
stabilization of the adduct that displaces the equilibrium toward the covalent carbamoyl-
enzyme. Stabilization is also important to prevent carbamoyl-enzyme hydrolysis, which
proceeds through an initial desulfonation for certain β-lactamases.9-11 Compared to the first-
generation inhibitors, avibactam inhibits a wider range of β-lactamases (class A, C, and some
2
ACS Paragon Plus Environment

Page 3 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
class D enzymes). In particular, avibactam inhibits β-lactamases that are fully insensitive to
inhibition by clavulanate, such as Klebsiella pneumoniae carbapenemases (KPC) and
Mycobacterium abscessus BlaMab. Avibactam does not effectively inhibit zinc-dependent
metallo-β-lactamases (class B enzymes ) since it binds only weakly to most tested enzymes and
it is slowly hydrolyzed in certain cases.12,13 In parallel to DBOs, unrelated β-lactamase
inhibitors based on a cyclic boronic acid pharmacophore are currently developed, including
vaborbactam,14,15 recently approved by the FDA in combination with meropenem.16,17
Discovery of additional β-lactamase inhibitors, including DBOs and boronates, is needed to
broaden inhibition spectra, in particular to include inhibition of class B and D enzymes. In
addition, mutations in β-lactamase genes have emerged in vitro and under treatment by the
ceftazidime-avibactam combination.18-21 Interestingly, the potency of vaborbactam was
reported to be less affected than that of avibactam in strains producing KPC-2 carbapenemase
variants that confer resistance to the ceftazidime-avibactam combination.17,22
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
Figure 1. Structure and mode of action of antibiotics and β-lactamase inhibitors. (1A) β-
lactamase-mediated resistance to imipenem. (1B) Structure of 1 and 2nd generation β-
st
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 46
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
lactamase inhibitors based on β-lactam and diazabicyclooctane (DBO) scaffolds, respectively.
(
1C) Mode of action of a β-lactamase inhibitor (clavulanate), which prevents imipenem
hydrolysis enabling inactivation of the penicillin-binding protein (PBP) targets. (1D)
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
Avibactam-mediated -lactamase inhibition by reversible formation of a carbamoyl-enzyme.
Functionalization of the DBO scaffold has been extensively explored to improve their β-
lactamase inhibition spectra.23-31 In addition, introduction of side chains was reported to afford
dual inhibitors that act both on β-lactamases and on essential penicillin-binding proteins
2
3
required for cell wall peptidoglycan polymerization. Since DBOs are chemically challenging
molecules in terms of their structures, we recently developed a versatile method using copper-
catalyzed alkyne-azide cycloaddition (CuAAC) to introduce 1,2,3-triazole-containing
32
substituents at the C
2
position of the DBO scaffold. Here, we report modification of our
synthetic scheme (Scheme 1), which enabled synthesis of a series of compounds in sufficient
amount for biological evaluation of a panel of five representatives of class A, C, and D β-
lactamases. This evaluation involved both the determination of kinetic parameters for β-
lactamase inhibition and evaluation of antibacterial activity against isogenic strains producing
specific β-lactamases.
Scheme 1. Strategies to access functionalized DBOs
RESULTS AND DISSCUSSION
Chemical synthesis. We recently reported the synthesis of 2-azido-diazabicyclooctane 4
32
starting from commercially available protected oxopyrrolidine 1 (Scheme 1). This procedure
4
ACS Paragon Plus Environment

Page 5 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
proceeds through the synthesis of the alcohol intermediate 3 and provides access to the key
azido-intermediate 4 for the copper(I)-catalyzed Huisgen-Sharpless cycloaddition reaction33
with an overall yield of 4%. Here, we propose a modification of our approach starting with the
carboxylic acid DBO derivative 2, which has recently been made commercially available. DBO
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
2
was activated with iso-butylchloroformate and directly reduced with NaBH to afford the
4
corresponding alcohol 3 (Scheme 1). By a classical strategy, the activation of 3 with
mesylchloride followed by a nucleophilic substitution in the presence of sodium azide provided
azido derivative 4.
Scheme 2. a) CuSO
i. Pd/C, H
4
30 mol %, sodium ascorbate 60 mol %, THF/H
2
O (3/1), rt, overnight; b)
2
, MeOH, rt, 48 h, ii. SO
3
-Pyr, pyridine, rt, overnight, iii. DOWEX Na+.
The functionalization of 4 was then explored with various alkynes to afford a series of new
DBOs (Table 1). The rationale behind the study was to validate the synthetic approach for
substituents containing various functional groups. Briefly, alkynes 5a, 5b, 5c, 5d, 5e, 5h, and
5j were commercially available whereas 5f, 5g, and 5i were prepared according to standard
procedures (See supporting information). Compound 4 and alkynes 5a-j were subjected to
CuAAC conditions using CuSO
4
(30 mol %) and sodium ascorbate (60 mol %) in THF/H O
2
overnight at room temperature to afford compounds 6a-j in 30 to 97% yield (Scheme 2). A
wide range of alkynes was compatible with this procedure including compounds containing
aromatic rings, such as pyridine (5a) and phenyl (5b, 5i), Si, O, or N heteroatoms (5c-f), a
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 46
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
protected amine (5g-i), or a carboxylic acid function (5j). The deprotection of the benzyl group
of compounds 6a-j was performed by hydrogenolysis followed by sulfation with sulfur trioxide
pyridine complex. The purification of the final compounds 7a-j has been optimized by elution
+
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
through DOWEX Na followed by the dissolution of the residue in ethanol and filtration. This
additional step was helpful for an efficient purification of DBOs over HPLC and allowed
increasing the overall sulfation yield (from 10% for compounds 7a and 7c according to the
3
2
previous method to 21% and 14%, respectively, in the current study). The final deprotection
step for Boc-containing DBOs (7g, 7h, and 7i) was performed by treatment with TFA in DCM
at 0°C for 1 h (Scheme 3). Unprotected DBOs 8g, 8h, and 8i were isolated in 36, 34 and 17%
yield after HPLC purification, respectively.
Scheme 3. a) TFA, DCM, 0°C, 1 h; 8g (36%); 8h (34%); 8i (17%).
To get access to unsubstituted triazoles 10a and 10b (Scheme 4), a new strategy has been
designed. The alcohol 3 was activated with mesylchloride and treated with 1H-1,2,3-triazole
3
4
under basic conditions to afford regioisomers 9a and 9b in 44 and 37% yield, respectively.
After separation of the two regioisomers by chromatography, each compound was submitted to
hydrogenolysis conditions to remove the benzyl group followed by the sulfation step and
+
elution on DOWEX Na , as described above.
6
ACS Paragon Plus Environment

Page 7 of 46
Journal of Medicinal Chemistry
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
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
Scheme 4. a) i. MsCl, NEt
3
, DMAP, DCM, 0°C, 1h, ii. 1H-1,2,3-triazole, tBuOK, MeCN,
reflux, 15h; 9a (44%); 9b (37%); b) i. Pd/C, H
2
, MeOH, rt, 48h, ii. SO -Pyr, pyridine, rt,
3
+
overnight, iii. DOWEX Na ; 10a (27%); 10b (8%).
Table 1. Structure of alkynes used for CuAAC, resulting triazole-containing DBOs, and final
sulfated compounds
Alkyne 5a-j
Compounds 6a-j
Yield
Compounds 7a-j
Yield
[%]
[
%]
5
5
a
6a
6b
6c
N
86
7a
7b
7c
N
21
19
14
N
N
N
N
N
N
N
N
N
N
N
O
OBn
O
OSO Na
3
b
N
N
85
30
N
N
N
N
N
N
N
N
O
OBn
O
OSO3Na
5
c
N
N
Si
Si
Si
N
N
N
N
N
N
N
N
O
OBn
O
OSO3Na
5d
6d
6e
6f
N
86
87
90
7d
7e
7f
N
13
18
8
O
N
N
N
O
N
N
N
O
N
N
O
OBn
O
OSO3Na
5
e
N
N
N
N
N
N
N
N
N
N
N
N
N
O
OBn
O
OSO3Na
5f
N
N
N
N
N
N
N
N
N
N
N
O
N
N
O
O
OBn
O
O
3
OSO Na
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 46
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
5g
6g
6h
N
N
93
97
7g
7h
N
N
30
BocHN
N
N
BocHN
N
N
NHBoc
N
N
O
OBn
O
OSO3Na
5
h
BocN
N
53
23
BocN
N
N
N
N
N
N
BocN
N
N
N
O
OBn
O
OSO3Na
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
5
i
6I
6j
HN
N
91
75
7i
7j
HN
N
N
N
N
N
N
BocHN
N
BocHN
O
O
HN
N
N
O
OBn
O
3
OSO Na
O
BocHN
5
j
N
N
N
17
HO2C
N
HO2C
N
N
N
N
O
N
N
O
OBn
O
OSO3Na
OH
Rationale for cell-based assays of DBOs. We chose a panel of five β-lactamases that are
clinically relevant, belong to different structural classes, and differ by their hydrolysis
3
spectrum. KPC-2 (class A) and OXA-48 (class D) hydrolyze carbapenems (carbapenemases),
which are often the last resort antibiotics against multi-drug-resistant bacteria. CTX-M-15 and
AmpC from Enterobacter cloacae were chosen as representatives of class A extended-spectrum
β-lactamases and of class C cephalosporinases (preferential hydrolysis of cephems). TEM-1 is
a wide-spread representative of penicillinases (preferential hydrolysis of penams). Genes
encoding these five β-lactamases were cloned under the control of an inducible promoter into
the vector pTR99k and introduced in Escherichia coli Top10. The efficacy of the DBOs was
tested by determining the minimal inhibitory concentration (MIC) of β-lactams in the presence
or absence of a fixed concentration of inhibitor (15 µM corresponding to 4 µg/ml for
avibactam). Since the TEM-1, KPC-2, CTX-M-15, AmpC, and OXA-48 β-lactamases display
different substrate spectra, preliminary experiments were performed to choose a suitable
antibiotic to compare the efficacy of the inhibitors (Figure 2).
8
ACS Paragon Plus Environment

Page 9 of 46
Journal of Medicinal Chemistry
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
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
Figure 2. Structure of β-lactam antibiotics used as substrates to determine the activity of β-
lactamases TEM-1, KPC-2, CTX-M-15, AmpC, and OXA-48. The drugs belong to the cephem,
monobactam, penam, or carbapenem classes of β-lactams, as indicated in parenthesis.
For each of the five chosen β-lactam/β-lactamase combinations, β-lactamase production led to
large MIC increases, ranging from 256 to 4,096 fold. For the cefamandole/TEM-1 combination
for example, production of TEM-1 in the E. coli Top10 host resulted in a 1,024-fold increase
in the MIC of cefamandole (from 2 to 2,048 µg/ml) (Table 2). Such a high dynamic range offers
the possibility to compare inhibitors with various efficacies leading to partial inhibition of the
β-lactamases and various MICs in the 2 to 2,048 µg/ml range. Complete in vivo inhibition of
TEM-1 requires very efficacious DBOs able to lead to the 1,024-fold decrease in the MIC
needed to restore full susceptibility. For such inhibitors, the MICs of cefamandole against E.
coli TOP10 and its isogenic TEM-producing derivative should be the same (2 µg/ml). For
TEM-1 and cefamandole, this is the case for the first-generation inhibitor clavulanate (Table
2
).
Table 2. Inhibition of penicillinase TEM-1 by DBOs and clavulanate
R
MIC of cefamandole (µg/ml)
Carbamoylation
N
Without[b]
efficacy
With
N
-1 -1
O
OSO3Na
(M s )
TEM-1
TEM-1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 46
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
R =
no inhibitor
NA
2
2
2,048
2
Avibactam
Relebactam
88,000 ± 2,000
43,000 ± 1,000
1,400 ± 100
2
2
4
4
8
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
7a
7b
4,700 ± 1,000
16
7c
370 ± 50
830 ± 50
2
2
64
16
7d
7e
780 ± 80
2
32
7f
1,200 ± 100
2,800 ± 400
1,700 ± 200
2
2
2
8
7g
128
256
7h
7i
3,700 ± 500
2
256
8
8
g
h
3,100 ± 200
3,600 ± 300
4
2
4
4
[a]
[a]
8i
940 ± 150
2
32
[a]
7j
720 ± 60
2
2
32
16
10a
2,600 ± 300
10b
1,200 ± 100
NA
2
2
8
2
Clavulanate
NA, not applicable as clavulanate and DBOs have different modes of action. Kinetics constants
were deduced from a minimum of six progress curves obtained in a minimum of two
independent experiments. Reported MICs are the medians from five biological repeats obtained
in two independent experiments. [a] Zwitterionic species have been tested. [b] Avibactam alone
was moderately active (MIC = 32 µg/ml) against E. coli Top10 harboring the vector pTRC-99k
but none of our DBOs displayed activity in the absence of β-lactam (MIC >128 µg/ml).
1
0
ACS Paragon Plus Environment

Page 11 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Rationale for enzymatic inhibition assays of DBOs. Genes encoding soluble fragments of
TEM-1, KPC-2, CTX-M-15, AmpC, and OXA-48 were cloned into the pET-TEV vector. β-
lactamases were produced in E. coli BL21 and purified from clarified lysates by metal affinity
and size-exclusion chromatography. For in vitro evaluation of inhibition efficacy, purified β-
lactamases were incubated with DBOs at various concentrations with a fixed concentration of
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
7
the chromogenic cephalosporin nitrocefin (Figure 3). Estimates of the inhibition efficacy were
provided by the ratio of the kinetic constants k
constant k for the chemical step of the reaction (carbamoylation) and the dissociation constant
for the preceding formation of a non-covalent complex. Since k and K could not be
2
i
and K , which takes into account the first order
2
K
i
2
i
independently determined the changes in efficacy might arise from either or both tighter binding
and placement of the groups in a more reactive conformation in the active site.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 46
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
Figure 3. Inhibition of β-lactamases by DBOs. (A) Reaction scheme and definition of rate
constants. v
i
and v are the initial and final velocities of the reactions. (B) Time-dependent
s
inhibition of KPC-2 by 7b and 10b. KPC-2 (2.5 nM) was incubated with nitrocefin (100 µM)
and various concentrations of 7b and 10b (20-90 µM or 30-90 µM, respectively). Equation 1
was fitted to progress curves to obtain the pseudo-first order rate constant kobs. Representative
results from one experiment are shown. (C) Determination of carbamoylation rate constant
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
k
2
i
/K . kobs was plotted as a function of the DBO concentration. Equation 2 was fitted to data to
determine the value of the k
2
/K ratio and of k-2. Kinetic constant k-2 was low for all DBOs (See
i
supplementary Table S1). Kinetics constants were deduced from a minimum of six progress
curves obtained in a minimum of two independent experiments.
Efficacy of inhibition of penicillinase TEM-1 by DBOs. The two reference DBOs, avibactam
and relebactam, afforded 1,024- and 512-fold reductions in the MIC of cefamandole indicating
that inhibition of TEM-1 was complete or nearly complete, respectively. Our synthetic DBOs
offered the possibility to investigate the impact of aromatic rings, carboxylic acids, amines, and
the corresponding Boc-protected compounds on the efficacy of DBOs. Two amine-containing
compounds, 8g and 8h, were as active as relebactam, also producing a 512-fold reduction in
the MIC of cefamandole. A 256-fold reduction was observed for compounds 7a and 7f, which
contain a pyridine or a morpholine group in their side chain, respectively. The in vivo efficacy
of the remaining compounds was lower, in particular that of Boc-protected compounds 7g, 7h,
and 7i (4- or 8-fold reduction in the MIC). The in vitro efficacy (k
2
/K ) of the latter compounds
i
-1 -1
was in the 1,600 to 3,700 M s range. These values are similar to those of DBOs 8g, 8h, and
-
1 -1
8
i (3,100 to 3,700 M s range) that afforded a 512-fold reduction in the MICs of cefamandole.
The difference between the antibacterial activity of the two series of compounds (7g-7i versus
g-8i) may therefore result from a poor penetration of the Boc-containing compounds though
the outer membrane of the tested strain. To further investigate the impact of the
8
1
2
ACS Paragon Plus Environment

Page 13 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
functionalization of the triazole ring, we synthesized regioisomers 10a and 10b containing
unsubstituted triazoles. The inactivation efficacy of these compounds (k
2
/K
i
of 2,600 and 1,200
-
1 -1
M s , respectively) was similar to that of compounds containing a substituted triazole ring (up
-
1
-1
-1 -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
to 4,700 M s ), but ca. 19- and 9-fold lower than that of avibactam (88,000 M s ) and
-1 -1
relebactam (43,000 M s ). These results indicate that the triazole linkage is disfavored in
comparison to the carboxamide of avibactam or to the amide of relebactam. None of the tested
side chains fully compensated this effect in vitro. Of note, inhibition of TEM-1 by avibactam
and 10b was reversible with similar off-rate constants (Supplementary Figure S1).
Efficacy of carbapenemase inhibition by DBOs. KPC-2 was fully inhibited by the two
reference DBOs, avibactam and relebactam, affording a 4,096-fold reduction in the MIC of
aztreonam used as the reporter drug for this β-lactamase (Table 3). Clavulanate was inactive as
3
5
expected. Among the DBOs synthesized in this study, 7a, 7f, 8g, and 8h were the most active
in vivo, affording a 64-fold reduction in the MIC of aztreonam. However, the MIC of aztreonam
in the presence of these inhibitors (MIC = 32 µg/ml) was much higher than that obtained with
avibactam and relebactam (MIC = 0.5 µg/ml) indicating that KPC-2 retained substantial
hydrolytic activity when the inhibitors were added to the culture medium. The difference in the
MICs conveyed by 7a, 7f, 8g, and 8h in comparison to avibactam and relebactam correlated
-1 -1
with a >20-fold lower inhibition efficacy in vitro (k
2
/K in the 120 to 550 M s range for 7a,
i
-
1
-1
7
f, 8g, and 8h versus 26,000 and 10,000 M s for avibactam and relebactam, respectively).
Comparisons of the data obtained with DBOs containing unsubstituted triazole rings (10a and
-1 -1
1
0b; k
2
/K of 350 and 280 M s , respectively) indicate that the side chain of three DBOs, 7a,
i
7
b, and 7i had a favorable impact on the inactivation efficacy (k
2
/K
of 550, 930, and 1,100 M-
i
1
s-1, respectively). Several side chains were poorly tolerated by KPC-2, including a
-
1
-1
-1 -1
trimethylsilyl in 7c (k
2
/K = 97 M s ), a methoxy in 7d (18 M s ), and a carboxylate in 7j
i
-1 -1
(
70 M s ).
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 46
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 3. Inhibition of class A β-lactamases by DBOs and clavulanate
Inhibition of KPC-2
Inhibition of CTX-M-15
MIC of amoxicillin
(µg/ml)
Without With
R
MIC of aztreonam
N
(µg/ml)
N
Carbamoylation
efficacy
Carbamoylation
efficacy
O
OSO3Na
Without
With
CTX-M-15
CTX-M-
15
R =
no inhibitor
KPC-2
KPC-2
-1
-1
-1 -1
(
M s )
(M s )
NA
130,000 
10,000
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
NA
0.5
2,048
0.5
8
2
2,048
2
Avibactam
26,000 
,000
0.25
1
Relebacta
m
10,000 
1,000
0.25
0.5
960 
80
2
16
7
a
550 
0.25
0.25
32
64
1,700 
2
8
32
64
1
0
400
7b
930 
6,900 
2,000
40
7
c
97 
8
18 
2
0.5
0.5
512
512
64
400 
10
120 
40
29 
5
8
8
4
8
8
8
1,024
256
7d
7e
120 
0.5
256
50
7
f
190 
0.25
0.25
0.25
32
100 
10
130 
10
230 
30
64
2
0
7
7
g
h
540 
512
2,048
1,024
4,096
60
830 
90
7i
1,100 
00
0.5
512
2,400 
200
8
768
2
8
g
120 
10
340 
30
0.25
0.25
32
32
88 
13
350 
60
4
4
128
64
[
a]
8h
[a]
8
7
i
j
180 
0.25
0.25
64
150 
4
2
128
96
[a]
3
3
10
70
256
110 
10

8
10a
10b
350
 40
280
0.5
0.25
0.5
512
512
512
120 
10
280 
10
4
8
8
512
64
2

30
Clavulanat
e
NA
NA
NA, not applicable as clavulanate and DBOs have different modes of action. Kinetics constants
were deduced from a minimum of six progress curves obtained in a minimum of two
independent experiments. Reported MICs are the medians from five biological repeats obtained
in two independent experiments. [a] Zwitterionic species have been tested.
1
4
ACS Paragon Plus Environment

Page 15 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Efficacy of extended-spectrum β-lactamase inhibition by DBOs. Inhibition of CTX-M-15
was tested with amoxicillin as the reporter drug (Table 3). Avibactam fully inhibited CTX-M-
15 in vivo (MIC = 2 µg/ml) but CTX-M-15 retained partial activity in bacteria exposed to
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
relebactam (MIC = 16 µg/ml). In comparison with unsubstituted DBOs 10a and 10b (k
2
/K of
i
-1 -1
1
20 and 280 M s , respectively), the pyridine and phenyl groups present in 7a, 7b, and 7i
-
1
-1
enhanced CTX-M-15 inactivation (k
2
/K of 1,700, 6,900, and 2,400 M s , respectively).
i
However, these compounds displayed >18-fold lower in vitro efficacies in comparison to
avibactam and, accordingly, 7a and 7b did not fully restore the antibacterial activity of
amoxicillin (MIC of 32 and 64 µg/ml, respectively). Relebactam was less active than 7a and
-
1 -1
7b in vitro (960 versus 1,700 and 6,900 M s ). Accordingly, relebactam failed to fully inhibit
CTX-M-15 in vivo (MIC = 16 µg/ml). DBO 7i only afforded an 8-fold reduction in the MIC of
amoxicillin in agreement with the proposed negative impact of the Boc group on permeability
described above. Of note, four DBOs (avibactam, relebactam, 7a and 7j) slightly increased (4-
fold) the activity of amoxicillin against the E. coli TOP10 host, which does not produce CTX-
M-15. These results suggest that these DBOs may inhibit PBP targets in addition to the β-
lactamase. This would imply that the DBOs and amoxicillin may inhibit the targets in a
cooperative manner leading to a higher antibacterial activity in comparison to amoxicillin alone.
Of note, inhibition of class B PBP2 has been documented for several DBOs.23,24
Inhibition of class C and D β-lactamases by DBOs. Production of AmpC in the E. coli Top10
host led to a 2,048-fold increase in the MIC of cefotaxime (from 0.125 to 256 µg/ml). In vivo
inhibition of AmpC (class C) was partial for avibactam (MIC = 1 µg/ml) and nearly complete
for relebactam (MIC = 0.25 µg/ml). The DBOs synthesized in this study achieved moderate
MIC reduction and their in vitro inhibition efficacy was ca. 100-fold lower than that of the
reference DBOs (avibactam and relebactam) (Table 4). The largest fold reductions in the MICs
were obtained with 10a and 10b containing an unsubstituted triazole ring (from 256 to 32 and
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 46
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
1
6 µg/ml, respectively). Accordingly, these two compounds were the most active in vitro (200
-
1
-1
and 210 M s ) indicating that none of the side chains had a positive impact on inhibition
efficacy. DBOs 7b and 7i were as active in vitro as DBOs 10a and 10b containing unsubstituted
triazole rings. The side chain of other compounds had a negative impact leading to 5- to 13-
fold lower inhibition efficacies. Avibactam was the only DBO that significantly reduced the
MIC of ertapenem against E. coli Top10 producing class D OXA-48 although the drug retained
substantial antibacterial activity (MIC = 0.5 µg/ml versus < 0.03 µg/ml in the absence of OXA-
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
4
4
8 production). In vitro evaluation indicated that four DBOs (7b, 7g, 7h, and 7i) inhibited OXA-
8 with efficacies similar to those of compounds 10a and 10b, which contained unsubstituted
-
1
-1
triazole rings (k
2
/K ranges of 110 to 210 versus 140 to 150 M s , respectively). The side
i
chains of other DBOs, including that of relebactam, were detrimental since no inhibition was
detected at the highest concentration that was tested (100 µM).
Table 4. Inhibition of class C (AmpC) and class D (OXA-48) β-lactamases by DBOs and clavulanate
Inhibition of AmpC
MIC of
Inhibition of OXA-48
R
MIC of ertapenem
µg/ml)
N
cefotaxime
µg/ml)
(
N
(
O
OSO3Na
Carbamoylation
efficacy
Carbamoylation
efficacy
R =
no inhibitor
Without
AmpC
0.125
With
AmpC
256
Without
With
OXA-48
-1
-1
-1 -1
(M s )
(M s )
OXA-48
< 0.03
NA
NA
8
Avibactam
Relebactam
18,000 ± 1,000
15,000 ± 2,000
0.125
1
980 ± 60
< 0.03
< 0.03
0.5
8
0.125
0.25
No inhibition
7a
85 ± 7
200 ± 10
24 ± 1
15 ± 5
20 ± 8
31 ± 3
0.125
0.25
128
128
64
No inhibition
110 ± 30
< 0.03
< 0.03
< 0.03
< 0.03
< 0.03
< 0.03
8
8
7b
7c
0.125
0.125
0.125
0.125
No inhibition
No inhibition
No inhibition
No inhibition
16
16
8
7d
128
64
7e
7f
128
8
1
6
ACS Paragon Plus Environment

Page 17 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7g
51 ± 8
29 ± 6
0.125
0.125
0.125
128
128
128
210 ± 50
210 ± 30
150 ± 40
< 0.03
< 0.03
< 0.03
16
16
16
7h
7i
180 ± 20
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
8g
16 ± 2
26 ± 2
28 ± 13
0.125
0.125
0.25
128
128
64
No inhibition
No inhibition
No inhibition
< 0.03
< 0.03
< 0.03
8
8
4
[
a]
8h
[a]
8i
[
a]
7j
99 ± 9
200 ± 20
210 ± 20
NA
0.25
0.125
0.125
0.0625
64
32
No inhibition
150 ± 30
140 ± 20
NA
< 0.03
< 0.03
< 0.03
< 0.03
8
8
8
8
10a
10b
16
Clavulanate
128
NA, not applicable as clavulanate and DBOs have different modes of action. No inhibition, no
inhibition at the highest inhibitor concentration that was tested (100 µM). Kinetics constants
were deduced from a minimum of six progress curves obtained in a minimum of two
independent experiments. Reported MICs are the medians from five biological repeats obtained
in two independent experiments. [a] Zwitterionic species have been tested.
Evaluation of antibacterial activity against clinical isolates. The capacity of DBOs to restore
the activity of β-lactams against β-lactam-producing clinical isolates of enterobacteria was
tested for a subset of the compounds synthesized in this study (Supplementary Table S2 and
S3). The data obtained with clinical isolates were in agreement with the inhibition profiles
determined by in vitro inhibition of β-lactamases and antibacterial evaluation in isogenic strains
of E. coli (Tables 2 , 3 , and 4). In particular, the most active compounds, such as 7a, restored
the activity of β-lactams against clinical isolates producing TEM-1, KPC-2, and CTX-M-15,
alone or in combination.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 46
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
Role of asparagine 132 in the carbamoylation efficacy of KPC-2 and CTX-M-15 by DBOs.
The crystal structure of KPC-2 and CTX-M-15 acylated by avibactam revealed hydrogen
interaction between the oxygen of the avibactam carboxamide group and the amine in the
carboxamide group of Asn at position 132 of the β-lactamases (Ambler numbering) (Figure
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
4
).10, 36,37 To assess the role of this interaction, we further characterized derivatives of KPC-2
132
and CTX-M-15 harboring an N G substitution in the conserved SDN motif. This substitution
was previously shown to impair (>1,000-fold) the efficacy of carbamoylation of KPC-2 and
CTX-M-15 by avibactam whereas hydrolysis of nitrocefin was moderately (≤ 3 fold) affected.38
132
In this study, we compared the impact of the N G substitution on the carbamoylation efficacy
of the β-lactamases by avibactam, 10b (unsubstituted triazole), and 7b (phenyl-substituted
triazole) (Table 5).
Figure 4. Hydrogen interaction between the carboxamides of avibactam and Asn at position
1
32 of β-lactamases (Ambler numbering).
132
Table 5. Impact of the N G substitution on the cabamoylation efficacy (k
2
i
/K ) of KPC-2 and
CTX-M-15 by DBOs
-1 -1
k
2
/K
i
(M s ) for indicated DBO
β-lactamase
Avibactam
26,000 ± 1,000
53 ± 3
130,000 ± 10,000
70 ± 6
10b
280 ± 30
31 ± 6
280 ± 10
6.6 ± 2.0
7b
930 ± 40
21 ± 3
6,900 ± 200
140 ± 10
KPC-2
KPC-2 N G
CTX-M-15
1
32
132
CTX-M-15 N G
1
8
ACS Paragon Plus Environment

Page 19 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In comparison to avibactam, the substitution had a lower impact on the efficacy of
carbamoylation of KPC-2 and CTX-M-15 by 10b and 7b. The latter compounds were much
less active than avibactam against the wild-type enzymes but as active as avibactam against
1
32
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
KPC-2 and CTX-M-15 harboring N G. These observations suggest that the hydrogen
interaction between avibactam and N132 is critical for efficacious carbamoylation of KPC-2 and
CTX-M-15 by avibactam and that the absence of this interaction in the case of triazole-
containing DBOs could be largely responsible for their lower efficacy in comparison to
avibactam. The modest k
2
/K
i
reductions observed for 10b and 7b following introduction of
1
32
132
N G could be accounted for by additional roles of N or a weak interaction of the latter
residue with the triazole ring. The role of the hydrogen interaction involving N132 is also
highlighted by comparison of relebactam and compound 8h that contain the same side chain
connected to the DBO scaffold by an amide or triazole, respectively.
CONCLUSIONS
Efficacies of the 17 DBOs for inhibition of TEM-1, KPC-2, CTX-M-15, AmpC, and OXA-48
were recapitulated in a heat map (Figure 5A) to highlight key differences. Compounds 10a and
1
0b, containing unsubstituted triazole rings, were less active than the reference DBOs
(
avibactam and relebactam) against TEM-1, KPC-2, CTX-M-15, and AmpC. Loss of the
hydrogen interaction between the carboxamide of avibactam and conserved residue N132 is
likely to be responsible for reduced activity that could not be fully compensated by any of the
substituents introduced in the triazole ring. In comparison to 10a and 10b, two DBOs, 7a and
7
b, displayed improved inhibition efficacies, in particular against the KPC-2 carbapenemase
and the CTX-M-15 extended-spectrum β-lactamase, indicating that the pyridine and
phenylgroups present in 7a and 7b had a positive impact on the inhibition of these class A β-
lactamases. Several side chains had a deleterious effect, in particular the methoxy and carboxyl
groups present in 7d and 7j. In vivo antibacterial susceptibility data were recapitulated in a heat
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 46
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
map presented in Figure 5B. The most active DBOs in vitro (7a and 7b) were also among the
most active compounds for restoring the activity of β-lactams against KPC-2- or CTX-M-15-
producing E. coli strains.
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
A
B
Figure 5. Heat maps highlighting differences in (A) the efficacy of in vitro inhibition of β-
lactamases (Log10 of k /K ) and (B) the antibacterial activity (Log of the fold reduction in the
2
i
2
MICs). *: No inhibition at the highest inhibitor concentration that was tested (100 µM). The
color code shows the activities relative to the most active compound for each β-lactamase.
Darker color indicates higher activity.
The correlation between antibacterial activity and in vitro inhibition efficacies did not appear
to apply to all DBOs, in particular for DBOs containing protected amines, which appeared to
be less active in vivo than expected from their in vitro activity. This was explored by plotting
the in vitro efficacy of the DBOs against the fold reduction of the MICs they afforded (Figure
6
A). Plotting the complete set of data, i.e. the inhibition of five β-lactamases by 17 DBOs,
revealed a positive correlation between the Log10 of the inhibition efficacy (Log10 of k
2
/K ) and
i
2
0
ACS Paragon Plus Environment

Page 21 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the Log
2
of the fold reduction in the MIC of the reporter drugs (Spearman correlation coefficient
-
14
of 0.72; p = 5.3 10 ). DBOs harboring a Boc group in their side chain clustered below the
regression line indicating that their in vitro efficacy was not translated into antibacterial activity
due to impaired penetration through the outer membrane (Figure 6B). The Spearman correlation
coefficient calculated without Boc-containing compounds was higher than that calculated with
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
-
21
the entire set of data (0.87; p = 4.7 10 ). In contrast, amine-containing compounds clustered
above the regression line indicating a facilitated access to the periplasm (Figure 6C).
Figure 6. Correlation between the efficacy of DBOs in reducing MICs and in inhibiting β-
lactamases. The trend line was obtained by linear regression. (A) Complete set of data. (B) The
regression line was calculated by omitting Boc-containing DBOs (highlighted with large
circles). (C) Clustering of amine-containing DBOs above the regression line (highlighted with
large circles).
To summarize, we have developed an efficient strategy based on copper(I)-catalyzed Huisgen-
Sharpless cycloaddition reaction for the synthesis of new DBOs. To achieve this goal, a
modification of the original synthesis of azido derivative 4 has been realized. Our methodology
is compatible with a wide range of functional groups for the modification of the C2 position of
the DBO scaffold. The synthetic route provides a versatile approach to explore the chemical
space for novel interactions with the catalytic cavity of potential targets.3,23,24,32 Although the
first series of triazole-containing compounds reported in this study did not lead to inhibitors
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 46
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
with improved characteristics for inhibition of β-lactamases of Gram-negative bacteria. The
reduced efficacy could be due to the absence of hydrogen bonding with a conserved asparagine
residue at position 132. This could only be partially compensated by additional interactions
involving the triazole substituents.
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
EXPERIMENTAL SECTION
General procedures. Reactions were carried out under argon atmosphere and solvents were
dried using standard methods and distilled before use. Unless otherwise specified, materials
were purchased from commercial suppliers and used without further purification. TLC was
performed using Merck commercial aluminum sheets coated with silica gel 60 F254. Compounds
were detected by charring with phosphomolibdic acid in ethanol followed by heating.
Purification was performed by flash chromatography on silica gel (60 Å, 180-240 mesh; Merck)
or by preparative high-performance liquid chromatography (HPLC) using Shimadzu
Prominence system with a Zorbax Extend-C18 prepHT column (150 x 21.2 mm, 5 µm;
Agilent). Compounds were eluted with a linear gradient (from 100% of H
CH CN) that was applied between 5 and 30 min at a flow rate of 15 ml/min. Products were
detected by UV absorption at 214 nm. NMR spectra were recorded on Bruker spectrometers
AM250, Avance II 500 and Avance III HD 4000). Chemical shifts (δ) are reported in ppm and
2
O to 100% of
3
(
referenced to the residual proton or carbon resonance of the solvents: CDCl
3
(δ 7.26), D O (δ
2
1
13
4
.79) or (CD
3
)
2
SO (δ 2.50) for H and CDCl
3
(δ 77.16) or (CD
3
)
2
SO (δ 39.52) for C. Signals
1
13
were assigned using 1D ( H and C) and 2D (HSQC, COSY, and HMBC) spectra. NMR
coupling constants (J) are reported in hertz (Hz). High-resolution mass spectroscopy (HRMS)
was recorded with an ion trap mass analyzer under electrospray ionization (ESI) in the negative
or positive ionization detection mode. HRMS was performed using Thermo Scientific LTQ
Orbitrap XL and Bruker MaXis II ETD spectrometers. The purity of final compounds (≥95%)
was etablished by analytical HPLC, which was performed on a Shimadzu Prominence system
2
2
ACS Paragon Plus Environment

Page 23 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with a Hypersil® BDS C18 column (150 x 4.6 mm, 5 µm) or an Agilent Zorbax Extend C18
column (250 x 4.6 mm, 5 µm) with UV detection at 214 and 220 nm. Optical rotations were
measured with a sodium lamp (589 nm) at 20°C on a Perkin Elmer polarimeter.
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
Compound (4). Isobutyl chloroformate (6.26 ml, 47.1 mmol) was added dropwise at 0°C to a
solution of (2S,5R)-6-(benzyloxy)-7-oxo-1,6-diazabicyclo[3.2.1]octane-2-carboxylic acid 2
(
6.5 g, 23.5 mmol) and triethylamine (3.64 ml, 25.9 mmol) in THF (75 ml). The solution was
stirred at 0°C for 2 h and NaBH (2.72 g, 70.6 mmol) was gradually added. The reaction mixture
4
was stirred for 3 additional hours at 0°C. Water (110 ml) and ethyl acetate (110 ml) were slowly
added and the heterogeneous mixture was stirred for 30 min at room temperature. The organic
layer was washed with brine, dried over MgSO , and concentrated under reduced pressure to
4
afford (2S,5R)-6-(benzyloxy)-2-(hydroxymethyl)-1,6-diazabicyclo[3.2.1]octan-7-one 3, which
was used in the next step without further purification. Triethylamine (9.94 ml, 70.6 mmol),
DMAP (4.36 g, 35.3 mmol) and Mesyl Chloride (2.76 ml, 35.3 mmol) were added at 0°C to a
solution of crude product (23.5 mmol) in DCM (71 ml). The reaction mixture was stirred at 0°C
for 2 h. DCM was then added and the organic layer was washed with brine, dried over MgSO ,
4
and concentrated under vacuum to afford ((2S,5R)-6-(benzyloxy)-7-oxo-1,6-
diazabicyclo[3.2.1]octan-2-yl)methyl methanesulfonate, which was used in the next step
without further purification. Sodium azide (7.69 g, 117 mmol) was added to a solution of
(
(2S,5R)-6-(benzyloxy)-7-oxo-1,6-diazabicyclo[3.2.1]octan-2-yl)methyl
methanesulfonate
(23.5 mmol) in DMF (95 ml). The solution was stirred at 90°C for 20 h. After being cooled to
room temperature, ethyl acetate was added and the organic layer was washed with brine, dried
over MgSO , and concentrated under reduce pressure. Purification by flash chromatography
4
using cyclohexane/ethyl acetate (7/3) as the eluent gave compound 4 as a white solid (2.4 g,
1
3
5% over 3 steps). H NMR (500 MHz, CDCl
3
) δ 7.43-7.33 (m, 5H, H10, 11, 12), 5.04 (d, J = 11.5
Hz, 1H, H
8
), 4.89 (d, J = 11.5 Hz, 1H, H
8
), 3.61-3.56 (m, 1H, H ), 3.53 (dd, J = 12.5, 7.4 Hz,
1
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 46
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
1
H, H
7
), 3.35-3.32 (m, 2H, H4, 7), 2.97 (s, 2H, H
5
), 2.06-2.02 (m, 1H, H
3
), 2.01-1.95 (m, 1H,
). C NMR (125 MHz, CDCl ) δ 169.7 (C ),
), 53.3
1
3
H
2
), 1.63-1.55 (m, 1H, H
3
), 1.54-1.49 (m, 1H, H
2
3
6
136.0 (C
), 44.1 (C
found 288.14523. [α]
9
), 129.3 (2C, C10), 128.8 (C12), 128.6 (2C, C11), 78.3 (C
), 20.5 (C ), 20.0 (C ). HRMS calculated for C14
: -53.2° (7.8 mg/ml, MeOH).
8
), 58.6 (C
4
), 56.6 (C
1
+
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
(
C
7
5
2
3
H
18
N
5
O
2
[M+H] : 288.14605;
D
General procedure for Copper(I)-catalyzed azide-alkyne cycloaddition reaction. To a
solution of azido 4 in THF, were successively added alkyne 5 (2 eq), sodium ascorbate (0.6 eq,
in water) and CuSO
room temperature. Ethyl acetate was added and the organic layer was washed with brine, dried
over MgSO , and concentrated under reduced pressure. The crude product was purified by flash
4
(0.3 eq, in water). The heterogeneous mixture was stirred overnight at
4
chromatography to afford product 6.
General procedure for introduction of sodium sulphite. 10 wt. % Pd/C (1 eq) was added to
a solution of 6 in MeOH and the reaction mixture was stirred under H for 48 h at room
2
temperature. Palladium was removed by filtration through celite and the filtrate concentrated.
Sulfur trioxide pyridine complex (6 eq) was added to a solution of deprotected compound in
pyridine and the reaction mixture was stirred for 2 h at room temperature. Additional SO -
3
pyridine complex (2 eq) was added, stirred overnight at room temperature, and pyridine was
removed under reduced pressure. The crude product was solubilized in water, filtered, eluted
on Dowex-Na resin with H O, and lyophilized. The residue was dissolved in EtOH, filtered,
2
and concentrated under vacuum. HPLC purification afforded product 7.
General procedure for Boc deprotection. TFA (12 eq) was added dropwise at 0°C to a
solution of 7 (1 eq) in DCM. The reaction mixture was stirred at 0°C for 1 h and concentrated
under vacuum. HPLC purification afforded product 8.
2
4
ACS Paragon Plus Environment

Page 25 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound (6a). Following the general procedure for CuAAC, starting from compound 4 (200
mg, 0.70 mmol) and 3-ethynylpyridine 5a (144 mg, 1.39 mmol), and using DCM/MeOH (96/4)
as eluent for flash chromatography purification, compound 6a was obtained as a colorless oil
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
(
234 mg, 86%). H NMR (500 MHz, CDCl
3
) δ 9.01 (bs, 1H, H14), 8.57 (bd, J = 3.8 Hz, 1H,
H13), 8.17 (dt, J = 7.9, 2.0 Hz, 1H, H11), 8.05 (s, 1H, H
8
), 7.42-7.33 (m, 6H, H12, 17, 18, 19), 5.02
(
d, J = 11.5 Hz, 1H, H15), 4.88 (d, J = 11.5 Hz, 1H, H15), 4.62 (dd, J = 14.3, 8.3 Hz, 1H, H
7
),
), 2.99-2.92
), 1.65-1.59
),
4.56 (dd, J = 14.3, 6.5 Hz, 1H, H
7
), 3.92-3.87 (m, 1H, H
), 2.06-1.99 (m, 1H, H
). C NMR (125 MHz, CDCl ) δ 169.4 (C
1
), 3.38-3.36 (m, 1H, H
), 1.73-1.67 (m, 1H, H
4
3
(
(
m, 2H, H
5
), 2.12-2.07 (m, 1H, H
3
2
13
m, 1H, H
2
3
6
), 149.4 (C13), 147.3 (C14), 145.3 (C
9
135.8 (C16), 133.2 (C11), 129.4 (2C, C17), 128.9 (C19), 128.7 (2C, C18), 126.8 (C10), 123.9 (C12),
1
20.5 (C
8
), 78.4 (C15), 58.4 (C
4
), 56.7 (C
1
), 52.1 (C
7
), 43.8 (C
5
), 20.5 (C
2
), 19.8 (C
3
). HRMS
+
calculated for C21
MeOH).
H
23
N
6
O
2
[M+H] : 391.18825; found: 391.18886. [α]
D
: -17.9° (8.0 mg/ml,
Compound (6b). Following the general procedure for CuAAC, starting from compound 4 (200
mg, 0.70 mmol) and phenylacetylene 5b (154 µl, 1.40 mmol), and using cyclohexane/ethyl
acetate (4/6) as eluent for flash chromatography purification, compound 6b was obtained as a
1
colorless oil (232 mg, 85%). H NMR (500 MHz, CDCl
3
) δ 7.97 (s, 1H, H ), 7.83 (d, J = 7.2
8
Hz, 2H, H11), 7.43-7.32 (m, 8H, H12, 13, 16, 17, 18), 5.03 (d, J = 11.5 Hz, 1H, H14), 4.88 (d, J = 11.5
Hz, 1H, H14), 4.63-4.54 (m, 2H, H
7
), 3.91-3.86 (m, 1H, H
), 2.03-1.96 (m, 1H, H
). C NMR (125 MHz, CDCl ) δ 169.3 (C ), 148.2 (C
1
), 3.36-3.35 (m, 1H, H
4
), 2.97-2.90
), 1.66-1.60
(
(
m, 2H, H
5
), 2.10-2.06 (m, 1H, H
3
2
), 1.72-1.68 (m, 1H, H
3
13
m, 1H, H
2
3
6
9
), 135.8 (C15), 130.3 (C10),
129.4 (2C, C16), 129.0 (2C, C12), 128.9 (C18), 128.7 (2C, C17), 128.5 (C13), 126.0 (2C, C11),
1
20.3 (C
8
), 78.4 (C14), 58.4 (C
4
), 56.7 (C
1
), 52.2 (C
7
), 43.9 (C
5
), 20.4 (C
2
), 19.8 (C
3
). HRMS
+
calculated for C22
DMSO).
H
24
N
5
O
2
[M+H] : 390.19300; found: 390.19165. [α]
D
: -44.5° (5.2 mg/ml,
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 46
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
Compound (6c). Following the general procedure for CuAAC, startingfrom compound 4 (200
mg, 0.70 mmol) and ethynyltrimethylsilane 5c (388 µl, 2.80 mmol), and using
cyclohexane/ethyl acetate (3/7) as eluent for flash chromatography purification, compound 6c
1
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
was obtained as a colorless oil (82 mg, 30%). H NMR (250 MHz, CDCl
3
) δ 7.64 (s, 1H, H ),
8
7
.42-7.32 (m, 5H, H13, 14, 15), 5.01 (d, J = 11.5 Hz, 1H, H11), 4.86 (d, J = 11.5 Hz, 1H, H11), 4.55
(
d, J = 7.3 Hz, 2H, H
7
), 3.85-3.75 (m, 1H, H
1
), 3.34-3.30 (m, 1H, H
), 1.72-1.62 (m, 1H, H
) δ 169.6 (C ), 147.3 (C
), 56.8 (C
), -1.0 (3C, C10). HRMS calculated for C19
: -24.4° (5.6 mg/ml, MeOH).
4
), 2.95-2.82 (m, 2H, H
5
),
),
2.09-1.97 (m, 1H, H
3
), 1.96-1.85 (m, 1H, H
2
3
), 1.62-1.50 (m, 1H, H
2
13
0
.30 (s, 9H, H10). C NMR (125 MHz, CDCl
3
6
9
), 135.9 (C12), 129.4
(
3C, C
8
and C13), 128.9 (C15), 128.7 (2C, C14), 78.4 (C11), 58.5 (C
4
1
7
), 51.5 (C ), 44.0
(C
5
), 20.3 (C ), 19.9 (C
2
3
H
28
N
6
O
2
Si [M+H]+:
3
86.20123; found: 386.20053. [α]
D
Compound (6d). Following the general procedure for CuAAC, starting from compound 4 (200
mg, 0.70 mmol) and methyl propargyl ether 5d (118 µl, 1.40 mmol), and using DCM/MeOH
(
96/4) as eluent for flash chromatography purification, compound 6d was obtained as a
1
colorless oil (215 mg, 86%). H NMR (500 MHz, CDCl
3
) δ 7.70 (s, 1H, H ), 7.42-7.33 (m, 5H,
8
H
14, 15, 16), 5.02 (d, J = 11.5 Hz, 1H, H12), 4.87 (d, J = 11.5 Hz, 1H, H12), 4.57 (s, 2H, H10), 4.55-
.48 (m, 2H, H ), 3.81 (qd, J = 7.4, 4.2 Hz, 1H, H ), 3.40 (s, 3H, H11), 3.34 (q, J = 3.0 Hz, 1H,
), 2.93 (bd, J = 11.9 Hz, 1H, H ), 2.88 (d, J = 11.9 Hz, 1H, H
4
7
1
H
4
5
5
), 2.09-2.03 (m, 1H, H
3
), 1.97
1
3
(
dq, J = 15.1, 7.4 Hz, 1H, H
2
), 1.70-1.63 (m, 1H, H
3
), 1.60-1.54 (m, 1H, H
2
). C NMR (125
MHz, CDCl3) δ 169.4 (C
15), 123.0(C
20.4 (C ), 19.8 (C
58.218771. [α] : -27.7° (6.6 mg/ml, MeOH).
6
), 145.8 (C ), 135.9 (C13), 129.4 (2C, C14), 128.9 (C16), 128.7 (2C,
9
C
8
), 78.4 (C12), 66.1 (C10), 58.5 (C11), 58.4 (C
4
), 56.7 (C
1
), 52.0 (C
7
), 43.9 (C ),
5
+
2
3
). HRMS calculated for C18
H
24
N
5
O
3
[M+H] : 358.18791; found:
3
D
Compound (6e). Following the general procedure for CuAAC, starting from compound 4 (150
mg, 0.52 mmol) and 3-dimethylamino-1-propyne 5e (112 µl, 1.04 mmol), and using
2
6
ACS Paragon Plus Environment

Page 27 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DCM/MeOH (9/1) + 1% NH
4
OH as eluent for flash chromatography purification, compound
1
6
e was obtained as a pale yellow oil (169 mg, 87%). H NMR (500 MHz, CDCl
3
) δ 7.65 (s, 1H,
H
8
), 7.38-7.28 (m, 5H, H14, 15, 16), 4.97 (d, J = 11.5 Hz, 1H, H12), 4.83 (d, J = 11.5 Hz, 1H, H12),
), 3.79-3.74 (m, 1H, H ), 3.56 (s, 2H, H10), 3.36-3.34 (m, 1H, H ), 2.95-
), 2.22 (s, 6H, H11), 2.03-1.97 (m, 1H, H ), 1.95-1.87 (m, 1H, H ), 1.68-1.62
). C NMR (125 MHz, CDCl ) δ 169.4 (C ), 143.3 (C ),
135.9 (C13), 129.4 (2C, C14), 128.9 (C16), 128.7 (2C, C15), 124.2 (C ), 78.4 (C12), 58.4 (C ), 56.9
), 19.8 (C ). HRMS calculated for
: -24.7° (7.5 mg/ml, MeOH).
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
4
2
.54-4.44 (m, 2H, H
.87 (m, 2H, H
), 1.55-1.49 (m, 1H, H
7
1
4
5
3
2
1
3
(
m, 1H, H
3
2
3
6
9
8
4
(
C
1
), 53.9 (C10), 52.0 (C
7
), 44.5 (2C, C11), 43.7 (C
5
), 20.4 (C
2
3
+
C
19
H
27
N
6
O
2
[M+H] : 371.21955; found: 371.21900. [α]
D
Compound (6f). Following the general procedure for CuAAC, starting from compound 4 (80
mg, 0.28 mmol) and compound 5f (70 mg, 0.56 mmol), and using DCM/MeOH (96/4) as eluent
for flash chromatography purification, compound 6f was obtained as a colorless oil (104 mg,
1
9
0%). H NMR (500 MHz, CDCl
3
) δ 7.61 (s, 1H, H
8
), 7.37 – 7.28 (m, 5H, H15,16,17), 4.96 (d, J
), 4.44
), 3.64 (t, J = 4.7 Hz, 4H, H12), 3.60 (s, 2H,
), 2.45 (t, J = 4.7 Hz, 4H, H11), 2.03-1.98 (m, 1H, H ),
). C NMR (125 MHz,
), 135.7 (C14), 129.2 (2C, C15), 128.7 (C17), 128.5 (2C, C16),
=
11.5 Hz, 1H, H13), 4.83 (d, J = 11.5 Hz, 1H, H13), 4.49 (dd, J = 14.2, 8.1 Hz, 1H, H
7
(dd, J = 14.2, 6.9 Hz, 1H, H
7
), 3.78-3.73 (m, 1H, H
1
H
10), 3.33 (bs, 1H, H
.95-1.87 (m, 1H, H
CDCl ) δ 169.3 (C ), 144.5 (C
), 78.2 (C13), 66.8 (2C, C12), 58.2 (C
4
), 2.88 (s, 2H, H
5
3
1
3
1
2
), 1.66-1.60 (m, 1H, H
3
), 1.54-1.48 (m, 1H, H
2
3
6
9
1
4
4
23.0 (C
3.7 (C
8
4
), 56.7 (C
1
), 53.6 (C10), 53.4 (2C, C11), 51.7 (C ),
7
+
5
), 20.3 (C
2
3
), 19.6 (C ). HRMS calculated for C21
H
29
N
6
O
3
[M+H] : 413.22957; found:
13.22753. [α] : -17.3° (4.4 mg/ml, MeOH).
D
Compound (6g). Following the general procedure for CuAAC, starting from compound 4
(200 mg, 0.70 mmol) and compound 5g (217 mg, 1.40 mmol), and using cyclohexane/ethyl
acetate (1/9) as eluent for flash chromatography purification, compound 6g was obtained as a
1
colorless oil (289 mg, 93%). H NMR (500 MHz, CDCl
3
) δ 7.62 (s, 1H, H ), 7.39-7.30 (m, 5H,
8
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 46
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
H
16, 17, 18), 4.98 (d, J = 11.5 Hz, 1H, H14), 4.84 (d, J = 11.5 Hz, 1H, H14), 4.51 (dd, J = 14.2, 8.1
Hz, 1H, H
1H, H ), 3.33 (bs, 1H, H
.60 (m, 1H, H ), 1.54-1.48 (m, 1H, H
69.3 (C ), 155.9 (C11), 145.9 (C ), 135.8 (C15), 129.3 (2C, C16), 128.8 (C18), 128.6 (2C, C17),
22.3 (C ), 79.7 (C12), 78.2 (C14), 58.3 (C
), 19.7 (C ). HRMS calculated for C22
: -20.5° (5.4 mg/ml, MeOH).
7
), 4.44 (dd, J = 14.2, 6.9 Hz, 1H, H
7
), 4.34 (d, J = 5.9 Hz, 2H, H10), 3.79-3.74 (m,
), 1.97-1.89 (m, 1H, H ), 1.67-
), 1.41 (s, 9H, H13). C NMR (125 MHz, CDCl ) δ
1
4
), 2.89 (s, 2H, H ), 2.05-1.99 (m, 1H, H
5
3
2
13
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
1
3
2
3
6
9
8
4
), 56.7 (C
1
), 51.7 (C
7
), 43.8 (C ), 36.3 (C10), 28.4
5
+
(3C, C13), 20.3 (C
2
3
H
31
N
6
O
4
[M+H] : 443.24068; found:
4
43.23941. [α]
D
Compound (6h). Following the general procedure for CuAAC, starting from compound 4 (200
mg, 0.70 mmol) and 1-boc-4-ethynylpiperidine 5h (293 mg, 1.40 mmol), and using
DCM/MeOH (96/4) as eluent for flash chromatography purification, compound 6h was
1
obtained as a colorless oil (336 mg, 97%). H NMR (500 MHz, CDCl
3
) δ 7.37-7.27 (m, 5H,
), 4.10
), 2.88 (bs, 5H, H5, 10, 12), 2.01-1.98 (m, 3H, H3,
), 1.57-1.51 (m, 3H, H2, 11), 1.42 (s, 9H, H15)*H
) δ 169.2 (C ), 154.7 (C13),
H18, 19, 20), 4.96 (d, J = 11.5 Hz, 1H, H16), 4.82 (d, J = 11.5 Hz, 1H, H16), 4.49 (bs, 2H, H
7
(
bs, 2H, H12), 3.76 (bs, 1H, H ), 3.32 (s, 1H, H
1
4
11), 1.94-1.87 (m, 1H, H
2
), 1.68-1.62 (m, 1H, H
3
8
1
13
not visible on the H NMR spectrum. C NMR (125 MHz, CDCl
3
6
1
35.7 (C17), 129.1 (2C, C18), 128.7 (C20), 128.5 (2C, C19), 79.4 (C14), 78.1 (C16), 58.3 (C
4
), 56.5
), 19.6
(C
1
), 52.1 (C
7
), 43.6 (3C, C
5
and C12), 33.6 (C10), 31.4 (2C, C11), 28.4 (3C, C15), 20.4 (C
2
(
[
C
3
) *C
8
and C
9
not visible on the 13C NMR spectrum. HRMS calculated for C26
H
37
N
6
O
4
+
M+H] : 497.28763; found: 497.28723. [α]
D
: -6.3° (6.3 mg/ml, MeOH).
Compound (6i). Following the general procedure for CuAAC, starting from compound 4 (90
mg, 0.31 mmol) and 5i (181 mg, 0.63 mmol) and using DCM/MeOH (96/4) as eluent for flash
chromatography purification, compound 6i was obtained as a pale yellow solid (163 mg, 91%).
1H NMR (500 MHz, CDCl
) δ 8.35 (bd, J = 12.5 Hz, 1H, NH), 7.85 (s, 1H, H ), 7.68 (d, J =
3
8
8
.1 Hz, 2H, H11), 7.57 (d, J = 8.2 Hz, 2H, H12), 7.40-7.31 (m, 5H, H22, 23, 24), 5.31 (bs, 1H, NH),
2
8
ACS Paragon Plus Environment

Page 29 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
.00 (d, J = 11.5 Hz, 1H, H20), 4.86 (d, J = 11.5 Hz, 1H, H20), 4.56 (dd, J = 14.2, 8.0 Hz, 1H,
H
7
), 4.50 (dd, J = 14.3, 6.9 Hz, 1H, H
(bs, 1H, H ), 2.94 (s, 2H, H ), 2.60 (t, J = 5.5 Hz, 2H, H15), 2.07-2.02 (m, 1H, H
m, 1H, H ), 1.71-1.64 (m, 1H, H
MHz, CDCl ) δ 170.1 (C14), 169.5 (C
2C, C22), 128.9 (C24), 128.7 (2C, C23), 126.4 (3C, C11 and C13), 120.2 (2C, C12), 119.9 (C
79.7 (C18), 78.3 (C20), 58.4 (C ), 56.7 (C ), 51.9 (C ), 43.9 (C ), 37.6 (C15), 36.6 (C16), 28.5 (3C,
19), 20.3 (C ), 19.8 (C ). HRMS calculated for C30
76.29582. [α] : -13.5° (4.0 mg/ml, MeOH).
7
), 3.86-3.81 (m, 1H, H
1
), 3.49-3.45 (m, 2H, H16), 3.35
4
5
3
), 2.00-1.92
1
3
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
(
2
3
), 1.60-1.54 (m, 1H, H
2
), 1.42 (s, 9H, H19). C NMR (125
), 138.2 (C10), 135.7 (C21), 129.3
),
3
6
), 156.5 (C17), 147.8 (C
9
(
8
4
1
7
5
+
C
2
3
H
38
N
7
O [M+H] : 576.29344; found:
5
5
D
Compound (6j). Following the general procedure for CuAAC, starting from compound 4 (200
mg, 0.70 mmol) and 4-pentynoic acid 5j (137 mg, 1.40 mmol), and using DCM/MeOH (9/1)
as eluent for flash chromatography purification, compound 6j was obtained as a pale yellow
1
solid (202 mg, 75%). H NMR (500 MHz, CDCl
3
) δ 7.57 (s, 1H, H ), 7.41-7.33 (m, 5H, H15, 16,
8
1
7
), 5.01 (d, J = 11.5 Hz, 1H, H13), 4.87 (d, J = 11.5 Hz, 1H, H13), 4.53-4.44 (m, 2H, H
3.77 (m, 1H, H ), 3.35 (bd, J = 2.8 Hz, 1H, H ), 3.03 (t, J = 6.4 Hz, 2H, H10), 2.95-2.88 (m, 2H,
), 2.75 (t, J = 6.2 Hz, 2H, H11), 2.07-2.01 (m, 1H, H ), 1.94 (dq, J = 15.1, 7.5 Hz, 1H, H ),
). C NMR (125 MHz, CDCl ) δ 175.8 (C12), 169.5
), 135.8 (C14), 129.4 (2C, C15), 128.9 (C17), 128.7 (2C, C16), 122.0 (C ), 78.4
), 19.8 (C ).
: -23.8° (7.9
7
), 3.82-
1
4
H
5
3
2
1
3
1
.69-1.63 (m, 1H, H
(C ), 146.8 (C
13), 58.4 (C
3
), 1.58-1.52 (m, 1H, H
2
3
6
9
8
(
C
4
), 56.7 (C
1
), 51.9 (C
7
), 43.9 (C
5
), 33.5 (C11), 21.0 (C10), 20.3 (C
2
3
+
HRMS calculated for C19
mg/ml, MeOH).
H
24
N
5
O
4
[M+H] : 386.18283; found: 386.18228. [α]
D
Compound (7a). Following the general procedure for the introduction of sodium sulphite,
compound 7a was obtained as a white powder (18 mg, 21%) starting from compound 6a (83
1
mg, 0.21 mmol). H NMR (500 MHz, D
2
O) δ 8.81 (bs, 1H, H14), 8.50 (bd, J = 4.8 Hz, 1H, H13),
8
.44 (s, 1H, H ), 8.12 (bd, J = 8.0 Hz, 1H, H11), 7.52 (dd, J = 7.9, 5.0 Hz, 1H, H12), 4.97 (dd, J
8
2
9
ACS Paragon Plus Environment