Structure-Activity Relationship for the 4(3H)-Quinazolinone Antibacterials

Article abstract of DOI:10.1021/acs.jmedchem.6b00372

We recently reported on the discovery of a novel antibacterial (2) with a 4(3H)-quinazolinone core. This discovery was made by in silico screening of 1.2 million compounds for binding to a penicillin-binding protein and the subsequent demonstration of antibacterial activity against Staphylococcus aureus. The first structure-activity relationship for this antibacterial scaffold is explored in this report with evaluation of 77 variants of the structural class. Eleven promising compounds were further evaluated for in vitro toxicity, pharmacokinetics, and efficacy in a mouse peritonitis model of infection, which led to the discovery of compound 27. This new quinazolinone has potent activity against methicillin-resistant (MRSA) strains, low clearance, oral bioavailability and shows efficacy in a mouse neutropenic thigh infection model.

Full text of DOI:10.1021/acs.jmedchem.6b00372

Subscriber access provided by NEW YORK UNIV
Article
Structure-Activity Relationship for the 4(3H)-Quinazolinone Antibacterials
Renee Bouley, Derong Ding, Zhihong Peng, Maria Bastian, Elena Lastochkin, Wei Song, Mark
A Suckow, Valerie A. Schroeder, William R. Wolter, Shahriar Mobashery, and Mayland Chang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00372 • Publication Date (Web): 18 Apr 2016
Downloaded from http://pubs.acs.org on April 19, 2016
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
StructureꢀActivity Relationship for the 4(3H)ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Quinazolinone Antibacterials
Renee Bouley†, Derong Ding†, Zhihong Peng†, Maria Bastian†, Elena Lastochkin†, Wei Song†,
Mark A. Suckow‡, Valerie A. Schroeder‡, William R. Wolter‡, Shahriar Mobashery†*, Mayland
Chang†*.
†Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556,
USA. ‡Freimann Life Sciences Center and Department of Biological Sciences, University of
Notre Dame, Notre Dame, IN 46556, USA
KEYWORDS Methicillinꢀresistant Staphylococcus aureus, antibiotics, penicillinꢀbinding
proteins, structureꢀactivity relationship
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 41
1
2
3
4
ABSTRACT
5
6
7
8
9
We recently reported on the discovery of a novel antibacterial (2) with a 4(3H)ꢀquinazolinone
core. This discovery was made by in silico screening of 1.2 million compounds for binding to a
penicillinꢀbinding protein and the subsequent demonstration of antibacterial activity against
Staphylococcus aureus. The first structureꢀactivity relationship for this antibacterial scaffold is
explored in this report with evaluation of 77 variants of the structural class. Eleven promising
compounds were further evaluated for in vitro toxicity, pharmacokinetics, and efficacy in a
mouse peritonitis model of infection, which led to the discovery of compound 27. This new
quinazolinone has potent activity against methicillinꢀresistant (MRSA) strains, low clearance,
oral bioavailability, and shows efficacy in a mouse neutropenic thigh infection model.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
2

Page 3 of 41
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
There exists an urgent need for novel classes of antibiotics for treatment of infection by
antibioticꢀresistant bacteria.1 The ESKAPE panel of bacteria, comprised of Enterococcus
faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acetinobacter baumanii, Pseudomonas
aeruginosa, and Enterobacteriaceae species—the underlined first letters for the names of the
genera makes the acronym—account for the majority of these infections.2, 3 Among these
organisms, methicillinꢀresistant S. aureus (MRSA) alone accounts for nearly half of the deaths
attributed to antibioticꢀresistant infections.4 MRSA is essentially resistant to the entire class of βꢀ
lactam antibiotics, which used to be the mainstays of clinical treatment for S. aureus infections.5,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
Newer variants of βꢀlactams, antibiotics such as ceftaroline and ceftobiprole, exhibit antiꢀ
MRSA activities, however, they require intravenous infusions every 8 to 12 hours.7
This broad resistance of MRSA to the βꢀlactam class is achieved through the acquisition of an
additional penicillinꢀbinding protein (PBP), designated PBP2a.8 This enzyme is able to evade
inhibition by the βꢀlactams, as it maintains a closed activeꢀsite conformation that is regulated by
allostery.9, 10 We have shown that ceftaroline has a unique affinity for PBP2a due to its ability to
bind nonꢀcovalently to the allosteric site at a distance of 60 Å from the active site.10 This binding
leads to conformational changes that culminate in the opening of the active site, predisposing it
to covalent inhibition by another molecule of ceftaroline.10 Resistance to this antibiotic has
already emerged, which is linked to the critical role that this allostery makes to the mechanism of
action.11ꢀ13 Whereas other antibiotics for treatment of MRSA have been introduced to the
clinic,14 only the oxazolidinones, linezolid and tedizolid, are orally bioavailable.15 Resistance to
all of these drugs has already emerged, and indeed it will to any other antibiotic.16ꢀ19 Hence,
discovery of new antibiotics with novel mechanisms of action remains a high priority.20
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 41
1
2
3
4
5
6
7
8
9
We recently reported the discovery of the 4(3H)ꢀquinazolinones from in silico screening of the
PBP2a active site (1VQQ)21 using a 1.2ꢀmillion ZINC compound library.22 Highꢀranking
compounds were screened for antibacterial activity against the ESKAPE panel, which led to the
discovery of compound 1, which displayed good activity against S. aureus (ATCC 29213) with a
minimalꢀinhibitory concentration (MIC) of 2 ꢁg/mL.22 Quinazolinone 2 was subsequently
identified as a lead compound by synthesis of a focus library of analogs, which displayed in vivo
efficacy in an MRSA mouse peritonitis model and showed good oral bioavailability.22 This
compound works by inhibiting PBP1 and PBP2a in MRSA, consistent with the mechanism of
action of βꢀlactams, which often inhibit more than one PBP.23 In an intriguing observation, we
documented that compound 2 also binds to the allosteric site of PBP2a, as does ceftaroline.22 So,
notwithstanding the distinct structures of ceftaroline and the lead quinazolinone, they mimic each
other in their respective mechanisms of action. The quinazolinone class of antibacterials is of
considerable interest as they are nonꢀβꢀlactam antibacterials that inhibit PBPs, but also that they
are orally bioavailable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Structures of quinazolinones 1 and 2 are given. The sites for structure diversification
under SAR1, SAR2 and SAR3 are highlighted by the colored boxes. Numbering of carbons on
ring 3 is shown.
ACS Paragon Plus Environment
4

Page 5 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We report herein the first structureꢀactivity relationship (SAR) of this class of antibacterials by
systematically varying the quinazolinone structure in rings 1, 2, and 3 (Figure 1). Each derivative
was screened for in vitro antibacterial activity by determining MICs against our panel of
bacterial strains; however we specifically aimed to improve the activity for S. aureus. Several
derivatives that displayed good activity in vitro were selected for testing for in vitro toxicity,
pharmacokinetic (PK) properties in mice, and efficacy in mouse models of infection. A
subsequent lead compound was identified from these select derivatives and in vivo efficacy was
further demonstrated in the neutropenic thigh infection model.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Synthesis. Variations of the structure of the hit quinazolinone were made on ring 1 (SAR1),
ring 2 (SAR2), and ring 3 (SAR3), as shown in Figure 1. A general threeꢀstep synthetic route
was used to synthesize 72 quinazolinone derivatives (Scheme 1), and an additional 5 compounds
were purchased from ChemDiv. For syntheses, anthranilic acid (I) was cyclized by heating in
triethyl orthoacetate and the solution was cooled to ꢀ20 °C to crystallize intermediate II in high
purity. Intermediate II was dissolved in glacial acetic acid by heating and a substituted aniline or
amine was added and the mixture was allowed to reflux for 4ꢀ6 h to afford intermediate III. To
obtain the final compounds, various aldehydes were allowed to react with III in refluxing acetic
acid. Further synthetic transformations of the final compounds could then be performed to
expand the structural space. To obtain the amine or the functionalized amine (e.g., 17 and 41ꢀ43),
the nitro analog was first prepared (e.g., 1 or 40), which was subsequently reduced in the
presence of tin (II) chloride in ethanol. The free aniline was obtained (41) and acylated with
acetic anhydride or mesyl chloride in pyridine to give compounds 42 and 43, respectively. To
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 41
1
2
3
4
5
6
7
8
9
obtain Oꢀacetylated compounds 37 and 38 the corresponding alcohol intermediate underwent
reaction with the respective aldehyde in the presence of acetic anhydride. In order to reduce the
doubleꢀbond linker of compound 19, hydrogenation on palladium on carbon was performed to
yield compound 24.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. The general threeꢀstep
synthetic route for the quinazolinones
prepared in this report is outlined.
Structureꢀactivity relationship. The
SAR of the quinazolinone class of
antibacterials
was
investigated
by
introducing variations on rings 1, 2, and 3
(Figure 1) and screening the final
compounds for activity against the
ESKAPE panel. These compounds
displayed activity against the ESKAPE
Gramꢀpositive organism S. aureus and
modest activity (MIC ≥ 8 ꢁg/mL) against E. faecium, but did not have activity (MIC ≥ 16
ꢁg/mL) against the Gramꢀnegative organisms. The MICs against S. aureus ATCC 29213, a
quality control methicillinꢀsusceptible strain, were used to determine the SAR. Thirty variations
on ring 1 were prepared to generate SAR1, with ring 2 as either 3ꢀhydroxyphenyl (series SAR1a)
or 3ꢀcarboxyphenyl (series SAR1b), as shown in Figure 2. When ring 1 was a phenyl ring, para
ACS Paragon Plus Environment
6

Page 7 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substitution (1) was found to be superior to meta (3) or ortho (4). In addition, small electronꢀ
withdrawing groups seemed to be favored; nitro (1), fluoro (5), chloro (8), nitrile (15), and
alkynyl (16) groups all showed potent activities, with compounds 15 and 16 being the best with
MICs of 0.03 ꢁg/mL and 0.003 ꢁg/mL, respectively. However, a trifluoromethyl group (9) was
not active, while the electronꢀdonating groups methyl (10) and ethyl (11) were well tolerated, but
isopropyl (12) was not. Introduction of hydrogenꢀbonding moieties such as hydroxyl (13), amino
(17), or carboxylic acid (18) were not tolerated. Substituting the hydroxyl group with a methoxyl
(14) restored activity, indicating that hydrogenꢀbond acceptors might be allowed. In addition, an
unsubstituted benzene ring was tolerated (19). A few heterocycles were explored, including a
1,3ꢀbenzodioxole (20) and 4ꢀpyridine (21), both of which maintained activity (MIC 2 µg/mL).
However a furan ring (22) was not tolerated. Replacement of the benzene ring (19) with a
cyclohexyl one (23) resulted in an eightꢀfold reduction in the MIC. The necessity for the doubleꢀ
bond linker was also explored. The reduction of the double bond to give 24 resulted in an
increase in activity, however shortening the linker length was not tolerated (25 and 26). A second
series, SAR1b was prepared with 3ꢀcarboxyphenyl as ring 2. An alkyne (27) was found to be the
most potent substituent in vitro with an MIC of 0.03 ꢁg/mL versus 2 ꢁg/mL for the
corresponding nitrile (2), the lead compound previously disclosed. Substitution at the terminal
alkyne was also tolerated (28 and 29). Introduction of halogens within the aromatic ring was
tolerated (31 and 32), however a second halogen substituent only afforded a twoꢀfold
improvement in the MIC compared to the parent (30). Finally, a mesyl group (33) was found to
be inactive, consistent with the results observed for the corresponding carboxylic acid (18).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 41
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Antibacterial activities of the 4(3H)ꢀquinazolinones derivatives at ring 1. The MICs
(ꢁg/mL) were determined for S. aureus ATCC 29213; red text for inactive compounds (MIC ≥
16 ꢁg/mL) and blue for active compounds (MIC ≤ 8 ꢁg/mL).
Three series were prepared for SAR2, while maintaining ring 1 as 4ꢀnitrophenyl, 4ꢀfluorophenyl,
or 4ꢀcyanophenyl (SAR2a, SAR2b, and SAR2c respectively) with a total of 26 variations, as
shown in Figure 3. With ring 2 as the phenyl group, meta (1) and ortho (35) substitutions were
equally active, but para (36) was generally not tolerated. We chose to move forward with meta
substituents in order to be consistent with the SAR1 studies. Acetylation (37 and 38) or
methylation (39) of the hydroxyl group resulted in loss of activity, suggesting that a hydrogenꢀ
bond donor might be favored at this position. This was further confirmed by the restoration of
activity by reducing the nitroaromatic group (40) to the corresponding aniline (41). Acylation of
the aniline to an Nꢀacetyl (42) and Nꢀmesyl group (43) improved activity, with the mesyl group
ACS Paragon Plus Environment
8

Page 9 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
being superior in vitro. The amide 44 was moderately active and the hydroxyethyl amide 45 was
twoꢀfold more active. Interestingly, adding a methylene group linker decreased activity for the
carboxylic acid 46 but improved potency for Oꢀacetyl derivative 47 and Nꢀacetylated 53.
Removal of the aromatic ring and replacement with the 3ꢀacetoxylpropyl group in derivative 60
resulted in a complete loss of activity. However, replacing the benzene ring with a pyrazole (61)
was well tolerated. Overall, hydrophilic groups were favored for R2 and the specific placement of
hydrogenꢀbond donors and acceptors proved important for the potency. The most potent in vitro
compound of SAR2 achieved an MIC of 0.004 ꢁg/mL with an N-mesyl group (54), which
contained a hydrogenꢀbond donor one atom away from the ring and two hydrogenꢀbond
acceptors three atoms away. There is some flexibility in the placement of these groups, as shown
by compound 53, which also had a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
very good MIC of 0.015 ꢁg/mL. But
the hydrogenꢀbond donors and
acceptors were each shifted over by
one atom. In addition, bulkier groups
were tolerated, as demonstrated by
compounds 51, 57, and 63.
Figure 3. Antibacterial activities of the
4(3H)ꢀquinazolinones derivatives at
ring 2. As in Figure 2, the MICs
(ꢁg/mL) were determined for S. aureus
ATCC 29213; red text for inactive
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 41
1
2
3
4
compounds (MIC ≥ 16 ꢁg/mL) and blue for active compounds (MIC ≤ 8 ꢁg/mL).
5
6
7
8
9
Pyridine derivatives of ring 3, in which ring 2 was kept as 3ꢀhydroxyphenyl, 3ꢀcarboxyphenyl,
or 3-carbomylphenyl (series SAR3a, SAR3b, and SAR3c, respectively) and substitution at the
C6 (SAR3d) and C7 (SAR3e) position, were explored for SAR3. Replacing ring 3 with a
pyridine was tolerated when ring 2 was 3ꢀhydroxyphenyl (64ꢀ66), however this replacement was
not tolerated when ring 2 was 3ꢀcarboxyphenyl (67ꢀ69) or 3-carbomylphenyl (70ꢀ72).
Substitution at the C6 position of ring 3 with a bromo or hydroxyl group (73 and 74,
respectively) was not tolerated. A similar
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
trend was observed with substitution at
the C7 position (75ꢀ77), with a loss of
activity. In general, variations and
substitutions on ring 3 resulted in reduced
activity.
Figure 4. Antibacterial activities of the
4(3H)ꢀquinazolinones derivatives at ring
3. As in Figure 2, the MICs (ꢁg/mL)
were determined for S. aureus ATCC
29213; red text for inactive compounds
(MIC ≥ 16 ꢁg/mL) and blue for active
compounds (MIC ≤ 8 ꢁg/mL).
ACS Paragon Plus Environment
10

Page 11 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Mouse peritonitis model. Selected compounds were evaluated for in vivo efficacy using the
mouse peritonitis model of infection with MRSA (ATCC 27660). This model of infection was
chosen because it is relatively easy to perform, the end points are simple (death or survival), and
the model has been validated with marketed antibiotics.24, 25 This model was used as the upꢀfront
animal model to identify compounds demonstrating efficacy in vivo, which could then be further
evaluated. Two doses of the quinazolinones were given intravenously (iv) at 5 mg/kg. We
evaluated the efficacy of compounds 5, 15, 23, 24, 27, 30, 42, 43, 50, 51, 52, 54, 57, 61, and 66
at 5 mg/kg to drive selection of compounds with better efficacy than compound 2, which showed
an ED50 (effective dose that results in 50% of the mice surviving) of 9.4 mg/kg in the mouse
peritonitis model of MRSA infection.22 Of the compounds tested, only compound 27 showed
>50% efficacy at 5 mg/kg.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fast pharmacokinetics (PK). In order to understand the lack of or diminished in vivo efficacy
of quinazolinones that had exhibited potent in vitro activity against living bacteria, we assessed
the PK properties of several of the compounds using fast PK studies, in which a limited number
of animals (n = 2 per time point) and time points (5 time points) were utilized. This strategy
significantly reduced the number of plasma samples for analysis by ultraꢀperformance liquid
chromatography (UPLC) and allowed for the estimation of systemic exposure (area under the
curve, AUC) and clearance (CL). We carried out the fast PK studies after the mouse peritonitis
model of infection, since the former required development of bioanalytical methods to determine
the blood levels of the quinazolinones, while the latter was less resource intensive. Eleven
compounds were selected in order to understand how various R1 and R2 groups affected the PK
properties (Table 1). Compound 15 with a 4ꢀCN in R1 and 3ꢀOH in R2 had high CL of 186
mL/min/kg, resulting in low systemic exposure of 26.8 µg·min/mL (Table 1). Substitution of the
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 41
1
2
3ꢀOH in R2 for 3ꢀCO2H in compound 2 reduced CL to 6.9 mL/min/kg, with a concomitant
increase in systemic exposure of 1180 µg·min/mL. The high CL of compound 15 was attributed
to the propensity of phenols to undergo phase II metabolism and subsequent excretion. When R1
contained 4ꢀF and R2 had a 3ꢀOH (compound 5) or 3ꢀCO2H (compound 30), no significant
differences in CL (10.7 µg·min/mL vs. 7.8 mL/min/kg) or AUC (466 µg·min/mL at a dose of 5
mg/kg vs. 2563 µg·min/mL at a dose of 20 mg/kg) were observed. Likewise, no differences in
CL (6.9 vs. 7.8 µg·min/mL) or AUC (1180 µg·min/mL at a dose of 10 mg/kg vs. 2560
µg·min/mL at a dose of 20 mg/kg) were observed when R2 contained 3ꢀCO2H and R1 had 4ꢀCN
(compound 2) or 4ꢀF (compound 30). Replacement of 3ꢀCO2H in R2 with 3ꢀNHAc or 3ꢀNHMs
increased CL (compare compound 30 vs. 42 and 43; compound 2 vs. 52 and 54). To reduce the
CL of compound 52, 3ꢀNHAc was replaced with 3ꢀCH2NHAc (compound 53), which resulted in
a 2ꢀfold reduction of CL and 2ꢀfold increase in AUC (after dose adjustment). These studies
revealed that compound 27 had the best PK properties and potent in vitro and in vivo activities.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
12

Page 13 of 41
Journal of Medicinal Chemistry
1
2
3
4
Table 1. In vitro and in vivo evaluation of select quinazolinones
5
6
7
8
9
Fast PK Parameters
MIC
R1
R2
HepG2 IC50
(ꢁg/mL)
Hemolysis
(%)a
b
Compound
S. aureus
(µg/mL)
2
0.25
0.03
0.03
1
AUC0ꢀlast
CLc
(mL/min/kg)
(R-benzyl)
(R-benzyl)
Dose (mg/kg)
(ꢁg·min/mL)
1180
466
2d
5
4ꢀCN
4ꢀF
4ꢀCN
4ꢀC≡CH
4ꢀF
3ꢀCO2H
3ꢀOH
3ꢀOH
3ꢀCO2H
3ꢀCO2H
3ꢀNHAc
3ꢀNHMs
3ꢀNH2
62.9
16.4
53.6
82.0
95.5
46.6
66.0
77.2
51.0
99.0
33.4
70.2
<1
<1
<1
<1
<1
3.6
<1
3.3
<1
<1
<1
<1
10
5
5
10
20
10
10
10
5
6.9
11
186
6.4
7.8
88
28
25
131
35
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
27
15
27
30
42
43
50
52
53
54
57
1540
2560
284
819
204
38
142
31
425
4ꢀF
4ꢀF
1
0.125
0.5
0.25
0.015
0.004
1
4ꢀCN
4ꢀCN
4ꢀCN
4ꢀCN
4ꢀCN
3ꢀNHAc
3ꢀCH2NHAc
3ꢀNHMs
3ꢀCONHC2H4OH
10
5
10
162
29
aThe percent hemolysis was determined at a concentration of 64 ꢁg/mL using TritonꢀX as positive control. PK parameters after a single iv dose: AUC0ꢀlast
=
b
area under the curve from 0 min to the last quantifiable time point, cCL = clearance. dData reproduced from Bouley et al. for comparison.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 41
1
2
3
4
5
6
7
8
9
In vitro cytotoxicity. In order to evaluate the potential cytotoxicity of the quinazolinones, the
XTT assay26 using HepG2 cells and hemolysis assays were performed. The resulting IC50 values
from the XTT assay are one measure of cytotoxicity of the compounds to eukaryotic cells. The
hemolysis of red blood cells was used as another measure of toxicity, primarily as a means for
flagging of potential membraneꢀactive compounds. Both assays were performed for each
compound that was tested in mice (Table 1). Compound 5 displayed relatively high toxicity to
HepG2 cells (IC50 = 16.4 ꢁg/mL), but no significant hemolytic activity at 64 ꢁg/mL (> 120ꢀfold
MIC), which was > 30ꢀfold the HepG2 IC50. Whereas compound 50 displayed lower cytotoxicity
to HepG2 cells (IC50 = 77.2 ꢁg/mL), it caused some hemolysis of red blood cells at a
concentration of 64 ꢁg/mL.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 15 of 41
Journal of Medicinal Chemistry
1
2
3
4
Table 2. Activities of select quinazolinones against a Gramꢀpositive panel of bacteria
5
6
7
MIC (ꢁg/mL)
2a
vancomycin linezolid
8
9
Strain
15
27
30
50
52
54
S. aureus ATCC 29213b
S. aureus ATCC 27660c
S. aureus NRS100 (COL)d
S. aureus NRS119d
S. aureus NRS120d
S. aureus VRS1e
2
8
0.03
0.5
0.03
0.06
0.25
0.06
0.5
1
2
8
2
8
8
1
0.5
0.5
4
0.25
1
0.004
0.25
0.5
1
1
2
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
8
0.25
0.125
0.125
0.5
1
2
2
4
0.25
0.25
4
0.5
2
32
32
2
8
4
0.25
0.5
2
16
2
0.5
8
512
64
16
S. aureus VRS2f
0.06
0.03
4
0.125 0.125
2
S. epidermis ATCC 35547
1
0.125 0.008 0.125 0.125
0.06
0.125
1
S. haemolyticus ATCC
1
0.125 0.008
0.25
1
0.25
0.125
2
2
29970
S. oralis ATCC 9811
>128 >128
>128 >128
>128 >128
128
128
128
128
128
>128
>128
>128
>128
>128
>128
>128
>128
>128
0.5
0.5
1
1
1
1
S. pyogenes ATCC 49399
B. cereus ATCC 13061
>128
B. licheniformis ATCC
>128 >128
128
>128
>128
>128
>128
0.5
1
12759
E. faecalis ATCC 29212b
E. faecalis 99g
>128
>128
128
64
>128
128
>128
>128
>128
>128
16
128
64
>128
128
2
2
1
2
128
0.5
E. faecium NCTC 7171
>128 >128
>128
>128
>128
>128
aData reproduced from Bouley et al. for comparison. bQuality control strain for susceptibility testing; cmecA
positive, resistant to methicillin, oxacillin, and tetracycline; dmecA positive, resistant to ciprofloxacin, gentamicin,
oxacillin, penicillin, and linezolid; evancomycinꢀresistant MRSA (vanA) clinical isolate from Michigan;
fvancomycinꢀresistant MRSA (vanA) clinical isolate from Pennsylvania; gvancomycinꢀresistant clinical isolate.
Activity against Gramꢀpositive panel of bacteria. Six antibacterials from Table 1 (15, 27,
30, 50, 52, and 54) were tested for antibacterial activity against a panel of Gramꢀpositive
organisms and were compared to compound 2, vancomycin, and linezolid (Table 2).
Quinazolinones mostly display activity against Staphylococcal species. Poor activity (MIC ≥ 16
µg/mL) was observed against a vancomycinꢀresistant E. faecalis strain. Importantly, compound
27 displayed MICs ≤0.5 µg/mL for all S. aureus strains tested, including vancomycinꢀ and
linezolidꢀresistant strains.
Minimalꢀbactericidal concentration (MBC). To determine whether compound 27 was
bactericidal or bacteriostatic, its MBC was determined. The MBC was defined as the minimum
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 41
1
2
concentration of compound required to achieve ≥ 99.9% reduction in the bacterial counts.27
Compound 2 was included as a comparator and linezolid as control. We used three strains of S.
aureus: ATCC 29213 (a qualityꢀcontrol, methicillinꢀsusceptible strain), ATCC 27660 (a
methicillinꢀresistant strain used in the mouse peritonitis infection model), and NRS70 (a clinical
MRSA strain used in the mouse neutropenic thigh model). Both compounds 2 and 27 show >4ꢀ
fold difference between MIC and MBC, indicating that the quinazolinones are bacteriostatic, as
is linezolid (Table 3).27 At higher concentrations, they exhibit bactericidal activity (lowered the
initial inoculum of 1×106 CFU/mL to ≤ 500 CFU/mL). Since the MICs for antibacterial 27 were
very low, the MBC values in the three cases that were evaluated in Table 3 were not
prohibitively high (≥ 1 µg/mL). Hence, this compound will exhibit bactericidal activity at
concentrations ≥1 µg/mL.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3. MBCs for compounds 2 and 27
MBC (µg/mL)
MIC (µg/mL)
Strain
Linezolid
Linezolid
2
27
2
27
S. aureus ATCC 29213
S. aureus ATCC 27660
S. aureus NRS70
8
0.125
16
16
8
2
0.03
2
2
2
64
32
1
8
2
0.06
0.03
0.5
Full PK studies. A full PK study in mice (n = 72 mice) was conducted with compound 27
after a single 10 mg/kg iv and oral (po) dose. This compound was converted to its sodium salt by
the procedure published for 222 and this form was used for all mouse studies due to its superior
water solubility (1.5 mg/mL). Concentrations were sustained above MIC (0.03 ꢁg/mL) for 24 h
(the last time point taken) when administered iv and for 30 h after po administration (Figure 5).
This compound showed a similar rate of clearance compared to 2, 6.4 mL/min/kg versus 6.9
mL/min/kg (<10% of hepatic blood flow) and slightly higher systemic exposure (1540
ACS Paragon Plus Environment
16

Page 17 of 41
Journal of Medicinal Chemistry
1
2
3
4
µg·min/mL versus 1180 µg·min/mL). However, the volume of distribution (Vd) was 10ꢀfold
5
6
7
8
higher, 3.58 L/kg compared to 0.30 L/kg for 2 (Table 4), indicating that compound 27 had better
distribution to tissues than compound 2. In addition, the peak concentration (Cmax) of compound
27 after po administration was 2.7 µg/mL and was achieved at 0.5 h, twoꢀfold higher than that of
compound 2, which was attained at 1 h. The absolute oral bioavailability of compound 27 was
37%, while that of compound 2 was 50%. Given that the MIC values of compound 27 are one to
two orders of magnitude smaller than those of compound 2 and had higher systemic exposure,
we anticipated that compount 27 would be more efficacious in vivo than compound 2.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Full pharmacokinetic parameters of compound 27 compared to compound 2 at a dose of
10 mg/kg
a
b
c
f
Compound
(route)
Cmax
Tmax
d (h)
CLe
(mL/min/kg)
Vd
Fh
(%)
AUC0ꢀlast
AUC0ꢀ∞
g
t1/2
(µg·min/mL)
(µg·min/mL)
(µg/mL)
(L/kg)
t
1/2α = 3.6 min
t1/2β = 6.5 h
t1/2α = 2.4 h
27 iv
27 po
2 ivi
1540
1570
ꢀ
ꢀ
0.5
ꢀ
6.4
ꢀ
3.58
ꢀ
37
ꢀ
552
582
2.7
ꢀ
ꢀ
t1/2β = 6.6 h
t1/2α = 35.2 min
t1/2β = >20 h
1180
1460
6.9
0.30
t1/2α = 1.9 h
t1/2β >20 h
2 poi
408
738
1.3
1
ꢀ
ꢀ
50
aAUC0ꢀlast = area under the curve from 0 min to the last quantifiable time point, bAUC0ꢀ∞ = area under the curve from
0 min extrapolated to infinity, cCmax = maximum concentration, dTmax = time at which maximum concentration is
achieved, eCL = clearance, Vd = volume of distribution, gt1/2α = halfꢀlife of distribution and t1/2β = halfꢀlife of
f
elimination, hF = oral bioavailbility. iData for compound 2 reproduced from Bouley et al. for comparison.
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 41
1
2
3
4
5
6
100
10
po
iv
7
8
9
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.1
0.01
0
5
10
15
20
25
30
Time (h)
Figure 5. Pharmacokinetics of 27 in ICR mice after a single iv or po dose at 10 mg/kg (n = 3 per
time point).
Plasma protein binding. We ascertained protein binding in human plasma for compound 27
by equilibrium dialysis, as high plasma protein binding can decrease the in vivo efficacy of an
otherwise potent in vitro compound.28 Compound 27 showed plasma protein binding of 99.6 ±
0.04%%, which was higher than the plasma protein binding of the previous lead quinazolinone 2
(96.5 ± 0.7%).22
Mouse neutropenic thigh infection model. Compound 27 was evaluated in the mouse
neutropenic thigh infection model. This model is more clinically relevant for S. aureus infections
and allows us to be better able to quantitatively compare compound 27 to compound 2 and
linezolid. Mice were rendered neutropenic by injection with cyclophosphamide, then MRSA was
administered intramuscularly to one thigh.29 Starting one hour after infection the compounds
were dosed subcutaneously at 20 mg/kg three times a day (every 8 h) and the infected and
uninfected thighs and plasma were harvested at 24 h. Linezolid (10 mg/kg) was used as a
ACS Paragon Plus Environment
18

Page 19 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
positive control and the vehicle as a negative control. The dosing regimen was chosen in order to
maximize the time above MIC against the clinical MRSA strain used: NRS70. Compounds 27
and linezolid showed statistically significant lowering of the bacterial load for NRS70 compared
to the negative control (Figure 6). Whereas compound 2 was not able to achieve statistically
significant efficacy (p > 0.05) in this mouse model of infection. Additionally, the levels of
compound 2, 27, or linezolid were quantified in plasma and uninfected thighs (n = 8 mice per
group). At 24 h after infection (7 h after the last dose) levels in plasma were: 2.43 ± 1.64 ꢁg/mL
for compound 27, 1.13 ± 1.07 ꢁg/mL for compound 2, and not quantifiable for linezolid. The
corresponding levels in the thighs were: 0.22 ± 0.17 ꢁg/mL for compound 27, 0.024 ± 0.023
ꢁg/mL for compound 2, and not quantifiable for linezolid. Levels of compound 27 in thighs were
9ꢀfold higher than those of compound 2, indicating better tissue distribution as reflected in its
higher Vd.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. In vivo efficacy of compound 27 in
2 27
Vehicle Compound 2 Compound 27 Linezolid
12
a mouse neutropenic thigh infection model.
The number of colonyꢀforming units (CFU)
per gram of thigh tissue was determined at 24
h after infection and averaged for each group
***
***
10
8
**
*
(n
=
8
mice). The initial inoculum
6
corresponded to approximately 5.13 log10
(CFU/g). Error is expressed as the standard
error of the mean (SEM) and a Mannꢀ
Whitney Uꢀtest was used to determine
statistical significance. Groups that were
4
2
0
NRS70
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 41
1
2
statistically significant from another are shown: p < 0.05, **p < 0.01, or ***p < 0.001.
3
4
*
5
6
7
8
9
PBP antisenseꢀmediated sensitization. For further mechanistic support for select
quinazolinones, an antisense technology developed by Roemer and colleagues was utilized.30
Four MRSA strains with expression of the genes for PBP1, PBP2, PBP2a, and PBP3 under
antisense control were tested for susceptibility to the compounds. In the presence of xylose the
antisense gene is transcribed and interferes with the translation of the corresponding protein,
causing expression levels of the protein to be greatly reduced. Thus, the strain becomes
increasingly sensitive to the action of an inhibitor of the corresponding protein as the xylose
concentration increases. An increased susceptibility to compound 27 was observed under xylose
induction for the PBP1 and PBP2a antisense strains, but not for PBP2 or PBP3, consistent with
our earlier report with compound 2. Compound 27 was compared to ceftaroline as a positive
control, in which an increase in susceptibility was observed for PBP1, PBP2, and PBP2a, as
expected.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
We have previously disclosed a lead 4(3H)ꢀquinazolinone (2) that demonstrated good in vitro
activity against MRSA strains, good PK properties and showed efficacy when dosed iv in the
mouse peritonitis model of infection.22 Exploration of the SAR for the 4(3H)ꢀquinazolinones is
presented here with 77 analogs screened for antibacterial activity, of which 44 displayed activity
(MIC ≤ 4 µg/mL) against S. aureus. Phenyl rings with para substituents were favored in ring 1,
substituents such as fluoro (5), chloro (8), nitrile (15), and alkyne (16) had the best activity.
ACS Paragon Plus Environment
20

Page 21 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Bulky groups such as isopropyl (12) and hydrogenꢀbond donors (13, 17) were not tolerated.
Replacing the phenyl moiety with a pyridine (21) and cyclohexyl (23) was tolerated, but a furan
(22) was inactive. Additionally, the doubleꢀbond linker was investigated; saturation (24) was
tolerated, but shortening the linker (25 and 26) was not. For ring 2, hydrogenꢀbond donors in the
meta position such as hydroxyl (15), amino (41), Nꢀacetyl (42), Nꢀmesyl (43), seemed to be
favored. Replacing the hydroxyl with an Oꢀacetyl (37) or methoxy (39) was not tolerated.
Bulkier groups were also tolerated (51, 57ꢀ59, 62ꢀ63) and replacing the phenyl moiety with a
pyrazole (61) was allowed. A carboxyl substituent in ring 2 was less active than the parent
hydroxyl in all cases, however these compounds demonstrated the best PK properties (2, 27, 30).
A few variations of ring 3 were explored including a pyridine ring (64ꢀ72) and substitutions at
the C6 (73 and 74) and C7 (75ꢀ77) positions, however in most cases these were not tolerated.
Eleven compounds (5, 15, 27, 30, 42, 43, 50, 52, 53, 54, and 57) were chosen based on their in
vitro activity to be evaluated for in vitro toxicity, PK properties, in vivo efficacy, and activity
against a panel of Gramꢀpositive organisms. In the mouse peritonitis model of infection,
compound 27 showed superior efficacy than compound 2. Compound 27 has low clearance, 10ꢀ
fold higher volume of distribution than compound 2, and 2ꢀfold higher Cmax after po
administration than compound 2. Thus, compound 27 demonstrated better efficacy in the mouse
neutropenic thigh MRSA infection model compared to 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Quinazolinones are nonꢀβꢀlactam antibacterials targeting PBPs for inhibition. The lead
quinazolinone of this study, compound 27, exhibits exquisite antibacterial activity (MIC 0.03
µg/mL) and may have favorable pharmacodynamics. The excellent MIC values are reflective of
avid binding of the critical PBPs in S. aureus. As the antisense methodology documented that
more than one PBP is targeted for inhibition, as is the case with βꢀlactam antibiotics, the MIC
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 41
1
2
3
4
5
6
values are a composite of the concentration that achieves this inhibition. These nonꢀβꢀlactam
antibacterials hold great promise in addressing the clinical need for infections by S. aureus.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL METHODS
Syntheses. The synthetic procedures for the eleven compounds evaluated in Table 1 are
detailed below. These methods are representative for those used for the preparation of other
derivatives. Purity of the final compounds was verified to be ≥95 %, as determined by HPLC.
2-Methyl-4H-benzo[d][1,3]oxazin-4-one (II). Anthranilic acid (I, 20 g, 146 mmol) was
dissolved in triethyl orthoacetate (45 mL, 245 mmol) and refluxed for 2 h. The product was
crystallized by cooling the reaction mixture on iced water for 4 h. The needle crystals were
filtered from the reaction mixture and washed liberally with hexanes to give the product (17 g,
1
72% yield). H NMR (500 MHz, CDCl3) δ 2.47 (s, 3H), 7.50 (t, J = 7.38 Hz, 1H), 7.54 (d, J =
7.98 Hz, 1H), 7.80 (t, J = 7.18 Hz, 1H), 8.18 (d, J = 7.78 Hz, 1H). 13C NMR (126 MHz, CDCl3)
δ 21.59, 116.84, 126.59, 128.42, 128.66, 136.77, 146.61, 159.89, 160.45. HRMS (m/z): [M +
H]+, calcd for C9H8NO2, 162.0550; found, 162.0555.
3-(3-Hydroxyphenyl)-2-methylquinazolin-4(3H)-one (IIIa). Compound II (2.1 g, 13.0 mmol)
and 3ꢀaminophenol (1.5 g, 13.7 mmol) were suspended in 5 mL glacial acetic acid. The
suspension dissolved completely upon heating. The reaction was refluxed for 4 h, cooled, and 10
mL water was added to the reaction mixture. The resulting precipitate was filtered, was washed
with water, cold ethanol and hexanes to give the product (2.2 g, 67% yield). 1H NMR (300 MHz,
CDCl3) δ 2.29 (s, 3H), 6.61 (s, 1H), 6.73 (d, J = 7.85 Hz, 1H), 6.85 (d, J = 8.37 Hz, 1H), 7.37 (t,
J = 8.13 Hz, 1H), 7.50 (t, J = 7.89 Hz, 1H), 7.69 (d, J = 8.13 Hz, 1H), 7.80 (t, J = 8.37 Hz, 1H).
13C NMR (126 MHz, CDCl3) δ 24.18, 115.54, 117.57, 118.79, 120.57, 126.92, 127.17, 127.35,
ACS Paragon Plus Environment
22

Page 23 of 41
Journal of Medicinal Chemistry
1
2
131.22, 135.28, 138.01, 147.65, 154.76, 158.90, 163.15. HRMS (m/z): [M + H]+, calcd for
C15H13N2O2, 253.0972; found, 253.0946.
3
4
5
6
7
8
9
3-(3-Carboxyphenyl)-2-methylquinazolin-4(3H)-one (IIIb). The compound was prepared
according to the procedure described for IIIa and as reported by Bouley et al.22
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(E)-3-(3-Carboxyphenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one (2). The compound was
prepared according to the procedure described previously by Bouley et al.22
(E)-3-(3-Hydroxyphenyl)-2-(4-fluorostyryl)quinazolin-4(3H)-one (5). Compound IIIa (0.25 g,
1 mmol) was suspended in 5 mL glacial acetic acid and dissolved upon heating, to which 4ꢀ
fluorobenzaldehyde (0.11 mL, 1 mmol) was added. The reaction was refluxed overnight (18 h)
and 10 mL water was added to the cooled reaction mixture. The resulting precipitate was filtered
and washed with water, followed by cold ethanol and hexanes to give the product (0.14 g, 40%
1
yield). H NMR (500 MHz, DMSOꢀd6) δ 6.32 (d, J = 15.55 Hz, 1H), 6.83 (s, 1H), 6.84 (d, J =
7.78 Hz, 1H), 6.96 (d, J = 8.17 Hz, 1H), 7.22 (m, 2H), 7.40 (t, J = 7.98 Hz, 1H), 7.44 (m, 2H),
7.53 (t, J = 8.17 Hz, 1H), 7.75 (d, J = 8.17 Hz, 1H), 7.87 (m, 2H), 8.12 (d, J = 7.78 Hz, 1H), 9.92
(s, 1H). 13C NMR (126 MHz, DMSOꢀd6) δ 116.55, 116.75, 116.92, 117.00, 120.00, 120.57,
121.35, 127.18, 127.18, 127.85, 130.35, 130.41, 131.11, 132.25, 135.45, 138.15, 138.56, 148.05,
151.98, 159.03, 161.80, 162.52. HRMS (m/z): [M + H]+, calcd for C22H16FN2O2, 359.1190;
found, 359.1186.
(E)-3-(3-Hydroxyphenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one (15). The compound was
1
prepared according to the procedure described for 5 (0.62 g, 84% yield). H NMR (500 MHz,
DMSOꢀd6) δ 6.51 (d, J = 15.55 Hz, 1H), 6.85 (m, 2H), 6.97 (d, J = 7.98 Hz, 1H), 7.39 (t, J =
7.78 Hz, 1H), 7.55 (m, 3H), 7.77 (d, J = 7.78 Hz, 1H), 7.81 (d, J = 8.37 Hz, 2H), 7.88 (t, J = 7.18
Hz, 1H), 7.89 (d, J = 15.55 Hz, 1H), 8.12 (d, J = 7.98 Hz, 1H), 9.94 (s, 1H). 13C NMR (126
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 41
1
2
3
4
5
6
7
8
9
MHz, DMSOꢀd6) δ 111.52, 115.87, 116.40, 118.61, 119.30, 120.81, 123.47, 126.50, 126.95,
127.29, 128.08, 130.42, 132.92, 134.80, 136.60, 137.65, 139.38, 147.17, 150.84, 158.32, 160.99.
HRMS (m/z): [M + H]+, calcd for C23H16N3O2, 366.1237; found, 366.1254.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(E)-3-(3-Carboxyphenyl)-2-(4-ethynylstyryl)quinazolin-4(3H)-one (27). The compound was
prepared according to the procedure described for 2 and purified by silica gel column
chromatography (0.377 g, 13% yield). The sodium salt form was prepared by dissolving the
compound in 20% ethyl acetate/ 80% acetone and reacting with 1 eq sodium hexylethanoate at 0
°C for 30 min. The reaction mixture was concentrated in vacuo and the resulting yellow solid
was thoroughly washed with ethyl acetate to obtain the final compound in 75% yield (0.260 g).
1H NMR (500 MHz, DMSOꢀd6) δ 4.32 (s, 1H), 6.34 (d, J = 15.55 Hz, 1H), 7.35 (d, J = 8.37 Hz,
2H), 7.44 (d, J = 8.37 Hz, 2H), 7.49 (d, J = 7.58 Hz, 1H), 7.55 (t, J = 7.98 Hz, 1H), 7.58 (t, J =
7.78 Hz, 1H), 7.78 (d, J = 8.17 Hz, 1H), 7.87 (m, 3H), 8.05 (d, J = 7.78 Hz, 1H), 8.13 (d, J =
13
7.98 Hz, 1H). C NMR (126 MHz, DMSOꢀd6) δ 82.70, 83.24, 120.66, 121.04, 122.84, 126.51,
126.81, 127.31, 127.83, 129.98, 130.12, 132.33, 132.39, 133.49, 134.90, 135.21, 137.21, 137.99,
147.36, 151.04, 161.37, 166.58. HRMS (m/z): [M + H]+, calcd for C25H17N2O3, 393.1234; found,
393.1250. HRMS (m/z): [M + Na]+, calcd for C25H16N2NaO3, 415.1053; found, 415.1054.
(E)-3-(3-Carboxyphenyl)-2-(4-fluorostyryl)quinazolin-4(3H)-one (30). The compound was
1
prepared according to the procedure described for 2 (0.53 g, 76% yield). H NMR (500 MHz,
DMSOꢀd6) δ 6.25 (d, J = 15.55 Hz, 1H), 7.19 (m, 2H), 7.46 (m. 2H), 7.55 (d, J = 8.17 Hz, 1H),
7.73 (m, 2H), 7.77 (d, J = 8.17 Hz, 1H), 7.87 (m, 2H), 8.02 (s, 1H), 8.13 (m, 2H). 13C NMR (126
MHz, DMSOꢀ d6) δ 115.91, 116.20, 119.77, 120.59, 126.48, 126.66, 127.21,129.82, 129.93,
130.06, 131.39, 131.44, 132.34, 133.49, 134.84, 137.25, 137.79, 147.39, 151.15, 161.35, 166.55.
HRMS (m/z): [M + H]+, calcd for C23H16N2O3 , 387.1139; found, 387.1140.
ACS Paragon Plus Environment
24

Page 25 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
3-(3-Nitrophenyl)-2-methylquinazolin-4(3H)-one (IIIc). The compound was prepared
according to the procedure described for IIIa and recrystallized by dissolving in hot ethanol and
7
8
9
1
then allowing the solution to cool (2.4 g, 47% yield). H (500 MHz, DMSOꢀd6) δ 2.14 (s, 3H),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7.54 (t, J = 7.18 Hz, 1H), 7.67 (d, J = 8.13 Hz, 1H), 7.88 (m, 2H), 7.99 (d, J = 7.89 Hz, 1H), 8.10
13
(d, J = 7.89 Hz, 1H), 8.38 (d, J = 8.13 Hz, 1H), 8.50 (s, 1H). C NMR (126 MHz, DMSOꢀd6) δ
24.13, 120.41, 124.05, 124.25, 126.32, 126.56, 126.74, 130.99, 134.79, 135.66, 138.95, 147.34,
148.54, 161.43. HRMS (m/z): [M + H]+, calcd for C15H12N3O3, 282.0873; found, 282.0901.
(E)-3-(3-Nitrophenyl)-2-(4-fluorostyryl)quinazolin-4(3H)-one (40). The compound was
1
prepared according to the procedure described for 5 (1.87 g, 68% yield). H (500 MHz, DMSOꢀ
d6) δ 6.31 (d, J = 15.35 Hz, 1H), 7.18 (t, J = 8.97 Hz, 2H), 7.55 (m, 3H), 7.78 (d, J = 8.17 Hz,
1H), 7.89 (m, 3H), 7.95 (d, J = 7.78 Hz, 1H), 8.13 (d, J = 7.98 Hz, 1H), 8.43 (d, J = 8.18 Hz,
1H), 8.51 (s, 1H). 13C NMR (126 MHz, DMSOꢀd6) δ 116.01, 116.18, 119.84, 120.70, 124.42,
124.89, 126.64, 126.86, 127.39, 130.32, 130.39, 131.15, 131.53, 131.56, 135.12, 136.11, 138.19,
138.34, 148.73, 151.16, 161.54, 162.06, 164.03. HRMS (m/z): [M + H]+, calcd for C22H15FN3O3,
388.1092; found, 388.1117.
(E)-3-(3-Aminophenyl)-2-(4-fluorostyryl)quinazolin-4(3H)-one (41). Compound 40 (1.80 g,
4.6 mmol) and tin (II) chloride dihydrate (5.24 g, 23.2 mmol) were suspended in 50 mL ethanol
to which 1 mL concentrated HCl was added. The solution was refluxed for 3 h and then allowed
to stir overnight at 25 °C. The solution was concentrated in vacuo and ethyl acetate was added to
completely dissolve the residue. The ethyl acetate solution was then neutralized with a NaOH
solution and washed with saturated potassium fluoride. The mixture was then filtered through a
bed of celite to remove the tin precipitate. The resulting filtrate was washed three times with a
saturated KF solution and the organic layer was dried with MgSO4 and concentrated in vacuo.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 41
1
2
3
4
5
6
7
8
9
The residue was then purified by silica gel chromatography to give the product (0.54 g, 32%
yield). 1H (500 MHz, DMSOꢀd6) δ 5.46 (s, 2H), 6.38 (d, J = 15.55 Hz, 1H), 6.50 (d, J = 7.58 Hz,
1H), 6.54 (s, 1H), 6.73 (d, J = 8.17 Hz, 1H), 7.23 (m, 3H), 7.43 (m, 2H), 7.52 (t, J = 7.18 Hz,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
1H), 7.74 (d, J = 7.78 Hz, 1H), 7.78 (m, 2H), 8.12 (d, J = 7.98 Hz, 1H). C NMR (126 MHz,
DMSOꢀd6) δ 113.50, 114.48, 115.42, 116.05, 116.23, 119.95, 120.62, 126.45, 126.55, 127.12,
129.57, 129.64, 129.99, 131.58, 131.60, 134.69, 137.33, 137.53, 147.32, 150.05, 151.35, 161.03,
163.77. HRMS (m/z): [M + H]+, calcd for C22H17FN3O, 358.1350; found, 358.1360.
(E)-3-(3-Acetamidophenyl)-2-(4-fluorostyryl)quinazolin-4(3H)-one (42). Compound 41 (0.15
g, 0.42 mmol) was dissolved in 4 mL pyridine, to which acetic anhydride was added (0.21 g, 2.1
mmol). The reaction was stirred for 3 h at 25 °C and then refluxed for 1 h. The reaction was then
poured over ice in 2M HCl and the resulting precipitate was filtered out and washed with cold
water to give the product (0.15 g, 83 % yield). 1H (500 MHz, DMSOꢀd6) δ 2.06 (s, 3H), 6.30 (d,
J = 15.55 Hz, 1H), 7.12 (d, J = 7.38 Hz, 1H), 7.22 (t, J = 8.77 Hz, 1H), 7.46 (m, 2H), 7.55 (m,
2H), 7.73 (d, J = 6.58 Hz, 2H), 7.82 (d, J = 7.98 Hz, 1H), 7.90 (t, J = 6.78 Hz, 1H), 7.93 (d, J =
15.75 Hz, 1H), 8.13 (d, J = 7.98 Hz, 1H), 10.33 (s, 1H). 13C NMR (126 MHz, DMSOꢀd6) δ
24.09, 111.17, 116.12, 116.30, 118.45, 118.87, 119.70, 120.31, 123.21, 126.64, 127.10, 127.41,
130.03, 135.13, 136.70, 140.61, 141.61, 158.28, 160.81, 168.78. HRMS (m/z): [M + H]+, calcd
for C24H19FN3O2, 400.1456; found, 400.1440.
(E)-3-(3-Methylsulfonamidophenyl)-2-(4-fluorostyryl)quinazolin-4(3H)-one (42). Compound
41 (0.15 g, 0.42 mmol) was dissolved in 4 mL pyridine and cooled to 0 °C, to which mesyl
chloride (0.24 g, 2.1 mmol) was added. The reaction was stirred at 25 °C for 3 h and then poured
over ice in 2M HCl. The resulting precipitate was filtered out and washed with cold 2M HCl
followed by cold water to give the product (0.19 g, 99% yield). 1H (500 MHz, DMSOꢀd6) δ 3.05
ACS Paragon Plus Environment
26

Page 27 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(s, 3H), 6.31 (d, J = 15.75 Hz, 1H), 7.22 (m, 3H), 7.33 (s, 1H), 7.42 (d, J = 8.17 Hz, 2H), 7.48
(m, 2H), 7.58 (t, J = 7.98 Hz, 2H), 7.92 (d, J = 3.59 Hz, 2H), 7.99 (d, J = 15.75 Hz, 1H), 78.12
(m, 2H), 10.20 (s, 1H). 13C NMR (126 MHz, DMSOꢀd6) δ 54.50, 115.94, 116.11, 118.14,
119.37, 120.08, 120.27, 123.88, 126.52, 127.01, 127.27, 129.92, 129.99, 130.43, 135.05, 139.56,
141.42, 146.46, 160.54, 161.94. HRMS (m/z): [M + H]+, calcd for C23H19FN3O3S, 436.1126;
found, 436.1136.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(E)-3-(3-Nitrophenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one (49). The compound was
1
prepared according to the procedure described for 5 (4.8 g, 75% yield). H (500 MHz, DMSOꢀꢀ
d6) δ 6.54 (d, J = 15.35 Hz, 1H), 7.58 (t, J = 7.78 Hz, 1H), 7.68 (d, J = 8.17 Hz, 2H), 7.78 (m,
13
3H), 7.91 (m, 4H), 8.14 (d, J = 7.98 Hz, 1H), 8.43 (d, J = 8.17 Hz, 1H), 8.52 (s, 1H). C NMR
(126 MHz, DMSOꢀd6) δ 111.77, 118.80, 120.88, 123.48, 124.50, 124.94, 126.68, 127.24, 127.54,
128.73, 131.16, 132.89, 135.18, 136.10, 137.54, 137.97, 139.35, 147.37, 148.72, 150.76, 161.45.
HRMS (m/z): [M + H]+, calcd for C23H14N4NaO3, 417.0958; found, 417.0941.
(E)-3-(3-Aminophenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one (50). The compound was
prepared according to the procedure described for 41 (2.2 g, 37% yield). 1H (500 MHz, DMSOꢀ
d6) δ 5.47 (s, 2H), 6.51 (d, J = 7.78 Hz, 1H), 6.56 (m, 2H), 6.73 (d, J = 8.17 Hz, 1H), 7.21 (t, J =
7.98 Hz, 1H), 7.53 (m, 3H), 7.76 (d, J = 8.17 Hz, 1H), 7.82 (d, J = 8.37 Hz, 2H), 7.87 (t, J = 6.98
Hz, 1H), 7.90 (d, J = 15.55 Hz, 1H), 8.12 (d, J = 7.98 Hz, 1H). 13C NMR (126 MHz, DMSOꢀd6)
δ 111.50, 113.50, 114.57, 115.42, 118.61, 120.79, 123.52, 126.48, 126.92, 127.27, 128.00,
130.00, 132.92, 134.73, 136.48, 137.34, 139.43, 147.14, 150.04, 150.89, 160.91. HRMS (m/z):
[M + H]+, calcd for C23H17N4O, 365.1397; found, 365.1367.
(E)-3-(3-Acetamidophenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one (52). The compound was
prepared according to the procedure described for 42 (0.19 g, 85% yield). 1H (500 MHz, DMSOꢀ
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 41
1
2
3
4
d6) δ 3.16 (s, 3H), 6.50 (d, J = 15.55 Hz, 1H), 7.13 (d, J = 7.38 Hz, 1H), 7.56 (m, 4H), 7.71 (d, J
5
6
7
13
= 7.98 Hz, 2H), 7.80 (m, 3H), 7.91 (m, 2H), 8.14 (d, J = 7.98 Hz, 1H), 10.22 (s, 1H). C NMR
8
9
(126 MHz, DMSOꢀd6) δ 24.23, 111.73, 118.77, 119.17, 119.75, 120.90, 123.50, 126.67, 127.22,
127.52, 128.34, 130.14, 133.06, 135.06, 137.03, 137.07, 139.50, 140.64, 147.33, 150.92, 161.24,
168.86. HRMS (m/z): [M + Na]+, calcd for C25H18N4NaO2, 429.1322; found, 429.1291.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(E)-3-(3-(Acetamidomethyl)phenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one
(53).
This
1
compound was prepared according to the procedure for 42 in 21% yield (0.17 g). H (500 MHz,
DMSOꢀd6) δ 1.83 (s, 3H), 4.32 (m, 2H), 6.45 (d, J = 15.5 Hz, 1H), 7.33 (d, J = 5.5 Hz, 2H), 7.44
(d, J = 8.0 Hz, 1H), 7.52 (m, 4H), 7.77 (m, 3H), 7.86 (m, 2H), 8.12 (dd, J1 = 1.0 Hz, J2 =8.0 Hz,
1H), 8.43 (t, J = 11.5 Hz, 1H). 13C NMR (100 MHz, DMSOꢀd6) δ 23.3, 42.4, 112.2, 119.2,
121.4, 124.0, 127.1, 127.6, 127.9, 128.0, 128.1, 128.7, 128.8, 130.2, 133.5, 135.5, 137.4, 137.5,
140.0, 142.3, 147.8, 151.4, 161.8, 170.0. HRMS (m/z): [M + H]+, calcd for C26H21N4O2,
421.1658; found, 421.1675.
(E)-3-(3-Methylsulfonamidophenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one
(54).
The
1
compound was prepared according to the procedure described for 42 (0.33 g, 77% yield). H
(500 MHz, DMSOꢀd6) δ 3.04 (s, 3H), 6.52 (d, J = 15.75 Hz, 1H), 7.21 (d, J = 7.58 Hz, 1H), 7.27
(s, 1H), 7.37 (d, J = 8.57 Hz, 1H), 7.59 (m, 4H), 7.81 (m, 3H), 7.90 (m, 2H), 8.15 (d, J = 7.98
Hz, 1H), 10.08 (s, 1H). 13C NMR (126 MHz, DMSOꢀd6) δ 111.60, 118.65, 119.73, 120.32,
120.76, 123.51, 125.24, 126.56, 127.12, 127.39, 128.22, 130.60, 132.90, 134.98, 136.89, 137.45,
139.39, 139.62, 147.19, 150.78, 161.10. HRMS (m/z): [M + Na]+, calcd for C24H18N4NaO3S,
465.0992; found, 465.0973.
(E)-3-(3-(2-Hydroxyethyl)carbomylphenyl)-2-(4-cyanostyryl)quinazolin-4(3H)-one
(57).
Compound 2 (0.19 g, 0.5 mmol) was dissolved in 12 mL THF, to which 2ꢀaminoethanol (0.037
ACS Paragon Plus Environment
28

Page 29 of 41
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
g, 0.6 mmol), and DIC (0.069 g, 0.55 mmol) were added. The mixture was stirred at room
temperature for 8 h and then 25 mL ethyl acetate was added and washed with a saturated NH4Cl
solution. The organic phase was separated and dried over Na2SO4, concentrated in vacuo, and
purified by silica gel column chromatography to give the product in 63% yield (0.14 g). 1H (400
MHz, DMSOꢀd6) δ 3.28 (m, 2H), 3.51 (q, J = 6.0 Hz, 2H), 4.75 (t, J = 5.6 Hz, 1H), 6.50 (d, J =
15.6 Hz, 1H), 7.57 (m, 3H), 7.63 (m, 2H), 7.81 (d, J = 8.4 Hz, 2H), 7.89 (m, 3H), 8.04 (m, 1H),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
8.14 (m, 1H), 8.59 (t, J = 5.6 Hz, 1H). C NMR (100 MHz, DMSOꢀd6) δ 42.3, 59.6, 111.6,
118.6, 120.6, 126.5, 127.1, 127.4, 127.9, 128.1, 128.2, 129.7, 131.6, 132.8, 135.0, 136.0, 136.7,
137.0, 139.2, 147.2, 150.8, 161.2, 165.2. HRMS (m/z): [M + H]+, calcd for C26H21N4O3,
437.1608; found, 437.1635.
HPLC analysis. Compounds were analyzed for purity using a PerkinElmer Series 200
Chromatography System 524 (PerkinElmer, Waltham, MA), equipped with a LC pump, UV/Vis
detector, with an autosampler. The samples were diluted in acetonitrile (20 ꢁM) and analyzed on
a Zorbax RXꢀC8 5.0 ꢁm, 4.6 mm x 250 mm analytical column (Agilent Technologies, Santa
Clara, CA). Isocratic elution at 1.0 mL/min of 20% water containing 0.1% TFA and 20%
acetonitrile containing 0.1% TFA for 6 min with UV detection from 349ꢀ351 nm.
Screening for interference compounds. Compounds were screening for presence of known
pan assay interference compounds (PAINS) using the PAINSꢀRemover.31 Compounds were
input as a SMILES string and all 77 compounds passed the filter.
Bacterial strains. E. faecium NCTC 7171 (ATCC 19734), S. aureus ATCC 29213, K.
pneumonia ATCC 700603, A. baumannii ATCC 17961, P. aeruginosa ATCC 17853, E.
aerogenes ATCC 35029, E. coli ATCC 25922, S. aureus ATCC 27660, S. epidermis ATCC
35547, S. haemolyticus ATCC 29970, S. oralis ATCC 9811, S. pyogenes ATCC 49399, B.
ACS Paragon Plus Environment
29