Novel Inhibitors of Staphyloxanthin Virulence Factor in Comparison with Linezolid and Vancomycin versus Methicillin-Resistant, Linezolid-Resistant, and Vancomycin-Intermediate Staphylococcus aureus Infections in Vivo

Article abstract of DOI:10.1021/acs.jmedchem.7b00949

Our previous work (Wang et al. J. Med. Chem. 2016, 59, 4831-4848) revealed that effective benzocycloalkane-derived staphyloxanthin inhibitors against methicillin-resistant Staphylococcus aureus (S. aureus) infections were accompanied by poor water solubility and high hERG inhibition and dosages (preadministration). In this study, 92 chroman and coumaran derivatives as novel inhibitors have been addressed for overcoming deficiencies above. Derivatives 69 and 105 displayed excellent pigment inhibitory activities and low hERG inhibition, along with improvement of solubility by salt type selection. The broad and significantly potent antibacterial spectra of 69 and 105 were displayed first with normal administration in the livers and hearts in mice against pigmented S. aureus Newman, Mu50 (vancomycin-intermediate S. aureus), and NRS271 (linezolid-resistant S. aureus), compared with linezolid and vancomycin. In summary, both 69 and 105 have the potential to be developed as good antibacterial candidates targeting virulence factors.

Full text of DOI:10.1021/acs.jmedchem.7b00949

Subscriber access provided by GRIFFITH UNIVERSITY
Article
Novel Inhibitors of Staphyloxanthin Virulence Factor in Comparison with
Linezolid and Vancomycin versus Methicillin-Resistant, Linezolid-Resistant
and Vancomycin-Intermediate Staphylococcus aureus Infections in vivo
Shuaishuai Ni, Hanwen Wei, Baoli Li, Feifei Chen, Yifu Liu, Wenhua Chen,
Yixiang Xu, Xiaoxia Qiu, Xiaokang Li, Yanli Lu, Wenwen Liu, Linhao Hu, Dazheng
Lin, Manjiong Wang, Xinyu Zheng, Fei Mao, Jin Zhu, Le-Fu Lan, and Jian Li
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00949 • Publication Date (Web): 07 Sep 2017
Downloaded from http://pubs.acs.org on September 8, 2017
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 57
Journal of Medicinal Chemistry
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 57
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
Novel Inhibitors of Staphyloxanthin Virulence Factor in
Comparison with Linezolid and Vancomycin versus
Methicillin-Resistant, Linezolid-Resistant and
Vancomycin-Intermediate Staphylococcus aureus Infections in
vivo
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, †
a, †
a, †
b
a
a
Shuaishuai Ni , Hanwen Wei , Baoli Li , Feifei Chen , Yifu Liu , Wenhua Chen ,
a
a
a
a
a
a
Yixiang Xu , Xiaoxia Qiu , Xiaokang Li , Yanli Lu , Wenwen Liu , Linhao Hu ,
a
a
a
a
a
b,
Dazheng Lin , Manjiong Wang , Xinyu Zheng , Fei Mao , Jin Zhu , Lefu Lan *, Jian
a,
Li *
a
Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China
University of Science and Technology, Shanghai 200237, China
b
State Key Laboratory of Drug Research, Shanghai Institute of Material Medical,
Chinese Academy of Sciences, Shanghai 201203, China
†
These authors contributed equally to this work.
*
To whom correspondence should be addressed: jianli@ecust.edu.cn or
llan@simm.ac.cn.
1
ACS Paragon Plus Environment

Page 3 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ABSTRACT
Our previous work (Wang, et al. J. Med. Chem. 2016, 59, 4831ꢀ4848) revealed
that effective benzocycloalkaneꢀderived staphyloxanthin inhibitors against
methicillinꢀresistant Staphylococcus aureus (S. aureus) infections were accompanied
by poor water solubility and high hERG inhibition and dosages (preꢀadministration).
In this study, ninetyꢀtwo chroman and coumaran derivatives as novel inhibitors have
been addressed for overcoming deficiencies above. Derivatives 69 and 105 displayed
the excellent pigment inhibitory activities and low hERG inhibition, along with the
improvement of solubility by salt type selection. The broad and significantly potent
antibacterial spectra of 69 and 105 were displayed firstly with normal administration
in the livers and hearts in mice against pigmented S. aureus Newman, Mu50
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
(vancomycinꢀintermediate S. aureus) and NRS271 (linezolidꢀresistant S. aureus),
compared with linezolid and vancomycin. In summary, both 69 and 105 have the
potential to be developed as good antibacterial candidates targeting virulence factors.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 57
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
Drugꢀresistant bacteria have become one of the most serious health issues for
global public health. Staphylococcus aureus (S. aureus) is a major Gramꢀpositive
pathogen that causes a series of diseases, from mild external infections to severe organ
bacteremia, with rapid evolution of antibiotic resistance under the abuse of antibiotic
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
ꢀ4
drugs. Methicillinꢀresistant S. aureus (MRSA)，the bestꢀknown pathogen, was
5
6ꢀ9
reported first in 1961. Other drugꢀresistant MRSA strains had been identified since ;
in particular, the emergence of linezolidꢀresistant S. aureus (LRSA) and
vancomycinꢀresistant S. aureus (VRSA) in hospitals have proved challenging to
10ꢀ11
currently existing drugs since they have limited efficacy in these strains.
As issued
by the US Centers for Disease Control and Prevention (CDC) in 2015, more than 23
thousand people died and more than 2 million got sick per year due to MRSA
12
infections in the United States. In 2017, the World Health Organization (WHO)
grouped and stratified methicillinꢀresistant and vancomycinꢀintermediate and
ꢀresistant S. aureus in the high tier on the global priority list of antibioticꢀresistant
13
bacteria. Consequently, the drugꢀresistant forms of MRSA strains have shrugged off
the most powerful antibiotics. To decelerate the spread of antibiotic resistance, it is
extremely urgent to facilitate the next class of chemotherapeutics with novel
mechanisms of action.
The function of bacterial virulence factors is to strengthen the growth and
1
4
survivability of S. aureus under stressful environments. Targeting bacterial virulence
factors is a magnetic approach to combatting pathogenesis without exerting
immediate lifeꢀorꢀdeath pressure on the target bacterium, offering a promising
3
ACS Paragon Plus Environment

Page 5 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
opportunity to creating new antibacterial molecules to prevent and treat infectious
15,16
diseases.
Staphyloxanthin (STX), known as a golden carotenoid pigment
distributed in S. aureus, the structure of which contains numerous conjugated double
bonds to resist the reactive oxygen species (ROS) produced by neutrophils and
macrophages, has been deemed an important virulence factor with antioxidant
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
1
7ꢀ19
properties.
Therefore, blocking the STX biosynthetic pathway is an effective
2
0ꢀ25
therapeutic strategy for complicated S. aureus infections.
Eric Oldfield and
colleagues reported compound 1 (BPHꢀ652, Figure 1) in 2008, which could block the
biosynthesis of STX by inhibition of the S. aureus dehydrosqualene synthase (CrtM)
26ꢀ28
enzyme.
Diapophytoene desaturase (CrtN), another essential enzyme for
catalyzing three sequential reactions during the biosynthesis of STX, was identified
by our group as an appealing druggable target against pigmented S. aureus
29
infections.
A previous study by our group revealed that compound 2 (NTF, Figure 1), an
FDAꢀapproved antifungal agent, was capable of blocking the STX biosynthesis
29
pathway of three MRSA strains by inhibiting CrtN. Based on the structure of 2,
compound 3 (Figure 1) with benzofuranyl attached was investigated as a better CrtN
inhibitor in vitro by replacing the naphthyl of NTF, which only exhibited slightly
30
greater efficacy than 2 in vivo. Combined with its high inhibition of hERG
potassium channels, 3 had been restrained for development forward. Subsequently, a
series of new benzocycloalkane derivatives was synthesized, among which more
compounds showed remarkable pigment inhibitory activities than benzofuranyl
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 57
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
derivatives. In particular, benzocycloalkane derivative 4 (Figure 1) displayed
31
excellent bioactivity in vitro and in vivo. However, low water solubility, high hERG
inhibition and administration at high dosages in vivo limited further development of 4.
Figure 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
Herein, our work focused on the design, synthesis and evaluation of a CrtN
inhibitor with new scaffolds, which not only demonstrated excellent activity against
MRSA in vitro and in vivo at lower dosages, but also overcame the disadvantages of
water solubility and hERG inhibition. Additionally, linezolid and vancomycin, the
lastꢀresort antimicrobial agents, were introduced for the first time as positive control
drugs into a murine model of S. aureus abscess formation to evaluate the effectiveness
of our new compounds intuitively.
DESIGN AND SYNTHESIS
We had found that benzocycloalkane derivatives possessed potent pigment
inhibitory activity in vitro and in vivo. However, the poor water solubility of the
benzocycloalkane skeleton, along with strong hERG inhibition, limited their further
development. As shown in Figure 2, benzocycloalkane derivative 5 exhibited more
significant potency (IC50 = 1.9 nM) than 4 (IC50 = 4.1 nM) in a previous report but
31
was blocked by poor water solubility (0.69 mg/mL) and strong hERG inhibition
(IC = 3.2 ꢁM, Table 4), though it was valuable for further evaluation as a lead
5
0
compound. Primarily, structural modification a general strategy was utilized to
promote the solubility in this study, due to its conciseness and efficiency. Additionally,
in order to improve its water solubility and maintain its potency with fewer changes
5
ACS Paragon Plus Environment

Page 7 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
simultaneously, it is rational to insert a hydrophilic oxygen into the cycloalkane of 5.
Next, we designed and synthesized 24 by replacing benzocycloalkane with a chroman
segment, which exhibited excellent pigment inhibitory activity (IC50 = 4.6 ± 0.2 nM)
and improved water solubility (13.2 mg/mL, about 20 folds increase). Emboldened by
the initial consequences of 24, compound 62, containing the coumaran segment as the
other new skeleton, was synthesized, presenting comparable activity and solubility
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
(IC50 = 2.0 ± 0.2 nM, 14.6 mg/mL), and the strategy of scaffold hopping is obviously
responsible. Then, we proposed that anchoring core structures (chroman segment and
coumaran segment) and introducing comprehensive modifications in other parts of
the structure of 24/62 were the logical process to decrease hERG inhibition and
maintain pigment inhibitory activity and solubility of derivatives, as shown in Figure
2
. Therefore, the two series were identical in structural modification strategies,
including: (1) incorporating hydrogen, cycloalkyls, heteroaryls and substituted aryls in
3
region A; (2) replacing the Nꢀmethyl group with various steric alkyl groups R
(hydrogen, ethyl and isopropyl); and (3) changing the allyl linker with other fragments.
These modifications generated ninetyꢀtwo derivatives (11-86, 89-94 and 100-109) in
total combining either Series B or C.
Figure 2
The synthesis routes for the preparation of derivatives 11-86 were outlined in
Scheme 1. The reduction of commercially available substituted cinnamaldehyde with
sodium borohydride yielded intermediates 6a-p, along with a bromination reaction by
phosphorus tribromide to generate intermediates 7a-q. The imidization of
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 57
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
coumaranꢀ5ꢀaldehyde with methylamine was further reduced by sodium borohydride,
yielding intermediate 10a. Then, target compounds 49, 52, 54, 56-60, 62, 64, 66,
7
0ꢀ71, 76-78, and 82 were achieved by coupling 10a with cinnamyl bromide 7a-q.
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
Two successive stages, including the MizorokiꢀHeck reaction and deprotection in
32
hydrochloric acid, were utilized to synthesize substituted cinnamaldehyde 8a-u,
which were difficult to obtain from commercial resources. Intermediates 9a-u, and
target derivatives 50, 51, 53, 55, 61, 63, 65, 67-69, 72-75, 79-81 and 83-86 were
prepared using the same synthetic routes, as described above. In parallel, the
preparation involved a nucleophilic aromatic substitution reaction between
3
ꢀpropanediol and 2,5ꢀdibromoꢀfluorobenzene to yield intermediate 10b, followed by
a bromination reaction to generate intermediate 10c. The core ring scaffold 10d was
constructed by cyclization of 10c with nꢀBuLi as the base at ꢀ70 °C, followed by N,
3
3
Nꢀdimethylformamide in nꢀBuLi at ꢀ70 °C again to yield intermediate 10e. Finally,
intermediate 10f and derivatives 11-48 were achieved.
Scheme 1
Derivatives 89-94 were prepared as described in scheme 2. Intermediates 87a and
87b were synthesized from 10e and coumaranꢀ5ꢀaldehyde, respectively, with
hydroxylamine hydrochloride using sodium hydroxide as the base at 40 °C, followed
by hydrogenation with palladium hydroxide. Then, derivatives 89 and 90 were
achieved by coupling 9j with the two intermediates 87aꢀb. Coumaranꢀ5ꢀaldehyde and
10e were imidized with ethylamine or isoꢀpropylamine, followed by reduction with
7
ACS Paragon Plus Environment

Page 9 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
sodium cyanoborohydride to yield 88aꢀd. Then, target derivatives 91ꢀ94 were
achieved by coupling 88aꢀd with 9j.
Scheme 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
Scheme 3 outlines the synthetic strategy used for the synthesis of derivatives
1
00-109. The synthesis of intermediate 95a was initiated from 8j via the Wittig
reaction. The reduction of 95a with diisobutyl aluminum hydride yielded intermediate
5b, and followed by a bromination reaction to generate intermediate 95c. The
reduction of 8j with sodium borohydride reduction to afford 96a, followed by H
9
2
34
under the catalysis of Pd/C, yielded intermediates 96b. 96b, which underwent
bromination reaction by carbon tetrabromide to generate 96c. After the Wittig reaction
of
methylꢀ(1ꢀmethylꢀtriphenylꢀphosphoranylidene)ꢀacetate
with
4
ꢀphenylꢀbenzaldehyde, ester 97a was obtained. The product was then reduced with
diisobutyl aluminum hydride to yield intermediate 97b, and followed by bromination
to generate 97c. Intermediate 98a was generated from 4ꢀphenylꢀbenzaldehyde by
Wittig reaction and then underwent esterification, followed by reduction to
35
intermediate 98b, and then the bromination of 98b with phosphorous tribromide to
98c. The starting material 4ꢀphenylꢀiodobenzene was reacted with
OꢀTBDMSꢀ2ꢀpropynyꢀ1ꢀol via Sonogashira coupling, and the product 99a was
converted to intermediate 99b by further deprotection. Then, the bromination of 99b
was performed with phosphorous tribromide to 99c. The key intermediates 95c-99c
were reacted with 10a or 10f according to a procedure described previously to obtain
derivatives 100-109.
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 57
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
Scheme 3
RESULTS AND DISCUSSION
In total, ninetyꢀtwo novel chroman and coumaran derivatives (11-86, 89-94 and
00-109) were designed and synthesized. Their chemical structures are shown in
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
Tables 1ꢀ3. These derivatives were synthesized through the routes outlined in
Schemes 1ꢀ3, and the details of the synthetic procedures and structural
characterization are described in the Supporting Information. All of the derivatives
were confirmed to have ≥ 95% purity (Table S1, Supporting Information), and were
identified with nonꢀPAINS on the web at http://fafdrugs3.mti.univparisꢀdiderot.fr/
36
recommended by editors from the ACS (American Chemical Society).
In Vitro Pigment Inhibitory Activities of Derivatives 11-86, 89-94 and 100-109
Initial studies involved exploration around region A and systemic variation in R
substitution for pigment inhibitory activities against S. aureus Newman. The results
are summarized in Table 1. In general, substitution with various types of groups at the
phenyl ring could remarkably affect pigment inhibitory activities, whereas methyl,
cycloalkyls or heteroaryls displayed disappointing results. Previous studies
preliminarily recognized that the electronꢀwithdrawing groups and electronꢀdonating
31
groups at the phenyl ring might have minor effects on activity. Herein, to investigate
the SARs of the two new class of scaffolds systemically, more electronꢀwithdrawing
(
fluoro, chloro, bromo, trifluoromethyl, difluoromethyl, nitro, phenyl, formate, cyan)
and electronꢀdonating groups (methyl, methoxyl, ethoxyl, tꢀbutyl) were introduced at
the phenyl ring. These results of modifications clearly demonstrated that there was no
9
ACS Paragon Plus Environment

Page 11 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
relationship between potency and electronic effect, whereas the substituted positions
at the phenyl ring substantially affected the activity. Subsequently, the substituted
positions on the phenyl ring obviously affect the activity (Table 1). The substitution at
the paraꢀposition showed more potency than the other two positions (19/57-31/69 at
paraꢀposition vs 32/70-37/75 at orthoꢀposition vs 38/76-43/81 at metaꢀposition).
Relatively, a suitable bulk of substituents can significant increase inhibitory activity
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
(57 vs 58 vs 59, 60 vs 61 vs 62 vs 63). The substitution at the paraꢀposition with
bromo, difluoromethyl, trifluoromethyl and phenyl (21, 23, 24, 31, 62 and 69)
provided significant improvements in activities.
Next, benefit from good performance after structure modification, we obtained
enough excellent derivatives for challenging the solubility and hERG inhibitory
activity. On one hand, eighteen derivatives, inhibitory potency at the singleꢀdigit
nanomolar level, were selected to explore the water solubility. As shown in Table 4, in
general, compared with benzocycloalkane compounds, chroman and coumaran
derivatives exhibited better solubility. New derivatives with the same substitutents as
31
previous reported compounds (Figure S5) significantly increased the solubility, at
least ~ 10 folds elevated more than 4, 5 and their analogs (20, 21, 24 vs 4, 4a, 4b vs 5,
5
a, 5b), whereas there was no significant difference between chroman and coumaran
derivatives (17 vs 55, 23 vs 61, 24 vs 62, 31 vs 69 and 45 vs 83), and these results
further demonstrated the feasibility of scaffold hopping. However, it must be
mentioned that the hydrophilic property of substitutents at the phenyl ring also
affected the solubility of compounds. Compared with the derivatives with hydrophilic
1
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 57
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
substitutents, derivatives with lipophilic substitutents (17, 22, 31, 55 and 69) restricted
the solubility (all less than 7 mg/mL). On the other hand, fifteen compounds (pigment
inhibition: IC50 ≤ 6 nM) were screened for investigating the hERG inhibition. As
opposed to the benefits of hydrophilic substitutents on solubility, they disclosed
passive influence in regard to hERG inhibition, while lipophilic substitutents exhibited
positive effects. As shown in Table 4, obviously, derivatives 17, 55, 31, 69 and 30
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
(hERG inhibition, IC50 > 6 ꢁM in whole) with lipophilic substitutents exhibited
superior than the other ones (hERG inhibition, IC50 < 4 ꢁM in whole). Even so,
different from the elevation of solubility by means of many mature methods, the
methods about decreasing hERG inhibition were currently occupied by chemical
modification, which forced us to give priority to derivatives with weaker hERG
inhibition for further explorations. Moreover, we had suggested that the problematic
solubility could be resolved adequately by combining structural modification with salt
type selection, and salt type selection as the other strategy was established and
discussed below. Therefore, 31 and 69 were selected to further investigate due to their
potent pigment inhibitory activity (IC50 < 4 nM), weak hERG inhibition (IC50 > 10 ꢁM)
and fair water solubility (Table 4). More coincidence, both of them contained
paraꢀbiphenyl in their respective scaffolds, so we aimed to identify other potent
inhibitors easier.
Table 1, Table 4
Analysis of the data in Table 2 revealed that once the Nꢀmethyl group was
removed (89, 90) or converted into an ethyl or isopropyl group (91-94), the pigment
11
ACS Paragon Plus Environment

Page 13 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibition activities abated (IC50 > 1000 nM). As shown in Table 3, when the vinyl
linker was replaced with a butadienyl, derivative 105 (IC50 = 3.0 ± 0.4 nM) evenly
displayed the most potency in the other scaffold. Fortunately, the hERG inhibition
IC50 value of 105 was greater than 40 ꢁM, which was better than that of 31 (hERG
inhibition IC50 = 32.1 ꢁM, Table 4). All results showed that 105 was more eligible as
the candidate compound. Elimination of the double bonds (101, 104, 106 and 109) led
to a loss of pigment inhibitory activities (IC50 > 1000 nM). Subsequently, the
introduction of an additional methyl moiety deprived the activities of derivatives (102,
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
1
07, IC > 1000 nM). Finally, the introduction of a branched methyl on a vinyl linker
50
resulted in a decrease in pigment inhibitory activity (IC50 = 17.3 ± 0.2 nM by 103 and
IC50 = 23.5 ± 0.8 nM by 108).
Tables 2-3
Structure-Activity, -Solubility and -hERG Inhibition Relationship
On the basis of the structural features and pigment inhibitory activities data, the
SARs are summarized in detail. (1) According to the modification of region A, the
potency improved obviously in the substituted phenyl >> heterocyclic = cycloalkyl.
Regarding the substituted phenyl, the electronic effect of the substituent was limited
to the pigment inhibitory activity, whereas the substituted position was fatal; notably,
the paraꢀposition was the best substituted position at the phenyl ring. The suitable size
of the substituent also affected the potency obviously. On the basis of these obtained
results and hERG inhibition, the paraꢀbiphenyl substituents represented more potency
than the others. (2) In the studied set replacing the Nꢀmethyl group with various steric
1
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 57
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
alkyl groups (hydrogen, ethyl and isopropyl), the size of the substituted group was
crucial to the potency, and the Nꢀmethyl group was necessary for pigment inhibitory
activities. (3) Multiallyl derivatives were synthesized by inserting one vinyl, and all of
the derivatives showed excellent inhibitory activities, whereas other modifications
showed that the unsubstituted allyl linker was critical for achieving high potency. (4)
Based on the results of solubility and hERG inhibition, we found that the strategy of
inserting hydrophilic oxygen into scaffolds remarkably increased the solubility of
derivatives by comparing with the previous compounds, whereas lipophilic
substituents at the phenyl ring weakened the efficiency but forcibly reduced hERG
inhibitory activity.
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
Target CrtN Determination of Chroman and Coumaran Derivatives.
Benzocycloalkane derivatives were determined to be CrtN inhibitors in our
31
previous research. To confirm whether the target enzyme was the same as coumaran
and chroman derivatives, we conducted identical HPLC experiments at 286 nm for
the analysis of 4,4’ꢀdiapophytoene (the product of CrtM and the substrate of CrtN).
As seen in Figure 3B, both the retention time and UV absorption spectra of
4,4′ꢀdiapophytoene were measured by monitoring the HPLC expression of CrtM in E.
coli, which was found in the similar HPLC expression of wildꢀtype S. aureus
Newman (Figure 3C). Moreover, this HPLC peak disappeared in the carotenoid
extracts of the CrtM mutant (Figure 3D) and was strengthened in the CrtN mutant
(Figure 3E). It was reassuring to note that this peak belonged to 4,4’ꢀdiapophytoene.
Comparing the HPLC peaks of 69 and 105ꢀtreated wildꢀtype S. aureus Newman
1
3
ACS Paragon Plus Environment

Page 15 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
Figure 3GꢀH) with NTFꢀtreated Newman (Figure 3F), these profiles were not only
identical to each other, but they were also similar to those of CrtM in E. Coli and the
CrtM mutant. Consequently, all of the results demonstrated that CrtN was the target of
coumaran and chroman derivatives 69 and 105.
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 3
In Vitro Enzymatic and Pigment Inhibitory Activities of Four Representative
Derivatives
29
In Vitro CrtN Enzymatic Inhibitory Activities. According to previous protocol,
selected derivatives in Table 5 (31, 62, 69 and 105) with pigment inhibitory activities
IC50 < 4 nM in S. aureus Newman were evaluated for enzymatic inhibitory activities
in vitro. The results showed that four derivatives displayed excellent enzymatic
inhibitory activity of CrtN at submicromolar concentrations, though the enzymatic
activities were less effective than pigment inhibition due to the accumulation of
compounds in the cytoplasm of S. aureus and three sequential reactions using CrtN in
29ꢀ31
the biosynthesis of STX, as we speculated in our previous papers.
derivatives 69 (IC50 = 181.3 ± 34.3 nM) and 105 (IC50 = 179.6 ± 29.5 nM) were
superior than 4 (IC50 = 320 nM)
In Vitro Pigment Inhibitory Activities against MRSA. Next, two community
In particular,
.
37ꢀ38
acquired MRSAs, USA400 MW2 and USA300 LAC,
along with other four
hospitalꢀacquired MRSAs, Mu50 (vancomycinꢀintermediate resistance, VISA/MRSA),
3
9ꢀ40
and NRS271 (linezolidꢀresistance, LRSA/MRSA),
were used to demonstrate the
broad spectra efficiency of derivatives 31, 62, 69 and 105. Pigment inhibition results
1
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 57
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
of these MRSA strains are shown in Table 5, in which derivative either 69 or 105
showed better inhibitory activities in Mu50 (IC50 = 0.33 ± 0.3 nM, IC50 = 0.36 ± 0.1
nM, respectively) and NRS271 (IC50 = 1.7 ± 0.2 nM, IC50 =0.4±0.1 nM, respectively)
in their respective series. For demonstrating the broad spectrum of 69 and 105, NF65Y
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
(vancomycinꢀintermediate
resistance,
VISA/MRSA)
and
LRSA56
(linezolidꢀresistance, LRSA/MRSA) were introduced into the assay, the results
showed that two compounds displayed excellent potency. In contrast to the antibiotic,
incubation with derivatives 69 or 105 did not affect the growth of S. aureus strains
(Newman strain and three MRSA strains) at 0.2 mM. (Figure S1, Supporting
Information).
Table 5
Effects of 69 and 105 on Sensitizing S. aureus to Immune Clearance.
Hydrogen Peroxide Killing Assay. Having confirmed that derivatives 69 and 105
displayed excellent potency in vitro, we then evaluated their effects on sensitizing S.
aureus and three other MRSA strains (USA400 MW2, USA300 LAC, and Mu50)
with pigment or nonꢀpigmented for immune clearance as in the previous report,
2
9
including the protocol of hydrogen peroxide killing and human whole blood killing.
As shown in Figure 4, after incubation with derivatives 69 or 105 (1 ꢁM, respectively),
nonꢀpigmented S. aureus cells were more vulnerable to be killed by 1.5% H O
2
2,
compared with the untreated S. aureus (mock) (survival rate, 2.6% vs 3.3% vs 33.3%).
In parallel, the survival rate of the known antioxidant Nꢀacetylcysteine (NAC)ꢀtreated
S. aureus cells were more elevated than the mock, as expected (60.7% vs 33.3%).
1
5
ACS Paragon Plus Environment

Page 17 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Similarly, the survival percentages of the three MRSA strains were reduced by factors
of ∼ 10 (2.3% vs 24.3% by 69, 2.6% vs 24.3% by 105, in Mu50), ∼ 10 (2.0% vs
2
4.0% by 69, 2.3% vs 24.0% by 105, in USA300 LAC), and ∼10 (2.6% vs 26.3% by
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
9, 2.5% vs 26.3% by 105, in USA400 MW2), respectively. The survival rates of the
three NACꢀtreated MRSA strains tended to be higher (50.7% vs 24.3% in Mu50,
4.7% vs 24.0% in USA300 LAC, 57.0% vs 26.3% in USA400 MW2). All of the
results above proved that the addition of H (with strong oxidation) exerted impacts
5
2
O
2
on the MRSA strain survival and the pigment definitely acted as a protective
antioxidant.
Human Whole Blood Killing Assay against S. aureus. Firstly, incubating time and
bacterial survival relationship was explored owing to the reproduce ability difference
of strains in different time stages. Five time intervals were established throughout the
assay, and associated with two hours in each interval. The results manifested that the
bacterial survival of 69 or 105 treatment groups were lowest during the third interval
(Figure S2, Supporting Information). On the basis of these results we defined six
hours as the standard incubating time in whole blood killing assay. Subsequently, the
other experiment was performed to compare the effects of compoundꢀtreated S.
aureus with nonꢀtreated bacteria by human whole blood killing, as in previous
research. As shown in Figure 5, the survival rates of the untreated S. aureus Newman
and the other three MRSA strains were much higher, compared to the
compoundꢀtreated ones (69, 105 and mock in order, 0.28% vs 0.63% vs 8.67% in
1
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 57
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
Newman, 0.46% vs 0.45% vs 9.33% in Mu50, 0.12% vs 0.23% vs 12.0% in USA300
LAC, and 0.20% vs 0.53% vs 20.67% in USA400 MW2).
Meanwhile, CrtN mutant S. aureus (depigment S. aureus) was introduced to
incubate with/without our compounds in human blood, for determining whether our
compounds stimulated mammalian immune response or not. This assay was made up
of six groups, nonꢀtreated Newman, nonꢀtreated CrtN mutant S. aureus, 69ꢀtreated
Newman, 69ꢀtreated CrtN mutant S. aureus, 105ꢀtreated Newman and 105ꢀtreated
CrtN mutant S. aureus. As shown in Figure S3 in Supporting Information, although
the nonꢀtreated CrtN mutant S. aureus group represented more potency compared with
nonꢀtreated Newman (survival rate, 0.9% vs 9.3%) in whole blood killing, its
efficiency had no significant difference with 69/105ꢀtreated Newman (0.9% vs 1.1%
vs 0.8%), and testified that nonꢀpigment S. aureus itself was actually vulnerable to
immune clearance. Similarly, compared with nonꢀtreated CrtN mutant S. aureus group,
there were no more effects exhibited by 69/105ꢀtreated groups against CrtN mutant S.
aureus (0.9% vs 0.76% vs 0.93%), which demonstrated the efficiency of inhibiting
unpigmented S. aureus failed to be strengthened with/without the treatment of our
compounds. Furthermore, we suggested that compounds 69/105 had no effect on the
stimulation of mammalian immune response in human blood killing assay.
Consequently, all above results were indicated that derivatives 69 and 105 exclusively
sensitized S. aureus to immune clearance in human blood.
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
Figures 4-5
In Vivo Pigment Inhibitory Activities of Derivatives 69 and 105
1
7
ACS Paragon Plus Environment

Page 19 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Derivatives 69 and 105 had inspiring activities in vitro, and to test their potential
7
pigment inhibitory activities in vivo, mice were challenged with 3.5×10 CFU
Newman bacteria to evaluate the efficiencies of 69 and 105 on affecting the outcome
of S. aureus (Figure S4, Supporting Information). The untreated group died out within
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
3
7
6
days, whereas 69 and 105 exhibited the excellent protective efficiencies, resulting in
3% and 67% animal survival respectively. As time went on to the sixth day, over
0% of the compoundꢀtreated mice were alive. These initial investigations clearly
proved that the in vivo 69 and 105 treatment weakened the virulence of S. aureus
Newman.
Next, four different drug regimens were applied, first two different doses, both
0.4 mg/bid/4.5 d (20 mg/kg/dose, 180 mg/kg in total dose) and 0.1 mg/bid/4.5 d (5
mg/kg/dose, 45 mg/kg in total dose) with pretreatment (preꢀadministration with drugs
or derivatives 24 hours in advance before the infection of S. aureus strains) were set in
30ꢀ31
previous works.
Then, both 0.4 mg/bid/3.5d (20 mg/kg/dose, 140 mg/kg in total
dose) and 0.1 mg/bid/3.5d (5 mg/kg/dose, 35 mg/kg in total dose) with normal
treatment (administered 6 hours after the infection of S. aureus strains) were set to
simulate the real situation of infection and treatment. Linezolid (LZD) and
vancomycin (VAN) were used for two positive controls at a dosage of 0.4 mg/bid/4.5d
(20 mg/kg/dose, 180 mg/kg in total dose) with pretreatment.
In Vivo Effects of Derivatives 69 and 105 on Attenuating the Virulence of S.
7
aureus Newman. After infection with 2.3 × 10 colonyꢀforming units (CFU) of S.
aureus Newman bacteria via retroꢀorbital injection, the mice were sacrificed at ~108h.
1
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 57
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
We then measured bacterial survival in the hearts and livers. As shown in Figure 6A,
in general, each case showed a significant reduction (P < 0.001 and more than a 95%
decrease in surviving bacteria), compared to nonꢀtreatment group (6.18 log10 CFU) in
hearts. In addition, 69 in the lowꢀdosage case (5 mg/kg/dose) in normal treatment
exhibited the same protective effects as the two positive cases with at high dosages
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
(20 mg/kg/dose) in pretreatment (4.58 log10 CFU by 69 vs 4.62 log10 CFU by
vancomycin vs 4.186 log10 CFU by linezolid). As shown in Figure 6B, firstly,
compared with the nonꢀtreatment group (6.65 log10 CFU), the staphylococcal loads of
all of the compoundꢀtreated groups were decreased in the livers with statistical
significance (P < 0.001, except 105 treatment in 20 mg/kg/dose). The activities of 69
treatment groups were comparable to these in the 105 treatment groups. Apart from
being inconsistent with the relationship of the dosageꢀactivity (0.4 mg/bid vs 0.1
mg/bid), there was no significant difference between preꢀadministration and normal
administration of 69 or 105 in the respective groups. In addition, the bacterial survival
rates of the highꢀdosage cases were slightly elevated relative to the two positive
groups, and they demonstrated a comparative clearance (all group significance P >
0.05). To our surprise, both 69 treatment and 105 treatment with low dosages in the
normal group (5 mg/kg/dose, 2.12 log10 CFU reduction by 69 treatment, 1.80 log10
CFU reduction by 105 treatment, respectively) had a more than 98% decrease in
surviving bacteria, especially the 69 treatment, which showed a 99.3% decrease in
surviving bacteria, comparable to the LZDꢀtreatment group (1.74 log10 CFU reduction,
a 98.2% decrease in surviving bacteria) and the VANꢀtreatment group (1.77 log10 CFU
1
9
ACS Paragon Plus Environment

Page 21 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
reduction, a 98.3% decrease in surviving bacteria). These results demonstrated clearly
that in vivo 69 treatment excellently attenuated the pathogenicity of S. aureus
Newman in contrast to the positive groups, especially the effects of 69 treatment at
lowꢀdosages in the normal treatment group. Moreover, the lack of significant
difference between the preꢀadministration and normal administration groups indicated
that our derivatives were closer to practical application.
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 Effects of Derivatives 69 and 105 on Attenuating the Virulence of Mu50
MRSA. Next, to determine whether the effects of derivatives 69 and 105 could also
be reflected in multiꢀdrug resistant MRSA strains in vivo, a vancomycinꢀintermediate
S.aureus (VISA, Table S3, Supporting Information), Mu50 infection model was
established. The data in Figure 7A revealed that derivativeꢀtreatment groups
significantly decreased Mu50 staphylococcal loads in the hearts. All of them
corresponded to a more over 95.0% decrease in surviving bacteria (significance P <
0.001), while the derivativeꢀtreatment groups were also more efficacious than the two
positive groups, VAN and LZD, by 0.96 and 1.07 log10 CFU, corresponding to 89.1%
and 91.5% decreases, respectively. In general, all of the regimens with 69 represented
excellent potency without significant differences, whereas enormous differences were
shown between the preꢀadministration and normal administration cases in the 105
treatment groups. Obviously, 69 and 105 treatment with low dosages in normal
administration groups displayed better activity among the respective groups again
(2.39 log10 CFU reduction by 69 treatment, 2.32 log10 CFU reduction by 105
2
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 57
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
treatment, more than 99.0% decrease in surviving bacteria, respectively). As shown in
Figure 7B, compared with the nonꢀtreatment group, the effects of attenuating the
virulence of Mu50 MRSA in the livers of the derivativesꢀtreatment group indicated
more potency (all groups’ significance P < 0.001, more than a 98% decrease in
surviving bacteria), whereas the two positive groups performed better than most of the
derivativesꢀtreatment groups (2.75 log10 CFU reduction by VAN treatment and 3.19
log10 CFU reduction by LZD treatment, corresponding to 99.8% and 99.9% decreases,
respectively.). The therapeutic efficiency between 69 treatment at lowꢀdosage in
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
normal treatment by 3.15 log CFU and the two positive groups were at comparable
10
levels, with a remarkable ∼99.9% decrease in surviving bacteria.
Figure 7
In Vivo Effects of Derivatives 69 and 105 on Attenuating the Virulence of NRS271
MRSA. To assess the spectrum of activities of 69 and 105, we evaluated their
antibacterial activity against NRS271 strains in vivo (linezolidꢀresistant S.aureus,
LRSA, Table S3, Supporting Information). As shown in Figure 8A, although the value
of the LZD group was superior to the nonꢀtreatment group (significance P < 0.001
reduction, a 91.7% decrease in surviving bacteria), it still displayed weak inhibitory
activity, compared with the other groups in the experiment. In contrast to the
LZDꢀtreatment group treated with NRS271, the derivativeꢀtreatment groups always
exhibited significant effects in comparison with nonꢀtreatment group in hearts (all
compoundꢀtreatment group significance P < 0.001, over 93% decrease in surviving
bacteria). In addition, either 69 or 105 in the lowꢀdosage case represented more
2
1
ACS Paragon Plus Environment

Page 23 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
efficacious results (significance difference with the LZDꢀtreatment group P < 0.01, ~
2
.4 log10 CFU reduction, a remarkable ∼ 99.5% decrease in surviving bacteria,
respectively), and they were not only at comparative levels to the VANꢀtreatment
group (2.6 log10 CFU reduction, a ∼99.8% decrease in surviving bacteria), but they
also had greater potency than the other groups. As shown in Figure 8B, similar to the
effects in hearts, the strains were weakened continuously in the LZDꢀtreatment group
in livers. However, our treatment groups still showed excellent performance. In
particular, four 105ꢀtreatment groups significantly decreased their NRS271
staphylococcal loads in the livers by 2.13 log10 CFU at least (more than a 99.3%
decrease in surviving bacteria), whereas the LZDꢀtreatment group led to a reduction
by only 1.19 log10 CFU, making it less efficacious than 105 treatment. In particular,
the bacterial survival rates of 105 treatment in lowꢀdosage cases demonstrated
comparative clearance to the VANꢀtreatment group, while the 69ꢀtreatment groups
maintained a comparative level with LZD treatment and a worse level than the
VANꢀtreatment groups.
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
In summary, two STX biosynthesis inhibitors, either 69 or 105, was capable of
decreasing the bacterial survival in the abscess formation model of S. aureus Newman,
MRSA and LRSA strains. Moreover, some of the 69 or 105 treatment groups were
more efficacious than the two positive groups, especially 69 treatment with a low
dosage in normal treatment. Furthermore, derivative 69 was slightly more successful
than 105 in comparing the two under the same conditions. In addition, the experiment
demonstrated that there was no significant difference between the pretreatment
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 57
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
regimens with normal treatment. To our surprise, the lowꢀdosage groups had excellent
performance, compared with the highꢀdosage cases in general. The speculated reasons
are provided as follows: (1) according to the pigment inhibitory activity of derivative
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
(
IC < 4 nM), the high dosage was more than sufficient for its purpose; and (2) high
50
dosages possibly weakened the ability of immune clearance in vivo, which
destabilized the situations of dosage and activity.
Figure 8
In Vivo Rat Pharmacokinetics Parameters of 69
Since its potency and therapeutic profiles, candidate 69 as the representative was
evaluated to rat pharmacokinetic (PK) in vivo. The pharmacokinetic parameters were
administrated with either IV injection of a single 5mg/kg dose or PO administration of
a single 10 mg/kg dose. As a result, candidate 69 exhibited fair pharmacokinetic
properties (F = ~7%) after oral administration of a 10 mg/kg dose to rats, and its
pharmacokinetic data are summarized in Table 6. Additionally, pharmacokinetic
analysis about injectable administration revealed that 69 had a Cmax of 2044ng/mL and
t1/2 of 2.3 h.
Table 6
In Vitro Antifungal Activities of 69 and 105
Because 69 and 105 was derived from mother nucleus structure of 2, a drug used
in antifungals, it was rational to investigate whether our derivatives inherited the
antifungal activity. As shown in Table 7, three fungus strains and three firstꢀline drugs
were chosen to proceed with the in vitro assay. Compared with the three positive
2
3
ACS Paragon Plus Environment

Page 25 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
groups, 69 and 105 exhibited weak activities against all three dermatophytes. This
result partly demonstrated our derivatives to be promising selective antibacterial
candidates.
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 7
Salt Type Selection
Although overall solubility of chroman and coumaran derivatives significantly
increased by inserting a hydrophilic oxygen into the cycloalkane, the optimized
candidates 69 (4.4 mg/mL) and 105 (3.8 mg/mL) still were not more soluble than
benzocycloalkane 4. Herein, salt type selection was set up for eliminating this flaw.
As shown in Table 8, ten different acids were introduced to salify with either 69 or
105 free base. Notably, phosphate 69b and 105b represented remarkable improvement
of solubility (16.2 mg/mL and 12.9 mg/mL, respectively).
Table 8
Conclusion
In this study, two series of derivatives (chroman and coumaran), ninetyꢀtwo in
total, were synthesized, and we evaluated their activity against wildꢀtype S. aureus
Newman and several MRSA strains systematically. According to the results of the
pigment inhibitory activities against S. aureus Newman, unambiguous SARs were
obtained. Four derivatives (31, 62, 69 and 105) were selected and subjected to further
evaluation of their outstanding pigment inhibitory activities. Finally, derivatives 69
and 105 were carefully selected from respective series since they exhibited favorable
characteristics in vitro, including hERG activities (IC50 = 11.3 ꢁM and IC50 > 40 ꢁM,
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 57
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
respectively), enzymatic activities and pigment inhibitory activities of six other MRSA
strains. The experiments of bacterial growth assays and immune clearance proved that
6
2 2
9 and 105 could sensitize S. aureus strains to be killed by H O or human blood,
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
neither interfering the bacterial growth nor stimulating mammalian immune response.
Derivatives 69 and 105 also displayed a broad and significantly potent antibacterial
spectrum in vivo against pigmented S. aureus Newman, Mu50 (VISA/MRSA) and
NRS271 (LRSA/MRSA). By means of the four different drug regimens established,
6
9 treatment with lowꢀdosages in the normal treatment group (5 mg/kg/dose, 35
mg/kg in total dose) showed comparable effects to the two positive groups (linezolid
and vancomycin) in the liver in Newman and Mu50 and they even more effective than
in the two positive groups in the heart in Mu50, whereas 105 treatment with
lowꢀdosage cases (5 mg/kg/dose) possessed excellent inhibitory activities compared to
vancomycin in the hearts and livers in NRS271. Additionally, 69 and 105 exhibited
weak activities against fungals, and 69 exhibited the average pharmacokinetic
efficiency. Furthermore, phosphate 69b and 105b disclosed remarkable solubility by
salt type selection. The favorable in vitro and in vivo activities of 69 and 105,
combined with their superior solubility and hERG safety profiles, made them
promising drug candidates. Further evaluation of these two drug candidates is
currently under way.
EXPERIMENTAL SECTION
General Chemistry
2
5
ACS Paragon Plus Environment

Page 27 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Reagents and solvents were obtained from commercial suppliers at high quality
and were used without further purification. TLC was performed on a HSGF 254
(
150ꢀ200 ꢁm thickness; Yantai Huiyou Co., China). UV light and I
monitor synthetic progress. Column chromatography was performed on silica gel
200ꢀ300 mesh), eluted with ethyl acetate and petroleum ether. NMR spectra data
2
were used to
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
(
were obtained on a Bruker AMXꢀ400 NMR using TMS as an internal standard.
1
Chemical shifts were provided in parts per million. H NMR data were reported from
the aspect of multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m =
multiplet and br = broad), coupling constant (Hz) and integrated value. Lowꢀ and
highꢀresolution mass spectral (MS) data were acquired with electron spray ionization
(ESI) produced by a Finnigan MATꢀ95 and LCQꢀDECA spectrometer. The purity of
each compound (> 95%) was determined by HPLC on an Agilent 1100 with a
quaternary pump and diodeꢀarray detector (DAD). The melting points of each
compound were determined on an SGW Xꢀ4 melting point apparatus.
General Procedure A for Preparation of Salts.
Taking 69 free base as an example, to a solution of oily derivative 69 free base
(100 mg) in ethyl ether (10 mL) stirred at room temperature with hydrochloride gas or
1
.1 equivalent of other acids for 1 min. After stirring for 15 min, the solvent was
removed by rotary evaporation and the residue was suspended in ethyl
acetate/petroleum ether (1:100, v/v, 10 mL) for an additional hour of agitation. The
precipitate was filtrated and washed with petroleum ether to obtain the final
compound in the form of hydrochloride or other acids. All other final derivatives
2
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 57
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
underwent through this process to yield an amorphous, solid form or oil.
Spectroscopic data reported below are in their hydrochloride form (Supporting
Information).
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
General Procedure B for Preparation of Salts.
To a solution of 69 or 105 (10 mmol) in EtOAc was treated with the solution of
acid (10 mmol) in EtOH in batches at room temperature, the mixture was then
allowed to stir at 80 ºC for an additional 1 hour, the solvent was removed by rotary
evaporation and the residue was suspended in ether for an additional hour of agitation.
The precipitate was filtrated and washed with ether to obtain the final salt,
respectively.
Biological
Pigment Inhibition Assay. Newman bacteria were cultured in TSB (4 mL) medium at
37 ºC for 48 h with inhibitor compounds dissolved in DMSO and diluted to a set of
concentrations, in duplicate. Three milliliters of bacterial cultures were centrifuged
and washed twice with 0.01 M phosphateꢀbuffered saline (PBS) and resuspended in
methanol to extract pigment. The absorbance value was determined at 450 nm on a
NanoDrop 2000c (Thermo Scientific) spectrophotometer. IC50 values were calculated
using Graphpad Prism software, version 5.0. The IC50 values of the USA300 LAC,
USA400 MW, Mu50, NRS271, NF65Y and LRSA56 strains were determined in the
same manner.
CrtN Enzyme Inhibition Assay. Diapophytoene was purified from
diapophytoeneꢀproducing E. coli BL21 (DE3) /pET28a :: crtM extracted with acetone.
2
7
ACS Paragon Plus Environment

Page 29 of 57
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Eight milligrams of diapophytoene were mixed with 24 mg of phosphatidylcholine
3
(SigmaꢀAldrich) in 200 ꢁL of CHCl to prepare the diapophytoene emulsion. The
mixture was spinꢀdried and incubated with 2 mL of 0.02 M HEPES buffer (pH = 7.5)
followed by sonication in ice water to obtain the homogeneous emulsion. For the
preparation of CrtN lysate, E. coli BL21 (DE3)/pET28a :: crtN was subꢀcultured into
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
1
000 mL of LB broth supplemented with 50 ꢁg/mL kanamycin to achieve an OD600
Eppendorf, BioPhotometer plu) of ~ 0.1 and was grown to an OD600 of ~ 0.5. The
expression of 6HisꢀCrtN protein was induced with 0.5 mM
(
isopropylꢀβꢀDꢀthiogalactoside (IPTG) at 16 °C overnight. The cells were harvested,
and the pellets were suspended in 30 mL HEPES buffer and lysed at 4 °C by
sonication.
The enzyme activity was determined in triplicate, with a total of 700 ꢁL of the
following: 50 ꢁL of diapophytoene emulsion, 70 ꢁL of different concentrations of
compounds or mock (ddH O), 3.5 ꢁL of FAD stock solution (10 mM), and 300 ꢁL of
2
CrtN lysate (~1.41 mg of CrtN, as estimated by western blotting using a known
concentration of the purified 6HisꢀcrtN protein)，followed by 0.02 M HEPES buffer
(pH 7.5) to 700 ꢁL. The tests were proceeded in an anaerobic atmosphere by adding a
final concentration of 20 U/mL glucose oxidase (SigmaꢀAldrich, G2133), 20000
U/mL catalase (SigmaꢀAldrich, C1345), and 2 mM glucose as an oxygenꢀtrapping
system. The reaction mixture was started by adding the lysate and incubating it
overnight at 37 ºC and then stopped by methanol. The pigments were extracted twice
against 700 ꢁL of chloroform. The organic phase was combined, concentrated, and
2
8
ACS Paragon Plus Environment