Disulfide Bond-Containing Ajoene Analogues As Novel Quorum Sensing Inhibitors of Pseudomonas aeruginosa

Article abstract of DOI:10.1021/acs.jmedchem.6b01025

Since its discovery 22 years ago, the bacterial cell-to-cell communication system, termed quorum sensing (QS), has shown potential as antipathogenic target. Previous studies reported that ajoene from garlic inhibits QS in opportunistic human pathogen Pseudomonas aeruginosa. In this study, screening of an in-house compound library revealed two sulfur-containing compounds which possess structural resemblance with ajoene and inhibit QS in bioreporter assay. Following a quantitative structure-activity relationship (SAR) study, 25 disulfide bond-containing analogues were synthesized and tested for QS inhibition activities. SAR study indicated that the allyl group could be replaced with other substituents, with the most active being benzothiazole derivative (IC₅₀ = 0.56 μM). The compounds were able to reduce QS-regulated virulence factors (elastase, rhamnolipid, and pyocyanin) and successfully inhibit P. aeruginosa infection in murine model of implant-associated infection. Altogether, the QS inhibition activity of the synthesized compounds is encouraging for further exploration of novel analogues in antimicrobial drug development.

Full text of DOI:10.1021/acs.jmedchem.6b01025

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Disulfide bond-containing ajoene analogs as novel
quorum sensing inhibitors of Pseudomonas aeruginosa
July Fong, Mingjun Yuan, Tim Holm Jakobsen, Kim T. Mortensen, May Margarette Salido Delos
Santos, Song Lin Chua, Liang Yang, Choon-Hong Tan, Thomas Eiland Nielsen, and Michael Givskov
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01025 • Publication Date (Web): 15 Dec 2016
Downloaded from http://pubs.acs.org on December 16, 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 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Disulfide bondꢀcontaining ajoene analogs as novel
quorum sensing inhibitors of Pseudomonas
aeruginosa
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
July Fong1, Mingjun Yuan1, Tim Holm Jakobsen2, Kim T. Mortensen 3, May Margarette Salido
Delos Santos1, Song Lin Chua1, Liang Yang1,4*, Choon Hong Tan1,5, Thomas E. Nielsen1,2,
Michael Givskov1,2*
1Singapore Centre for Environmental Life Sciences Engineering (SCELSE), Nanyang
Technological University, 60 Nanyang Drive, Singapore 637551, Singapore
2 Costerton Biofilm Center, Department of Immunology and Microbiology, University of
Copenhagen, 2200 København N, Denmark
3 Department of Chemistry, Technical University of Denmark, Kemitorvet, 2800 Kgs. Lyngby,
Denmark
4School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive,
Singapore 637551, Singapore
5Division of Chemistry and Biological Chemistry, School of Physical and Mathematical
Sciences, Nanyang Technological University, 21 Nanyang Link, Singapore 637371, Singapore
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 45
1
2
3
4
ABSTRACT
5
6
7
8
9
Since its discovery 22 years ago, the bacterial cellꢀtoꢀcell communication system, termed quorum
sensing (QS), has shown potential as antiꢀpathogenic target. Previous studies reported that ajoene
from garlic inhibits QS in opportunistic human pathogen Pseudomonas aeruginosa. In this study,
screening of an inꢀhouse compound library revealed two sulfurꢀcontaining compounds which
possess structural resemblance with ajoene and inhibit QS in bioreporter assay. Following a
quantitative structureꢀactivity relationship (SAR) study, 25 disulfide bondꢀcontaining analogs
were synthesized and tested for QS inhibition activities. SAR study indicated that the allyl group
could be replaced with other substituents, with the most active being benzothiazole derivative
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 = 0.56 ꢀM). The compounds were able to reduce QSꢀregulated virulence factors (elastase,
rhamnolipid and pyocyanin), and successfully inhibit P. aeruginosa infection in murine model of
implantꢀassociated infection. Altogether, the QS inhibition activity of the synthesized
compounds is encouraging for further exploration of novel analogs in antimicrobial drug
development.
INTRODUCTION
Infectious diseases are the direct cause of 16 million deaths annually.1 Among those, an
estimated 65% fatalities are associated with biofilm infections, where a chronic inflammatory
phase with increased resistance towards antibiotic treatments are often observed.2 It has been
reported that bacteria in the biofilm mode are orders of magnitude more tolerant to antibiotics in
compared with their planktonic counterparts.3 For two decades, academia has proposed a
paradigm shift in the development of new antimicrobial agents which includes antiꢀvirulence and
antiꢀpathogenic strategies.4 According to our previous work, “antiꢀpathogenic” refers to an
ACS Paragon Plus Environment
2

Page 3 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
antibacterial strategy that doesn’t kill bacteria per se but rather attenuates virulence and the
capacity to cause diseases.5 Most commonly available antibiotics primarily focus on the
bactericidal properties which inherently stimulate bacterial evolution.6 Instead of targeting
components essential for growth, the antiꢀpathogenic strategy relies on attenuating cascades of
bacterial gene expression leading to pathogenicity (i.e. microbial attachment, invasion, and
biofilm formation, secretion of toxins, chemical signaling and immune evasion, etc.).7
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
For some pathogenic bacterial species, it has been shown that central processes in biofilm
development are regulated by quorum sensing (QS), in particular those involved in shielding the
biofilm bacteria from antibiotics and phagocytic neutrophils.8 In P. aeruginosa, the two main QS
system are regulated by Nꢀacyl homoserine lactones (AHLs)ꢀbased signaling molecules las and
rhl systems. In the las system, LasI synthase directs the synthesis of Nꢀ(3ꢀoxododecanoyl)ꢀLꢀ
homoserine lactone (3ꢀoxoꢀC12ꢀHSL), which triggers the expression of LasR to induce virulence
genes such as lasB, apr and toxA.9 The LasIR system also positively regulates RhlIR system,
which directs the synthesis of NꢀbutanoylꢀLꢀhomoserine lactone (C4ꢀHSL) and enhances the
transcription of lasB and rhamnolipid synthesis genes rhlAB.10 The rhl system also directs the
synthesis of rhamnolipid, pyocyanin and hydrogen cyanide synthase.11 The third intercellular
signal is 3ꢀheptylꢀ3ꢀhydroxyꢀ4ꢀquinolone, which is also referred as Pseudomonas quinolone
signal (PQS). PQS system provides regulatory link and intertwined between the two AHLꢀbased
QS systems, where LasR and RhlR additionally regulate the PQS system.12 On top of that, it also
autoinduces its own production via interaction with transcriptional regulator PqsR and positively
modulates several virulence genes expression.12a Altogether, the QS signals control 5ꢀ10% of the
P. aeruginosa genome through its complex yet finely regulated mechanism.13 Recently, QS
inhibitors (QSI) have been proposed as a promising agent for controlling persistent infections.4b
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 45
1
2
3
4
Deletion of one or more QS genes in P. aeruginosa result in lower virulence factors as compared
5
6
to the wild type and render biofilm bacteria more susceptible towards antibiotic treatment.4a
7
8
9
Till date, nonꢀnative AHLs remain the most extensively studied class of synthetic QSIs,14 with
several SAR studies been carried out around AHLs backbone and halogenated furanones been
reported.14ꢀ15 However, the HSL is susceptible to hydrolysis at physiological pH and degradable
by mammalian lactonases,16 and the halogenated furanones suffer from inherent reactivity.17 The
resulting side products are either inactive against QS or toxic, thus limiting the use of such
molecules as pharmaceutical agents.18 The exploration of new compound classes that are
structurally distinctive from native AHLs is needed to ultimately deliver clinical candidates.
In the present study, we report a SAR study on novel QSIs identified through the screening of inꢀ
house chemical libraries against a wellꢀestablished QS reporter strain P. aeruginosa PAO1ꢀlasBꢀ
gfp.19 Four of these compounds were discovered to inhibit lasB, rhlA, and pqsA expression as
judged from bioreporter assay based on expression of green fluorescence protein (GFP) in P.
aeruginosa PAO1 (lasBꢀgfp; rhlAꢀgfp; pqsAꢀgfp). Moreover, we also discovered important
scaffolds of the asymmetric disulfide that could improve the bioactivities to subꢀmicromolar
range. The structure of these small organosulfur compounds had been previously reported as
potent modulator of several mammalian thiol and seleniumꢀcontaining enzymes.20 In the case of
mammalian thioredoxinꢀ1 (Trx), selective inhibition of an asymmetric disulfide could be
achieved via cysteine residues on the active sites.21 Nevertheless, to our knowledge, no
precedence of QSIs based on this class of compounds has ever been reported. A small focused
library consisting of compounds carrying disulfide bonds was then constructed to provide
insights into features important for bioactivities. As potential QSI candidates, we also tested the
most promising compounds on QSꢀactivated virulence phenotypes in P. aeruginosa, including
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
4

Page 5 of 45
Journal of Medicinal Chemistry
1
2
3
4
elastase, pyocyanin and rhamnolipid production. In addition, a mouse implantꢀbiofilm infection
5
6
experiment was performed to evaluate the bioactivities of compounds in vivo.
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
RESULTS AND DISCUSSION
Screening of Quorum Sensing Inhibitors. Four hundred and eighty compounds were
screened against reporter strain, PAO1ꢀlasBꢀgfp for the QS inhibition activities. Induction of the
lasB gene (encoding elastase) is mainly controlled by LasR and PAIꢀ1 (the Las QS system),22
hence it is a good indicator for LasR activity. In this reporter system, a decrease in GFP
fluorescence would indicate the presence of an antagonist of 3ꢀoxoꢀC12ꢀHSL, which inhibits the
expression of lasB. Growth of the reporter strain (optical density of 600 nm (OD600)) was
monitored to ensure that the compounds would not affect the growth rate (Supporting
Information Figure S1). The GFP expression was normalized by dividing the GFP values with
the growth measured at the respective time points. Four compounds were found to inhibit lasBꢀ
gfp expression in a doseꢀdependent manner with micromolar IC50 values without affecting
growth (Figure 1B).
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 45
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 1. (A) Natural AHL signal molecules used by P. aeruginosa. (B) Structure of hit compounds 1 – 4, natureꢀ
derived ajoene 5 and iberin 6 with antagonistic activity lowering the expression of PAO1ꢀlasBꢀgfp.
Compounds 1 ꢀ 4 were further tested with the rhlAꢀgfp and pqsAꢀgfp bioreporters in PAO1.
Both rhlA and pqsA are the first genes of the rhl operon and pqs operon that code for the
production of rhamnolipid and PQS molecule.23 To ensure that the compounds indeed target
corresponding QS genes and not the GFP, we also tested the compounds with PAO1ꢀgfp as
control and did not observe any reduction in the fluorescence output signals (Supporting
Information Figure S2). Promising results were obtained for compound 2 with all three reporter
strains (Figure 2), which indicated the importance of an allyl disulfide bond in the structure.
ACS Paragon Plus Environment
6

Page 7 of 45
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
Figure 2. Doseꢀdependent inhibition curves of compound 2 incubated with the following QS monitors PAO1ꢀlasBꢀ
gfp, rhlAꢀgfp, and pqsAꢀgfp. The experiments were done in triplicate manner.
IC50 values were calculated from the doseꢀresponse curves when inhibition started to occur,
usually in the range from 4ꢀ8 hours. The IC50 values represent half of the concentration required
to inhibit the biosynthesis rate of GFP due to AHL induction.24 Promising results were obtained
from our first screening with IC50 values in the low micromolar range (< 40
potent inhibitory activity was seen for compound 2 with an IC50 value at 0.98
bioreporter strain (Figure 3).
ꢀ
M). The most
ꢀM for lasBꢀgfp
Figure 3. IC50 values of compound 2 (A) 0.98 ꢁM for PAO1ꢀlasBꢀgfp, (B) 2.03 ꢁM for PAO1ꢀrhlAꢀgfp, (C) 0.33
ꢁM for PAO1ꢀpqsAꢀgfp. IC50 calculation was based on three biological replicates and performed using Graphpad
Prism 6 software.
Structural-Activity Relationship (SAR) Study. As shown in Figure 1B, two of the six
compounds (1 and 2) with good activity in the QS reporter screenings have disulfide (SꢀS) bonds
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 45
1
2
3
4
5
6
7
8
9
with different heterocyclic functional groups. Disulfide compounds are known to exhibit a wide
range of bioactivities, such as antitumor, enzymeꢀinhibitory, antibiotic and antifungal
properties.25 Ajoene, one of the active ingredients extracted from garlic also contains a disulfide
moiety. Previous report by Jakobsen et al. indicated that ajoene can inhibit a subset of QSꢀ
controlled virulence genes in P. aeruginosa and successfully clear infection in a pulmonary
mouse model.26 Based on our knowledge, no synthetic compounds carrying a disulfide backbone
have been identified as QSIs, aside from the natural derivative ajoene. With facile synthetic
transformations, a series of compounds with disulfide backbone were synthesized for a SAR
study.
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
Compounds 1 and 2 share the same allylic disulfide bond as a structural feature with ajoene,
thus it was hypothesized that the bioactivities of these compounds are related to the allylꢀ
disulfide functional group. To test this hypothesis, initial chemical transformation started with
the replacement of allyl group with different aliphatic and aromatic groups while retaining the
benzothiazole disulfide moiety. There are various approaches available for the synthesis of
disulfide compounds, and DDQ oxidation was selected for this study (Scheme 1). The reaction
proceeded smoothly and relatively fast to give the benzothiazoleꢀfunctionalized disulfide
compounds 7 ꢀ 19 in moderate yields. During the synthesis, 1.5 eq. of DDQ was necessary to
ensure full conversion of the starting materials. In most cases, homoꢀcoupling of 2ꢀ
mercaptobenzothiazole was observed as the major sideꢀreaction.
For the modification of the benzothiazole aromatic sideꢀchain, the effects of heteroatoms other
than S (i.e. O and N) in the benzothiazole ring or simple benzyl group were investigated. In
addition, substituents with diverse electronic properties were examined to evaluate the effect on
bioactivity. The synthesis started with the oxidation of diallyl disulfide with slow addition of mꢀ
ACS Paragon Plus Environment
8

Page 9 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CPBA (metaꢀchloroperbenzoic acid). The resulting intermediate allicin was directly treated with
heterocyclic thiols to give compounds 2, 20 ꢀ 26 (Scheme 1). Lastly, several analogs of
monosulfide were synthesized to study the importance of disulfide functional group. The
monosulfides 27 ꢀ 30 were readily synthesized through alkylation of 2ꢀmercaptobenzothiazole
with various halides (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 1. Preparation of unsymmetrical disulfidesa
aReagents and conditions: (i) DDQ, CH2Cl2, 0 °C; (ii) mꢀCPBA, CH2Cl2, 0 °C; (iii) RSH, MeOH
*Compounds isolated as minor products.
Scheme 2. Synthesis of 2-mercaptobenzothiazole derivativesa
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 45
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
aReagents and conditions: (i) RBr, K2CO3 or NEt3, H2O; (ii) CuI, 1,10ꢀphenanthroline, 1ꢀchloroꢀ4ꢀiodobenzene,
K2CO3, DMF, 120 oC.
QSI activities of the novel synthetic S-S disulfides bond compounds. The synthesized
compounds were tested for their QS inhibitory activity against the P. aeruginosa PAO1ꢀlasBꢀgfp
strain and compared with the natural product QSIs iberin and ajoene. Twenty of the synthesized
compounds inhibited the expression of lasB gene in P. aeruginosa without affecting the growth
of bacteria (Table 1).
From the preliminary screening with the P. aeruginosa PAO1ꢀlasBꢀgfp strain, it could be
deduced that the bioactivity was not strictly contingent with the presence of allyl group. For
example, the saturated analog 10 displayed excellent activity towards inhibiting the expression of
lasB gene. Interestingly, the pꢀchlorophenyl derivative 7 improves the bioactivity as compared to
phenyl or pꢀmethoxyphenyl analogs (8, 9). However, the cyclohexyl 11 and benzothiazole
moieties in compounds 16-19 were not efficient due to the hydrophobicity and bulky groups.
Based on the structure of compound 7, we further explored the presence of different functional
groups on the benzothiazole core while retaining the pꢀchlorophenyl moiety. This heterocycle is
present in many compounds that display a range of pharmacologically interesting properties,
such as anticancer, antifungal, antiꢀinflammatory, analgesic and antipyretic activities.27
Compounds 12, 14 and 15 displayed excellent bioactivities with IC50 values in the subꢀ
micromolar range (0.56 – 0.72 µM). For those compounds without the benzothiazole functional
group such as 31 and 32, lasB expression was not observed when tested at 10 µM. Based on the
ACS Paragon Plus Environment
10

Page 11 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
experimental results, it could be deduced that the presence of a benzothiazole (or analogous
benzoxazole) system was essential for antagonist activity towards expression of the lasB gene in
P. aeruginosa.
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
To confirm that the absence of disulfide bond will render the QSI ineffective, several
analogues without the disulfide bonds (27 ꢀ 30) were synthesized and tested against PAO1ꢀlasBꢀ
gfp. Starting material bearing the 2ꢀmercaptobenzothiazole moiety was also included as part of
this miniꢀSAR study. Although there have been numerous reports on the bioactivities of
benzothiazole,28 we did not observe any significant reduction of lasB with the P. aeruginosa
PAO1ꢀlasBꢀgfp. The results confirm our hypothesis that the presence of disulfide bonds is
important for bioactivities.
In summary, our initial studies indicate that both disulfide and benzothiazole moieties are
highly important for QSI activity, as determined by the reduction in lasB expression. Most of the
compounds synthesized showed encouraging results with QSI activity higher than that of ajoene
(Table 1). The IC50 values calculated for ajoene was 15 µM and 50 µM for lasBꢀgfp and rhlAꢀ
gfp reporter strains.26 By simply changing the substituents to pꢀchlorophenyl moiety, the
synthetic ajoeneꢀderived analogs displayed excellent QSI activity without affecting the growth of
bacteria.
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 45
1
2
3
4
Table 1. Summary of QSI activity against PAO1-lasB-gfpa
5
6
7
8
IC50 (uM)
IC50 (uM)
Compound
1*
Inhibitor Structure
Compound
16
Inhibitor Structure
S
N
N
S
S
14.48 ±
3.63
2.28 ±
0.11
H2N
S
N
S
S
S
9
N
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
O
S
N
S
S
O
S
S
N
45.38 ±
7.71
2*
0.98 ±
0.14
17
S
S
N
N
Ph
S
N
N
S
N
N
S
S
S
N
16.9 ±
0.99
31*
32*
ꢀ
ꢀ
18
19
S
S
S
O
CF3
S
N
F3C
S
N
12.13 ±
1.44
O
S
S
O
S
S
O
3.45 ±
1.40
5**
6**
15 ±
2.64
20
21
S
S
S
C
N
MeO
Cl
O
S
S
O
4.08 ±
1.25
31 ±
7.63
N
S
S
N
S
S
Cl
S
S
20.92 ±
4.29
7
8
0.56 ±
0.11
22
23
S
N
O
S
S
S
S
3.58 ±
0.62
1.34 ±
0.36
S
S
S
N
N
Cl
S
OMe
S
S
N
0.78 ±
0.07
9
1.66 ±
0.24
24
25
S
S
N
S
S
S
S
1.12 ±
0.24
10
2.20 ±
0.46
S
S
N
N
F3C
S
S
S
S
1.35 ±
0.22
11
12
6.00 ±
1.27
26
27
S
S
N
N
O
S
Cl
S
N
S
N
ꢀ
ꢀ
0.72 ±
0.12
S
S
S
Cl
Cl
S
N
S
N
13
3.25 ±
0.61
28
S
S
S
S
Cl
S
ꢀ
ꢀ
14
15
0.68 ±
0.12
29
30
S
S
N
N
S
F3C
S
Cl
S
0.58 ±
0.08
S
Cl
S
N
N
aIC50 values calculations were based on three biological replicates and performed using Graphpad Prism 6
software. *Compounds from inꢀhouse chemical library. **QSIs isolated from natural sources (5: ajoene from garlic;
6: iberin from horseradish)26, 29
ACS Paragon Plus Environment
12

Page 13 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Effect of QSIs on P. aeruginosa virulence. Based on the experimental results, compounds 2
and 7 displayed excellent inhibition activity on genes that are expressed in response to QS signal
molecules (lasB, rhlA, and pqsA). In this study, the three virulence factors (elastase, pyocyanin
and rhamnolipid) were quantified to assess the effects of these lead compounds on virulence.
Elastase or LasB protease is the major virulence factor of P. aeruginosa and capable to cause
extensive tissue damage during infection in the human host.30 Rhamnolipids have the properties
of biosurfactants and are known as key component in the protection against innate immunity.31
Rhamnolipid is believed to be prerequisite for development of chronic infections in CF patients,
whereby production of rhamnolipids has led to killing of polymorphonuclear (PMN) leukocytes
of host organism.31a This virulence phenotype is directly controlled by RhlR, which governs
expression of the essential enzymes responsible for rhamnolipid biosynthesis (rhlAB).32
Pyocyanin is a blue redoxꢀactive secondary metabolite produced by P. aeruginosa and one of the
key virulence factors that plays a major role in cystic fibrosis (CF) patients.33 It has been shown
that a high level of pyocyanin (27 µg/mL) is present in the sputum of CF patients infected with
P. aeruginosa.33ꢀ34 Recent studies have shown the roles of pycoyanin in P. aeruginosaꢀmediated
pathogenesis.35 Pyocyanin was shown to be critical for lung infection in mice by mediating tissue
damage and necrosis during infection.34
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
To further evaluate the QSI activities, the abilities of the lead compounds on elastase,
rhamnolipid and pyocyanin were tested. The elastase assay was done using a standard enzymatic
assay to measure the elastase activities of P. aeruginosa cultures. The ꢁlasIꢁrhlI double mutant
of PAO1 was used as negative control, where rhamnolipid production and elastolysis were
almost absent.22 Elastase production was found to be suppressed by compounds 2 and 7 within 3ꢀ
hours cultivation period, which was comparable to the levels of PAO1 ꢁlasIꢁrhlI (Figure 5A).
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 45
1
2
3
4
5
6
7
8
9
There was also apparent reduction in pyocyanin level in the treated samples (Figure 5B).
Rhamnolipid production was also reduced to almost 50% as compared to the wild type PAO1. In
QS mutant strain, no significant rhamnolipid production was detected (Figure 5C). These results
showed that compounds suppressing QS genes indeed lowered the production of virulence
factors.
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 5. Effects of compounds 2 and 7 on virulence. (A) Elastase activity, (B) pyocyanin production, and (C)
rhamnolipid production. Compounds were tested at final concentration of 10 µM. PAO1 ꢁlasIꢁrhlI was used as
negative control. Error bars are means ± SDs. * = p < 0.05, ** = p < 0.01, *** = p < 0.001, Student’s t test.
Effect of serum albumin on QSI activity of the novel compounds. Serum albumin is one of
the most abundant proteins in blood plasma and often interferes and binds to drug molecules. To
investigate the QSI effects of compounds in the presence of serum albumin, compounds were
incubated with bovine serum albumin (BSA) up to 50 mg/mL. For this assay, P. aeruginosa with
the lasBꢀgfp reporter strain was used as bioreporter. No apparent reduction in QSI activity of
compounds 2 and 7 was observed at 10
ꢀM concentration with BSA concentration up to 50
mg/mL (Supporting Information Figure S3).
Cytotoxicity of the novel compounds. Murine macrophage RAW264.7 cell line was used to
evaluate toxicity. As compared to DMSO, compounds 2 and 7 showed cytotoxic effects only at
higher concentrations of 100 µM. No cytotoxic effect was observed at concentration of 10 µM
ACS Paragon Plus Environment
14

Page 15 of 45
Journal of Medicinal Chemistry
1
2
3
4
and lower (Figure 6). This indicates that compounds 2 and 7 were not toxic to eukaryotic cells at
5
6
QSIꢀrelevant concentrations and could be used for subsequent mouse experiments.
7
8
9
25000
20000
15000
10000
5000
0
100 ꢁM
10 ꢁM
1 ꢁM
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 ꢁM
0.01 ꢁM
Figure 6. Cell toxicity assay. Macrophages were treated with DMSO control and compounds 2 and 7 at various
concentrations (100 µM, 10 µM, 1 µM, 0.1 µM, and 0.001 µM). Data represent the average of three measurements.
Error bars are means ± SDs
.
Mouse implant infection model. Biofilms formed on the surfaces of infected foreign bodies
like caterers, implants, and heart valves further increase the susceptibility to infections and
resistance towards conventional antibiotics treatment.36 To assess the QS effects of the identified
compounds in vivo, a murine model for an implantꢀassociated infection was conducted.37 This
assay would allow us to study the efficacy of compounds used in vivo as it mimics the foreign
bodyꢀrelated infections.
The QS activity of compounds appears to weaker the biofilm formation on implants and hence
facilitate host immune system in clearing the biofilm infections. There was apparent reduction in
bacteria counts for treated groups, which is consistent with the physical appearance and wellꢀ
being of the mice. Treatment of mice with compounds 2 and 7 after 2 h of implantation was able
to significantly reduce the development of bacterial biofilms on the implants as well as to
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 45
1
2
3
4
5
6
prevent bacterial dissemination to the spleen (Figure 7). We are further optimizing the dosing
strategy for applying these QSIs in treatment of infections in combination with antibiotics.
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 7. Foreign body mouse model of infection. Each mouse was challenged with a silicone implant coated with
P. aeruginosa that would develop into biofilms. 2 h.p.i, the two groups of mice were treated with compounds 2 and
7 at 10 µM. Mice were sacrificed on 24 h.p.i and the bacterial loads on (A) implants and (B) spleen were quantified.
The statistical difference in CFU/mL was tested by MannꢀWhitney U test using GraphPad Prism 6 software. ** = p
< 0.01
.
CONCLUSION
The discovery of small molecules as QSIs is an attractive approach to develop resistantꢀrobust
drugs and new antimicrobial treatments. Toward this goal, many research groups have tried to
identify and develop QSIs from natural sources as well as in silico study. In this study, we
screened an inꢀhouse chemical library with P. aeruginosa PAO1ꢀlasBꢀgfp bioreporter strain.
Four hit compounds were identified and found to have doseꢀdependent inhibition of QSꢀrelated
gene expression (lasB, rhlA and pqsA). Sulfurꢀcontaining compounds appear as interesting class
of molecules capable of inhibiting QS, since they have little structural similarity to AHLs and yet
possess excellent QSI properties. We performed quantitative structureꢀactivity relationship
(QSAR) studies to extend the library and assess their effects of QS modulation, with the main
ACS Paragon Plus Environment
16

Page 17 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
focus on virulence factors. The results presented here, from highꢀthroughput screening (HTS) to
SAR studies, could provide important scaffolds for future chemical development, in both design
and discovery of novel QSIs.
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
We have shown that 10 µM compounds 2 and 7 efficiently inhibit lasꢀrhlꢀpqs system in a
doseꢀdependent manner. Among the synthesized compounds, compound 7 has the lowest IC50
values at 0.56 µM for PAO1ꢀlasBꢀgfp bioreporter. IC50 values on rhlAꢀgfp was 3.49 µM and 5.63
µM for pqsAꢀgfp bioreporter strain. The difference in IC50 values of compound 7 against the
three QS systems suggests that these compounds might not target a specific QS pathway. Instead,
our compounds might interact with multiple targets involved in the QS process. We also showed
that the compounds could reduce the production of important QSꢀcontrolled virulence factors in
wildꢀtype P. aeruginosa. Elastase, pycoyanin and rhamnolipid levels were reduced upon addition
of compounds 2 and 7 at 10 µM. Several potent synthetic QSIs have been identified to date,
however, only few of these inhibit QS regulated genes in the subꢀmicromolar range.18, 24, 38
Lastly, we evaluated the effectiveness of our compounds in vivo. The compounds were not
eliminated by BSA binding up to 50 mg/mL and found to be nonꢀtoxic in murine macrophage
cell lines at 10 µM. Treatment of the P. aeruginosa foreignꢀbody implants with compounds 2
and 7 showed significant bacterial clearance effect, compared to the nonꢀtreated control group.
Our data showed encouraging results for our compounds to be developed as potential QSI and
antivirulence drugs for treatment of P. aeruginosa infections in future.
It is yet unclear on the underlying mechanism and mode of inhibition of our identified
compounds. Previously, microarray and transcriptomic analysis of ajoene indicated that the
inhibition occurs at the posttranscriptional level, where ajoene targets small regulatory RNAs in
the lower part of the QS hierarchy.26 As the compounds have structural similarity with ajoene, it
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 45
1
2
3
4
5
6
7
8
9
could be speculated that our compounds have similar working mechanism to ajoene. Future work
in our laboratory is aiming to provide molecular mechanisms of these compounds for the QS
inhibition, and exploring other chemical scaffolds as potential QSꢀinterfering compounds in the
treatment of bacterial infections.
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 SECTION
1. Biological Assays. Bacterial strains and media. Bacterial strains and plasmids used in this
study are described in Table 2. Bacteria were cultured in LuriaꢀBertani (LB) broth (1% tryptone,
0.5% yeast extract, 0.5% NaCl) and ABTGC (Bꢀmedium (0.1% MgCl2, 0.1% CaCl2, 0.1%
FeCl3) supplemented with 10% A10, 0.2% glucose and 0.2% casamino acids). Media were
solidified with 1.5% Bacto Agar (Difco) and 300 ꢂg/mL Carbenicillin was used when
appropriate. Kings medium (MilliQ water supplemented with proteose peptone (20 g/L),
potassium sulphate (10 g/L), magnesium chloride, anhydrous (1.640 g/L) and glycerol (10%
v/v)) were used for pyocyanine quantification assay.
Table 2. Strains and plasmid used in this study
Strains or plasmids
Strains
Relevant genotype and/or characteristicsa
PAO1
ATCC Pseudomonas aeruginosa.39
PAO1ꢀgfp
GFPꢀtagged wild type Pseudomonas aeruginosa.40
PAO1 lasBꢀgfp
PAO1 containing lasBꢀgfp(ASV) translational
reporter fusion.39
PAO1 ꢁlasIꢁrhlI
Plasmid
Gma; PAO1 lasI and rhlI mutant.4b
prhlAꢀgfp
Gma/Carba; pUCPNotIꢀbased plasmid carrying
RlhRꢀregulated
rhlAꢀgfp(ASV)
translational
ACS Paragon Plus Environment
18

Page 19 of 45
Journal of Medicinal Chemistry
1
2
fusion.41
3
4
5
6
7
ppqsAꢀgfp
Gma/Carba pUCP22NotIꢀbased plasmid carrying
;
pqsAꢀgfp(ASV) transcriptional fusion.40
8
9
aDescription of the strains’ antibiotic resistance. Gm, gentamicin; Carb, carbenicillin resistance.
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
1.1. Screening of Quorum Sensing Inhibitors. All screening compounds were acquired from
the Nanyang Chemical Library, Chemistry and Biological Chemistry Department, School of
Physical and Mathematical Science at Nanyang Technological University in Singapore. The
1
13
purity of all hit compounds was assessed via H and C NMR spectroscopy. Compounds were
then weighed and dissolved in DMSO to make 10 mM stock solutions, before transferring to 96ꢀ
well plates for subsequent screening steps.
1.2. P. aeruginosa QS inhibition assays.24 The first batch of screening involved 480
compounds prepared in 96ꢀwell microtiter plate (Nunc, Denmark) at concentration of 10 mM in
100% DMSO. Test compounds were then mixed with ABTGC medium and serial diluted to give
final concentration of 20 ꢀM in the first well. An overnight culture of PAO1ꢀlasBꢀgfp strain
(grown in LB medium at 37 °C, 200 rpm) was diluted in ABTGC medium to a final optical
density at 600 nm (OD600) of 0.02 (2.5 x 108 CFU/mL). An equal amount of the bacterial
suspension was added to the wells to reach final inhibitor concentration of 10
ꢀM. DMSO
control (0.1% final concentration) and blank control were used. The microtiter plate was
incubated at 37 °C in Tecan Infinate 200 Pro plate reader (Tecan Group Ltd, Männedorf,
Switzerland) to measure the cell density (OD600) and GFP fluorescence (excitation at 485 nm,
emission at 535 nm) with 15ꢀmin intervals for at least 12 hours. The inhibition assay for all test
compounds and controls were done in triplicate manner.
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 45
1
2
3
4
5
6
7
8
9
For P. aeruginosa Rhl and Pqs inhibition assay, similar method was performed as the LasB
inhibition assay. The bacterial strains and plasmids used for respective test are shown in Table 2.
1.3. Elastase assay. Overnight culture of P. aeruginosa wild type and ꢁlasIꢁrhlI mutant strain
were diluted in 5 mL of ABTGC medium to a final optical density at 600 nm (OD600) of 0.01.
Compounds were added to give final concentration of 10 ꢁM. The cultures were grown for 24
hours at 37 °C, shaking condition (200 rpm). 0.8 mL culture supernatants were sampled and
centrifuged at 10,000 rpm for 10 mins. Elastase activity was measured by using
EnzChekElastase assay kit (Invitrogen, USA), which consisted of BODIPY fluorophore (FL)ꢀ
labeled DQ elastin conjugate and yields fluorescent fragments upon cleavage by elastase
enzyme.24 Fluorescence signal was recorded every 10 mins with Tecan Infinate 200 Pro plate
reader (Tecan Group Ltd, Männedorf, Switzerland; excitation 490 nm, emission 520 nm) for
three hours.
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
1.4. Pyocyanin Quantification Assay. Overnight culture of P. aeruginosa wild type and
ꢁlasIꢁrhlI mutant strain was standardized to OD600 of 0.1 and diluted 100 times into 25 mL of
fresh Kings Medium in 250 mL flask. Compounds were added to give final concentration of 10
ꢂM. The cultures were grown for 48 hours at 37 °C, shaking condition (200 rpm). The cultures
were monitored every 24 hours to observe any colour changes from light yellow to greenish
yellow solution in untreated flask. Mutant strain ꢁlasIꢁrhlI was used as negative control. The
final cell density was measured at 600 nm (OD600) using Tecan Infinate 200 Pro plate reader
(Tecan Group Ltd, Männedorf, Switzerland). The cultures were then centrifuged for 10 mins at
10,000 rpm and 7.5 mL of the supernatants were transferred into new Falcon tubes. Pyocyanin
extraction was conducted by extracting with chloroform (3 mL) and 0.2 M HCl (1.5 mL). The
top aqueous layer of HCl containing pyocyanin was pipetted into microtiter plate and absorbance
ACS Paragon Plus Environment
20

Page 21 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
was measured at 520 nm. The data was normalized by dividing OD520 reading with the final
OD600 values.
7
8
9
1.5. Rhamnolipid Quantification Assay. Rhamnolipid production was quantified using
Orcinol according to original protocol by Koch et al. with modifications.42 Subculture of P.
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
aeruginosa was diluted 100 times from overnight culture into fresh ABTGC medium (OD600
=
0.1). Treated cultures were added with test compounds to give final concentration of 10 ꢁM. The
cultures were grown for 24 hours at 37 °C, shaking condition (200 rpm). The final cell density
was measured at 600 nm (OD600) using Tecan Infinate 200 Pro plate reader (Tecan Group Ltd,
Männedorf, Switzerland). Supernatants were collected by centrifuging at 10,000 rpm for 10 mins
and extracted with diethyl ether (twice). Organic fractions were concentrated to yield white
solids. It was then reꢀsuspended in deionized water and added with 0.19% (w/v) orcinol in 50%
H2SO4. The resulting mixture was incubated at 80 °C for 30 mins to give yellowꢀorange solution.
After cooling to room temperature, the absorbance was measured at 421 nm and results were
normalized with final OD600 values.
1.6. Effects of serum albumin on QSI activity of the compounds. Bovine serum albumin
was dissolved in ABTGC medium to a concentration of 100 mg/mL. In a 96ꢀwell microtiter dish
(Nunc, Denmark), 200 µL of the serum albumin was added to the first row and 2ꢀfold serial
dilution was made to the rest of the rows. Compounds were then added into each well to give
final concentration of 20 µM. An overnight culture of PAO1ꢀlasBꢀgfp strain (grown in LB
medium at 37 °C, shaking condition) was diluted in ABTGC medium to a final optical density at
600 nm (OD600) of 0.02. An equal amount of the bacterial suspension was added to the wells to
reach final inhibitor concentration of 10 ꢀM. After incubation for 18 hours, GFP/OD readings
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 45
1
2
3
4
5
6
were monitored with Tecan Infinate 200 Pro plate reader (Tecan Group Ltd, Männedorf,
Switzerland) at 37 °C. Similar protocol was done as described in previous section.
7
8
9
1.7. Cytotoxicity assay. The murine macrophage RAW264.7 cell line was cultured in
Dulbecco’s modified Eagle’s medium (DMEM) (Life Technologies), supplemented with 10%
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
5
fetal bovine serum (Gibco) at 37 °C and 5% CO2. For experiment, 1 × 10 RAW264.7
macrophages were first grown in each well of 96ꢀwell culture plates (Nunc, US) for 16 hrs. The
cells were then washed with phosphateꢀbuffered saline (PBS) and treated with varying
concentrations of compound (0, 0.01, 0.1, 1, 10, 100 µM) in 200 µL DMEM. The cells were
incubated with the compound at 37 °C and 5% CO2 for 16 hrs. Alamar Blue Cell Viability
Reagent (Life Technologies) was added to the final concentration of 25 µM to each well of cells.
Cells were incubated at 37 °C and 5% CO2 for 3 hrs. If the cells are alive, there will be a color
change of Alamar Blue from blue to red, whereas dead cells are unable to convert the dye to red.
OD595 of each well was measured with Tecan Infinate 200 Pro plate reader (Tecan Group Ltd,
Männedorf, Switzerland). Experiments were performed in triplicates, and the results are shown
as the mean±s.d.
1.8. Implant infection model. Female BALB/c mice were purchased from InVivos Singapore
at 8 weeks of age. Mice are maintained on water and standard mouse chow ad libitum for 7 days
before the experiment. All animal experiments were conducted in compliance with the guide for
the Care and Use of laboratory animals (National Research Council) under Nanyang
Technological University Institutional Animal Care and Use Committee (IACUC) protocol
number ARF SBS/NIEꢀA0192, which was approved by the Nanyang Technological Universityꢀ
IACUC. The experiment was conducted as previously reported.37, 43 Silicone tubes (inner
diameter 4.0 mm, outer diameter 6.0 mm, wall thickness 1.0 mm) were cut to a length of 4 mm.
ACS Paragon Plus Environment
22

Page 23 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The implants were placed in 20 mL 0.9% NaCl in the Erlenmeyer flasks with overnight culture
of P. aeruginosa (OD600 of 0.1) and incubated for 18 h at 37 °C, shaking condition (110 rpm).
Mice were anesthetized by subcutaneous injections in the groin area with 100 µL cocktail of
Ketamine HCl and Xylazine HCl (2 mL Ketamine HCl (100 mg/mL), 0.5 mL Xylazine HCl (20
mg/mL) and 7.5 mL saline). The implant was inserted into the right side of the peritoneal cavity
near the groin area. The drugs were prepared by dissolving compounds in 96% EtOH to give
final concentration of 10 mM before further diluted 1000 times in a 20% vehicle solution [(2ꢀ
hydroxypropyl)ꢀβꢀcyclodextrin dissolved in 0.9% NaCl] to a final concentration of 10 µM. 200
µL of this solution was then injected at the site of implantation after 2 h of infection. In the
control group, only the vehicle solution was given. One day post infection, the mice were
sacrificed to remove the implants and also the spleens to investigate the spread of systemic
infection. CFU/mL determination of implants and spleens was conducted as follows: implants
were placed in Eppendorf tube containing 1 mL 0.9% NaCl and sonicated in ice water bath for
10 mins (5 mins of degassing followed by 5 mins of sonication) to disrupt the biofilm. The
spleens were removed from the mouse peritoneum and homogenized with homogenizer (Bioꢀ
Gen Pro200, Pro Scientific) to disrupt the tissues and release any bacterial cells. Next, the
samples were serially diluted, plated on LB plates and incubated at 37 °C overnight. CFU/mL
was calculated by multiplying the average number of colonies by the dilution factor and divided
by the volume in 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
1
2. Chemical Synthesis. General Information. H NMR spectra were recorded using a Bruker
AVꢀ300 (300 MHz), BrukerAvance III 400 (400 MHz) and Bruker AMX500 (500 MHz)
spectrometer. 13C NMR spectra were recorded using Bruker AVꢀ300 (75.6 MHz) and
BrukerAvance III 400 (100 MHz) spectrometer. Chemical shifts (ppm) were recorded with
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 45
1
2
3
4
5
6
7
8
9
residual solvent peak as internal reference. Mass spectrometric data were recorded on a
ThermoFinnigan PolarisQ MS / ThermoFinnigan LCQ Fleet MS. The values were reported in the
unit of mass to charge ratio (m/z). Purification of products via column chromatography was done
on Merck 60 (0.040 – 0.063 mm mesh silica gel). Thin layer chromatography (TLC) was
performed using Merck 60 F254 silica gel plates. Visualization was carried out with UV lamp
and KMnO4 stain. Melting points (mp) were recorded on Buchi Bꢀ54 melting point apparatus
and uncorrected. Purity of all final compounds was 95% and higher. Purity of compounds was
tested with elemental analysis and HPLC. Elemental analysis was done using Vario MICRO
cube elemental analyzer, Elementar Analysensysteme GmbH, Hanau Germany. If necessary,
purity of compounds was determined by high performance liquid chromatography (HPLC) using
Shimadzu LCꢀ20AD workstation with SPDꢀM20A photodiode array detector. The column was
Venusil MP C18(2) (Agela Technologies), 5 µm particle size (250 mm x 4.6 mm), eluent
CH3CN/H2O with 0.1% trifuluoroacetic acid, with a flow rate of 1 mL minꢀ1 and detection at 260
nm; column temperature of 25 °C.
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
2.1. Synthesis of unsymmetrical disulfide via oxidation with DDQ44: 2ꢀ
mercaptobenzothiazole (1.0 mmol) and thiol (1.5 mmol) were dissolved in 3 mL of
dichloromethane (DCM) at 0 °C. DDQ (1.0 mmol) was then slowly added in portions to the
solution and the reaction mixture was stirred for 0.5 – 1 hour on an ice bath. Progress of the
reaction was monitored with TLC and LCMS. Upon completion, solvents were removed via
rotary evaporation, and the crude compounds were purified via column chromatography (ethyl
acetate:hexane, 1:50) to give the desired disulfide reaction products.
1
2-((4-chlorophenyl)disulfanyl)benzothiazole (7). Yellow solid; 40% yield; mp 59 °C. H
NMR (400 MHz, CDCl3):
δ
7.90 (d, J = 8.2 Hz, 1H), 7.79 (d, J = 8.0 Hz, 1H), 7.61 – 7.54 (m,
ACS Paragon Plus Environment
24

Page 25 of 45
Journal of Medicinal Chemistry
1
2
2H), 7.48 – 7.42 (m, 1H), 7.39 – 7.28 (m, 3H); 13C NMR (100 MHz, CDCl3) δ 170.8, 154.7,
135.9, 134.9, 133.5, 130.6, 129.6, 129.4, 129.3, 126.5, 125.0 , 122.4, 121.2; MS (ESI, m/z) calc
for [M+H]+ = 310.84, found [M+H]+ = 310.06, [M+Na]+ = 333.10. Elemental analysis calc (%)
for C13H8ClNS3: C 50.39, H 2.60, N 4.52, found: C 50.45 H 2.92 N 4.39.
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
1
2-(phenyldisulfanyl)benzothiazole (8). Yellow oil; 36% yield. H NMR (300 MHz, CDCl3):
δ 7.89 (d, J = 8.1 Hz, 1H), 7.82 – 7.74 (d, 8.1, 1H), 7.64 (dt, J = 4.3, 2.4 Hz, 2H), 7.50 – 7.40 (m,
1H), 7.39 – 7.27 (m, 4H); 13C NMR (75.6 MHz, CDCl3): δ 174.5, 155.0, 135.9, 135.0, 129.4
(2C), 129.0 (2C), 128.5, 126.3, 124.7, 122.3, 121.2; MS (ESI, m/z) calc for [M+H]+ = 276.40,
found [M+H]+ = 276.10. HPLC analysis: retention time = 21.280 min; peak area, 99.21%;
eluent CH3CN/H2O with 0.1% TFA over 30 min with a flow rate of 1 mL minꢀ1 and detection at
260 nm; column temperature, rt.
2-((4-methoxyphenyl)disulfanyl)benzothiazole (9). White solid; 15% yield; mp 60 °C. 1H
NMR (300 MHz, CDCl3): δ 7.89 (d, J = 8.1 Hz, 1H), 7.81 (d, J = 7.3 Hz, 1H), 7.67 – 7.59 (m,
2H), 7.48 – 7.40 (m, 1H), 7.39 – 7.31 (m, 1H), 6.90 – 6.83 (m, 2H), 3.79 (s, 3H); 13C NMR (100
MHz, CDCl3): δ 175.2, 161.0, 155.1, 136.1, 133.4, 126.3, 124.6, 122.5, 121.1, 115.0 (2C), 55.4.
HPLC analysis: retention time = 21.973 min; peak area, 99.62%; eluent CH3CN/H2O with 0.1%
TFA over 30 min with a flow rate of 1 mL minꢀ1 and detection at 260 nm; column temperature,
rt.
1
2-(propyldisulfanyl)benzothiazole (10). Light yellow oil; 54% yield. H NMR (300 MHz,
CDCl3): δ 7.92 – 7.73 (m, 2H), 7.38 (dddd, J = 8.5, 7.9, 7.3, 1.3 Hz, 2H), 2.99 – 2.87 (m, 2H),
1.87 – 1.70 (m, 2H), 1.03 (t, J = 7.3 Hz, 3H); 13C NMR (100 MHz, CDCl3): δ 173.2, 155.2,
135.9, 126.2, 124.5, 122.1, 121.1, 41.5, 22.4, 13.1; MS (ESI, m/z) calc for [M+H]+ = 242.38,
found [M+H]+ = 242.09. HPLC analysis: retention time = 20.917 min; peak area, 99.17%; eluent
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 45
1
2
CH3CN/H2O with 0.1% TFA over 30 min with a flow rate of 1 mL minꢀ1 and detection at 260
nm; column temperature, rt.
3
4
5
6
7
8
9
1
2-(cyclohexyldisulfanyl)benzothiazole (11). Light yellow oil; 76% yield. H NMR (300
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
MHz, CDCl3): δ 7.82 (ddd, J = 21.6, 8.1, 0.6 Hz, 2H), 7.52 – 7.19 (m, 1H), 3.03 (tt, J = 10.9, 3.7
Hz, 1H), 2.19 – 2.04 (m, 2H), 1.88 – 1.72 (m, 2H), 1.61 (dd, J = 10.0, 4.4 Hz, 1H), 1.55 – 1.39
(m, 2H), 1.28 (qdd, J = 11.4, 8.5, 2.7 Hz, 3H); 13C NMR (100 MHz, CDCl3): δ 172.5, 155.2,
135.8, 126.2, 124.5, 122.1, 121.7, 50.8, 32.6 (2C), 26.0 (2C), 25.4. HPLC analysis: retention
time = 24.277 min; peak area, 97.01%; eluent CH3CN/H2O with 0.1% TFA over 30 min with a
flow rate of 1 mL minꢀ1 and detection at 260 nm; column temperature, rt.
2-((4-chlorophenyl)disulfanyl)-6-ethoxybenzo[d]thiazole (12). White solid; 15% yield; mp
1
96 °C. H NMR (300 MHz, CDCl3) δ 7.77 (d, J = 9.0 Hz, 1H), 7.60 – 7.50 (m, 2H), 7.34 – 7.28
(m, 2H), 7.23 (d, J = 2.5 Hz, 1H), 7.02 (dd, J = 9.0, 2.5 Hz, 1H), 4.06 (q, J = 7.0 Hz, 2H), 1.43 (t,
13
J = 7.0 Hz, 3H); C NMR (75 MHz, CDCl3): δ 166.5, 156.9, 149.0, 137.4, 134.7, 133.8, 130.5
(2C), 129.5 (2C), 123.0, 115.9, 104.8, 64.2, 14.8; MS (ESI, m/z) calc for [M+H]+ = 354.90,
found [M+H]+ = 354.08. Elemental analysis calc (%) for C15H12ClNOS3: C 50.91 H 3.42 N 3.96.
Found C 50.70 H 3.52 N 3.88.
6-chloro-2-((4-chlorophenyl)disulfanyl)benzo[d]thiazole (13). White solid; 10% yield; mp
1
155 °C. H NMR (300 MHz, CDCl3) δ 7.78 (t, J = 5.8 Hz, 2H), 7.57 (d, J = 8.5 Hz, 2H), 7.40
(dd, J = 8.8, 1.8 Hz, 1H), 7.33 (d, J = 8.5 Hz, 2H); 13C NMR (100 MHz, CDCl3): δ 170.8, 153.2,
134.7, 133.2, 131.5, 130.2 (2C), 129.5 (2C), 126.9, 122.8, 120.6; MS (ESI, m/z) calc for [M+H]+
= 345.29, found [M+H]+ = 345.08. HPLC analysis: retention time = 24.736 min; peak area,
95.28%; eluent CH3CN/H2O with 0.1% TFA over 30 min with a flow rate of 1 mL minꢀ1 and
detection at 260 nm; column temperature, rt.
ACS Paragon Plus Environment
26

Page 27 of 45
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2-((4-chlorophenyl)disulfanyl)-6-methylbenzo[d]thiazole (14). White solid; 25% yield; mp
100 °C. 1H NMR (300 MHz, CDCl3) δ 7.72 – 7.61 (m, 2H), 7.60 – 7.52 (m, 2H), 7.35 – 7.27
(m, 2H), 7.16 (dd, J = 8.2, 1.0 Hz, 1H), 2.47 (s, 3H); 13C NMR (75 MHz, CDCl3): δ 170.4,
155.2, 136.6, 133.7, 132.9, 130.5 (2C), 129.5 (2C), 126.5, 122.5, 120.7, 21.5; MS (ESI, m/z) calc
for [M+H]+ = 324.87, found [M+H]+ = 324.11. Elemental analysis calc (%) for C14H10ClNS3: C
51.92 H 3.11 N 4.32. Found C 51.54 H 3.37 N 4.12.
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
2-((4-chlorophenyl)disulfanyl)-6-(trifluoromethyl)benzo[d]thiazole (15). White solid; 40%
yield; mp 86 °C. 1H NMR (300 MHz, CDCl3) δ 8.08 (s, 1H), 7.96 (d, J = 8.6 Hz, 1H), 7.68 (dd,
13
J = 8.6, 1.4 Hz, 1H), 7.62 – 7.52 (m, 2H), 7.38 – 7.29 (m, 2H); C NMR (75 MHz, CDCl3): δ
174.9, 156.9, 135.9, 135.3, 133.0, 130.8 (2C), 129.7 (2C), 127.1, 125.1), 123.5 (d, J = 3.4 Hz),
122.6, 118.8 (d, J = 4.2 Hz); MS (ESI, m/z) calc for [M+H]+ = 378.84, found [M+H]+ = 378.14.
HPLC analysis: retention time = 23.499 min; peak area, 95.71%; eluent CH3CN/H2O with 0.1%
TFA over 30 min with a flow rate of 1 mL minꢀ1 and detection at 260 nm; column temperature,
rt.
1,2-bis(benzo[d]thiazol-2-yl)disulfane (16). Yellowish white solid; 26% yield; mp 182 °C. 1H
NMR (400 MHz, CDCl3): δ 7.67 (d, J = 7.9 Hz, 1H), 7.38 (t, J = 7.7 Hz, 1H), 7.28 (dd, J = 15.9,
7.9 Hz, 2H); 13C NMR (75 MHz, CDCl3): δ 171.0, 156.4, 136.1, 125.7, 123.8, 123.1, 122.1,
119.4; MS (ESI, m/z) calc for [M+H]+ = 333.47, found [M+H]+ = 333.13.
1
1,2-bis(6-ethoxybenzo[d]thiazol-2-yl)disulfane (17). White solid; 5% yield; mp 132 °C. H
NMR (300 MHz, CDCl3) δ 7.81 (d, J = 9.0 Hz, 1H), 7.21 (d, J = 2.5 Hz, 1H), 7.04 (dd, J = 9.0,
13
2.5 Hz, 1H), 4.05 (q, J = 7.0 Hz, 2H), 1.43 (t, J = 7.0 Hz, 3H); C NMR (100 MHz, CDCl3) δ
164.1, 157.2, 148.7, 137.8, 123.3, 116.2, 104.7, 64.2, 14.8; MS (ESI, m/z) calc for [M+H]+ =
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 45
1
2
421.58, found [M+H]+ = 421.14. Elemental analysis calc (%) for C18H16N2O2S4: C 51.40 H 3.83
N 6.66. Found C 51.34 H 4.19 N 6.39.
3
4
5
6
7
8
9
1
1,2-bis(6-methylbenzo[d]thiazol-2-yl)disulfane (18). White solid; 5% yield; mp 180 °C. H
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
NMR (500 MHz, CDCl3) δ 7.74 (s, 2H), 7.74ꢀ7.60 (m, 1H), 7.19 (d, J = 8.1 Hz, 1 H), 2.48 (s,
13
3H); C NMR (100 MHz, CDCl3): δ 169.5, 156.4, 136.8, 126.9, 122.7, 120.8 (2C), 21.5; MS
(ESI, m/z) calc for [M+H]+ = 361.53, found [M+H]+ = 361.16. HPLC analysis: retention time =
14.300 min; peak area, 98.01%; eluent CH3CN/H2O with 0.1% TFA over 30 min with a flow rate
of 1 mL minꢀ1 and detection at 260 nm; column temperature, rt.
1,2-bis(6-(trifluoromethyl)benzo[d]thiazol-2-yl)disulfane (19). White solid; 45% yield; mp
152 °C. 1H NMR (300 MHz, CDCl3) δ 8.05 (dd, J = 10.6, 4.7 Hz, 2H), 7.72 (dd, J = 8.6, 1.4 Hz,
1H); 13C NMR (75 MHz, CDCl3): δ 171.0, 156.4, 136.1, 127.9, 125.7, 123.8 (q, J = 3.4 Hz),
123.1, 122.1, 119.4; MS (ESI, m/z) calc for [M+H]+ = 469.47, found [M+H]+ = 469.18.
Elemental analysis (%) calc (%) for C16H6F6N2S4: C 41.02 H 1.29 N 5.98. Found C 41.05 H 1.71
N 5.84.
2.2. General procedure for the synthesis of allicin45: A solution of mCPBA (0.48 mmol) in
DCM was added to a solution of diallyl disulfide (0.45 mmol) in 3 mL DCM at 0 °C. The
reaction mixture was stirred in ice bath for 4 hours and monitored with TLC and LCMS. Upon
completion, reaction mixture was washed two times with sodium bicarbonate and two times with
water. The combined aqueous layer was then collected and washed two times with ether. The
combined organic layer was dried with sodium sulfate. Solvent was removed under reduced
pressure and crude compounds were purified via column chromatography (ethyl acetate:hexane,
1:50).
ACS Paragon Plus Environment
28

Page 29 of 45
Journal of Medicinal Chemistry
1
2
1
3
4
5
6
7
8
9
S-allyl prop-2-ene-1-sulfinothioate. Yellow oil; 15% yield; H NMR (300 MHz, CDCl3): δ
6.03 – 5.82 (m, 1H), 5.53 – 5.11 (m, 2H), 3.96 – 3.66 (m, 2H); MS (ESI, m/z) calc for [M+H]+ =
163.26, found [M+H]+ = 163.84.
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
2.3. Synthesis of allyl disulfide via oxidation of mCPBA. Allicin (1 mmol) was slowly added
into the thiol (0.5 mmol.) in MeOH at room temperature. The reaction mixture was stirred at
room temperature for 0.5 – 1 hour and monitored with TLC and LCMS. Upon completion, the
solvent was removed under reduced pressure. The crude products were purified with column
chromatography.
1
2-(allyldisulfanyl)benzothiazole (2). Yellow oil; 90% yield; H NMR (300 MHz, CDCl3): δ
7.86 (d, J = 7.9 Hz, 1H), 7.82 – 7.72 (m, 1H), 7.48 – 7.37 (m, 1H), 7.36 – 7.24 (m, 1H), 5.88
13
(ddt, J = 17.2, 9.9, 7.4 Hz, 1H), 5.20 (dd, J = 12.7, 11.8 Hz, 2H), 3.57 (d, J = 7.4 Hz, 2H); C
NMR (100 MHz, CDCl3): δ 172.7, 155.1, 135.8, 131.5, 126.3, 124.61, 122.2, 121.2, 120.5,
42.31; MS (ESI, m/z) calc for [M+H]+ = 240.37, found [M+H]+ = 240.95. Elemental analysis
calc (%) for C10H9NS3: C 50.18 H 3.79 N 5.85. Found C 50.24 H 4.06 N 5.67.
1
2-(allyldisulfanyl)-5-methoxybenzooxazole (20). Yellow sticky oil; 31% yield; H NMR
(300 MHz, CDCl3) δ 7.34 (dd, J = 16.2, 8.9 Hz, 1H), 7.14 (dd, J = 15.9, 2.5 Hz, 1H), 6.99 – 6.77
(m, 1H), 5.97 – 5.66 (m, 1H), 5.34 – 5.00 (m, 2H), 3.99 (dd, J = 7.4, 0.9 Hz, 1H), 3.84 (s, 3H),
3.62 (d, J = 7.4 Hz, 1H); 13C NMR (75.6 MHz, CDCl3): δ 165.3, 157.5, 147.1, 142.7, 131.5,
120.1, 113.1, 110.4, 102.6, 55.9, 42.1; MS (ESI, m/z) calc for [M+H]+ = 254.33, found [M+H]+
= 254.06. Elemental analysis calc (%) for C11H11NO2S2: C 52.15 H 4.38 N 5.53. Found C 52.15
H 4.63 N 5.50.
1
2-(allyldisulfanyl)-6-chlorobenzooxazole (21). Yellow oil; 40% yield; H NMR (300 MHz,
CDCl3): δ 7.87 – 7.81 (m, 1H), 7.44 (ddd, J = 8.3, 7.2, 1.4 Hz, 1H), 7.34 (ddd, J = 8.4, 7.5, 1.3
ACS Paragon Plus Environment
29