Amphiphilic Tobramycin-Lysine Conjugates Sensitize Multidrug Resistant Gram-Negative Bacteria to Rifampicin and Minocycline

Article abstract of DOI:10.1021/acs.jmedchem.6b01742

Chromosomally encoded low membrane permeability and highly efficient efflux systems are major mechanisms by which Pseudomonas aeruginosa evades antibiotic actions. Our previous reports have shown that amphiphilic tobramycin-fluoroquinolone hybrids can enhance efficacy of fluoroquinolone antibiotics against multidrug-resistant (MDR) P. aeruginosa isolates. Herein, we report on a novel class of tobramycin-lysine conjugates containing an optimized amphiphilic tobramycin-C12 tether that sensitize Gram-negative bacteria to legacy antibiotics. Combination studies indicate the ability of these conjugates to synergize rifampicin and minocycline against MDR and extensively drug resistant (XDR) P. aeruginosa isolates and enhance efficacy of both antibiotics in the Galleria mellonella larvae in vivo infection model. Mode of action studies indicate that the amphiphilic tobramycin-lysine adjuvants enhance outer membrane cell penetration and affect the proton motive force, which energizes efflux pumps. Overall, this study provides a strategy for generating effective antibiotic adjuvants that overcome resistance of rifampicin and minocycline in MDR and XDR Gram-negative bacteria including P. aeruginosa.

Full text of DOI:10.1021/acs.jmedchem.6b01742

Subscriber access provided by University of Newcastle, Australia
Article
Amphiphilic tobramycin-lysine conjugates sensitize multidrug
resistant Gram-negative bacteria to rifampicin and minocycline
Yinfeng Lyu, Xuan Yang, Sudeep Goswami, Bala Kishan Gorityala, Temilolu Idowu,
Ronald Domalaon, George G Zhanel, Anshan Shan, and Frank M. Schweizer
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01742 • Publication Date (Web): 14 Apr 2017
Downloaded from http://pubs.acs.org on April 17, 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 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Amphiphilic tobramycinꢀlysine conjugates sensitize multidrug resistant Gramꢀ
negative bacteria to rifampicin and minocycline
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
2
2
2
2
Yinfeng Lyu , Xuan Yang , Sudeep Goswami , Bala Kishan Gorityala , Temilolu Idowu ,
2
3
*1
*2,3
Ronald Domalaon , George G. Zhanel , Anshan Shan and Frank Schweizer
1
Institute of Animal Nutrition, Northeast Agricultural University, Harbin, Heilongjiang, 150030, P.R.
China
2
Department of Chemistry, University of Manitoba, Winnipeg, MB, R3T 2N2, Canada
3
Department of Medical Microbiology/Infectious Diseases, University of Manitoba, Winnipeg, MB,
R3T 1R9, Canada
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 62
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
Abstract
Chromosomallyꢀencoded low membrane permeability and highly efficient efflux systems are major
mechanisms by which Pseudomonas aeruginosa evades antibiotic actions. Our previous reports have
shown that amphiphilic tobramycinꢀfluoroquinolone hybrids can enhance efficacy of fluoroquinolone
antibiotics against multidrugꢀresistant (MDR) P. aeruginosa isolates. Herein, we report on a novel class
of tobramycinꢀlysine conjugates containing an optimized amphiphilic tobramycinꢀC12 tether which
sensitize Gramꢀnegative bacteria to legacy antibiotics. Combination studies indicate the ability of these
conjugates to synergize rifampicin and minocycline against MDR and extensivelyꢀdrug resistant (XDR)
P. aeruginosa isolates and enhance efficacy of both antibiotics in the Galleria mellonella larvae in vivo
infection model. Mode of action studies indicate that the amphiphilic tobramycinꢀlysine adjuvants
enhance outer membrane cell penetration and effect the proton motive force which energize efflux pumps.
Overall, this study provides a strategy for generating effective antibiotic adjuvants that overcome
resistance of rifampicin and minocycline in MDR and XDR Gramꢀnegative bacteria including P.
aeruginosa.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ACS Paragon Plus Environment

Page 3 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The global health crisis caused by the resurgence of multidrug resistant bacteria strains, once believed
to have been defeated, have called for an urgent need to rethink the principle of antibacterial drug
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
discovery, and the judicious deployment of our current arsenal. An FDA incentive of accelerated
2
development and review process for breakthrough therapies encourages the need to find alternative or
better use of currently approved drugs instead of developing entirely new scaffolds. The Generating
3
Antibiotics Incentives Now (GAIN) Acts of 2012 that seek to extend the exclusivity of new antibiotics
further stimulates the need to consolidate on the valuable knowledge of existing antibacterial scaffolds as
a way out of the current attrition of drug candidates. Among the most recalcitrant bacteria, typified by the
4
ESKAPE pathogens, Pseudomonas aeruginosa, an opportunistic pathogen that commonly affects
immunocompromised patients, is particularly infamous for its highly sophisticated intrinsic and acquired
5,6
resistance machineries. Compared to other bacteria, P. aeruginosa displays low membrane permeability
which limits the penetration of most antibiotics into the cell and a highly efficient membraneꢀassociated
efflux system of broad substrate specificity that significantly reduces bioaccumulation of drugs within its
7, 8
cytosol. The reduced intracellular concentration further promotes the activation of secondary adaptive
resistance mechanisms (such as overexpression of efflux pump proteins and a variety of sensor kinases)
9
that renders it completely refractory to treatment. In addition, resistance often emerges when antibiotics
10,11
are administered as monotherapy,
thus combination therapy being the preferred choice in the
Although debatable, the argument for combination drug treatment
12,13
treatment of complicated infections.
1
4
is premised on the largeꢀscale genetic interaction networks between targets. The use of two or more
antibiotics that impact multiple targets simultaneously, or adjuvants that aid the action of legacy
antibiotics can indeed extend the antimicrobial space as well as mitigate the development of antibiotic
11
resistance. Unfortunately, rational combination regimens and correlative synergistic mechanisms have
remained largely unexplored and clinical benefits are yet to be demonstrated. Since the outer membrane
of P. aeruginosa remains a major impediment to the influx of antibiotics, we and others have been
investigating the effects of adjuvants that perturb this ‘impermeable’ layer, and consequently synergize
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the activities of other antibiotics. For instance, Evotec AG and Spero Therapeutics are currently
developing a polymyxinꢀbased antimicrobial peptide SPRꢀ741 as a potentiator to overcome outer
15
membrane impermeability of legacy antibiotics against Gramꢀnegative pathogens in clinical trials . Our
group has recently shown that tobramycinꢀfluoroquinolone hybrids interact with both the outerꢀ and
innerꢀmembranes of P. aeruginosa resulting in enhanced cell penetration and reduced efflux by
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
6, 17
dissipating the proton motive force (PMF) that drive efflux pumps in P. aeruginosa.
Mode of action
studies indicate that the function of the tobramycin moiety in these hybrids is limited to a membraneꢀ
destabilizing effect of the outer membrane that results in selfꢀpromoted uptake of the hybrid and/or the
antibiotic. In contrast, the function of the fluoroquinolone moiety in the hybrid is less clear but the
adjuvantꢀ and antibacterial properties appear to be correlated to the hydrophobic nature and membrane
1
6,17
destabilizing effect of the fluoroquinolone group.
Amphiphilic aminoglycosides (AAGs) that combine
aminoglycosides such as tobramycin with alkyl or other hydrophobic groups have also been previously
reported to show improved activity and different modes of action in killing pathogens, compared to their
18
constituent parent antibiotics.
For example, some amphiphilic tobramycin derivatives were
1
9ꢀ21
demonstrated to primarily target the bacterial membrane,
as well as possess immumodulatory
2
2
properties that closely resemble that of the natural host defense peptides. To put all these findings into
context, we hypothesized that appending a lipophilic membraneꢀactive component to a tobramycin vector
will confer an adjuvantꢀlike property that can revive the efficacies of clinical antibiotics against resistant
pathogens in a similar fashion as previously reported for tobramycinꢀfluoroquinolone hybrid antibiotics.
To preserve the amphiphilic nature of the membraneꢀactive tobramycin derivatives, it is imperative to
carefully evaluate the hydrophobic nature of the moiety to be attached. Recently, Haldar and coꢀworkers
described a series of ultrashort antibacterial lysineꢀbased peptoid mimics that contain facial segregation of
2
3
positivelyꢀcharged Lꢀlysine, hydrophobic aromatic core, and alkyl chain. These molecules were reported
to facilitate the attraction of compounds to bacterial surfaces and permeate cell membrane, and displayed
23
some promising antimicrobial properties. We reasoned that the inherent amphipathic nature of these
peptoids, combined with their antibacterial properties could be amplified by linking to a tobramycinꢀ
4
ACS Paragon Plus Environment

Page 5 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
16,17
based vector as previously reported for tobramycinꢀfluoroquinolone hybrid antibiotics.
In the current
study, we designed a series of tobramycin analogs 1 – 3 by conjoining an amphiphilic peptoid mimic 4 to
the Cꢀ5 position of tobramycin, with varied alkyl tether, to investigate the adjuvant property of the
resultant conjugates in combination with commonly used antibiotics (Figure 1). The synergistic effects of
the tobramycinꢀlysine conjugates in combination with various classes of antibiotics, against 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
(
wildꢀtype and clinical isolates) were determined using checkerboard study. The emergence of bacterial
resistance was compared between single agent and combination therapy for 25 bacterial passages, and
Galleria mellonella infection worm model was further used to assess in vivo synergistic benefits of the
optimal drug combination that protects P. aeruginosaꢀchallenged larvae. Hemolysis and cytotoxicity
assays were carried out to ascertain toxicity against mammalian cells, and prokaryotic membraneꢀ
compound interactions were studied to gain mechanistic insights on possible modes of action.
RESULTS
Synthesis. The reference compound ultrashort peptoid mimic 4 was synthesized by reductive
amination of aromatic aldehyde 5 with dodecylamine 6 generating secondary amine 7 which after
coupling to diꢀBocꢀprotected lysine produced protected lysineꢀbased peptoid 8. Deprotection of the Bocꢀ
23
protecting groups with TFA afforded ultrashort lysine peptoid 4 as previously described (Scheme 1).
The synthesis of tobramycinꢀlysine conjugates 1 – 3 was achieved by preparing amphiphilic tobramycin
derivatives (5 steps), followed by a singleꢀstep reductive amination conjoining and a final deprotection
(
Scheme 2). Preparation of the tobramycin derivatives commenced by protecting the amines on
commercially available tobramycin with Boc anhydride, followed by silylation of the hydroxyl groups
with a bulky protecting group such as TBDMSCl to give 9. This is to ensure the more hindered Cꢀ5
position of the Bocꢀ and TBDMSꢀprotected tobramycin intermediate, the desired point of alkylation, is
unprotected. This is the preferred position because Cꢀ5ꢀmodified tobramycin derivatives retained
24
antibacterial activity and superior adjuvant properties against Gramꢀnegative bacteria like P.
16, 17
aeruginosa.
The Cꢀ5 hydroxyl group of 9 was subsequently alkylated with 1,nꢀdibromoalkane (n = 4,
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 62
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
8
, 12), under phaseꢀtransfer catalytic conditions, to afford alkylated tobramycin analogs 10a–c. The
terminal bromoꢀgroup of intermediate 10a–c was then displaced by an azido nucleophile under anhydrous
conditions to give 11a–c, followed by reduction to free amine 12a–c via catalytic hydrogenation. This
free amine was successively reacted with commerciallyꢀavailable 10ꢀchloroꢀ9ꢀanthracenaldehyde via
reductive amination, to afford intermediate 13a–c with secondary amine, which was then coupled to diꢀ
Bocꢀprotected Lys to produce protected amphiphilic tobramycin amides 14a–c. The final deprotection
step involved the removal of Boc and TBDMS protecting groups using MeOH:HCl (3:2, v/v) to afford the
desired target compounds 1 – 3. The final compounds were characterized by NMR, mass spectrometry,
and reverseꢀphase HPLC, with > 95 % purity (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
Antimicrobial Activity. The antimicrobial activities of reference peptoid mimic (compound 4) and
tobramycin hybrids 1 – 3 against a panel of Gramꢀnegative and Gramꢀpositive bacteria are presented as
the minimum inhibitory concentration (MIC) in Table 1. Reference peptoid mimic 4 with a C12
23
hydrophobe displayed weaker antimicrobial activity compared to the reported C10 peptoid, with MIC ≥
8
ꢁg/mL against all the strains tested in this study. For tobramycinꢀlysine conjugates, there was a positive
correlation between antimicrobial activity and the length of the carbon chain tether. Compound 3 with
C12 tether was the most potent analog of all the hybrids, and displays moderate activity against Gramꢀ
positive bacteria (MIC of 2 – 32 ꢁg/mL) but a relatively weak activity against Gramꢀnegative bacteria
(
MIC ≥ 16 ꢁg/mL). Further, the antiꢀpseudomonal activities of all compounds were evaluated against
wildꢀtype and seven clinical isolates of P. aeruginosa, including MDR, XDR, and colistinꢀresistant strains
Table 2). A similar trend of longer carbon chain displaying better activity was observed for drugꢀresistant
(
P. aeruginosa as well, suggesting C12 as the optimal tether length of all analogues tested. In addition, we
observed comparable activity of 3 against wildꢀtype and drugꢀresistant strains (MIC ranging from 8 to 64
ꢁg/mL) indicating that compound 3 is not greatly affected by resistance. Again, compound 4 did not
show potent activity against any of the clinical isolates tested, with MIC ≥ 64 ꢁg/mL. The MIC of
tobramycin increased from 0.25 ꢁg/mL in wildꢀtype P. aeruginosa to > 8 ꢁg/mL in drugꢀresistant strains,
particularly in P262 (MIC = 512 ꢁg/mL). Similar results were also observed for other tested antibiotics,
6
ACS Paragon Plus Environment

Page 7 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with the exception of colistin. It is however interesting to note that compound 3 displayed better activity
against P262 (MIC = 32 ꢁg/mL) than tobramycin, moxifloxacin, minocycline, rifampicin,
chloramphenicol, erythromycin, and trimethoprim (Table 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
Combination Study of Hybrids with Antibiotics. To assess the adjuvant properties of the hybrids,
checkerboard studies were performed to determine the synergistic effects of the most active hybrid 3 with
14 different antibiotics (cutting across all classes) against wild type P. aeruginosa PAO1. The fractional
inhibitory concentration (FIC) index, a numerical quantification of the interactions between antibiotics,
1
6
was calculated as previously described. FIC indices of ≤ 0.5, 0.5 ꢀ 4, and > 4 indicate synergy, no
2
5
interaction, and antagonism respectively. Compound 3 showed strong synergy with most antibiotics
tested against PAO1 except ceftazidime, colistin, meropenem and the aminoglycosides gentamicin,
kanamycin A and amikacin. The strongest potentiation was seen with novobiocin (FIC index = 0.071),
minocycline and rifampicin (FIC index = 0.094 for both) as shown in Table 3. Synergism with
minocycline and rifampicin was also observed with 2, but not with 1, 4, and tobramycin (Table S1). As
shown in Figure 2, the absolute MICs (the MIC of antibiotics in the presence of adjuvants at 4 ꢁg/mL) of
minocycline or rifampicin in combination therapy with hybrids was dramatically lower than
monotherapy, in particular with compound 3 where MICs of minocycline and rifampicin were reduced
from 8 ꢁg/mL and 16 ꢁg/mL in monotherapy to 0.25 ꢁg/mL (32ꢀfold potentiation) and 0.0625 ꢁg/mL
(
256ꢀfold potentiation), respectively. Thus, combinations of 3 with minocycline and rifampicin were
selected for further synergy studies against a panel of P. aeruginosa clinical isolates (Table 4). Compound
demonstrated strong synergy with minocycline and rifampicin across the clinical isolates panel (FIC
3
indices of 0.039 to 0.281), with the exception of rifampicin against P. aeruginosa 91433 (FIC index =
0.5). Since the breakpoints for minocycline and rifampicin against P. aeruginosa are not available (as
they are not conventional drugs for treating P. aeruginosa infections), the susceptible or intermediate
resistant breakpoints of minocycline against Acinetobacter spp. and that of rifampicin against
26
Enterococcus spp. reported by CLSI, were considered as interpretive MIC standards for this study. The
susceptible breakpoints of minocycline (MIC ≤ 4 ꢁg/mL) against Acinetobacter spp. were reached for all
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 62
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
minocyclineꢀresistant, MDR, or XDR P. aeruginosa isolates at 4 ꢁg/mL of 3. For rifampicin, susceptible
(
MIC ≤ 1 ꢁg/mL) or intermediate resistant (MIC = 2 ꢁg/mL) breakpoints against Enterococcus spp. were
reached in 5/7 rifampicinꢀresistant, MDR, or XDR P. aeruginosa isolates, with the exception of strain
P262 and #91433.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To gain insights into the membrane effects and relevant synergistic mechanism of 3, colistin, a
membraneꢀactive antibiotic, was tested in combination with five antibiotics against PAO1. Our results
indicated that colistin was also able to potentiate the activity of rifampicin and novobiocin (Figure 3), but
to a lesser extent than compound 3. To investigate the relevance of efflux pumps on the observed adjuvant
properties of 3, combination studies of minocycline and rifampicin each with compound 3 were carried
out in efflux pumpꢀmutated strains, PAO200 and PAO750. PAO200 is a MexABꢀOprM deletion strain
while PAO750 is an effluxꢀsensitive strain that lacks five different clinicallyꢀrelevant RND pumps
(
MexABꢀOprM, MexCDꢀOprJ, MexEFꢀOprN, MexJK, and MexXY) and the outer membrane protein
OpmH. Some of these pumps are homologues of broad substrate specificities that expel different classes
of antimicrobial agents and confer resistance on P. aeruginosa. For instance, the tripartite protein system
MexABꢀOprM, MexCDꢀOprJ, MexEFꢀOprN and MexXYꢀOprM allow the translocation of a wide variety
of substrates such as quinolones, chloramphenicol, trimethoprim, imipenem, and tetracyclines out of the
27
cell. As shown in Table 5, synergism was observed with rifampicin in both PAO200 (FIC index =
0.129) and PAO750 (FIC index = 0.156), but not with minocycline (FIC index > 0.5). The ability to
potentiate minocycline in PAO1 but not in effluxꢀdeficient strains corroborates the hypothesis that
tobramycinꢀlysine conjugate compromises the efficiency of efflux proteins. Rifampicin, which is not a
substrate for these pumps was however potentiated, suggesting a second mode of action consistent with
membrane permeabilization.
To study the spectrum of activity of 3 in combination with minocycline and with rifampicin, we
examined in vitro potency against other clinical isolates of highly pathogenic Gramꢀnegative bacteria,
including Acinetobacter baumannii, Enterobacter cloacae, and Klebsiella pneumoniae. Table 6
summarizes these results and indicates strong synergy of 3 with minocycline in other Gramꢀnegative
8
ACS Paragon Plus Environment

Page 9 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
species except AB031 and K. pneumoniae 110193 where the effect of 3 on minocycline is at best
marginally additive. Surprisingly, combination of 3 with rifampicin displayed potent synergy in all tested
isolates, with FIC indices < 0.25. The synergistic MIC of rifampicin in combination with 4 ꢁg/mL of 3
against rifampicinꢀresistant E. cloacae 117029 was 0.063 ꢁg/mL, which is 16ꢀfold lower than rifampicinꢀ
susceptible breakpoints of ≤ 1 ꢁg/mL.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Time-kill Curve. The kinetics of P. aeruginosa PAO1 killing as a function of time, using monoꢀ and
combinationꢀtherapy of minocycline, rifampicin, and compound 3, are shown in Figures 4 and S24.
Minocycline alone was not bactericidal even at 4 × MIC after 6 h, while rifampicin as a monotherapy was
bactericidal at 4 × MIC after 2 h of drug exposure (Figure S24). Compound 3 showed bactericidal activity
at 1 × MIC after 2 h, and more rapid killing was observed at 2 × MIC and 4 × MIC for only 30 and 10
mins antimicrobial exposure respectively (Figure S24). Minocycline, rifampicin, or 3 at subꢀinhibitory
concentrations were unable to suppress bacteria growth in monotherapy, even after 6 h exposure (Figure
4). However, upon combination with subꢀMIC (1/8 to 1/2 × MIC) of 3, in vitro bactericidal activities of
minocycline and rifampicin were both enhanced, yielding synergistic killing at subꢀMIC concentration
(
1/8 × MIC) after 90 mins of incubation (Figure 4).
Resistance Study. The ability of drug combinations to suppress resistance development was
determined using wildꢀtype P. aeruginosa PAO1. This assay was validated by demonstrating that the
MIC of colistin and tobramycin increased by 1024ꢀ and 256ꢀfold respectively over 25 serial passages,
while that of minocycline increased by 16ꢀfold (Figure 5). Upon combination with compound 3, the
emergence of resistance in minocycline was suppressed by 4ꢀfold while rifampicin did not promote
resistance either as monotherapy or in combination with 3 (Figure 5).
Mode of Action Studies. Outer Membrane Permeabilization. Since the negativelyꢀcharged outer
7
, 8
membrane of P. aeruginosa serves as the first barrier that prevents the uptake of antibiotics, the ability
of the amphiphilic cationic compounds to perturb this lipid bilayer was investigated in PAO1 using
28
carboxyfluorescein diacetate succinimidyl ester (CFDASE), a cell permeable dye. We reasoned that the
amphipathic nature of the hybrids might confer membrane effects similar to those of the host defense
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 62
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
peptides on them. The increased fluorescence induced by the compounds was calculated by subtracting
the fluorescence of negative control that was treated similarly but in the absence of drug, while 1% Triton
Xꢀ100 that exhibited the highest fluorescence compared to other treatments served as the positive control
(Figure 6). At the concentration tested (32 ꢁg/mL), the outer membrane permeabilization induced by 3
was slightly lower than that of colistin. In contrast, reference compound 4 displayed the weakest
membrane permeabilizing ability with the lowest fluorescence increase.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cytoplasmic Membrane Depolarization. Compounds that perturb the bacterial outer membrane can
potentially be trapped within the periplasmic space where they could interfere with the respiratory chain
on the cytoplasmic membrane to induce death. Indeed, this has been proposed as one of the mechanisms
2
9
by which polymyxin exerts its antibacterial effects, among many others. Depolarization of cytoplasmic
membrane can lead to loss of membrane potential, an important electrochemical gradient used by the
3
0
bacteria to maintain an active and functional efflux system. To investigate the effect of 3 on bacterial
cytoplasmic membrane, diSC ꢀ5, a membrane potentialꢀdependent probe, was used to study the
3
1
9
differential in fluorescence caused by membrane depolarization. A doseꢀdependent fluorescence
increasing was observed for all the tested antimicrobials in PAO1 (Figure S25). At 32 ꢁg/mL, colistin was
observed to depolarize the cytoplasmic membrane faster than other compounds in the first 300 secs, with
an accompanying decrease in fluorescence thereafter (Figure 7). Although reference compound 4
displayed similar properties as colistin, the decline in fluorescence seems to be specific for colistin.
Compound 3 however displayed the highest fluorescence up till 1200 secs, while only weak membrane
depolarization ability was observed for tobramycin at 32 ꢁg/mL, a 128ꢀfold higher value than its MIC.
Swimming Motility Assay. Swimming motility is a flagellumꢀdependent bacterial movement that is
governed by the respiratory chain on the cytoplasmic membrane. When cytoplasmic membrane potential
or proton motive force (PMF) is disrupted, the electron transfer across respiratory chain is inhibited,
3
1
resulting in a reduction of ATP synthesis which is essential for flagellar function. Previous studies have
implicated that amphiphilic tobramycinꢀmoxifloxacin hybrids can perturb the PMF resulting in reduced or
17
inefficient efflux. We therefore studied the effect of the hybrids 3 and 4 on the swimming motility of
1
0
ACS Paragon Plus Environment

Page 11 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
PAO1 by monitoring its movement on lowꢀviscosity swim plates (0.3 % agar, w/v). Surprisingly, 3 and 4
were observed to significantly constrict the swimming bacteria diameter (at 4 ꢁg/mL) relative to the
untreated control (Figure 8). Meanwhile, these compounds repressed bacterial swimming motility in a
doseꢀdependent manner, and the observed effects were superior to colistin at subꢀinhibitory
concentrations, while tobramycin was unable to inhibit bacterial motility with similar bacterial swimming
diameters to untreated control (Figure S26).
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
Hemolytic Activity and Cytotoxicity. A potential problem usually associated with membraneꢀactive
agents is their toxicity towards eukaryotic cells. The hemolytic properties of the hybrid molecules were
first examined using freshly collected pig erythrocytes. All hybrids demonstrated lower hemolytic
activities (< 20 %) relative to 4, which is highly toxic with 87 % hemolysis at the highest measured
concentration of 512 ꢁg/mL (Figure 9A). As to the structureꢀactivity relationships (SAR) between the
hybrids, a slight increase in hemolytic activity with increase in carbon length was evident. Compound 3
was also tested against human epithelial prostate (DU145) and breast (JIMTꢀ1) cancer cell lines, with
greater than 50 % viability at 20 ꢁM (25.2 ꢁg/mL), a six times higher concentration than effective
synergistic concentration (4 ꢁg/mL) in combination therapy (Figure 9B).
In Vivo Efficacy. To gain insights into the potential clinical benefits of compound 3, an in vivo
efficacy evaluation using Galleria mellonella infection model was initiated. The maximum tolerable dose
was first determined by injecting drugs alone at high concentrations (100 or 200 mg/kg), and the survival
rates scored for 4 days. As shown in Figure S27, 100 % survival was observed after 4 days in the group
that had been injected with 200 mg/kg of 3, indicating the relative safety of the compound to the worms at
this dose. Next, the ability of the drug or drug combinations to protect larvae from MDR P. aeruginosa
P262 infection was determined at single doses of 75 mg/kg in drug monotherapy or 12.5 + 12.5, 25 + 25,
37.5 + 37.5, or 75 + 75 mg/kg in drug combinations (Figure 10). 100 % mortality was observed in the
monotherapy of minocycline, rifampicin, and 3 at 75 mg/kg after 24 h, and in combination at lower doses.
However, combinations of minocycline or rifampicin with 3 at a high dose of 75 + 75 mg/kg both resulted
in 77 % survival after 24 h, demonstrating the ability of this compound to offer protection against
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 62
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
infection at very tolerable dose. Interestingly, it appears that combination therapy of rifampicin and 3 at
low dosage appears to be superior when compared to combinations of minocycline and 3. This is rather
surprising as the in vitro studies suggest a lower MIC for minocycline when compared to rifampicin. This
discrepancy may be related to the different pharmacokinetics of minocycline and rifampicin in the larvae
or the difference between bacteriostatic minocycline and bactericidal rifampicin (Figure S24).
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
DISCUSSION
Bacterial resistance can frequently emerge in antibiotic monotherapy due to the selective pressure that
1
0, 11
naturally separates out the resistant phenotypes.
Combination of two or more antimicrobials that can
impact multiple targets simultaneously is believed to be capable of suppressing drug resistance, as well as
3
2, 33
broaden the spectrum of activity of a treatment course than single agents.
In the last two years the
FDA has approved two new combination drugs Avycaz (ceftazidime + avibactam) and Zerbaxa
(
ceftolozane + tazobactam to combat MDR Gramꢀnegative infections. Ceftazidime/avibactam contains an
rd
older 3 generation cephalosporin ceftazidime, with avibactam a synthetic nonꢀβꢀlactam, βꢀlactamase
inhibitor that inhibits the activities of Ambler class A and C βꢀlactamases and some Ambler Class D
34
enzymes. Limited data suggest that the addition of avibactam does not improve the activity of
ceftazidime versus Pseudomonas aeruginosa. Ceftolozane is a novel cephalosporin with a chemical
structure similar to that of ceftazidime, with the exception of a modified sideꢀchain at the 3ꢀposition of the
3
5
cephem nucleus, which confers potent antipseudomonal activity. The addition of tazobactam extends
the activity of ceftolozane to include most ESBL producers but not P. aeruginosa. Nevertheless, effective
drug combinations often lead to inconclusive benefits of combination therapy over monotherapy during
3
6
metaꢀanalysis. Recent reports about the potentials of amphiphilic tobramycin analogues to permeabilize
19,20
cell membrane,
and our previous studies that demonstrated the intrinsic ability of tobramycinꢀ
fluoroquinolone hybrids to potentiate the antimicrobial activity of several classes of antibiotics against
16, 17
clinical P. aeruginosa isolates,
encouraged a further optimization of this promising scaffold for use as
adjuvants. In the current study, we prepared new amphiphilic tobramycin hybrids by taking advantage of
1
2
ACS Paragon Plus Environment

Page 13 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the membraneꢀactive peptoid 4 as a modulator. The antimicrobial properties of these derivatives alone
were assessed and demonstrated to be weaker compared to the parental tobramycin molecule (Table 1).
3
7
Although tobramycin is believed to induce pleiotropic mechanisms of action, the most acceptable
38
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
hypothesis suggests tobramycin permeates the outer membrane via a selfꢀpromoted uptake mechanism
and acts by impairing bacterial protein synthesis through irreversible binding to the 30S ribosomal
39
subunit. The differing activity between tobramycin and the newly synthesized hybrid molecules
suggests that the protein translation inhibitory effect is compromised by attachment of hydrophobic
1
6, 17
moieties to tobramycin as previously shown.
Furthermore, activity trend between the hybrids revealed
a correlation between antimicrobial potency and carbon chain length. The longer the carbon tether, the
better the antimicrobial efficacy of the compound. This is however not surprising as studies have shown
that high hydrophobicity facilitates penetration of membraneꢀactive compounds across the bacterial
4
0, 41
membrane.
The physicochemical properties necessary to navigate a complex membrane topology,
especially as represented in P. aeruginosa, is perhaps the principal reason for the varied activity of the
hybrid molecules. Although ultrashort and amphiphilic lysineꢀbased peptoid mimics were previously
2
3
reported to have promising activities against P. aeruginosa MTCC 424, our evaluation of compound 4
(different alkyl chain) against a panel of organisms revealed otherwise (Table 1). This may be due to the
slight change in the length of the alkyl chain (C10 to C12), and perhaps, the different bacterial strains
1
6
tested. The amphiphilic nature of 4, the reported properties of tobramycin hybrids, and the differential
activity of the new molecules based on carbon chain length, gave a clue on possible membrane effect of
these compounds. Thus, we investigated antimicrobial activities of the amphiphilic conjugates in
combination with other antibiotics, particularly against P. aeruginosa, the major nosocomial pathogen
and leading cause of infection in cystic fibrosis patients. Although most of the clinical isolates
investigated in this study were resistant to tobramycin and other antibiotics (with the exception of
colistin), the MIC values of the conjugates against these strains were similar to that of wildꢀtype P.
aeruginosa PAO1 (Table 2), suggesting that the targeting site and/or mechanism of action of the
conjugates was different than that of tobramycin.
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The ability of 3 to perturb the membrane was verified by its synergistic effect with vancomycin, a drug
that cannot pass through the outer membrane of P. aeruginosa due to its large size (M.W. = 1449.2).
Synergism was also observed for other antibiotics with different modes of action against P. aeruginosa,
the most prominent being novobiocin, minocycline and rifampicin. Importantly, combining 3 with
minocycline, or with rifampicin, can revive the antimicrobial activities of these antibiotics against MDR
and XDR P. aeruginosa isolates (Table 4). The uptake of tetracyclines, such as minocycline, is known to
be driven by transmembrane chemical gradient (∆pH) of PMF generated by the respiratory chain on the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
2
cytoplasmic membrane. The other component of PMF is electrical potential (∆Ψ), which is known to
43
drive aminoglycosides uptake. Bacteria control ∆Ψ and ∆pH exquisitely to maintain a constant value of
44
PMF, and disruption of either component is compensated for by a counteracting increase in the other.
When a compound disrupts ∆Ψ, antagonism effect will be observed in combination with aminoglycosides,
while synergism will show in combination with tetracyclines due to the compensatory increase of ∆pH.
Tetracyclines and aminoglycosides have therefore been used as two relevant antibiotics in combination
studies with other drugs to identify compounds that affect membrane PMF and specifically dissipate
31
either component of PMF. In this study, compound 3 displayed different synergistic effects with
minocycline (synergism) and aminoglycosides (no interaction), an observation that is consistent with
dissipation of ∆Ψ component of the PMF by 3. However, the expected antagonistic effect of 3 with
aminoglycosides was not observed, likely due to the membrane penetration induced by 3 that slightly
affected aminoglycosides uptake into bacterial cells. The effect of 3 on ∆Ψ was further corroborated by
the increased diSC ꢀ5 fluorescence (Figure 7) and repression of swimming motility controlled by this
3
parameter (Figure 8). Compounds that collapse the PMF are known to inhibit ATP synthesis and flagellar
4
5
motility, preventing or reducing swimming activity.
Membraneꢀassociated efflux is another major mechanism that prevents bioaccumulation of drugs
4
6
within the cytosol, thus preventing/reducing access of antibiotics to intracellular targets. Efflux pump
proteins localized in the cytoplasmic and outer membrane, and linked by a periplasmic membrane fusion
7
ꢀ9
protein (MFP), play a major role in intrinsic and acquired resistance of P. aeruginosa. The associated
1
4
ACS Paragon Plus Environment

Page 15 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
30
resistance is based on energyꢀdependent effluxes, which are usually driven by PMF. We envisaged that
the dissipation of ∆Ψ in PMF will prevent electron transport across the respiratory chain, thus inhibiting
ATP synthesis and ultimately affect efflux pump system. We therefore studied the effect of the conjugates
on efflux pumps using P. aeruginosa effluxꢀdeficient strains PAO200 and PAO750, and perhaps explain
the observed synergistic mechanism more exquisitely. The efflux pump knockꢀout decreased the MIC of
minocycline by 8ꢀfold (Table 5), confirming that minocycline is a substrate for these efflux pumps, which
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
47
is consistent with previous study. The synergism observed when 3 was combined with minocycline
against wildꢀtype PAO1 was not observed in PAO200 and PAO750, with FIC indices > 0.5, indicating
that the antimicrobial activity potentiation of minocycline by 3 is due to the inhibition of efflux pumps,
particularly the RND pumps. To validate this hypothesis, the synergistic effects of 3 with other known P.
aeruginosa efflux pump substrates were evaluated in PAO200 and PAO750, including chloramphenicol,
7
, 48ꢀ50
erythromycin, trimethoprim, and moxifloxacin.
The results showed weak synergy or additive effects
of 3 in these combinations (Table S2), corroborating the efflux pump inhibitory activity of 3. We posited
that the dissipation of electrochemical gradient across the cytoplasmic membrane affected respiratory
ATP production, thereby compromising efflux pump efficiency.
Surprisingly, rifampicin that is not substrate of the five efflux pumps investigated in this study
(MexABꢀOprM, MexCDꢀOprJ, MexEFꢀOprN, MexJK, and MexXY) was similarly strongly synergized
by 3 against PAO200 and PAO750 (Table 5). Rifampicin is known to kill bacteria by inhibiting RNA
51
synthesis after binding to DNAꢀdependent RNA polymerase. Both Gramꢀpositive and Gramꢀnegative
bacteria are similarly sensitive to rifampicin, with the higher MICs reported in Gramꢀnegative bacteria
5
2
due to its low penetration across the outer membrane. A combination study of rifampicin with colistin (a
wellꢀknown membrane permeabilizer) has demonstrated that perturbation of P. aeruginosa outer
5
3
membrane can indeed potentate the antimicrobial activity of rifampicin, thus confirming the results of
this study with 3 (Figure 3). Outer membrane perturbation is perhaps the reason why 3 was able to
synergize rifampicin against PAO200 and PAO750 despite not being a substrate for the pumps. It is
however clear that minocycline is more sensitive to PMF dissipation caused by compound 3 than simply
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 62
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
membrane penetration induced by colistin, as evident in the FIC indices > 0.5 shown in Table 5. These
results suggest that compound 3 is not just able to penetrate P. aeruginosa cell membrane like colistin,
but could also dissipate the cytoplasmic membrane, and compromise the efficient functioning of the
efflux systems. Although reference compound 4 similarly displayed cytoplasmic membrane
depolarization activity and suppressed bacterial swimming motility, it was to a lesser extent than 3 at the
same concentration (Figures 7 and 8), and did not display any synergistic effect in combination with
minocycline and with rifampicin (Table S1). The inability of 4 to potentiate minocycline like 3 despite its
ability to partially depolarize the cytoplasmic membrane as well may be due to its weak outer membrane
perturbation that prevents uptake of other antimicrobial agents. The simultaneous occurrence of both
phenomenon is indeed critical to the adjuvant properties of tobramycinꢀlysine conjugates.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In contrast to the bactericidal nature of rifampicin, minocycline is known to be only bacteriostatic,
which was evident in the timeꢀkill assay with constant bacterial cells number at all concentrations tested
54, 55
(
Figure S24), the observation that is consistent with previous studies.
However, the increased killing
efficiency of minocycline when used in combination with 3 is likely attributable to the effect of 3 on the
membrane. Attempts to select for resistance with combination of 3 and minocycline during 25 serial
passages resulted in a 4ꢀfold increase in MIC, as opposed to minocycline and tobramycin alone that had
16ꢀ and 256ꢀfolds increase respectively (Figure 5). Indeed, it is more difficult for bacteria to develop
resistance to simultaneouslyꢀacting drug combination, especially when one of the drug acts on
56ꢀ58
membrane
.
A major concern about membraneꢀacting and PMFꢀcollapsing agents is their toxicities towards
31
eukaryotic cells. To verify the safety of these compounds, the toxicities of the conjugates were evaluated
against pig erythrocytes and mammalian cancer cell lines. It was surprising to see a dramatic reduction of
the hemolytic activity of 4 when joint to tobramycin (Figure 9A). This is probably caused by changes to
5
9
the molecular amphipathy as previously seen for antimicrobial peptides. Moreover, combination therapy
would allow for reduced doses to be used, minimizing cytotoxicity, and 3 displayed negligible toxicity at
its effective synergistic concentration (≤ 4 ꢁg/mL). In the in vivo study, the high tolerance of Galleria
1
6
ACS Paragon Plus Environment

Page 17 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mellonella worms to 3 (100 % survival at 200 mg/kg after 96 h) further confirmed the safety of this
compound. Galleria mellonella injection model has been commonly used in accessing the in vivo efficacy
of antimicrobials against P. aeruginosa because it shares a high degree of structural and functional
6
0
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
homology to the immune systems of vertebrates with both cellular and humoral defenses. In contrast to
monotherapy, single dose combination of 3 (75 mg/kg) plus minocycline (75 mg/kg) or 3 (75 mg/kg) plus
rifampicin (75 mg/kg) effectively protected larvae from XDR P. aeruginosa P262 infection with more
than 75 % survival after 24 h, indicating the therapeutic potential of amphiphilic tobramycin as an
adjuvant to treat infection caused by XDR P. aeruginosa.
CONCLUSIONS
In this study, we demonstrated that amphiphilic tobramycinꢀlysine conjugates preserve many of the
known adjuvant properties of previously reported tobramycinꢀfluoroquinolone hybrids. From a medicinal
chemistry point, linking a tobramycin Cꢀ12 vector to an amphiphilic lysine conjugate enhances the outer
membrane destabilization effect of the amphiphilic lysine analog. As such our study suggests that a
tobramycinꢀC12 tether at Cꢀ5 position in tobramycin serves as an effective vector to promote delivery of
compounds through the outer membrane barrier of Gramꢀnegative bacteria with an optimized effect on P.
aeruginosa. However, the effect of the tobramycinꢀC12 tether appears to be not limited to the outer
membrane but also involves the cytoplasmic membrane. For instance, we provide evidence that
conjugates containing a tobramycinꢀC12 tether reduces the ∆Ψ component of the PMF located at the
cytoplasmic membrane. This leads to decreased activity of the efflux associated pumps but at the same
can lead to enhanced cytoplasmic uptake of agents which depend on the ∆ pH component of the PMF like
the tetracycline class of antibiotics. Overall, this study provides a promising strategy for generating
effective antibiotic adjuvants that overcome drug resistance in MDR Gramꢀnegative bacteria including P.
aeruginosa by carefully designing amphiphilic tobramycin tethers. The discovery that compound 3 can
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 62
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
potentiate several classes of antibiotics against resistant pathogens is set to expand the antimicrobial space
and optimize our usage of antibiotics in our current armamentarium.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL SECTION
1
13
Chemical Synthesis. General Information. NMR spectra ( H, C, DEPT, COSY, HSQC and HMBC)
1
13
were recorded on a Bruker Avance 500 spectrometer (500 MHz for H NMR, 126 MHz for C). All
reactions were monitored by analytical thinꢀlayer chromatography (TLC) on preꢀcoated silica gel plates
60 F254 (0.25 mm, Merck, Ontario, Canada), and the spots were visualized by ultraviolet light and/or by
staining with ninhydrin solution in nꢀbutanol. Mass spectrometry was carried by ESI analyses on a Varian
500 MS Ion Trap Mass Spectrometer, and MALDIꢀTOF on a Bruker Daltonics Ultraflex MALDI
TOF/TOF Mass Spectrometer. Chromatographic separations were performed on a silica gel column by
flash chromatography (Kiesel gel 40, 0.040ꢀ0.063 mm; Merck, Ontario, Canada). Yields were calculated
after purification. When reactions were carried out under anhydrous conditions, the mixtures were
maintained under nitrogen atmosphere. Analytical HPLC was performed on Hitachi LC system equipped
with autosampler, using Superspher 100 RPꢀ18 column and a detection wavelength of 260 nm. The purity
of final compounds determined by HPLC analysis were > 95 %. Detailed experimental procedures of the
intermediates BocꢀLys(Boc)ꢀOH, 10a-c, 11a-c and 12a-c were described in the Supporting Information.
The following data are for the intermediates after reductive amination and the final compounds tested in
the biological studies.
General Procedure A: Deprotection of Boc and TBDMS Groups for the Synthesis of Tobramycinꢀlysine
Conjugates (1ꢀ3). Compounds 14a–c (1 equiv) were treated with 40 % HCl in MeOH (~1.5 mL of solvent
per 0.01 mmol of intermediates 14a–c) at room temperature for 3 h. The reaction progress was monitored
by TLC (MeOH/CH Cl /NH OH, 6:4:3). At the end of the reaction, the mixture was concentrated under
2
2
4
reduced pressure to get the solid tobramycin conjugate as HCl salt. The crude was further purified via Cꢀ
8 reverseꢀphase flash column chromatography (eluted with deionized water) to afford analytically pure
compounds as yellow solid.
1
1
8
ACS Paragon Plus Environment

Page 19 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
General Procedure B: Reductive Amination for the Synthesis of 5ꢀOꢀ(alkylatedꢀ10ꢀaminomethylꢀ9ꢀ
chloroanthracene)ꢀ1,3,2′,6′,3′′ꢀpentaꢀNꢀ(tertꢀbutoxycarbonyl)ꢀ4′,2′′,4′′,6′′ꢀtetraꢀOTBDMSꢀtobramycin
(13a–c). Compounds 12a–c (1 equiv) and commercially available aromatic aldehyde 5 (1.2 equiv) were
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
dissolved in dry CH Cl :MeOH (1:1, v/v) (~20.0 mL of solvent per mmol of intermediates 12a–c) and
2
2
stirred at room temperature overnight. The resulting clear solution was cooled to 0 °C, sodium
borohydride (3 equiv) added and stirred at room temperature for 3 h. The solvents were then evaporated
and the crude reꢀdispersed in diethyl ether followed by the addition of 2 N NaOH. The mixture was
stirred at room temperature for 15 mins and the organic layer separated from aqueous phase, washed with
water and brine, and dried over anhydrous Na SO . The mixture was concentrated and purified by flash
2
4
column chromatography (eluted with CH Cl /MeOH from 300:1 to 30:1, v/v) to give the desired product
2
2
as yellow solid.
General Procedure C: Secondary Amide Coupling for the Synthesis of Compounds 14a–c. Bocꢀ
Lys(Boc)ꢀOH (1.5 equiv) dissolved in DMF (~30.0 mL of solvent per mmol of intermediates 13a–c) was
activated with DIPEA (3 equiv) and HBTU (1.5 equiv) at 0 °C for 15 mins and subsequently treated with
13a–c (1 equiv). The mixture was stirred at 0 °C to room temperature overnight. The reaction progress
was monitored by TLC (CH Cl /MeOH, 35:1), and at the end, the mixture was diluted with water and
2
2
extracted with EtOAc. The organic layer was washed with water and brine, dried over anhydrous Na SO ,
2
4
and concentrated. The resulting residue was purified by flash column chromatography (eluted with
CH Cl /MeOH from 300:1 to 30:1, v/v) to afford the desired compound as yellow solid.
2
2
Compound 1. Synthesized following general procedure A from 14a (18.6 mg, 0.009 mmol). The crude
material was purified by Cꢀ18 reverseꢀphase flash column chromatography using deionized water. The
1
product was isolated as a yellow solid (6.3 mg, 78 %). H NMR (500 MHz, Deuterium Oxide) δ 8.74 –
8.60 (m, 2H, anthracene), 8.33 – 8.24 (m, 2H, anthracene), 7.86 – 7.73 (m, 4H, anthracene), 5.99 (d, J =
1
2
1
2
1
1
5
5.4 Hz, 1H, NꢀCH H ꢀanthracene), 5.30 (d, J = 15.3 Hz, 1H, NꢀCH H ꢀanthracene), 5.14 (d, J = 2.5 Hz,
H, Hꢀ1′), 4.95 (d, J = 3.4 Hz, 1H, Hꢀ1″), 4.43 (t, J = 6.3 Hz, 1H, αꢀCH of Lys), 4.29 – 4.24 (m, 1H, Hꢀ
′), 4.03 – 3.94 (m, 2H, Hꢀ4, Hꢀ4′), 3.83 – 3.77 (m, 2H, Hꢀ6, Hꢀ2″), 3.65 – 3.46 (m, 5H, Hꢀ1, Hꢀ3, Hꢀ2′,
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 62
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
Hꢀ4″, Hꢀ5″), 3.39 – 3.30 (m, 2H, Hꢀ6′), 3.21 – 3.12 (m, 3H, Hꢀ5, Hꢀ3″, Hꢀ6″), 3.03 – 2.95 (m, 1H, Nꢀ
1
2
1
2
CH H CH ), 2.90 – 2.64 (m, 5H, NꢀCH H CH , εꢀCH of Lys, OꢀCH of linker), 2.59 – 2.54 (m, 1H, Hꢀ
2
2
2
2
2
), 2.39 – 2.32 (m, 1H, Hꢀ6″), 2.24 – 2.12 (m, 2H, Hꢀ3′), 1.97 – 1.82 (m, 3H, Hꢀ2, βꢀCH of Lys), 1.64 –
2
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
1.54 (m, 2H, δꢀCH of Lys), 1.45 – 1.31 (m, 2H, γꢀCH of Lys), 1.27 – 1.10 (m, 3H, CH of linker), 1.09 –
2 2 2
1
3
0.99 (m, 1H, CH of linker). C NMR (126 MHz, Deuterium Oxide) δ 169.34 (NCOCH), 131.18, 130.09,
2
1
8
5
28.13, 127.48, 127.40, 126.65, 125.62, 124.36, 100.79 (anomeric CHꢀ1″), 93.15 (anomeric CHꢀ1′),
2.41, 81.22, 77.20 (CHꢀ4), 75.41 (CHꢀ5′), 73.18, 72.80 (OꢀCH ꢀlinker), 68.44, 64.30, 63.42 (CHꢀ6),
2
8.41 (CH ꢀ6″), 54.46, 50.61 (αꢀCH of Lys), 49.36, 48.60, 47.65, 45.32 (NꢀCH CH ), 40.74 (NꢀCH ꢀ
2
2
2
2
anthracene), 38.91 (εꢀCH of Lys), 38.60 (CHꢀ6′), 30.90 (βꢀCH of Lys), 28.34 (CH ꢀ3′), 28.18 (CH ꢀ2),
2
2
2
2
26.55 (δꢀCH of Lys), 25.36, 24.45, 21.16; MALDIꢀTOFꢀMS: m/z calc′d for C H ClN O Na: 913.457,
2
43 67
8
10
+
found: 913.463 [M+Na] .
Compound 2. Synthesized following general procedure A from 14b (28 mg, 0.013 mmol). The crude
material was purified by Cꢀ18 reverseꢀphase flash column chromatography using deionized water. The
1
product was isolated as a yellow solid (9.4 mg, 75 %). H NMR (500 MHz, Deuterium Oxide) δ 8.49 –
8.44 (m, 2H, anthracene), 8.15 – 8.11 (m, 2H, anthracene), 7.71 – 7.62 (m, 4H, anthracene), 5.70 (d, J =
1
2
1
2
15.4 Hz, 1H, NꢀCH H ꢀanthracene), 5.35 – 5.28 (m, 2H, NꢀCH H ꢀanthracene, Hꢀ1′), 5.16 (d, J = 3.5 Hz,
1H, Hꢀ1″), 4.37 (t, J = 6.3 Hz, 1H, αꢀCH of Lys), 4.32 – 4.27 (m, 1H, Hꢀ5′), 4.15 (t, J = 9.8 Hz, 1H, Hꢀ4),
3
3
6
2
.96 – 3.89 (m, 3H, Hꢀ6, Hꢀ4′, Hꢀ2″), 3.83 – 3.76 (m, 4H, Hꢀ5, OꢀCH H of linker, Hꢀ4″, Hꢀ5″), 3.75 –
1 2
.72 (m, 1H, Hꢀ6″), 3.67 – 3.55 (m, 5H, Hꢀ1, Hꢀ3, Hꢀ2′, Hꢀ3″, OꢀCH H of linker), 3.53 – 3.49 (m, 1H, Hꢀ
1
2
1
2
″), 3.40 – 3.33 (m, 2H, Hꢀ6′), 2.98 – 2.90 (m, 1H, NꢀCH H CH ), 2.86 – 2.78 (m, 2H, εꢀCH of Lys),
2
2
1
2
.62 – 2.52 (m, 2H, Hꢀ2, NꢀCH H CH ), 2.25 – 2.20 (m, 2H, Hꢀ3′), 2.02 – 1.96 (m, 1H, Hꢀ2), 1.92 – 1.82
2
(m, 2H, βꢀCH of Lys), 1.64 – 1.57 (m, 2H, δꢀCH of Lys), 1.44 – 1.35 (m, 3H, γꢀCH of Lys, CH of
2
2
2
2
linker), 1.32 – 1.27 (m, 1H, CH of linker), 1.11 – 1.04 (m, 1H, CH of linker), 1.02 – 0.91 (m, 2H, CH
2
2
2
1
3
of linker), 0.83 – 0.63 (m, 4H, CH of linker), 0.62 – 0.53 (m, 3H, CH of linker). C NMR (126 MHz,
2
2
Deuterium Oxide) δ 169.14 (NCOCH), 131.04, 129.85, 127.86, 127.19, 127.09, 126.25, 125.25, 124.12,
01.39 (anomeric CHꢀ1″), 92.71 (anomeric CHꢀ1′), 82.04, 81.95, 76.97 (CHꢀ4), 75.59 (CHꢀ5′), 73.84 (Oꢀ
1
2
0
ACS Paragon Plus Environment

Page 21 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CH ꢀlinker), 73.18, 68.56, 64.70, 63.29, 59.12 (CH ꢀ6″), 54.72, 50.72 (αꢀCH of Lys), 49.76, 48.45, 47.37,
2
2
4
5.80 (NꢀCH CH ), 40.80 (NꢀCH ꢀanthracene), 38.94 (εꢀCH of Lys), 38.59 (CHꢀ6′), 30.72 (βꢀCH of
2 2 2 2 2
Lys), 29.19, 28.20, 28.15, 28.08, 28.02, 27.25, 26.49 (δꢀCH of Lys), 25.15, 24.86, 21.24; MALDIꢀTOFꢀ
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
MS m/z calc′d for C H ClN O Na: 969.519, found: 969.523 [M+Na] .
47
75
8
10
Compound 3. Synthesized following general procedure A from 14c (125 mg, 0.058 mmol). The crude
material was purified by Cꢀ18 reverseꢀphase flash column chromatography using deionized water. The
1
product was isolated as a yellow solid (42.3 mg, 73 %). H NMR (500 MHz, Deuterium Oxide) δ 8.69 –
8
.58 (m, 2H, anthracene), 8.30 – 8.24 (m, 2H, anthracene), 7.79 – 7.70 (m, 4H, anthracene), 5.87 (d, J =
1
2
1
2
1
5.4 Hz, 1H, NꢀCH H ꢀanthracene), 5.47 (d, J = 15.3 Hz, 1H, NꢀCH H ꢀanthracene), 5.43 (d, J = 2.6 Hz,
1H, Hꢀ1′), 5.22 (d, J = 3.6 Hz, 1H, Hꢀ1″), 4.40 (t, J = 6.3 Hz, 1H, αꢀCH of Lys), 4.32 – 4.28 (m, 1H, Hꢀ
5′), 4.21 (t, J = 9.8 Hz, 1H, Hꢀ4), 3.99 – 3.74 (m, 11H, Hꢀ5, Hꢀ6, Hꢀ2′, Hꢀ4′, Hꢀ2″, Hꢀ4″, Hꢀ5″, Hꢀ6″, Oꢀ
CH of linker), 3.67 – 3.57 (m, 3H, Hꢀ1, Hꢀ3, Hꢀ3″), 3.45 – 3.32 (m, 2H, Hꢀ6′), 3.03 – 2.96 (m, 1H, Nꢀ
2
1
2
1
2
CH H CH ), 2.89 – 2.79 (m, 2H, εꢀCH of Lys), 2.78 – 2.71 (m, 1H, NꢀCH H CH ), 2.58 – 2.54 (m, 1H,
2
2
2
Hꢀ2), 2.32 – 2.23 (m, 2H, Hꢀ3′), 2.03 – 1.86 (m, 3H, Hꢀ2, βꢀCH of Lys), 1.75 – 1.66 (m, 2H, CH of
2
2
linker), 1.65 – 1.59 (m, 2H, δꢀCH of Lys), 1.45 – 1.29 (m, 6H, γꢀCH of Lys, CH of linker), 1.25 – 1.18
2
2
2
(m, 2H, CH of linker), 1.13 – 0.94 (m, 4H, CH of linker), 0.91 – 0.83 (m, 2H, CH of linker), 0.78 –
2
2
2
13
0.71 (m, 1H, CH of linker), 0.68 – 0.59 (m, 3H, CH of linker), 0.58 – 0.50 (m, 2H, CH of linker). C
2
2
2
NMR (126 MHz, Deuterium Oxide) δ 169.21 (NCOCH), 131.30, 130.11, 128.10, 127.30, 127.18, 126.53,
25.43, 124.24, 101.39 (anomeric CHꢀ1″), 92.73 (anomeric CHꢀ1′), 81.93, 81.89, 76.82 (CHꢀ4), 75.66
CHꢀ5′), 73.84 (OꢀCH ꢀlinker), 73.22, 68.54, 64.77, 63.24, 59.27 (CH ꢀ6″), 54.76, 50.73 (αꢀCH of Lys),
1
(
2
2
4
9.79, 48.44, 47.34, 46.04 (NꢀCH CH ), 41.03 (NꢀCH ꢀanthracene), 38.94 (εꢀCH of Lys), 38.55 (CHꢀ6′),
2 2 2 2
30.74 (βꢀCH of Lys), 29.57 (CH of linker), 29.13, 29.00, 28.70, 28.56, 28.19, 28.16 (CH ꢀ3′), 27.98,
2
2
2
27.91 (CH ꢀ2), 27.32, 26.50 (δꢀCH of Lys), 25.40, 25.22, 21.28; MALDIꢀTOFꢀMS: m/z calc′d for
2
2
+
C H ClN O Na: 1025.582, found: 1025.586 [M+Na] .
5
1
83
8
10
LysꢀNꢀdodecylꢀ10ꢀaminomethylꢀ9ꢀchloroanthracene Trifluoroacetate (4). The BocꢀLys(Boc)ꢀNꢀalkylꢀ
aromatic compound 8 (22.8 mg, 0.031 mmol) was dissolved in CH Cl :TFA (2:1, v/v) (3 mL) and stirred
2
2
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 62
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
at room temperature for 1 h. The reaction was monitored by TLC (CH Cl /NH OH/MeOH, 5:1:1). At the
2
2
4
end of the reaction, the mixture was evaporated to dryness, and purified by Cꢀ18 reverseꢀphase flash
column chromatography (eluted with deionized water) to get analytically pure compound 4 as a yellow
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
solid (10.3 mg, 62 %). H NMR (500 MHz, Methanolꢀd ) δ 8.61 – 8.57 (m, 2H, anthracene), 8.43 – 8.38
4
1
2
(m, 2H, anthracene), 7.69 – 7.61 (m, 4H, anthracene), 6.10 (d, J = 15.3 Hz, 1H, NꢀCH H ꢀanthracene),
1
2
5
.45 (d, J = 15.3 Hz, 1H, NꢀCH H ꢀanthracene), 4.28 (t, J = 6.2 Hz, 1H, αꢀCH of Lys), 3.02 – 2.96 (m,
1
2
1
2
1
H, NꢀCH H CH ), 2.78 – 2.69 (m, 2H, εꢀCH of Lys), 2.68 – 2.60 (m, 1H, NꢀCH H CH ), 1.86 – 1.76
2
2
2
(
(
m, 2H, βꢀCH of Lys), 1.59 – 1.52 (m, 2H, δꢀCH of Lys), 1.48 – 1.39 (m, 2H, γꢀCH of Lys), 1.34 – 1.20
2 2 2
m, 10H, NꢀCH (CH ) CH ), 1.17 – 1.11 (m, 2H, NꢀCH (CH ) CH ), 1.09 – 1.03 (m, 2H, Nꢀ
2
2 10
3
2
2 10
3
1
3
CH (CH ) CH ), 0.97 – 0.77 (m, 9H, NꢀCH (CH ) CH ). C NMR (126 MHz, Methanolꢀd ) δ 168.34
2
2 10
3
2
2 10
3
4
(NCOCH), 131.61, 130.03, 128.40, 127.33, 126.75, 126.61, 125.23, 124.19, 50.46 (αꢀCH of Lys), 45.08
(
NꢀCH CH ), 39.48 (NꢀCH ꢀanthracene), 38.70 (εꢀCH of Lys), 31.64, 30.92 (βꢀCH of Lys), 29.27,
2 2 2 2 2
2
9.16, 29.06, 29.02, 28.82, 28.56, 28.37, 26.76 (δꢀCH of Lys), 25.95, 22.31, 21.25, 13.01 (CH CH );
2
2
3
+
MALDIꢀTOFꢀMS: m/z calc′d for C H ClN O: 538.356, found: 538.358 [M+H] .
33
49
3
Nꢀdodecylꢀ10ꢀaminomethylꢀ9ꢀchloroanthracene Hydrochloride (7). Dodecylamine 6 (11.9 mg, 0.065
mmol) and aromatic aldehyde 5 (19 mg, 0.078 mmol) were dissolved in dry CH Cl :MeOH (1:1, v/v) (3
2
2
mL) and stirred at room temperature overnight. The mixture was cooled to 0 °C, and exposed to sodium
borohydride (7 mg, 0.195 mmol) at room temperature overnight. The reaction mixture was subsequently
concentrated, reꢀdispersed in diethyl ether, and treated with 2 N NaOH (3 mL) at room temperature for
additional 15 mins. The organic layer was separated from the aqueous phase, washed with water and
brine, dried over anhydrous Na SO , concentrated, and purified by flash column chromatography (eluted
2
4
with CH Cl /MeOH from 300:1 to 30:1, v/v) to afford the desired compound as a yellow solid (19 mg, 70
2
2
1
%). H NMR (500 MHz, Chloroformꢀd) δ 8.60 – 8.55 (m, 2H, anthracene), 8.40 – 8.35 (m, 2H,
anthracene), 7.64 – 7.54 (m, 4H, anthracene), 4.70 (s, 2H, NHꢀCH ꢀanthracene), 2.91 – 2.84 (m, 2H, NHꢀ
2
CH CH ), 1.62 – 1.57 (m, 2H, NHꢀCH CH ), 1.37 ꢀ 1.24 (m, 18H, NHꢀCH CH (CH ) CH ), 0.89 (t, J =
2
2
2
2
2
2
2 9
3
13
6
.8 Hz, 3H, CH CH ). C NMR (126 MHz, Chloroformꢀd) δ 131.67, 130.70, 129.00, 128.66, 126.31,
2 3
2
2
ACS Paragon Plus Environment

Page 23 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
26.21, 125.54, 124.53 (anthracene), 50.71 (NHꢀCH CH ), 45.93 (NHꢀCH ꢀanthracene), 32.00, 30.17
2
2
2
(
NHꢀCH CH ), 29.75, 29.72, 29.70, 29.64, 29.44, 27.45, 22.77, 14.20 (CH CH ); ESIꢀMS: m/z calc′d for
2 2 2 3
+
C H ClN: 410.2, found: 410.1 [M+H] .
2
7
37
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
BocꢀLys(Boc)ꢀNꢀdodecylꢀ10ꢀaminomethylꢀ9ꢀchloroanthracene (8). BocꢀLys(Boc)ꢀOH (24 mg, 0.068
mmol) dissolved in DMF:CHCl (5:2, v/v) (3.5 mL) was activated with DIPEA (18 mg, 0.135 mmol) and
3
HBTU (26 mg, 0.068 mmol) at 0 °C for 15 mins, and treated with 7 (19 mg, 0.045 mmol). The mixture
was stirred at 0 °C to room temperature overnight, concentrated under reduced pressure, and the resulting
solution diluted in ethyl acetate. The mixture was washed with 0.5 M KHSO , water, and brine
4
successively, and dried over anhydrous Na SO . The organic layer was concentrated and purified by flash
2
4
column chromatography (eluted with CH Cl /MeOH from 300:1 to 100:1, v/v) to afford the desired
2
2
1
compound as a yellow solid (30 mg, 90 %). H NMR (500 MHz, Chloroformꢀd) δ 8.64 – 8.54 (m, 2H,
anthracene), 8.30 – 8.22 (m, 2H, anthracene), 7.64 – 7.51 (m, 4H, anthracene), 6.05 (d, J = 15.3 Hz, 1H,
1
2
1
2
NꢀCH H ꢀanthracene), 5.28 (d, J = 15.2 Hz, 1H, NꢀCH H ꢀanthracene), 4.57 – 4.51 (m, 1H, αꢀCH of
1
2
Lys), 3.12 – 2.95 (m, 2H, εꢀCH of Lys), 2.94 – 2.86 (m, 1H, NꢀCH H CH ), 2.77 – 2.66 (m, 1H, Nꢀ
2
2
1
2
CH H CH ), 1.63 – 1.56 (m, 2H, βꢀCH of Lys), 1.51 – 1.34 (m, 23H), 1.33 – 1.18 (m, 9H), 1.18 – 1.12
2
2
1
3
(m, 2H), 1.11 – 1.05 (m, 2H), 1.04 – 0.98 (m, 2H), 0.97 – 0.83 (m, 7H). C NMR (126 MHz,
Chloroformꢀd) δ 172.45 (NCOCH), 155.91, 155.49 (2 COOC), 131.66, 130.50, 128.51, 127.57, 126.74,
1
26.56, 125.78, 124.22, 50.28 (αꢀCH of Lys), 45.28 (NꢀCH CH ), 40.29 (εꢀCH of Lys), 39.54 (NꢀCH ꢀ
2 2 2 2
anthracene), 33.77 (βꢀCH of Lys), 31.90, 29.57, 29.51, 29.48, 29.39, 29.32, 29.25, 29.22, 28.84, 28.40 ꢀ
2
28.38 (CH ), 26.54, 22.67, 14.11 (CH CH ); ESIꢀMS: m/z calc′d for C H ClN O Na: 760.443, found:
3 2 3 43 64 3 5
+
7
60.485 [M+Na] .
5ꢀOꢀ(4ꢀbutylꢀ10ꢀAminomethylꢀ9ꢀchloroanthracene)ꢀ1,3,2′,6′,3′′ꢀpentaꢀNꢀ(tertꢀbutoxycarbonyl)ꢀ
4′,2′′,4′′,6′′ꢀtetraꢀOTBDMSꢀtobramycin (13a). Synthesized following general procedure B from 12a (33
mg, 0.022 mmol), aromatic aldehyde 5 (6 mg, 0.026 mmol) and sodium borohydride (2.5 mg, 0.066
mmol). The resulting residue was purified by flash column chromatography using CH Cl /MeOH (300:1
2
2
1
to 30:1, v/v) to afford the desired compound as a yellow solid (21 mg, 55 %). H NMR (500 MHz,
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 62
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
Chloroformꢀd) δ 8.58 – 8.53 (m, 2H, anthracene), 8.41 – 8.36 (m, 2H, anthracene), 7.62 – 7.55 (m, 4H,
anthracene), 5.29 – 5.13 (m, 2H, Hꢀ1′, Hꢀ1″), 4.71 (s, 2H, NHꢀCH ꢀanthracene), 4.30 – 4.06 (m, 3H), 3.85
2
–
3.17 (m, 15H), 2.84 (t, J = 7.1 Hz, 2H, NHꢀCH CH ), 2.51 – 2.40 (m, 1H), 2.03 – 1.96 (m, 1H), 1.66 –
2 2
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
1.55 (m, 4H, CH of linker), 1.55 – 1.51 (m, 1H), 1.48 – 1.29 (m, 45H), 1.10 – 1.01 (m, 1H), 0.94 – 0.81
2
1
3
(m, 36H, SiꢀCCH ), 0.14 – ꢀ0.05 (m, 24H, SiꢀCH ). C NMR (126 MHz, Chloroformꢀd) δ 155.64,
3
3
1
1
7
55.47, 154.78, 154.56, 154.27 (5 COOC), 131.73, 130.81, 128.94, 128.65, 126.35, 126.28, 125.45,
24.69, 97.77 (anomeric CHꢀ1″), 96.36 (anomeric CHꢀ1′), 85.81, 79.91, 79.39, 79.27, 79.19, 78.71,
5.25, 73.17, 72.60, 71.49, 67.99, 66.93, 63.16, 57.26, 50.89 (NHꢀCH CH ), 50.54, 48.89, 48.32, 45.81
2
2
(
NHꢀCH ꢀanthracene), 41.62, 35.98, 35.73, 28.62 ꢀ 28.22 (OꢀCCH ), 26.72, 26.10 ꢀ 25.78 (SiꢀCCH ),
2 3 3
18.44, 18.29, 18.06, 17.91 (4 SiꢀCCH ), ꢀ3.45, ꢀ3.80, ꢀ4.22, ꢀ4.87, ꢀ4.92, ꢀ5.06, ꢀ5.19, ꢀ5.22 (8 SiꢀCH );
3
3
+
ESIꢀMS: m/z calc′d for C H ClN O Si : 1721.9, found: 1721.6 [M+H] .
86
152
6
19
4
5ꢀOꢀ(8ꢀoctylꢀ10ꢀaminomethylꢀ9ꢀchloroanthracene)ꢀ1,3,2′,6′,3′′ꢀpentaꢀNꢀ(tertꢀbutoxycarbonyl)ꢀ
4′,2′′,4′′,6′′ꢀtetraꢀOTBDMSꢀtobramycin (13b). Synthesized following general procedure B from 12b (35
mg, 0.023 mmol), aromatic aldehyde 5 (7 mg, 0.028 mmol) and sodium borohydride (3 mg, 0.069 mmol).
The resulting residue was purified by flash column chromatography using CH Cl /MeOH (300:1 to 30:1,
2
2
1
v/v) to afford the desired compound as a yellow solid (28 mg, 70 %). H NMR (500 MHz, Chloroformꢀd)
δ 8.62 – 8.52 (m, 2H, anthracene), 8.42 – 8.33 (m, 2H, anthracene), 7.67 – 7.50 (m, 4H, anthracene), 5.26
–
4
2
1
5.12 (m, 2H, Hꢀ1′, Hꢀ1″), 4.71 (s, 2H, NHꢀCH ꢀanthracene), 4.31 – 4.21 (m, 1H), 4.21 – 4.13 (m, 1H),
2
.12 – 4.04 (m, 1H), 3.83 – 3.18 (m, 15H), 2.87 (t, J = 7.4 Hz, 2H, NHꢀCH CH ), 2.51 – 2.44 (m, 1H),
2
2
.03 – 1.98 (m, 1H), 1.61 – 1.56 (m, 2H, CH of linker), 1.55 – 1.51 (m, 1H), 1.51 – 1.37 (m, 47H), 1.35 –
2
13
.24 (m, 8H), 1.09 – 1.00 (m, 1H), 0.97 – 0.82 (m, 36H, SiꢀCCH ), 0.19 – ꢀ0.00 (m, 24H, SiꢀCH ).
C
3
3
NMR (126 MHz, CDCl ) δ 155.69, 155.53, 154.73, 154.56, 154.18 (5 COOC), 131.19, 130.76, 129.16,
3
128.66, 126.37, 125.55, 124.51, 97.86 (anomeric CHꢀ1″), 96.49 (anomeric CHꢀ1′), 85.75, 79.91, 79.38,
7
4
ꢀ
9.22, 79.08, 78.88, 75.29, 73.33, 72.66, 71.55, 68.00, 66.86, 63.12, 57.25, 50.64 (NHꢀCH CH ), 50.51,
2 2
8.91, 48.34, 45.79 (NHꢀCH ꢀanthracene), 41.67, 35.94, 35.67, 30.65, 30.04 (CH of linker), 29.62, 28.63
2
2
28.40 (OꢀCCH ), 27.50, 26.18 ꢀ 25.78 (SiꢀCCH ), 18.48, 18.32, 18.09, 17.91 (4 SiꢀCCH ), ꢀ3.42, ꢀ3.79, ꢀ
3
3
3
2
4
ACS Paragon Plus Environment

Page 25 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
.20, ꢀ4.88, ꢀ4.94, ꢀ5.07, ꢀ5.17, ꢀ5.22 (8 SiꢀCH ); ESIꢀMS: m/z calc′d for C H ClN O Si : 1778.0,
3
90 160
6
19
4
+
found: 1778.0 [M+H] .
ꢀOꢀ(12ꢀdodecylꢀ10ꢀaminomethylꢀ9ꢀchloroanthracene)ꢀ1,3,2′,6′,3′′ꢀpentaꢀNꢀ(tertꢀbutoxycarbonyl)ꢀ
5
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
4′,2′′,4′′,6′′ꢀtetraꢀOTBDMSꢀtobramycin (13c). Synthesized following general procedure B from 12c (116
mg, 0.072 mmol), aromatic aldehyde 5 (21 mg, 0.086 mmol) and sodium borohydride (8 mg, 0.216
mmol). The resulting residue was purified by flash column chromatography using CH Cl /MeOH (300:1
2
2
1
to 30:1, v/v) to afford the desired compound as a yellow solid (112 mg, 85 %). H NMR (500 MHz,
Chloroformꢀd) δ 8.61 – 8.51 (m, 2H, anthracene), 8.41 – 8.32 (m, 2H, anthracene), 7.65 – 7.51 (m, 4H,
anthracene), 5.25 – 5.12 (m, 2H, Hꢀ1′, Hꢀ1″), 4.70 (s, 2H, NHꢀCH ꢀanthracene), 4.32 – 4.22 (m, 1H), 4.21
2
– 4.13 (m, 1H), 4.11 – 4.04 (m, 1H), 3.82 – 3.19 (m, 15H), 2.86 (t, J = 7.3 Hz, 2H, NHꢀCH CH ), 2.51 –
2
2
2.43 (m, 1H), 2.04 – 1.98 (m, 1H), 1.61 – 1.55 (m, 2H, CH of linker), 1.55ꢀ1.51 (m, 1H), 1.50 – 1.36 (m,
2
4
7H), 1.34 – 1.22 (m, 16H), 1.11 ꢀ 1.02 (m, 1H), 0.97 – 0.82 (m, 36H, SiꢀCCH ), 0.19 – ꢀ0.02 (m, 24H,
3
13
SiꢀCH3). C NMR (126 MHz, Chloroformꢀd) δ 155.67, 155.51, 154.72, 154.55, 154.19 (5 COOC),
1
31.59, 130.71, 129.00, 128.65, 126.32, 126.25, 125.53, 124.51, 97.79 (anomeric CHꢀ1″), 96.54
anomeric CHꢀ1′), 85.71, 79.88, 79.35, 79.19, 79.04, 78.80, 75.28, 73.40, 72.68, 71.53, 68.01, 66.81,
63.09, 57.27, 50.67 (NHꢀCH CH ), 50.52, 48.95, 48.38, 45.90 (NHꢀCH ꢀanthracene), 41.65, 35.90, 35.65,
(
2
2
2
30.63, 30.13, 30.05, 29.72, 29.70, 29.66, 29.61, 28.63 ꢀ 28.40 (OꢀCCH ), 27.42, 26.19 ꢀ 25.78 (SiꢀCCH ),
3
3
1
8.47, 18.33, 18.09, 17.90 (4 SiꢀCCH ), ꢀ3.42, ꢀ3.80, ꢀ4.20, ꢀ4.88, ꢀ4.94, ꢀ5.07, ꢀ5.18, ꢀ5.23 (8 SiꢀCH3);
3
+
ESIꢀMS: m/z calc′d for C H ClN O Si : 1833.3, found: 1833.1 [M+H] .
94
168
6
19
4
Compound 14a. Synthesized following general procedure C from 13a (21 mg, 0.012 mmol), Bocꢀ
Lys(Boc)ꢀOH (6 mg, 0.018 mmol), DIPEA (5 mg, 0.036 mmol) and HBTU (7 mg, 0.018 mmol). After
column chromatography (eluted with CH Cl /MeOH from 300:1 to 30:1, v/v), the product was afforded
2
2
1
as a yellow solid (18.6 mg, 74 %). H NMR (500 MHz, Chloroformꢀd) δ 8.66 – 8.55 (m, 2H, anthracene),
1
2
8
.47 – 8.18 (m, 2H, anthracene), 7.71 – 7.54 (m, 4H, anthracene), 6.27 – 5.95 (m, 1H, NꢀCH H ꢀ
1
2
anthracene), 5.36 – 5.23 (m, 1H, NꢀCH H ꢀanthracene), 5.11 – 4.99 (m, 2H, Hꢀ1′, Hꢀ1″), 4.52 – 4.39 (m,
1
H, αꢀCH of Lys), 4.26 – 3.95 (m, 2H), 3.83 – 3.73 (m, 1H), 3.70 – 3.10 (m, 15H), 3.08 – 2.95 (m, 2H, εꢀ
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 62
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
CH of Lys), 2.95 – 2.68 (m, 2H, NꢀCH CH ), 2.55 – 2.35 (m, 1H), 2.01 – 1.91 (m, 1H), 1.54 – 1.52 (m,
2
2
2
2H, βꢀCH of Lys), 1.51 – 1.37 (m, 53H), 1.37 – 1.16 (m, 17H), 1.13 – 1.05 (m, 1H), 0.94 – 0.75 (m,
2
1
3
3
8H), 0.18 – ꢀ0.19 (m, 24H, SiꢀCH ). C NMR (126 MHz, Chloroformꢀd) δ 172.15 (NCOCH), 155.88,
3
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
155.80, 155.65, 155.21, 154.82, 154.49, 154.35 (7 COOC), 131.89, 131.77, 130.49, 128.54, 127.87,
127.32, 126.92, 126.65, 125.94, 125.79, 125.65, 124.67, 124.38, 123.70, 98.24 (anomeric CHꢀ1″), 96.55
(
anomeric CHꢀ1′), 86.13, 79.71, 79.38, 79.29, 79.24, 79.18, 78.92, 78.82, 75.56, 72.61, 72.30, 71.76,
7.59, 66.90, 63.15, 57.11, 50.53, 50.30 (αꢀCH of Lys), 49.27, 48.44, 45.45 (NꢀCH CH ), 41.47, 40.35 (εꢀ
6
2
2
CH of Lys), 39.55 (NꢀCH ꢀanthracene), 35.77, 35.57, 33.2 (βꢀCH of Lys), 29.67, 29.56, 28.72 ꢀ 28.30
2
2
2
(
OꢀCCH ), 27.34, 26.12 ꢀ 25.78 (SiꢀCCH ), 22.45, 18.38, 18.31, 18.08, 17.92 (4 SiꢀCCH ), ꢀ3.35, ꢀ3.74, ꢀ
3 3 3
4.25, ꢀ4.87, ꢀ5.03, ꢀ5.16, ꢀ5.22, ꢀ5.35 (8 SiꢀCH ); MALDIꢀTOFꢀMS: m/z calc′d for C H ClN O Si Na:
3
102 179
8
24
4
+
2071.173, found: 2071.203 [M+Na]
Compound 14b. Synthesized following general procedure C from 13b (28 mg, 0.016 mmol), Bocꢀ
Lys(Boc)ꢀOH (8 mg, 0.024 mmol), DIPEA (6 mg, 0.048 mmol) and HBTU (9 mg, 0.024 mmol). After
column chromatography (eluted with CH Cl /MeOH from 300:1 to 30:1, v/v), the product was afforded
2
2
1
as a yellow solid (28 mg, 85 %). H NMR (500 MHz, Chloroformꢀd) δ 8.66 – 8.56 (m, 2H, anthracene),
1
2
8.33 – 8.24 (m, 2H, anthracene), 7.71 – 7.50 (m, 4H, anthracene), 6.10 (d, J = 15.2 Hz, 1H, NꢀCH H ꢀ
1
2
anthracene), 5.33 – 5.27 (m, 1H, NꢀCH H ꢀanthracene), 5.25 – 5.19 (m, 1H, Hꢀ1′), 5.14 – 5.11 (m, 1H, Hꢀ
1″), 4.56 – 4.51 (m, 1H, αꢀCH of Lys), 4.30 – 4.21 (m, 1H), 4.20 – 4.12 (m, 1H), 4.10 – 4.02 (m, 1H),
1
2
3
.84 – 3.19 (m, 15H), 3.09 – 2.97 (m, 2H, εꢀCH of Lys), 2.96 – 2.86 (m, 1H, NꢀCH H CH ), 2.74 – 2.64
2
2
1
2
(
m, 1H, NꢀCH H CH ), 2.54 – 2.42 (m, 1H), 2.04 – 1.97 (m, 1H), 1.58 – 1.55 (m, 2H, βꢀCH of Lys),
2 2
1
.54 – 1.52 (m, 1H), 1.51 ꢀ 1.38 (m, 65H), 1.38 – 1.31 (m, 4H, γꢀCH of Lys, δꢀCH of Lys), 1.31 – 1.21
2 2
(m, 2H, CH of linker), 1.21 – 1.12 (m, 2H, CH of linker), 1.12 – 1.00 (m, 5H), 1.00 – 0.79 (m, 38H),
2
2
13
0.22 – ꢀ0.09 (m, 24H, SiꢀCH ). C NMR (126 MHz, Chloroformꢀd) δ 172.37 (NCOCH), 155.90, 155.74,
3
155.54, 155.42, 154.70, 154.56, 154.34 (7 COOC), 131.69, 130.49, 128.54, 127.70, 127.16, 126.76,
126.61, 125.79, 124.27, 123.92, 97.91 (anomeric CHꢀ1″), 96.44 (anomeric CHꢀ1′), 85.83, 79.91, 79.45,
79.38, 79.24, 79.18, 78.93, 78.78, 75.34, 73.23, 72.65, 71.57, 67.95, 66.87, 63.16, 57.24, 50.48, 50.31 (αꢀ
2
6
ACS Paragon Plus Environment

Page 27 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CH of Lys), 48.87, 48.32, 45.26 (NꢀCH CH ), 41.69, 40.28 (εꢀCH of Lys), 39.36 (NꢀCH ꢀanthracene),
2
2
2
2
35.92, 35.69, 33.69 (βꢀCH of Lys), 30.73, 30.03, 29.47, 29.28, 28.63 ꢀ 28.39 (OꢀCCH ), 26.95, 26.22 ꢀ
2 3
2
5.78 (SiꢀCCH ), 22.52, 18.47, 18.32, 18.10, 17.91 (4 SiꢀCCH ), ꢀ3.39, ꢀ3.77, ꢀ4.20, ꢀ4.87, ꢀ4.93, ꢀ5.08, ꢀ
3
3
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
5.16, ꢀ5.20 (8 SiꢀCH ); MALDIꢀTOFꢀMS: m/z calc′d for C H ClN O Si Na: 2127.206, found:
3 106 187 8 24 4
+
2127.219 [M+Na] .
Compound 14c. Synthesized following general procedure C from 13c (112 mg, 0.061 mmol), Bocꢀ
Lys(Boc)ꢀOH (31.9 mg, 0.092 mmol), DIPEA (24 mg, 0.183 mmol) and HBTU (35 mg, 0.092 mmol).
After column chromatography (eluted with CH Cl /MeOH from 300:1 to 30:1, v/v), the product was
2
2
1
afforded as a yellow solid (125 mg, 95 %). H NMR (500 MHz, Chloroformꢀd) δ 8.63 – 8.55 (m, 2H,
anthracene), 8.31 – 8.24 (m, 2H, anthracene), 7.64 – 7.53 (m, 4H, anthracene), 6.06 (d, J = 15.3 Hz, 1H,
1
2
1
2
NꢀCH H ꢀanthracene), 5.29 (d, J = 15.2 Hz, 1H, NꢀCH H ꢀanthracene), 5.24 – 5.19 (m, 1H, Hꢀ1′), 5.17 –
5
.12 (m, 1H, Hꢀ1″), 4.56 – 4.52 (m, 1H, αꢀCH of Lys), 4.31 – 4.21 (m, 1H), 4.20 – 4.12 (m, 1H), 4.11 –
4
.03 (m, 1H), 3.82 – 3.15 (m, 15H), 3.08 – 2.96 (m, 2H, εꢀCH of Lys), 2.95 – 2.85 (m, 1H, Nꢀ
2
1
2
1
2
CH H CH ), 2.76 – 2.67 (m, 1H, NꢀCH H CH ), 2.53 – 2.40 (m, 1H), 2.02 – 1.97 (m, 1H), 1.60 – 1.55
2
2
(
m, 2H, βꢀCH of Lys), 1.54 – 1.52 (m, 1H), 1.50 – 1.39 (m, 65H), 1.38 – 1.32 (m, 4H, γꢀCH of Lys, δꢀ
2 2
CH of Lys), 1.30 – 1.23 (m, 4H, CH of linker), 1.22 – 1.11 (m, 6H, CH of linker), 1.09 – 0.98 (m, 5H),
2
2
2
1
3
0.97 – 0.81 (m, 40H), 0.21 – ꢀ0.03 (m, 24H, SiꢀCH3). C NMR (126 MHz, Chloroformꢀd) δ 172.42
(
NCOCH), 155.89, 155.68, 155.66, 155.46, 154.72, 154.55, 154.43 (7 COOC), 131.67, 130.50, 128.52,
1
1
6
27.59, 127.14, 126.75, 126.57, 125.79, 124.23, 123.95, 97.82 (anomeric CHꢀ1″), 96.51 (anomeric CHꢀ
′), 85.74, 79.89, 79.76, 79.51, 79.36, 79.20, 78.94, 78.73, 75.29, 73.38, 72.68, 71.54, 68.03, 66.81,
3.09, 57.27, 50.51, 50.28 (αꢀCH of Lys), 48.94, 48.36, 45.28 (NꢀCH CH ), 41.68, 40.27 (εꢀCH of Lys),
2
2
2
39.52 (NꢀCH ꢀanthracene), 35.93, 35.64, 33.75 (βꢀCH of Lys), 30.68, 30.13, 29.75, 29.66, 29.53, 29.41,
2
2
29.26, 28.95, 28.62 ꢀ 28.37 (OꢀCCH ), 26.60, 26.25 ꢀ 25.77 (SiꢀCCH ), 22.55, 18.47, 18.32, 18.09, 17.90
3
3
(
4 SiꢀCCH ), ꢀ3.42, ꢀ3.80, ꢀ4.21, ꢀ4.89, ꢀ4.95, ꢀ5.08, ꢀ5.19, ꢀ5.23 (8 SiꢀCH ); MALDIꢀTOFꢀMS: m/z calc′d
3 3
+
for C H ClN O Si Na: 2183.298, found: 2183.302 [M+Na] .
110 195
8
24
4
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 62
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
Biological Activity Assays. Bacterial Isolates. Bacterial isolates were obtained as part of the
61
Canadian National Intensive Care Unit (CANꢀICU) study and Canadian Ward Surveillance
62, 63
(
CANWARD) studies
. The CANꢀICU study included 19 medical centres across Canada with active
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
ICUs. From September 2005 to June 2006, 4180 isolates represented in 2580 ICU patients were recovered
from clinical specimens including blood, urine, wound/tissue, and respiratory specimens (one pathogen
per cultured site per patient). Only “clinically significant” specimens from patients with a presumed
infectious disease were collected. The isolates obtained were shipped to the reference laboratory (Health
Sciences Centre, Winnipeg, Canada) on Amies charcoal swabs. Then isolates were subꢀcultured onto
appropriate medium and stocked in skim milk at ꢀ80 °C until subsequent MIC testing was carried out. The
quality control strains including Staphylococcus aureus ATCC 29213, methicillinꢀresistant S. aureus
(MRSA) ATCC 33592, Enterococcus faecalis ATCC 29212, Enterococcus faecium ATCC 27270,
Streptococcus pneumoniae ATCC 49619, Escherichia coli ATCC 25922, Pseudomonas aeruginosa
ATCC 27853, and Klebsiella pneumoniae ATCC 13883 were acquired from the American Type Culture
Collection (ATCC). The clinical strains, including methicillinꢀresistant Staphylococcus epidermidis
(
MRSE), CANꢀICU 61589 (cefazolin MIC > 32 ꢁg/mL), gentamicin resistant E. coli CANꢀICU 61714,
Amikacinꢀresistant E. coli CANꢀICU 63074 (MIC = 32 ꢁg/mL), gentamicin resistant P. aeruginosa CANꢀ
ICU 62584, Strenotrophomonas maltophilia CANꢀICU 62584, and Acinetobacter baumannii CANꢀICU
6
3169 were obtained from hospitals across Canada as a part of the CANꢀICU study. Methicillinꢀ
susceptible S. epidermidis (MSSE) CANWARDꢀ2008 81388 was obtained from the 2008 CANWARD
study, while gentamicinꢀresistant tobramycinꢀresistant ciprofloxacinꢀresistant [aminoglycoside modifying
enzyme aac(3)ꢀIIa present] E. coli CANWARDꢀ2011 97615, and gentamicinꢀresistant tobramycinꢀ
resistant P. aeruginosa CANWARDꢀ2011 96846 were obtained from the 2011 CANWARD study. In
addition, P. aeruginosa PAO1, P. aeruginosa P259ꢀ96918, P. aeruginosa P262ꢀ101856, P. aeruginosa
P264ꢀ104354, colistinꢀresistant P. aeruginosa 91433, colistinꢀresistant P. aeruginosa 101243, A.
baumannii AB027, A. baumannii AB030, A. baumannii AB031, A. baumannii 110193, Enterobacter
cloacae 117029, and Klebsiella pneumonia 116381 were kindly provided by Dr. George G. Zhanel. The
2
8
ACS Paragon Plus Environment

Page 29 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
efflux pumpꢀmutated strains, P. aeruginosa PAO200 and P. aeruginosa PAO750, were provided by Dr.
Ayush Kumar from University of Manitoba in Canada.
Multiꢀdrug resistance in P. aeruginosa was defined as concomitant resistance to 3 or more chemically
unrelated antimicrobial classes, while extensively drug resistant was defined as concomitant resistance to
5 or more chemically unrelated antimicrobial classes.
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
Antimicrobial Activity Assay. The antimicrobial activity of the compounds against a panel of bacteria
was evaluated by microliter dilution method in accordance with the Clinical and Laboratory Standards
Institute (CLSI) guidelines. The overnight bacterial culture was diluted in saline to 0.5 McFarland
turbidity, and then 1:50 diluted in MuellerꢀHinton broth (MHB) for inoculation. The antimicrobial agents
were 2ꢀfold serially diluted in MHB in 96ꢀwell plate and incubated with equal volumes of inoculum for
18 h at 37 °C. The lowest concentration that prevented visible bacterial growth was taken as the MIC for
each antimicrobial agent. The broth with or without bacterial cells was employed as positive or negative
controls, respectively.
Combination Studies with Different Antibiotics. FIC index was determined by setting up standard
6
4
checkerboard assay in 96ꢀwell plate as previously described. Each antibiotic to be tested was serially
diluted along the abscissa in MHB, while adjuvant was diluted along the ordinate to create a 10 × 7
matrix. The bacterial culture was prepared in MHB by 1:50 dilution from the 0.5 McFarland turbidity
culture in saline. The inoculum was added to each well of the plate and incubated for 18 h at 37 °C. After
®
the incubation, plates were read on EMax Plus microplate reader (Molecular Devices, Sunnyvale, CA,
USA). MIC was recorded as wells with the lowest concentration of drugs with no bacterial growth. The
FIC for each antibiotic was calculated as the concentration of the antibiotic for a well showing no growth
in the presence of adjuvant divided by the MIC for that antibiotic alone. The FIC for each adjuvant was
calculated as the concentration of the adjuvant for a well showing no growth in the presence of antibiotic
divided by the MIC for that adjuvant alone. The FIC index is the sum of the two FICs. Chemicalꢀ
2
9
ACS Paragon Plus Environment