Matej Zore1, Shella Gilbert-Girard2, Inés Reigada2, Jayendra Z Patel1, Kirsi Savijoki2, Adyary Fallarero2, Jari Yli-Kauhaluoma1. 1. Drug Research Program, Division of Pharmaceutical Chemistry and Technology, Faculty of Pharmacy, University of Helsinki, Viikinkaari 5 E, FI-00014 Helsinki, Finland. 2. Drug Research Program, Division of Pharmaceutical Biosciences, Faculty of Pharmacy, University of Helsinki, Viikinkaari 5 E, FI-00014 Helsinki, Finland.
Abstract
We recently identified fingolimod as a potent antibiofilm compound by screening FDA-approved drugs. To study if the antibacterial activity of fingolimod could be further improved and to explore in-depth structure-activity relationships, we synthesized 28 novel fingolimod derivatives and evaluated their efficacy against Staphylococcus aureus grown in planktonic/single cell and biofilms. The most effective derivatives were tested on preformed S. aureus biofilms and against Gram-negative bacteria Acinetobacter baumannii and Pseudomonas aeruginosa, using fingolimod as the reference compound. Seven derivatives were more effective against S. aureus, while five other derivatives showed improved activity against P. aeruginosa and/or A. baumannii, with no apparent change in cytotoxicity on human cells. The most interesting derivatives, compounds 43 and 55, displayed a broader spectrum of antibacterial activity, possibly exerted by the change of the para-hydrocarbon chain to a meta position for 43 and by an additional hydroxyl group for 55.
We recently identified fingolimod as a potent antibiofilm compound by screening FDA-approved drugs. To study if the antibacterial activity of fingolimod could be further improved and to explore in-depth structure-activity relationships, we synthesized 28 novel fingolimod derivatives and evaluated their efficacy against Staphylococcus aureus grown in planktonic/single cell and biofilms. The most effective derivatives were tested on preformed S. aureus biofilms and against Gram-negative bacteria Acinetobacter baumannii and Pseudomonas aeruginosa, using fingolimod as the reference compound. Seven derivatives were more effective against S. aureus, while five other derivatives showed improved activity against P. aeruginosa and/or A. baumannii, with no apparent change in cytotoxicity on human cells. The most interesting derivatives, compounds 43 and 55, displayed a broader spectrum of antibacterial activity, possibly exerted by the change of the para-hydrocarbon chain to a meta position for 43 and by an additional hydroxyl group for 55.
Bacterial infections
are one of the most important challenges of
modern medicine, as the emergence of multidrug-resistant strains increasingly
reduces the options of antibiotic treatment.[1] In particular, biofilm-forming organisms are responsible for an
important number of recalcitrant infections.[2,3] Biofilms,
defined as multicellular communities protected by a self-produced
matrix, are the most common phenotype of bacteria and are well known
for their high tolerance to antibiotic treatments.[4] They are particularly prevalent in device-related infections,
which are increasing in numbers alongside the augmenting use of medical
devices, and frequently cause the removal of the device.[5,6] For these reasons, bacterial biofilms represent a large burden on
the health care system and lead to significant morbidity and considerable
costs.Great efforts have been made in recent years to develop
new treatment
strategies and identify new antimicrobial compounds among natural
and synthetic products as well as approved drugs. Sphingosine (Figure ), a sphingoid base
naturally found in mammalian cells, has a well-reported antimicrobial
activity. This molecule plays a role in the immune protection of the
skin and the airways, with its depletion contributing to infection.[7−9] Sphingosine, along with other sphingolipids, has been tested against
various bacterial species and its potential as an antibacterial treatment
has been under evaluation in recent years.[10−13] Treatment with sphingosine by
inhalation or as a coating on endotracheal catheters has demonstrated
a protective effect against Pseudomonas aeruginosainfection in vivo.[14,15]
Figure 1
Structures
of sphingosine and fingolimod.
Structures
of sphingosine and fingolimod.Repurposing drugs as antimicrobials is an approach with many advantages,
namely, that a large amount of information about the drug’s
activity, pharmacokinetics, and safety is already available, facilitating
its development for a new purpose.[16] Recently,
we identified fingolimod as an antibacterial compound in a screening
of a library of FDA-approved drugs.[17] Fingolimod
is a sphingosine structural analogue that was initially synthesized
from the natural fungal product myriocin.[18] It was approved for the treatment of relapsing multiple sclerosis
in view of its immunomodulatory activity that causes a redistribution
of the lymphocyte population.[19] Fingolimod’s
antibacterial activity has been reported against a few bacterial species
and, like sphingosine, it has shown a protective effect against P. aeruginosain vivo.[14,17,20] As fingolimod and structurally
similar compounds have strong activity against different bacterial
species, we set out to synthesize fingolimod derivatives and evaluate
their antibacterial effects, with a focus on their activity against
biofilms. Fingolimod has been the focus of optimization efforts as
a potential anticancer drug,[21,22] but to our knowledge,
this is the first report of the synthesis and evaluation of fingolimod
derivatives as antibacterial compounds. The derivatives were first
tested against Staphylococcus aureus grown in planktonic and biofilm states, and then the active compounds
were further tested against preformed S. aureus biofilms and other pathogenic and clinically prevalent microorganisms
such as Acinetobacter baumannii and P. aeruginosa. We studied the cytotoxicity effects
of the most promising derivatives as well as their ability to interfere
with the bacterial central cell-to-cell communication system (quorum
sensing), a key mediator of biofilm formation and antibacterial resistance.
Finally, we evaluated the effect of fingolimod on S.
aureus biofilms formed in the presence of human neutrophils
to emulate the in vivo conditions of an infection
following the implantation of a medical device.
Results and Discussion
Chemistry
We designed and synthesized a library of
28 fingolimod derivatives by chemical modification of the aromatic
ring, the eight-carbon chain, and the 2-amino-1,3-diol polar head.
For the first set of derivatives (10–16, 25–29, 33), we focused
mainly on the modification of the hydrocarbon chain and left the rest
of the structure intact. We replaced the chain with additional aromatic
structures bearing various substituents. The exceptions in this set
are compounds 10 and 11, which compared
to fingolimod have shorter and longer carbon chains, respectively.
Based on the screening of the first set of 13 derivatives, we designed
a second set (42–45, 48–49, 54–55, 61–62, 79–83), and in this step, we focused on the modifications of the aromatic
ring and the ethylene linker connecting the aromatic ring and polar
head. Again, we left the amino diol moiety intact. All compounds in
this set have a hydrocarbon chain, consisting of either eight or ten
carbons.The four to six-step synthesis of all derivatives is
presented in Schemes –. A crucial step in the synthesis was the nucleophilic substitution
of halide or mesylate with diethyl acetamidomalonate in the presence
of cesium carbonate to obtain para-bromo N-acetylated
diester intermediates 2, 36–37, 51, 58, and 70–73, or N-acetylated diesterphenol intermediate 20.[23] Bromo intermediates were later used
for microwave-assisted Suzuki coupling with an appropriate boronic
acid and palladium catalyst (Schemes and ), while a phenol intermediate was used for O-alkylation with appropriate
alkyl bromide in the presence of potassium carbonate (Scheme ). The last step in the synthesis
of all derivatives was a reduction of the corresponding diester with
sodium borohydride and calcium chloride, or lithium aluminum hydride
to obtain N-acetylated diol intermediates, which were further hydrolyzed
with lithium hydroxide (aq.) to give the desired compounds.[24]
Scheme 1
Synthesis of Fingolimod Derivatives 10–16
and 42–45
Reagents and conditions: (i)
diethyl acetamidomalonate, Cs2CO3, MeCN, MW,
130 °C, 1 h; (ii) R3-B(OH)2 or R4-B(OH)2, Pd2(dba)3, SPhos, K2CO3, toluene/water (4:1), MW, 110 °C, 1.5–2
h; (iii) NaBH4, CaCl2, EtOH/water (4:1), rt,
overnight; and (iv) 2 M LiOH (aq.), MeOH/tetrahydrofuran (THF) (1:1),
reflux, 5 h.
Scheme 3
Synthesis of Derivatives 54–55 (A),
Pyridine Derivatives 61–62 (B), and
Quinoline and Naphthalene Derivatives 33 and 79–83 (C)
Reagents and conditions: (i)
diethyl acetamidomalonate, Cs2CO3, MeCN, or
DMF, MW, 130 °C, 1 h; (ii) n-octylboronic acid
or n-decylboronic acid, Pd2(dba)3, SPhos, K2CO3, toluene/water (4:1), MW, 110
°C, 1.5–8 h; (iii) NaBH4, CaCl2,
EtOH/water (4:1), rt, overnight or 1 M LiAlH4, THF, 0 °C
to rt, 2 h (for 80 and 81); and (iv) 2 M
LiOH (aq.), MeOH/THF (1:1), reflux, 5 h.
Scheme 2
Synthesis of Phenyl Ether Derivatives 25–29, 48, and 49
Reagents and conditions: (i)
MsCl, Et3N, dichloromethane (DCM), 0 °C, 2 h; (ii)
diethyl acetamidomalonate, Cs2CO3, MeCN, MW,
130 °C, 1 h; (iii) H2, 10% Pd/C, EtOH/THF (1:1), rt,
5 h; (iv) R-Br, K2CO3, N,N–dimethylformamide (DMF), 0 °C to rt, overnight;
(v) NaBH4, CaCl2, EtOH/water (4:1), rt, overnight;
(vi) 2 M LiBH4, THF, rt, overnight (for 25 and 27); and (vii) 2 M LiOH (aq.), MeOH/THF (1:1),
reflux, 5 h.
Synthesis of Fingolimod Derivatives 10–16
and 42–45
Reagents and conditions: (i)
diethyl acetamidomalonate, Cs2CO3, MeCN, MW,
130 °C, 1 h; (ii) R3-B(OH)2 or R4-B(OH)2, Pd2(dba)3, SPhos, K2CO3, toluene/water (4:1), MW, 110 °C, 1.5–2
h; (iii) NaBH4, CaCl2, EtOH/water (4:1), rt,
overnight; and (iv) 2 M LiOH (aq.), MeOH/tetrahydrofuran (THF) (1:1),
reflux, 5 h.
Synthesis of Phenyl Ether Derivatives 25–29, 48, and 49
Reagents and conditions: (i)
MsCl, Et3N, dichloromethane (DCM), 0 °C, 2 h; (ii)
diethyl acetamidomalonate, Cs2CO3, MeCN, MW,
130 °C, 1 h; (iii) H2, 10% Pd/C, EtOH/THF (1:1), rt,
5 h; (iv) R-Br, K2CO3, N,N–dimethylformamide (DMF), 0 °C to rt, overnight;
(v) NaBH4, CaCl2, EtOH/water (4:1), rt, overnight;
(vi) 2 M LiBH4, THF, rt, overnight (for 25 and 27); and (vii) 2 M LiOH (aq.), MeOH/THF (1:1),
reflux, 5 h.
Synthesis of Derivatives 54–55 (A),
Pyridine Derivatives 61–62 (B), and
Quinoline and Naphthalene Derivatives 33 and 79–83 (C)
Reagents and conditions: (i)
diethyl acetamidomalonate, Cs2CO3, MeCN, or
DMF, MW, 130 °C, 1 h; (ii) n-octylboronic acid
or n-decylboronic acid, Pd2(dba)3, SPhos, K2CO3, toluene/water (4:1), MW, 110
°C, 1.5–8 h; (iii) NaBH4, CaCl2,
EtOH/water (4:1), rt, overnight or 1 M LiAlH4, THF, 0 °C
to rt, 2 h (for 80 and 81); and (iv) 2 M
LiOH (aq.), MeOH/THF (1:1), reflux, 5 h.
Screening
of the Derivatives for Their Antibacterial Activity
against S. aureus
The antibacterial
activity of the synthesized compounds was first evaluated against S. aureus at a concentration of 50 μM (Supporting
Information, Figure S1). Compounds that
inhibited both the planktonic cells (turbidity and viability) and
the biofilm formation (viability and biomass) by at least 60% were
considered active. From the screening of the first set of 13 compounds
(Figure S1A), we identified only two active
compounds (11 and 15), which were then selected
for further characterization. Compound 11 had the strongest
inhibitory activity and was also the most structurally similar to
fingolimod, with only two carbons added to its hydrocarbon chain.
As opposed to the other derivatives with two aromatic rings, compounds 15 and 29 both possessed a short hydrocarbon
chain. Yet, only 15 displayed high enough activity to
be considered active. These results demonstrated the importance of
the hydrocarbon chain for the antibacterial and antibiofilm activity,
and its replacement led to decreased or lost inhibitory activity even
at the concentration of 50 μM. In addition, the length of the
alkyl chain proved important. Compound 10 had a shorter,
six-carbon chain and possessed only limited activity against S. aureus biofilms. As the only derivatives with
a measurable antibacterial activity were the ones with a hydrocarbon
chain, and since fingolimod and 11, the two compounds
with the longest chains, displayed the highest inhibitory activity,
the second step of optimization focused on derivative structures with
this feature.The antibacterial activity of the second set of
derivatives was also first studied against S. aureus at 50 μM (Figure S2B). Among these,
only pyridine derivative 61 did not show an inhibitory
activity comparable to fingolimod and was therefore discarded. This
finding was intriguing, as 62 had an identical structure,
with a longer hydrocarbon chain, and yet displayed a higher activity.
Including the two derivatives selected from the first screening (11 and 15), a total of 16 derivatives active
against S. aureus were selected for
further characterization.
Concentration–Response of the Active
Derivatives
To determine the minimal inhibitory concentration
(MIC), the active
derivatives were tested at different concentrations against S. aureus ATCC 25923. Here, the MIC is defined as
the lowest concentration where no growth was visible (resulting in
over 90% inhibition of turbidity and viability of the planktonic cells)
after 18 h of incubation. In addition, the compounds were tested in
post-exposure for their capacity to affect a preformed biofilm (compounds
added to 18-h-old biofilms and incubated for 24 h). Fingolimod was
previously found to have an MIC of 15 μM in pre-exposure with
some activity also against preformed biofilms.[17] The same combination of four measurements used for the
initial screening rounds (planktonic turbidity, planktonic viability,
biofilm viability, and biofilm total biomass) was used in both pre-
and post-exposure studies. Figure shows the inhibition of the viability of S. aureus biofilms by the derivatives in either pre-exposure
at 5–25 μM or post-exposure at 25–200 μM.
The viability of the biofilms is shown as a representative parameter,
and the other three measurements lead to similar conclusions regarding
the activity of the derivatives. The inhibition percentages and standard
deviation (SD) obtained for all measurements (viability and turbidity
of the planktonic cells, viability, and biomass of the biofilms) are
available as the Supporting Information (Table S1). In pre-exposure (Figure A), seven derivatives displayed higher inhibitory activity
than fingolimod. Four derivatives, 45, 79, 80, and 83, had an MIC of 10 μM,
and for some replicates, the MIC of 45 and 55 was as low as 5 μM. The derivatives 11 and 81 displayed an MIC varying from 10 to 15 μM (depending
on the replicate), and 62 and 82 had the
same MIC as fingolimod. Interestingly, wide variations in the antibacterial
activity were observed between very similar structures, and in some
cases, small structural changes led to a very drastic change of activity.
Figure 2
Inhibition
of the viability of S. aureus ATCC
25923 biofilms in pre-exposure (A) and post-exposure (B) after
exposure to fingolimod (Fgd) derivatives. The results are expressed
as the inhibition percentage ± SD in comparison to the untreated
control. The experiment was repeated three times, each with two replicates
per concentration.
Inhibition
of the viability of S. aureus ATCC
25923 biofilms in pre-exposure (A) and post-exposure (B) after
exposure to fingolimod (Fgd) derivatives. The results are expressed
as the inhibition percentage ± SD in comparison to the untreated
control. The experiment was repeated three times, each with two replicates
per concentration.Generally, a longer hydrocarbon
chain seemed beneficial as all
derivatives with a 10-carbon chain (11, 45, 49, 55, 62, 81, and 83) had an MIC either equal or lower than fingolimod,
except for 49. As an example, derivatives 44 and 45, which structurally differ only in the length
of the chain, had an MIC of 15–25 and 5–10 μM,
respectively. However, this did not apply to all comparable structures,
as 80 and 81 had very similar activity and
the compound with the shorter chain was slightly more potent. Compound 81, like fingolimod, was not fully soluble in dimethyl sulfoxide
(DMSO), which is likely to decrease its activity. Replacing the phenyl
ring in fingolimod with quinoline or naphthalene seemed to either
improve the activity or leave it unchanged. All derivatives with bicyclic
aromatic rings performed the same as fingolimod (81 and 82) or better (79, 80, and 83). On the other hand, replacing the phenyl ring with pyridine
negatively affected the inhibitory activity of derivative 61. However, the longer chain of compound 62 compensated
for this modification since the MIC was similar to fingolimod. A shorter
carbon linker between the aromatic ring and the polar head did not
seem to positively affect the activity in itself (44 has
lower activity than fingolimod), while not necessarily being correlated
with a lower activity as some of the most potent derivatives harbored
this modification (45, 79, and 83). Addition of another hydroxyl group next to the polar head decreased
the inhibitory activity (54), although a longer hydrocarbon
chain seems to compensate for this modification (55).
Phenyl ether derivatives 48 and 49 had lower
activity, indicating that the oxygen on the para position negatively
affects the antibacterial activity. The meta-isomer
(43) had slightly lower activity (MIC of 15–25
μM), suggesting that the position of the hydrocarbon chain plays
a minor role against S. aureus. A combination
of different structural changes in some derivatives might also affect
the antibacterial activity in a more complex way that would be difficult
to directly correlate to one or another change. In the case of 82 and 83, several modifications were made compared
to fingolimod; yet, the activity remained unchanged or even improved
(83).Very different results were obtained in post-exposure
(Figure B), as the
derivatives 44, 54, and 82 were
the strongest
inhibitors, affecting both the viability (Figure B) and total biomass (Table S1) of preformed biofilms by at least 70% at 100 μM.
Considering that preformed biofilms display a greater tolerance to
chemical agents, such a reduction is not negligible. Other derivatives
showed activity comparable to fingolimod (43, 48, 55, 62, 80, and 83) and caused an important inhibition of the preformed biofilms. As
opposed to pre-exposure, a longer hydrocarbon chain did not improve
the post-exposure activity. On the contrary, many derivatives with
this feature that showed potent activity in pre-exposure had an important
loss of activity in post-exposure. In addition, when comparing similar
compounds with shorter and longer chains, the compounds with the shorter
chain always performed better (e.g., fingolimod vs 11, 44 vs 45, etc.). A shorter linker to
the polar head also appeared a positive change (fingolimod vs 44), while a bicyclic aromatic ring did not seem to change
the post-exposure activity. The addition of a hydroxyl group (54, 55) also appeared to positively influence
the activity. Interestingly, the derivatives that performed best in
pre-exposure were quite different from those that did best in post-exposure,
underlining how each setting’s conditions are greatly important
in what type of structures will prove efficient.
Activity of
the Active Derivatives on Other Bacterial Strains
To verify
if the derivatives that performed best against S. aureus ATCC 25923 could also show activity against
other clinical strains, we tested six of the most effective derivatives
(45, 55, 79, 80, 82, and 83) against S.
aureus ATCC 12598 (pathogenic septic arthritis isolate)
and P2 (penicillin-resistant hip prosthetic implant isolate) at 5–10–15–25
μM as described above for planktonic and biofilm cells. The
MIC values obtained for these derivatives against each S. aureus strain are shown in Table . All of the inhibition results against the
clinical strains of S. aureus are available
as the Supporting Information in Table S2. All six derivatives and fingolimod showed a similar and even increased
activity in some cases against the clinical strains compared to S. aureus ATCC 25923. The derivatives 45 and 79 were the most potent, with an MIC between 5
and 10 μM, which exceeds MIC detected with the parent compound.
Table 1
Minimum Inhibitory Concentration (MIC)
of the 16 Selected Active Fingolimod Derivatives against all Strains
Testedd
The MIC is defined
as the lowest
concentration where no growth was visible (>90% inhibition of turbidity
and viability of the planktonic cells). The experiment was repeated
three times with two replicates per concentration.
When some inhibitory activity was
observed, but no MIC was reached, the MIC is indicated to be over
the highest concentration tested.
A dash indicates that the compound
was not tested on the strain.
“Inactive” indicates
that no activity was observed up to the highest concentration tested
(50 μM with A. baumannii and
200 μM with P. aeruginosa).
The MIC is defined
as the lowest
concentration where no growth was visible (>90% inhibition of turbidity
and viability of the planktonic cells). The experiment was repeated
three times with two replicates per concentration.When some inhibitory activity was
observed, but no MIC was reached, the MIC is indicated to be over
the highest concentration tested.A dash indicates that the compound
was not tested on the strain.“Inactive” indicates
that no activity was observed up to the highest concentration tested
(50 μM with A. baumannii and
200 μM with P. aeruginosa).In addition, we previously
reported that fingolimod shows strong
activity against A. baumannii (MIC
25 μM) and modest activity against P. aeruginosa.[17] These bacteria belong to the ESKAPE
pathogens (Enterococcus faecium, S. aureus, Klebsiella pneumoniae, A. baumannii, P.
aeruginosa, and Enterobacter species),
causing a majority of nosocomial infections and having a great ability
to evade treatment.[25] Here, we tested the
16 derivatives selected from the initial screening against A. baumannii at 10–15–25–50
μM and against P. aeruginosa at
50–100–150–200 μM using the same protocol
as above with necessary modifications, taking into account the Gram-negative
nature of the tested species. A. baumannii NCTC 13423 is a multidrug-resistant strain[26] and P. aeruginosa ATCC 9027 is a
nonvirulent antibiotic-sensitive clinical strain.[27] Fingolimod had previously shown the strongest activity
against this strain out of a few P. aeruginosa strains.[17] Derivatives were tested in
pre-exposure, and Table shows the MIC measured for each derivative. Where no MIC could be
obtained, the results for the highest concentration tested are shown.
All inhibition results are available as the Supporting Information
in Table S3. Four derivatives (43, 55, 62, and 82) displayed
a clearly increased activity with an MIC of 15 μM against A. baumannii. The derivatives that were found the
most effective against A. baumannii were not correlated to those detected as most effective against S. aureus in pre-exposure. The derivatives 55, 80, and 82 performed well on
both strains, whereas 43 and 62 displayed
improved activity only against A. baumannii. On the other hand, 11, 45, and 79, compounds with high efficacy against S.
aureus, were inactive at concentrations up to 50 μM
against A. baumannii. Interestingly, 43 and 82, which have the hydrocarbon chain at
a different position compared to fingolimod, showed an improved activity.
While a longer chain seemed beneficial against S. aureus, this was not necessarily the case against A. baumannii, and only two compounds with a 10-carbon chain (55 and 62) showed an improved activity. Other structural changes
either did not affect the activity or did so negatively.The
activity of the derivatives against P. aeruginosa was also different from what was observed against S. aureus and A. baumannii. Two derivatives, 43 and 54, displayed
a clearly improved activity with an MIC of 50–100 μM,
while fingolimod was not able to completely inhibit the bacteria at
a concentration up to 200 μM. The reported activity of fingolimod
and sphingosine against P. aeruginosa has been strongly variable depending on the strains tested and the
experimental conditions,[10,14,17] but as inhalation of fingolimod itself was shown to rescue susceptible
mice from a P. aeruginosa lung infection,
the derivatives 43 and 54 might be interesting
compounds for similar use.[14] For this species,
a longer hydrocarbon chain has systemically led to a lower activity,
as exemplified by 11, 45, and 55, in comparison with fingolimod, 44 and 54, respectively. A bicyclic aromatic ring has similarly reduced the
activity. As observed with A. baumannii, meta-isomer (43) also performed better
against P. aeruginosa, suggesting that
meta-substitution is preferable against Gram-negative species. The
addition of a hydroxyl group next to the polar head (54) also improved the inhibitory activity against P.
aeruginosa, a modification that also improved the
activity against S. aureus in post-exposure.
Other modifications did not seem to visibly affect the activity of
the compounds.When comparing the activity of the derivatives
against different
species, very different conclusions can be made between the Gram-positive
and Gram-negative species. Compounds with a longer hydrocarbon chain,
bicyclic aromatic ring, and/or shorter carbon linker on the right
side seem to increase the activity against S. aureus. Changing the position of the hydrocarbon chain or an additional
hydroxyl group next to the polar head seems to broaden the spectrum
of activity, as some of the derivatives bearing these modifications
have a good activity also against Gram-negative species. The derivative 82 displayed similar activity compared to fingolimod against S. aureus in pre-exposure and better activity against S. aureus in post-exposure and against A. baumannii. Also, 43, while having
good activity against S. aureus in
pre-exposure, was among the most active against the Gram-negative
strains.In a recent work, the mode of action of sphingosine
as an antibacterial
compound against S. aureus and P. aeruginosa was suggested to involve an interaction
between sphingosine’s protonated amino group and a negatively
charged protein, cardiolipin, in the bacterial membrane, causing clustering
of the protein and thus compromising membrane’s permeability.[13] The presence of the amino group in the structure
was proven necessary for the antibacterial activity of sphingosine.
Considering the structural similarity between fingolimod and sphingosine,
it is likely that fingolimod uses the same mechanism, although our
results also suggest that other parts of the structure, particularly
the long hydrocarbon chain, are important for these molecules’
antibacterial activity.
Activity of Selected Derivatives on Quorum
Sensing
Quorum sensing (QS) is a community-based communication
system that
is activated when bacteria reach a high enough density, at which point
QS induces community behaviors, such as the formation of a biofilm.[28] Inhibiting QS is therefore a potential antivirulence
approach to prevent biofilm formation and keep bacteria in their more
susceptible planktonic state.[29] The QS
system molecules vary from species to species, and in Gram-negative
bacteria, they are generally acyl-homoserine lactones (AHLs).[28]Chromobacterium violaceum can be used as a reporter for QS activation as its QS induces the
production of a violet pigment, violacein.[30] We previously reported that fingolimod could inhibit QS in C. violaceum.[17] To test
whether some of the derivatives here would show an improved QS inhibitory
activity, we tested selected derivatives that performed well on various
strains using C. violaceum as previously
described.[17,31]The derivatives were tested
against both the wild-type strain (ATCC 31532) and the CV026 mutant
strain that needs the addition of AHL to activate QS.[30] The use of both strains allows indicating if the QS inhibition
takes place upstream or downstream of the AHL synthesis, for instance,
in the case of quorum quenchers via binding to the AHL outside of
the cells or by affecting the AHL diffusion through the cell wall/membrane.
The viability of the bacteria was measured in parallel with a resazurin
staining to assess if the inhibitory activity resulted from genuine
anti-QS effects rather than bactericidal activity (Supporting information, Figure S2). Quercetin was used as a control for
QS inhibition (QSI), as its activity has been reported previously,[32] and azithromycin was used as a bactericidal
control. Figure shows
the inhibitory effects of the selected derivatives and the controls
on both the wild-type and the mutant strains. None of the derivatives
displayed a bactericidal effect on either strain of C. violaceum (Figure S2). On the other hand, they promoted strong anti-QS effects with at
least 60% inhibition at 15 μM on both strains. This indicates
that the QS signal is inhibited downstream of the release of the AHL,
potentially by quenching the signal or by competition with the AHL’s
receptor. The inhibitory profile of the derivatives did not differ
significantly from that detected with fingolimod. Intriguingly, 82 had a significantly better QSI activity against the mutant
strain (p = 0.006). Derivative 79 was
found to be an effective QSI, but displayed significantly lower QSI
activity than fingolimod against the mutant strain at 10 μM (p = 0.021) and the wild-type strain at 10 μM (p < 0.001), 15 μM (p = 0.018),
and 25 μM (p < 0.001). Although having a
seemingly strong QSI activity against the wild-type strain at 10 μM, 62 performed significantly worse than fingolimod at 15 and
25 μM (p = 0.016 and 0.036, respectively).
Since the QSI activity profiles of each derivative are highly similar,
clear conclusions as to what structural changes benefit the QSI activity
are hard to make.
Figure 3
Quorum sensing inhibitory activity of selected fingolimod
(Fgd)
derivatives at various concentrations against (A) C.
violaceum ATCC 31532 (wild type) and (B) C. violaceum CV026 (AHL-deficient mutant). The results
are expressed as the inhibition percentage ± SD as compared to
the untreated control. The experiment was repeated twice with two
replicates per concentration. The stars indicate the significance
of the difference between the derivatives and fingolimod, *: p < 0.05, **: p < 0.01, ***: p < 0.001.
Quorum sensing inhibitory activity of selected fingolimod
(Fgd)
derivatives at various concentrations against (A) C.
violaceum ATCC 31532 (wild type) and (B) C. violaceum CV026 (AHL-deficient mutant). The results
are expressed as the inhibition percentage ± SD as compared to
the untreated control. The experiment was repeated twice with two
replicates per concentration. The stars indicate the significance
of the difference between the derivatives and fingolimod, *: p < 0.05, **: p < 0.01, ***: p < 0.001.
Cytotoxicity of Selected
Derivatives
Fingolimod has
been reported to have cytotoxic effects on various cancer cell lines
with a CC50 value in the range of 5–20 μM.[33,34] To see if some of the structural changes made on the derivatives
could ameliorate this, the same six derivatives selected for the QSI
experiment were also tested on three cell lines, A549, HEp-2, and
HL-60. Table shows
the CC50 values obtained for each cell line after exposure
to the compounds. In contrast with previous reports, fingolimod had
a CC50 value above or close to the highest concentration
tested (40 μM), possibly due to different incubation times and
poor solubility. None of the derivatives displayed reduced cytotoxicity
in comparison to fingolimod, but 43 and 79 both displayed a lesser cytotoxic activity than the other derivatives
and were quite similar to fingolimod. All compounds, including fingolimod,
had CC50 at a higher concentration than their MIC against S. aureus and A. baumannii, offering a potential therapeutic window for antibacterial treatment.
While many derivatives have shown an increased antibacterial activity,
fingolimod also remains an interesting potential antibacterial compound
as it is less cytotoxic. In addition, it has been reported that only
the nonphosphorylated form of fingolimod is cytotoxic, while the phosphorylated
compound might even have a cytoprotective effect.[35,36] As fingolimod, and possibly some of its derivatives, become phosphorylated in vivo, it is possible that they would prove a safe treatment
option, something that could be confirmed by future in vivo studies.
Table 2
CC50 of Fingolimod and
Six Derivatives on Three Human Cell Lines (A549, HEp-2, and HL-60),
after a 24 h Incubation
CC50 (μM)a
A549
HEp-2
HL-60
fingolimod
>40
>40
36.1 ± 13.4
43
>40
>40
23.4 ± 3.2
45
27.4 ± 15.1
>40
3.2 ± 1
55
35.3 ± 5.1
33.9 ± 1.2
7.4 ± 6.9
62
27.6 ± 6.4
30.9 ± 6.2
3.6 ± 0.4
79
>40
>40
26.9 ± 1.6
82
27.5 ± 3
28.8 ± 1.8
6.5 ± 0.8
Calculated from the results of four
experiments with each two technical replicates per concentration (two
experiments for HL-60).
Calculated from the results of four
experiments with each two technical replicates per concentration (two
experiments for HL-60).
Effect
of Fingolimod on S. aureus in Coculture
with Human Neutrophils
As biofilms are particularly
prevalent in device-associated infection, a potential application
for fingolimod and its derivatives could be the protection of medical
devices against biofilm formation. Many factors can affect the activity
of a drug in vivo, for instance, the presence of
human cells and their by-products as well as inflammation and tissue
damage following the insertion of a medical device. The placement
of endotracheal tubes results in tissue injuries that induce the influx
of neutrophils, increasing their number by up to 10-fold in the trachea
after intubation.[37,38] This acute inflammatory response
may lead to the ineffectiveness of the innate immune system to clear
out planktonic bacteria, giving a perfect opportunity for biofilm
formation.[39] Fingolimod being the least
cytotoxic compound in our study, we used it in a proof-of-concept
model to show how it would protect a catheter made of clinically relevant
material, low-density polyethylene (LDPE), from biofilm formation
in a less simplified environment. We used a coculture system that
includes HL-60 cells differentiated into neutrophils and S. aureus ATCC 25923.[40]Figure shows
the effects of fingolimod (25 μM) on the prevention of S. aureus ATCC 25923 attachment on LDPE tubes after
a 24 h incubation. Fingolimod alone (in the absence of HL-60 cells)
significantly reduced the numbers of attached viable S. aureus (p = 0.013). Adding the
HL-60 cells to the system resulted in a significant reduction of attached S. aureus (p = 0.023). The addition
of fingolimod in combination with HL-60 cells also caused a significant
reduction of the bacterial attachment in comparison with the coculture
control (p = 0.042). Fingolimod and the neutrophils
seemed to act in synergy against the bacteria as the combination of
the two had a slightly stronger effect than fingolimod or the cells
alone. This showed that fingolimod, and quite probably its derivatives,
are still active against S. aureus and
has a protective activity on the catheter even in the presence of
cells.
Figure 4
Viable counts of adhered S. aureus ATCC 25923 on LDPE tubes, grown alone or in coculture with differentiated
HL-60 cells and exposed for 24 h to fingolimod at 25 μM. The
results are expressed as mean + SD of three technical replicates in
experiments repeated three times. * Indicates differences with the
monoculture control, # indicates differences with the coculture control
(*: p < 0.05, **: p < 0.01,
#: p < 0.05).
Viable counts of adhered S. aureus ATCC 25923 on LDPE tubes, grown alone or in coculture with differentiated
HL-60 cells and exposed for 24 h to fingolimod at 25 μM. The
results are expressed as mean + SD of three technical replicates in
experiments repeated three times. * Indicates differences with the
monoculture control, # indicates differences with the coculture control
(*: p < 0.05, **: p < 0.01,
#: p < 0.05).
Conclusions
In this work, we developed a library of 28 fingolimod
derivatives
and evaluated their antibacterial and antibiofilm activities. Compared
to fingolimod, seven compounds displayed a better activity against S. aureus, with an MIC below 15 μM. Five derivatives
were also more active against preformed S. aureus biofilms. On the other hand, five derivatives showed improved activity
against P. aeruginosa and/or A. baumannii. Some of the most interesting compounds
in our study include 43, the meta-isomer
of fingolimod, which was the most active against the Gram-negative
species and was one of the least cytotoxic derivatives. Derivative 55 had an increased activity against S. aureus in both pre- and post-exposure and against A. baumannii. The derivative 45 was the most active against S. aureus in pre-exposure while showing poor activity
in other conditions. By contrast, compound 54, while
having a lower MIC of 25 μM against S. aureus, had improved activity in post-exposure and good activity against P. aeruginosa. This study presents many modifications
of the structure of fingolimod that affected its antibacterial activity
in many interesting ways. While none of the tested derivatives displayed
lower cytotoxicity than fingolimod, many optimization efforts generally
do not lead to both an improvement of the desired activity and reduced
cytotoxicity. Nevertheless, the information gathered in this work
can guide further optimization toward less cytotoxic and more potent
antibacterial molecules. This family of structures, which have gained
more and more attention in recent years, have promising antibacterial
and QSI activities, among other potential therapeutic uses.
Experimental
Section
All reagents used in the synthesis were acquired
from Merck (Darmstadt, Germany), Fluorochem (Hadfield, United Kingdom),
Combi-Blocks (San Diego), and TCI (Tokyo, Japan) and used without
further purification. Moisture-sensitive reactions were conducted
using dry solvents in oven-dried (120 °C, >24 h) glassware
under
an inert argon atmosphere. The progress of chemical reactions was
monitored by thin-layer chromatography (TLC) on 0.2 mm Silica gel
60 F254 aluminum plates (Merck, Darmstadt, Germany) and
visualized by UV light (254/366 nm). Microwave reactions were conducted
with a Biotage Initiator + SP Wave Microwave Synthesizer (Uppsala,
Sweden). Column chromatography was carried out on automated Biotage
Isolera Spektra Systems with ACI and Assist (ISO-1SW Isolera One)
equipped with a variable UV–VIS (200–800 nm) photodiode
array (Uppsala, Sweden), using preloaded Biotage SNAP and Sfär
columns, and the indicated mobile phase gradient. Nuclear magnetic
resonance spectra (1H, 13C, and 19F NMR) were recorded on a Bruker Ascend 400 MHz—Avance III
HD NMR spectrometer (Bruker Corporation, Billerica, MA). Chemical
shifts (δ) are reported in parts per million (ppm) relative
to the residual NMR solvent signals: CDCl3 7.26 and 77.16
ppm, CD3OD 3.31 and 49.00 ppm, DMSO-d6 2.50 and 39.52 ppm, (CD3)2CO 2.05 and
29.84/206.26 for 1H and 13C NMR, respectively.
When necessary, two-dimensional NMR spectra (HSQC, HMBC, COSY) were
recorded to support structure determination. Multiplicities of peaks
are represented as s (singlet), d (doublet), t (triplet), q (quartet),
p (pentet), dd (doublet of doublets), m (multiplet), and bs (broad
singlet). Coupling constants J are quoted in hertz
(Hz). All spectra were processed for recorded FID files with MestReNova
12.0.1 software (Mestrelab Research, Santiago de Compostela, Spain).
The exact mass and purity (>95%) of all tested compounds were confirmed
by liquid chromatography–mass spectrometry (LC–MS) analyses
with a Waters Acquity1 UPLC system (Waters, Milford, MA) equipped
with an Acquity UPLC1 BEH C18 column (1.7 μm, 50 mm × 2.1
mm, Waters, Ireland), an Acquity PDA detector, and a Waters Synapt
G2 HDMS mass spectrometer (Waters, Milford, MA) via an electrospray
ionization (ESI) ion source in a positive mode. High-resolution mass
spectrometry (HRMS-ESI) data were reported for the molecular ions
[M + H]+.
General Procedure A: Diethyl acetamidomalonate
Alkylation
To a solution of diethyl acetamidomalonate (DEAM)
and bromide or
mesylate (typically 1.2–2 equiv) in anhydrous MeCN or DMF (0.1–0.2
M) was added cesium carbonate (Cs2CO3, 1.5 or
2.0 equiv), and the reaction mixture was irradiated in a microwave
reactor for 1 h at 130 °C. The progress of the reaction was monitored
by TLC using 50% EtOAc in n-heptane as a mobile phase.
The mixture was filtered and concentrated on a rotary evaporator.
The crude product was purified by silica gel column chromatography
with an increasing gradient of EtOAc (starting with 10–30%)
in n-heptane to afford the desired N-acetylated diester.
General Procedure B: Suzuki Coupling
Aryl bromide or
aryl chloride, boronic acid (1.2 equiv), 5 mol % of tris(dibenzylideneacetone)dipalladium(0)
(Pd2(dba)3), and 10 mol % of 2-dicyclohexylphosphino-2′,6′-dimethoxybiphenyl
(SPhos) were dissolved under argon in toluene/water (4:1; 0.1–0.2
M). When the starting material did not dissolve completely, a few
drops of DMF were added. The reaction mixture was bubbled with argon
for 1 min. Then, potassium carbonate (K2CO3,
3.0 equiv) was added, and the reaction mixture was irradiated in a
microwave reactor for 1.5–8 h at 110 °C. The progress
of the reaction was monitored by TLC using 50% EtOAc in n-heptane as a mobile phase. The mixture was filtered through Celite
and concentrated on a rotary evaporator. The crude product was purified
by silica gel column chromatography with an increasing gradient of
EtOAc in n-heptane, starting with 10 or 30% of EtOAc.
General Procedure C: O-Alkylation of Diethyl 2-Acetamido-2-(4-hydroxyphenethyl)malonate
(20)
To a cooled solution of diethyl 2-acetamido-2-(4-hydroxyphenethyl)malonate
(20) in anhydrous DMF was added anhydrous K2CO3 (2.0 equiv), and the mixture was stirred for 30 min
on an ice bath. Appropriate bromide (1.2 equiv) was dissolved in anhydrous
DMF, and the resulting solution was added dropwise to the reaction
at 0 °C. Stirring was continued for 30 min, after which the reaction
mixture was allowed to warm up to room temperature and stirred overnight.
The progress of the reaction was monitored by TLC using 50% EtOAc
in n-heptane as a mobile phase. The reaction was
quenched with water and extracted with EtOAc. Combined organic layers
were washed with brine, dried over anhydrous sodium sulfate (Na2SO4), filtered, and concentrated on a rotary evaporator.
The crude product was purified by silica gel column chromatography
with an increasing gradient of EtOAc in n-heptane,
starting with 20% of EtOAc.
General Procedure D: Diester Reduction and
the Subsequent Amide
Hydrolysis
To a solution of diester in EtOH/water (4:1),
sodium borohydride (NaBH4, 5.0 equiv) and calcium chloride
(CaCl2, 2.5 equiv) were added in portions over 1 h, and
the reaction mixture was stirred at room temperature overnight. The
progress of the reaction was monitored by TLC-NH2 using
5% MeOH in DCM as a mobile phase. If needed, excess quantities of
NaBH4 (5.0 equiv) and CaCl2 (2.5 equiv) were
added and the mixture was stirred for an additional 2–4 h.
The reaction was quenched with water or a 0.1 M solution of HCl in
water and extracted with either EtOAc or DCM. Combined organic layers
were washed with brine, dried over anhydrous Na2SO4, filtered, and concentrated on a rotary evaporator to give
an N-acetylated diol intermediate, which was used in the next step
without further purification. To a solution of N-acetylated diol intermediate
in MeOH/THF (1:1), a 2 M solution of lithium hydroxide (LiOH) in water
was added and the reaction mixture was refluxed 4–7 h. The
progress of the reaction was monitored by TLC-NH2 using
5% MeOH in DCM as a mobile phase. The mixture was cooled to room temperature
and portioned between water and EtOAc or DCM. The layers were separated,
and the aqueous layer was extracted with either EtOAc or DCM. The
combined organic layers were washed with brine, dried over anhydrous
Na2SO4, filtered, and concentrated on a rotary
evaporator. Silica gel column chromatography was performed with a
Biotage SNAP KP-NH column and an increasing gradient of MeOH in DCM,
starting with 2% of MeOH, to afford the desired amino diol compounds.
General procedure D was followed using 10 (70.0 mg, 0.160 mmol), NaBH4 (30.3 mg, 0.800
mmol, 5.0 equiv), CaCl2 (44.4 mg, 0.400 mmol, 2.5 equiv),
and EtOH/water (4:1, 2 mL); N-acetylated diol intermediate
(white solid, 62 mg), MeOH/THF (1 mL), and 2 M LiOH (aq., 2 mL). Compound 16 was obtained as a white solid (34 mg, 67%). 1H NMR (400 MHz, CD3OD): δ 7.32–7.10 (m, 9H),
6.41 (d, J = 15.9 Hz, 1H), 6.35–6.26 (m, 1H),
3.55–3.42 (m, 6H), 2.66–2.57 (m, 2H), 1.70–1.61
(m, 2H). 13C NMR (101 MHz, CD3OD): δ 143.0,
141.7, 136.6, 132.0, 129.6, 129.53, 129.51, 129.48, 127.2, 127.1,
66.5, 56.8, 40.3, 37.5, 30.1. HRMS-ESI (m/z): calcd for C20H26NO2 [M + H]+ 312.1964, found: 312.1966.
4-(Benzyloxy)phenethyl
Methanesulfonate (18)
To a cooled solution of
2-[4-(benzyloxy)phenyl]ethanol (17) (1.50 g, 6.57 mmol)
and triethylamine (Et3N; 1.37 mL,
9.86 mmol, 1.5 equiv) in anhydrous DCM (25 mL) was added methanesulfonyl
chloride (0.66 mL, 8.54 mmol, 1.3 equiv), and the reaction mixture
was stirred for 2 h. The mixture was diluted with DCM and washed with
brine. The organic layer was dried over anhydrous Na2SO4, filtered, and concentrated on a rotary evaporator. Column
chromatography, with an increasing gradient of EtOAc in n-heptane, starting with 20% of EtOAc, gave compound 18 as a white solid (1.72 g, 86%). 1H NMR (400 MHz, CDCl3): δ 7.47–7.36 (m, 4H), 7.36–7.30 (m,
1H), 7.18–7.12 (m, 2H), 6.97–6.91 (m, 2H), 5.06 (s,
2H), 4.38 (t, J = 6.9 Hz, 2H), 3.00 (t, J = 6.9 Hz, 2H), 2.83 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 158.0, 137.1, 130.2, 128.7, 128.1, 127.6, 115.3,
70.7, 70.2, 37.4, 34.9. HRMS-ESI (m/z): calcd for C16H18O4S [M + H]+ 306.0926, found: 306.0925.
To a solution of 19 (0.788
g, 1.84 mmol)
in EtOH/EtOAc (1:1, 20 mL) was added 10% palladium on carbon (0.196
g, 0.184 mmol, 0.1 equiv). The reaction mixture was stirred at room
temperature under a hydrogen atmosphere for 5 h, after which it was
filtered through Celite and concentrated on a rotary evaporator. The
resulting solid was washed with diisopropyl ether and dried in a vacuum
to give 20 as a white solid (547 mg, 88%). 1H NMR (400 MHz, CD3OD): δ 7.00–6.94 (m, 2H),
6.73–6.68 (m, 2H), 4.27–4.13 (m, 4H), 2.54–2.48
(m, 2H), 2.45–2.39 (m, 2H), 2.01 (s, 3H), 1.25 (t, J = 7.1 Hz, 6H). 13C NMR (101 MHz, CD3OD): δ 172.5, 169.2, 156.7, 132.9, 130.3, 116.2, 67.9, 63.4,
35.7, 30.2, 22.3, 14.3. HRMS-ESI (m/z): calcd for C17H24NO6 [M + H]+ 338.1604, found: 338.1603.
To a cooled solution of
2-(quinolin-2-yl)ethan-1-ol (30) (0.150 g, 0.866 mmol)
and Et3N (0.247 mL, 1.77 mmol,
2.0 equiv) in anhydrous DCM (4 mL) was added methanesulfonyl chloride
(0.122 mL, 1.06 mmol, 1.2 equiv), and the reaction mixture was stirred
for 3 h. The mixture was diluted with DCM and washed with water and
brine. The organic layer was dried over anhydrous Na2SO4, filtered, and concentrated on a rotary evaporator. Column
chromatography, with an increasing gradient of EtOAc in n-heptane, starting with 20% of EtOAc, gave 31 as a white
solid (78 mg, 37%). 1H NMR (400 MHz, CDCl3):
δ 8.12 (d, J = 8.4 Hz, 1H), 8.01 (d, J = 8.5 Hz, 1H), 7.81 (d, J = 8.1 Hz, 1H),
7.74–7.68 (m, 1H), 7.57–7.48 (m, 1H), 7.34 (d, J = 8.4 Hz, 1H), 4.81 (t, J = 6.4 Hz, 2H),
3.42 (t, J = 6.4 Hz, 2H), 2.93 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 157.2, 148.0, 136.8, 129.9,
128.9, 127.8, 127.2, 126.5, 122.0, 69.0, 38.1, 37.4.
A mixture of diisopropylamine (1.89 mL,
13.5 mmol, 3.0 equiv) and
anhydrous THF (2.0 mL) was cooled to −78 °C under an argon
atmosphere. N-Butyllithium (a 2.5 M solution in hexane,
5.04 mL, 12.6 mmol, 2.8 equiv) was slowly added and the mixture was
stirred for 30 min. Then, a solution of 6-bromoquinaldine (64, 1.00 g, 4.50 mmol) in dry THF (4.0 mL) was slowly added. After
30 min, diethyl carbonate (1.91 mL, 15.8 mmol, 3.5 equiv) was slowly
added and the mixture was stirred for 2 h. The reaction was quenched
with water and extracted with EtOAc. The combined organic layers were
washed with brine, dried over anhydrous Na2SO4, filtered, and concentrated on a rotary evaporator, to give compound 65 as an orange solid (1.28 g, 97%). 1H NMR (400
MHz, CDCl3): δ 8.04 (d, J = 8.5
Hz, 1H), 7.96 (d, J = 2.2 Hz, 1H), 7.93 (d, J = 8.9 Hz, 1H), 7.76 (dd, J = 9.0, 2.2
Hz, 1H), 7.46 (d, J = 8.5 Hz, 1H), 4.20 (q, J = 7.1 Hz, 2H), 4.02 (s, 2H), 1.27 (t, J = 7.1 Hz, 3H). 13C NMR (101 MHz, CDCl3): δ
170.4, 155.5, 146.6, 135.7, 133.2, 131.0, 129.7, 128.3, 122.8, 120.4,
61.4, 44.9, 14.3. HRMS-ESI (m/z):
calcd for C13H13NO2Br [M + H]+ 294.0130, found: 294.0129.
2-(6-Bromoquinolin-2-yl)ethan-1-ol
(66)
To a solution
of 65 (1.25 g, 4.24 mmol) in anhydrous THF (10 mL) was
added a 1 M solution of DIBAL-H in THF (12.7 mL, 12.7 mmol, 3.0 equiv)
dropwise over 30 min at −78 °C and under an argon atmosphere.
After the completion of the addition, the mixture was warmed to −10
°C and stirred for 3 h. The reaction was quenched with methanol.
Then, 10% Rochelle′s salt and EtOAc were added and the mixture
was stirred vigorously for 30 min. The layers were separated and the
water layer was further extracted with EtOAc. The combined organic
layers were washed with brine, dried over anhydrous Na2SO4, filtered, and concentrated on a rotary evaporator.
Column chromatography, with an increasing gradient of EtOAc (starting
with 30%) in n-heptane, gave compound 66 as yellow solid (0.848 g, 79%). 1H NMR (400 MHz, CDCl3): δ 8.00 (d, J = 8.3 Hz, 1H), 7.94
(d, J = 2.2 Hz, 1H), 7.88 (d, J =
8.9 Hz, 1H), 7.75 (dd, J = 9.0, 2.2 Hz, 1H), 7.29
(d, J = 8.5 Hz, 1H), 4.46 (bs, 1H), 4.14 (t, J = 5.5 Hz, 2H), 3.19 (t, J = 5.5 Hz, 2H). 13C NMR (101 MHz, CDCl3): δ 162.1, 146.1,
135.7, 133.2, 130.7, 129.8, 128.1, 122.9, 112.0, 61.3, 39.5. HRMS-ESI
(m/z): calcd for C11H11NOBr [M + H]+ 225.0024, found: 225.0025.
To a cooled solution of 66 (0.750 g, 2.96
mmol) and Et3N (0.62 mL, 4.45 mmol, 1.5 equiv) in anhydrous
DCM (20 mL) was added methanesulfonyl chloride (0.298 mL, 3.85 mmol,
1.3 equiv), and the reaction mixture was stirred for 3 h. The mixture
was diluted with DCM and washed with water and brine. The organic
layer was dried over anhydrous Na2SO4, filtered,
and concentrated on a rotary evaporator. Greyish solid (0.901 g, 92%). 1H NMR (400 MHz, CDCl3): δ 8.03 (d, J = 8.3 Hz, 1H), 7.97 (d, J = 2.2 Hz, 1H),
7.88 (d, J = 8.9 Hz, 1H), 7.77 (dd, J = 9.0, 2.2 Hz, 1H), 7.35 (d, J = 8.4 Hz, 1H), 4.80
(t, J = 6.5 Hz, 2H), 3.40 (t, J =
6.5 Hz, 2H), 2.94 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 157.7, 146.6, 135.8, 133.3, 130.7, 129.8, 128.3, 122.9,
120.3, 68.7, 38.0, 37.4. HRMS-ESI (m/z): calcd for C12H13NO3SBr [M + H]+ 329.9800, found: 329.9799.
6-Bromo-2-(bromomethyl)quinolone
(69)
To a solution of 6-bromoquinaldine (64, 1.00 g, 4.50
mmol) in MeCN (20 mL) was added tert-butyl hydroperoxide
(4.64 mL, 36.0 mmol, 8.0 equiv) and copper bromide (0.904 g, 6.30
mmol, 1.4 equiv), and the mixture was stirred at 70 °C for 8
h. Then, it was cooled to room temperature and portioned between water
and DCM. The layers were separated and the aqueous layer was extracted
with DCM. Residual bromide in the organic phase was quenched with
a saturated solution of sodium thiosulphate in H2O. The
layers were separated and the organic layer was washed with a saturated
solution of NH4Cl in H2O and brine, dried over
Na2SO4, filtered, and concentrated on a rotary
evaporator. Column chromatography, with an increasing gradient of
EtOAc (starting with 0%) in n-heptane, gave compound 69 as a white solid (358 mg, 26%). 1H NMR (400
MHz, CDCl3): δ 8.09 (d, J = 8.6
Hz, 1H), 7.98 (d, J = 2.2 Hz, 1H), 7.93 (d, J = 9.0 Hz, 1H), 7.79 (dd, J = 9.0, 2.2
Hz, 1H), 7.59 (d, J = 8.5 Hz, 1H), 4.69 (s, 2H). 13C NMR (101 MHz, CDCl3): δ 157.5, 146.3,
136.4, 133.6, 131.1, 129.7, 128.6, 122.2, 121.2, 34.2. HRMS-ESI (m/z): calcd for C10H8NBr [M + H]+ 299.9024, found: 299.9023.
To a cooled solution of 75 (85.0 mg, 0.183
mmol) in
anhydrous THF (2.0 mL) was added a 1 M solution of LiAlH4 in THF (0.37 mL, 0.37 mmol, 2.0 equiv) dropwise over 10 min, and
the mixture was stirred for 1 h on an ice bath. Then, it was warmed
up to room temperature and stirred for an additional 1 h. The reaction
was slowly quenched with water. Then, 10% Rochelle’s salt and
EtOAc were added and the mixture was stirred vigorously overnight.
The layers were separated and the water layer was further extracted
with EtOAc. The combined organic layers were washed with brine, dried
over anhydrous Na2SO4, filtered, and concentrated
on a rotary evaporator to give an N-acetylated diol intermediate as
a yellow oil (72 mg). The intermediate was dissolved in THF/MeOH (2
mL), 2 M LiOH (aq., 2 mL) was added, and the reaction mixture was
refluxed for 4 h. The reaction mixture was cooled to room temperature
and portioned between water and EtOAc. The layers were separated and
the aqueous layer was extracted with EtOAc. The combined organic layers
were washed with brine, dried over anhydrous Na2SO4, filtered, and concentrated on a rotary evaporator. Column
chromatography, using a Biotage SNAP KP-NH column and an increasing
gradient of MeOH in DCM, starting with 2% of MeOH, gave compound 80 as a pale yellow solid (19 mg, 29%). 1H NMR
(400 MHz, CD3OD): δ 8.19 (d, J =
8.4 Hz, 1H), 7.87 (d, J = 8.6 Hz, 1H), 7.66 (d, J = 1.9 Hz, 1H), 7.60 (dd, J = 8.7, 2.0
Hz, 1H), 7.44 (d, J = 8.5 Hz, 1H), 3.54 (q, J = 10.9 Hz, 4H), 3.04–2.96 (m, 2H), 2.80 (t, J = 7.5 Hz, 2H), 1.89–1.80 (m, 2H), 1.72 (p, J = 7.5 Hz, 2H), 1.43–1.22 (m, 10H), 0.92–0.84
(m, 3H). 13C NMR (101 MHz, CD3OD): δ 163.6,
147.2, 142.3, 138.3, 132.5, 128.4, 128.2, 127.3, 122.8, 66.3, 57.1,
36.7, 35.5, 33.2, 33.0, 32.5, 30.6, 30.38, 30.35, 23.7, 14.4. HRMS-ESI
(m/z): calcd for C22H35N2O2 [M + H]+ 359.2699,
found: 359.2698.
S. aureus ATCC 25923 was provided
by the Faculty
of Pharmacy of the University of Helsinki (Helsinki, Finland). S. aureus ATCC 12598 was kindly provided by Dr. Per
Saris from the Antimicrobials, probiotics and fermented food laboratory,
Faculty of Agriculture and Forestry, University of Helsinki (Helsinki,
Finland). S. aureus P2 was isolated
from a hip prosthetic implant at the Hospital Fundación Jiménez
Díaz (Madrid, Spain) and was kindly given by Dr. Ramón
Pérez-Tanoira. A. baumannii NCTC
13423 and C. violaceum NCTC 13278 (Tn5-mutant
CV026) were bought from the National Collection of Type Culture (NCTC)
(Salisbury, United Kingdom). P. aeruginosa ATCC 9027 and C. violaceum ATCC 31532
were bought from the American Type Culture Collection (ATCC) (Wesel,
Germany). At the start of all experiments, every strain (except for P. aeruginosa and C. violaceum) was first grown on tryptic soy agar (TSA) (Lab M Ltd., Heywood,
U.K.) overnight at 37 °C under aerobic conditions. Then, the
colonies were suspended in 5 mL of tryptic soy broth (TSB) (Lab M
Ltd., Heywood, U.K.) in a 50 mL Falcon tube (Greiner Bio-One, Kremsmünster,
Austria) and grown to exponential phase under aerobic conditions at
37 °C with shaking (220 rpm), until the culture reached a concentration
of approximately 1 × 108 colony-forming unit (CFU)
mL–1. The culture was diluted to 1 × 106 CFU mL–1 before starting the experiment.
The same was done with P. aeruginosa using Lennox broth (LB) and LB-agar (LBA) (Lab M Ltd., Heywood,
U.K.). For the QS inhibition analysis, C. violaceum was grown on LBA overnight at 27 °C, and the colonies were
used to start the experiment directly.
Compounds
Fingolimod,
with purity ≥98%, was
purchased from Carbosynth (Compton, U.K.). Fingolimod was dissolved
into dimethyl sulfoxide (DMSO) and sonicated for 5 min using a Branson
3800 ultrasonic bath (Cleanosonic, Richmond, VA) at 40 Hz.
Antibacterial
Activity Evaluation
The antimicrobial
activity was assessed in a similar way for the initial screening and
the dose–response evaluation against different bacterial strains.
With the exception of the post-exposure experiment, the compounds
were added to the wells along with the bacteria to test the ability
of compounds to prevent bacterial growth and biofilm formation (pre-exposure).
The compounds were diluted in DMSO and plated directly in 96-well
plates before adding the bacterial culture, and plates were incubated
for 18 h under aerobic conditions at 37 °C with shaking (220
rpm). The antibacterial activity of the compounds was assessed on
planktonic cells and biofilms separately using two different measurements
for each: optical density (turbidity) and resazurin reduction (viability)
on planktonic cells and resazurin followed by a crystal violet staining
assay (total biomass) on biofilms, as described in the following sections.For the post-exposure experiment, bacteria were grown with no compounds
in the same conditions for 18 h, after which the media was changed,
and compounds were added on the preformed biofilms. The plates were
then incubated in the same conditions for an additional 24 h before
proceeding to the staining assays.
Resazurin Staining
The resazurin staining was performed
according to the protocol previously optimized with a few modifications.[46] Briefly, the planktonic solution was first transferred
into a new 96-well plate and the OD of the planktonic solution was
measured using a Varioskan LUX Multimode microplate reader (Thermo
Scientific, Vantaa, Finland). Then, 10 μL of a 400 μM
resazurin (R7017, Sigma-Aldrich, St. Louis, MO) diluted in phosphate-buffered
saline (PBS) (Thermo Fisher Scientific, MA) was added to the wells
(for a final resazurin concentration of 20 μM). The plate was
incubated in the dark with shaking (220 rpm) for about 3–10
min at room temperature with S. aureus and A. baumannii (until the bacteria
controls turned pink) or for 20 min at 37 °C with P. aeruginosa. Fluorescence was measured at λex = 560 nm and λem = 590 nm using the top
optics of the Varioskan LUX Multimode microplate reader. The planktonic
plate was then discarded. The original challenge plate containing
the biofilms was washed once with PBS, and 200 μL of a 20 μM
resazurin solution was added to the wells. The plate was incubated
in the dark at 37 °C with shaking (220 rpm) for 30 min with S. aureus, 60 min with A. baumannii, and 90 min with P. aeruginosa. Fluorescence
was measured as described for the planktonic solution. The biofilms
were next treated with crystal violet staining.
Crystal Violet
Staining
After removing the resazurin
solution from the plate, biofilms were fixed with ethanol 100% for
15 min at room temperature. Ethanol was then removed, and the biofilms
were left to air-dry completely at RT. Crystal violet, diluted to
1:100 in deionized water from the 2.3% commercial solution (HT90132,
Sigma-Aldrich, St. Louis, MO), was added to the wells and, after a
5 min incubation with the dye, the biofilms were washed twice with
deionized water and left to air-dry for about 10 min. The dye was
solubilized in ethanol 100% for 1 h, and absorbance was measured at
595 nm using a Multiskan Sky microplate spectrophotometer (Thermo
Scientific, Vantaa, Finland).
Quorum Sensing Inhibition
Assay
The QS inhibitory activity
of the compounds was assessed as described previously.[31,47] Briefly, two QS reporter strains, C. violaceum ATCC 31532 and the violacein-negative, mini-Tn5 mutant of C. violaceum CV026 (NCTC 13278), were grown on LBA
(for CV026, the agar was supplemented with kanamycin at 100 μg
mL–1) overnight at 27 °C. Colonies were suspended
in PDYT (0.5% peptone, 0.3% d-glucose, 0.25% yeast extract,
0.05% l-tryptophan, w/v) to reach an OD600 of 0.02. The CV026
culture was supplemented with C6-HSL (N-hexanoyl-l-homoserine lactone) (10007896, Cayman Chemical, MI) at 0.5
μM to induce the QS-moderated synthesis of violacein. For each
strain, 200 μL of bacterial culture was added per well in two
identical 96-well plates containing the compounds (2 μL per
well for a final DMSO concentration of 1%). In each 96-well plate,
untreated cells were used as negative controls and quercetin (Q4951,
Sigma-Aldrich) and azithromycin (PZ0007, Sigma-Aldrich) were included
as positive controls for QS inhibition and cell viability (bactericidal
activity), respectively. The plates were incubated at 27 °C under
aerobic conditions (200 rpm) for 22 h.The synthesized and nonsoluble
violacein was collected from the first 96-well plate by centrifugation
(4000 rpm, for 15 min, 20 °C), and the supernatants were removed.
Violacein was dissolved in 100 μL per well of 96% (v/v) ethanol
and was separated from cells by centrifugation as described above.
The supernatant containing violacein was then transferred to a new
96-well plate, and the absorbance was measured at 595 nm using a Multiskan
Sky Microplate spectrophotometer. To detect any bactericidal effects,
the viability of the cells was measured in the second replica plate
by adding 10 μL of 400 μM resazurin in each well and incubating
the plate for 30 min in the dark at 27 °C (220 rpm). After centrifugation
of the plate to separate the cells from the solution, 100 μL
of each well were transferred into a new 96-well plate and the fluorescence
was recorded as described above using the Varioskan LUX multimode
plate reader.
Culture of Human Cells
The humanlung adenocarcinoma
epithelial cells A549, the cervical adenocarcinoma HeLa-derived human
epithelial type 2 (Hep-2) cells, and the HL-60 cells were all purchased
from the American Type Culture Collection (ATCC, Wesel, Germany).
The A549 cells were grown in Dulbecco’s modified Eagle’s
medium (DMEM) (BE12-707F, Lonza, Basel, Switzerland) supplemented
with 10% heat-inactivated fetal bovine serum (FBS) (Sigma-Aldrich,
St. Louis, MO), 2 mM glutamine (Sigma-Aldrich, St. Louis, MO), and
20 μg mL–1 gentamycin (Sigma-Aldrich, St.
Louis, MO). The HEp-2 cells were grown in DMEM (41966-029, Sigma-Aldrich,
St. Louis, MO), supplemented with 10% (v/v) heat-inactivated FBS and
20 μg mL–1 gentamycin. Both cell lines were
maintained at 37 °C in 5% CO2 in a humidified incubator.
HL-60 cells were grown and maintained in Roswell Park Memorial Institute
(RPMI) 1640 Medium (R8758, Sigma-Aldrich, St. Louis, MO), supplemented
with 10% (v/v) heat-inactivated FBS and 1% (v/v) penicillin/streptomycin
(Sigma-Aldrich, St. Louis, MO). The cells were maintained at 37 °C
in 5% CO2 in a humidified incubator. Before each experiment,
the HL-60 cells were differentiated into neutrophile-like cells by
culturing the cells for 6 days in the maintenance medium supplemented
with 100 mM of N,N–dimethylformamide
(DMF) (Sigma-Aldrich, St. Louis, MO).[48]
Cytotoxicity Assessment
The cytotoxicity of the compounds
was assessed on the three different human cell lines, A549, HEp-2,
and HL-60, using the redox-based resazurin staining assay. The cells
were seeded in clear 96-well plates at a density of 1 × 104 cells per well in a volume of 200 μL per well. The
adherent cells were let to attach for 24 h before changing the media
and adding the compounds (final concentration of 0.25% DMSO). HL-60
were counted and added to the plates at a density of 1 × 104 cells per well in a volume of 200 μL directly with
the compounds. Untreated cells were used as negative controls, and
cells treated with 1% TritonX-100 (Sigma-Aldrich, St. Louis, MO) were
used as positive controls for cytotoxicity. The cells were incubated
with the compounds for 24 h.For the resazurin assay, the media
was first removed, and the adherent cells were washed once with PBS.
Then, 200 μL of 20 μM resazurin was added per well and
cells were incubated for two more hours before the measurement of
fluorescence, as mentioned earlier. For HL-60, 10 μL of 400
μM resazurin was added directly to the wells and the cells were
incubated for an additional 4 h before the measurement of fluorescence.
Effect of Fingolimod on S. aureus in
Coculture with Differentiated HL-60 Cells
The coculture
system was as previously described.[40] Briefly,
the bacteria were prepared as described above and the concentration
was adjusted to 2 × 107 CFU mL–1 in RPMI 1640. The media of the differentiated HL-60 cells was refreshed
with RPMI 1640 24 h before the experiments to remove possible traces
of antibiotics. On the day of the experiment, the cells were counted
and adjusted to a concentration of 2 × 105 cells mL–1 in RPMI 1640. Each suspension (bacterial and differentiated
HL-60 cells, 500 μL of each) was added on 1-cm-long sections
of a sterilized fine bore LDPE tubing (Smiths Medical ASD, Minneapolis,
MN) in 24-well plates (Nunclon Δ surface, Nunc, Roskilde, Denmark).
Fingolimod was also added to the wells at a concentration of 25 μM
(0.25% DMSO). Coculture controls with no added compound were included,
as well as bacterial controls without differentiated HL-60 cells with
and without fingolimod. The plates were incubated for 24 h in the
same conditions as for cell maintenance. Then, the LDPE tubes were
gently washed with TSB to remove adhering planktonic cells and transferred
to Eppendorf tubes containing 1 mL of 0.5% (w/v) Tween 20-TSB solution.
Next, the tubes were sonicated in a water bath sonicator (Branson
Ultrasonics, Danbury, CT) for 5 min at 35 kHz. The tubes were vortexed
for 20 s prior to and after the sonication step. Serial dilutions
of the resulting bacterial suspensions were made from 10–1 up to 10–7 and plated on TSA plates.
Statistical
Analysis
In the QS inhibition assay, statistical
difference between the test compounds was calculated at each concentration
using a one-way analysis of variance (ANOVA) with a Welch correction
in the case of violation of the homogeneity of variance. A significant
difference was observed against the wild-type strain at 10 μM
[F(6, 9.26) = 10.71, p < 0.001], 15 μM [F(6,
21) = 5.87, p = 0.001], and 25 μM [F(6, 8.79)
= 9.51, p = 0.002] as well as against the mutant
strain at 10 μM [F(6, 9.25) = 12.64, p <
0.001]. A Dunnett’s post hoc test was used to detect statistical
differences between fingolimod and each of its derivatives. In the
bacteria-neutrophil coculture assay, significance was determined using
unpaired t-tests.
Authors: Grace E Martin; Ryan M Boudreau; Caroline Couch; Katrin Anne Becker; Michael J Edwards; Charles C Caldwell; Erich Gulbins; Aaron Seitz Journal: Biochimie Date: 2017-03-21 Impact factor: 4.079
Authors: Rizwan S Shaikh; Stefanie S Schilson; Stefan Wagner; Sven Hermann; Petra Keul; Bodo Levkau; Michael Schäfers; Günter Haufe Journal: J Med Chem Date: 2015-04-14 Impact factor: 7.446
Authors: Carol L Fischer; David R Drake; Deborah V Dawson; Derek R Blanchette; Kim A Brogden; Philip W Wertz Journal: Antimicrob Agents Chemother Date: 2011-12-12 Impact factor: 5.191
Authors: Yael Pewzner-Jung; Shaghayegh Tavakoli Tabazavareh; Heike Grassmé; Katrin Anne Becker; Lukasz Japtok; Jörg Steinmann; Tammar Joseph; Stephan Lang; Burkhard Tuemmler; Edward H Schuchman; Alex B Lentsch; Burkhard Kleuser; Michael J Edwards; Anthony H Futerman; Erich Gulbins Journal: EMBO Mol Med Date: 2014-09 Impact factor: 12.137
Figure 1
Scheme 1
Scheme 3
Scheme 2
Figure 2
Figure 3
Figure 4