Structural optimization and antibacterial evaluation of rhodomyrtosone B analogues against MRSA strains

Article abstract of DOI:10.1039/c8md00257f

Methicillin-resistant Staphylococcus aureus (MRSA) infections are well-known as a significant global health challenge. In this study, twenty-two congeners of the natural antibiotic rhodomyrtosone B (RDSB) were synthesized with the aim of specifically enhancing the structural diversity through modifying the pendant acyl moiety. The structure-activity relationship study against various MRSA strains revealed that a suitable hydrophobic acyl tail in the phloroglucinol scaffold is a prerequisite for antibacterial activity. Notably, RDSB analogue 11k was identified as a promising lead compound with significant in vitro and in vivo antibacterial activities against a panel of hospital mortality-relevant MRSA strains. Moreover, compound 11k possessed other potent advantages, including breadth of the antibacterial spectrum, rapidity of bactericidal action, and excellent membrane selectivity. The mode of action study of compound 11k at the biophysical and morphology levels disclosed that 11k exerted its MRSA bactericidal action by membrane superpolarization resulting in cell lysis and membrane disruption. Collectively, the presented results indicate that the novel modified RDSB analogue 11k warrants further exploration as a promising candidate for the treatment of MRSA infections.

Full text of DOI:10.1039/c8md00257f

MedChemComm
View Article Online
View Journal
RESEARCH ARTICLE
Structural optimization and antibacterial
evaluation of rhodomyrtosone B analogues
Cite this: DOI: 10.1039/c8md00257f
against MRSA strains†
Liyun Zhao,‡ac Hongxin Liu,‡ab Luqiong Huo,ac Miaomiao Wang,ac Bao Yang,a
b
a
a
Weimin Zhang, Zhifang Xu, Haibo Tan a and Sheng-Xiang Qiu
*
*
Methicillin-resistant Staphylococcus aureus (MRSA) infections are well-known as a significant global health
challenge. In this study, twenty-two congeners of the natural antibiotic rhodomyrtosone B (RDSB) were
synthesized with the aim of specifically enhancing the structural diversity through modifying the pendant
acyl moiety. The structure–activity relationship study against various MRSA strains revealed that a suitable
hydrophobic acyl tail in the phloroglucinol scaffold is a prerequisite for antibacterial activity. Notably, RDSB
analogue 11k was identified as a promising lead compound with significant in vitro and in vivo antibacterial
activities against a panel of hospital mortality-relevant MRSA strains. Moreover, compound 11k possessed
other potent advantages, including breadth of the antibacterial spectrum, rapidity of bactericidal action,
and excellent membrane selectivity. The mode of action study of compound 11k at the biophysical and
morphology levels disclosed that 11k exerted its MRSA bactericidal action by membrane superpolarization
resulting in cell lysis and membrane disruption. Collectively, the presented results indicate that the novel
modified RDSB analogue 11k warrants further exploration as a promising candidate for the treatment of
MRSA infections.
Received 17th May 2018,
Accepted 9th August 2018
DOI: 10.1039/c8md00257f
rsc.li/medchemcomm
of MRSA infections are very limited,11 and it has now been
declared a public-health imperative.3,10,12 Therefore, the dis-
1. Introduction
Infectious diseases are one of the major causes responsible
for human death worldwide, particularly in developing coun-
tries.1,2 Methicillin-resistant Staphylococcus aureus (MRSA) is
well-known as a pathogenic bacterium giving rise to terrible
infections, associated with high rates of morbidity and mor-
tality.3,4 MRSA infection is considered as one of the most
intractable diseases because of its emerging resistance,
exhibiting resistance to almost every commercially available
antibiotic, including β-lactam,5 macrolide,6 fluoroquino-
lone,7 aminoglycoside,8 tetracycline,9 and lincosamide anti-
biotics,9 with numerous resistance mechanisms.10 As a re-
sult, the therapeutic options for the treatment and control
covery of potential new agents to control multidrug-resistant
pathogens is crucial to the maintenance of public health in
the near future.13
In our continuous campaign to find novel antibiotics from
traditional Chinese medicinal plants or through synthesis to
counter the threat of MRSA bacteria,14–17 we have recently
reported two natural acylphloroglucinols (Scheme 1),
rhodomyrtone and rhodomyrtosone B (RDSB),15 which were
reputed to exhibit significant antibacterial activity against
Gram-positive pathogens, even including MRSA, with novel
mechanisms.18–20 However, the paucity of rhodomyrtone's
natural supply and its inconvenient total synthesis make the
prospects for structural optimization of rhodomyrtone seem
distinctly challenging and rather bleak.21 Although RDSB did
not attract any attention or become widely accepted by me-
dicinal scientists, even though the details of its chemical syn-
thesis have been well established, its comparable anti-
bacterial activity with that of rhodomyrtone and much more
accessible synthesis make it also an appealing lead com-
pound for new antibacterial drug discovery. In this regard, an
effort towards the design and synthesis of novel RDSB ana-
logues, which would facilitate convergent assembly in con-
junction with considerable antibacterial activity, seemed to
be worthy of pursuit.
a Key Laboratory of Plant Resources Conservation and Sustainable Utilization,
South China Botanical Garden, Chinese Academy of Sciences, Guangzhou, 510650,
People's Republic of China. E-mail: tanhaibo@scbg.ac.cn, sxqiu@scbg.ac.cn
b State Key Laboratory of Applied Microbiology Southern China, Guangdong
Provincial Key Laboratory of Microbial Culture Collection and Application,
Guangdong Open Laboratory of Applied Microbiology, Guangdong Institute of
Microbiology, Guangzhou 510070, People's Republic of China
c University of Chinese Academy of Sciences, Beijing 100049, People's Republic of
China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8md00257f
‡ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2018
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
Scheme 1 The structure of rhodomyrtone, RDSB, and their analogue 11k.
Herein, we describe our efforts, accompanied by a full ex-
ploration of the details of the synthesis of the RDSB ana-
logues, their antibacterial evaluations, lead structural optimi-
zation through extensive structure–activity relationship (SAR)
investigations, and their antibacterial mechanism.
acetylphloroglucinol substrates, tending to provide every de-
sired acyclic product 10 in excellent yields ranging from 88%
to 96%. Further intramolecular cyclization, enabled by
treating acyclic acetylphloroglucinols 10 with PTSA (1.0
equiv.) in refluxing toluene, was totally completed in 1 hour
and delivered all of the corresponding RDSB analogues 11 in
more than 80% yield.21 The structures of all of the new RDSB
analogues 11 were fully characterized and confirmed by
2. Results and discussion
2.1. Chemistry
1
HRMS, H NMR, and 13C NMR spectra.
The implementation of the chemical synthesis of RDSB ana-
logues 11 was commenced with the preparation of syncarpic
acid (7) by the previously established protocol with slight
modification, as shown in Scheme 2.21,22 Following this, the
syncarpic acid (7) could be readily accessed through a stan-
dard sequence of aluminum trichloride-catalyzed Friedel–
Crafts acylation, selective C-tetramethylation, and retro-
Claisen condensation. On treating the syncarpic acid (7) with
isovaleraldehyde, the proline-promoted Knoevenagel conden-
sation reaction would take place smoothly to provide the crit-
ical intermediate 9 within almost quantitative yield (95%
yield).22 The crucial coupling of intermediate 9 and the rela-
tive acetylphloroglucinols 5 was initiated at room tempera-
ture in THF under strong basic conditions. Satisfyingly, the
reaction conditions seemed to be compatible with a series of
2.2. Biological evaluation
2.2.1. Antibacterial activity of RDSB analogues 11. The
minimum inhibitory concentration (MIC) for RDSB ana-
logues 11 against MRSA cells (JCSC 4788) was evaluated by
means of the micro dilution susceptibility test, performed in
Mueller–Hinton broth (MHB) following CLSI guidelines and
using vancomycin as the positive control. The highest MIC
for the tested acylphloroglucinol analogues 11 was 256 μg
mL−1, indicating their potent antibacterial activities. More-
over, all of the synthesized compounds 11 were also tested
for their antibacterial activity against standard Staphylococcus
aureus (ATCC 6538). The results are listed in Table 1, and the
data were obtained with vancomycin as the standard. Results
Scheme 2 The synthesis of RDSB analogues 11.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2018

View Article Online
MedChemComm
Research Article
Table 1 In vitro antibacterial evaluation (MIC) of RDSB analogues 11a–11v against MRSA (JCSC 4788) and standard SA (ATCC 6538)
Comp.
R
MICMRSA μg mL−1
MICSA μg mL−1
Comp.
R
MICMRSA μg mL−1
MICSA μg mL−1
11a
11b
11c
11d
11e
11f
11g
11h
11i
11j
11k
1
—
H
CH3
256
128
4.0
2.0
2.0
4.0
1.0
1.0
0.5
0.5
0.25
1.0
>256
128
4.0
2.0
2.0
4.0
0.5
0.5
0.5
11l
n-C8H17
n-C9H19
n-C10H21
n-C11H23
n-C13H27
n-C15H31
c-C4H7
0.5
0.50
1.0
0.25
0.50
1.0
11m
11n
11o
11p
11q
11r
11s
11t
11u
11v
Van.
C2H5
n-C3H7
i-C3H7
n-C4H9
i-C4H9
n-C5H11
n-C6H13
n-C7H15
—
8.0
8.0
256
>256
8.0
4.0
4.0
4.0
2.0
1.0
128
>256
4.0
4.0
2.0
4.0
2.0
0.5
c-C5H9
c-C6H11
CH2Ph
C2H4Ph
—
0.25
0.25
1.0
from anti-MRSA screening demonstrated a clear structure–ac-
tivity relationship, wherein the lipophilic ability of the acyl
group tails had a significant influence on the potency of bac-
terial growth inhibition.
bition even at 128 μg mL−1. Moreover, a cyclic ring in the hy-
drophobic acetyl tail seemed not to significantly affect the
antibacterial activity, as clarified with 11r–11v.
Although the lipophilicity seemed to be the principal fac-
tor influencing the antibacterial activity, both the hydropho-
bicity and length of the acyl tail significantly modulate their
antibacterial potency, which was not linearly correlated with
their lipophilicity. These results were attributed to the pene-
trability of the RDSB analogues to diffuse into the MRSA
membrane to reach their antibacterial site. However, it was
clear that the lipophilicity might be the most important
physicochemical property for determining the antibacterial
activity of the RDSB analogues. For the more active RDSB
analogues, suitable lipophilicity is a prerequisite and the op-
timal length for the acetyl tail was found to be between 4
and 9 carbons. As shown in the context of the structure–ac-
tivity relationship investigation, the most active RDSB ana-
logue was found to be 11k (molecular weight = 470.27),
which showed profound anti-MRSA activity with an MIC
value as low as 0.25 μg mL−1 towards MRSA cells (JCSC
4788). Its activity was similar to that for the clinical drug
2.2.2. Structure activity relationship study of RDSB ana-
logues 11. Ketone motifs are usually considered to be one of
the most ubiquitous and intriguing functional groups in the
repertoire of medicinal chemistry, forming the basic active
moieties for various biologically meaningful natural prod-
ucts and pharmaceuticals. Therefore, an initial effort was
conducted to unravel the mystery of the role of the ketone
moiety in the RDSB analogues, which might subsequently
serve as an informative vehicle for the rational structure de-
sign of lead compounds. As expected, the compound 11b,
whose lipophilic ketone tail was replaced by an aldehyde
group, showed dramatically weak antibacterial activity even
at 128 μg mL−1, strongly indicating that the hydrophobic
acyl tail and the ketone moiety in the phloroglucinol core
are vital elements in maintaining the biological properties
(Table 1). In addition, compound 11a without the acetyl tail
was devoid of any noticeable activity against these same
strains at a concentration even up to 256 μg mL−1, further
consolidating this conclusion.
Biophysical processes play a significant role in the antimi-
crobial activity, especially for amphiphilic molecules.16,23,24
Thus, it could be readily anticipated that suitable tuning of
the amphiphilic property of the target molecule would result
in a critical influence on its biological activities. With this
scenario in mind, the effect on bactericidal activity of hydro-
phobic acetyl tails with different lengths was evaluated. The
anti-MRSA activity was found to linearly correlate with the en-
hancing lipophilicities of the acetyl tail in 11c–11k. Moreover,
when more lipophilic acetyl tails were induced in the RDSB
analogues 11g–11n, neither the length nor the lipophilicity of
the acetyl tails seemed to have a significant influence on the
anti-MRSA activity, giving rise to potent anti-MRSA activity
with the MIC values ranging from 0.25–1.0 μg mL−1, similar
to that of vancomycin (MIC = 1.0 μg mL−1). Notably, if the
length of the acetyl tails was more than 11 carbons, the more
lipophilic properties had a detrimental effect on the anti-
bacterial activity. In particular, 11q showed no bacterial inhi-
vancomycin (MIC
= =
1.0 μg mL−1, molecular weight
1485.71), an agent of last resort for the treatment of severe
MRSA infections; thus, finding alternative therapeutic op-
tions is critical to address the burden of vancomycin.
2.2.3. The antibacterial spectrum evaluation of RDSB ana-
logue 11k. Motivated by the success of 11k with promising
anti-MRSA activity, an effort to assess its antibacterial spec-
trum was further conducted against a broad array of clinically
relevant multidrug-resistant MRSA isolates and common bac-
teria. As expected, 11k successfully maintained its potent
antibacterial activity with MICs identical to or 0.5 to 2-fold
higher than that reported for MRSA JCSC 4788 against all of
the tested MRSA isolates ((JCSC 2172), (JCSC 3063), (JCSC
4469), (JCSC 4744), (NCTC 10442), (N 315), and (85/2082)),
and even against VRE (No. 151458137) at MIC = 2.0 μg mL−1
(Table 2), which strongly suggests that the RDSB analogue
11k can satisfactorily overcome its cross-resistance with com-
monly used clinical antibiotics (Table 2). Moreover, 11k also
exhibited potent activity against all other tested Gram-
positive bacterial strains such as S. aureus (ATCC 6538), S.
This journal is © The Royal Society of Chemistry 2018
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
Table 2 Antibacterial spectrum of 11k against various bacteria
Table 3 The anti-MRSA selectivity versus antiproliferative (IC50) activity
against RAW 264.7, MCF-7 and SF-268 cells
MIC (μg mL−1
)
RAW 264.7
MCF-7
SF-268
Bacteria
G+/G− RDSB 11k
Vancomycin
IC50 (μg mL−1
)
5.6 0.42
22.4
9.98 1.08
39.9
10.15 0.98
40.6
S. aureus (ATCC 29213)
S. aureus (ATCC 6538)
S. aureus (CMCC 26003)
MRSA (JCSC 4788)
MRSA (JCSC 2172)
MRSA (JCSC 3063)
MRSA (JCSC 4469)
MRSA (JCSC 4744)
MRSA (NCTC 10442)
MRSA (N315)
MRSA (85/2082)
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G+
G−
G−
1.0
1.0
0.50
1.0
0.50
0.50
0.50
0.50
1.0
1.0
1.0
0.50
0.50
1.0
0.50
0.25
0.25
0.50
0.25
0.25
0.25
1.0
1.0
0.50
1.0
1.0
Selectivity ratioa
a
Selectivity ratio = IC50/MIC against MRSA (JCSC 4788).
1.0
0.50
mL−1, respectively. These positive results indicated that 11k
could emerge as a promising candidate and merits further
consideration for the development of novel anti-MRSA thera-
peutic agents.
0.125 1.0
0.50
0.25
0.25
0.50
0.25
0.25
0.25
2.0
>32
>32
>32
>32
>32
>32
1.0
1.0
1.0
B. cereus (ATCC 10876)
P. acnes (ATCC 6919)
E. faecalis (ATCC 29212)
S. epidermidis (ATCC 12228)
VRE (No. 151458137)
B. polymyxa (GIM 1.467)
B. polymyxa (ATCC 842)
E. coli (ATCC 8739)
S. typhi (CMCC 44102)
0.50
0.50
0.50
0.50
>32
>32
>32
>32
>32
>32
>32
2.2.5. Time-kill assay of RDSB analogue 11k against MRSA
(JCSC 4788) cells. Glycopeptides such as vancomycin and oxa-
zolidinones such as linezolid are well-known as two corner-
stone clinical antimicrobials responsive for the treatment of
systemic MRSA infections, but both of them have unavoid-
able drawbacks influencing their clinical efficiency, namely
linezolid being a bacteriostatic agent and slow bactericidal
action for vancomycin,26 which usually result in intractable
problems for patient rehabilitation or clinical failure in many
cases.27,28 In order to confirm the bactericidal activity of the
new RDSB analogues, the most promising compound 11k
was further evaluated against MRSA (JCSC 4788) cells in com-
parison with the known bactericidal agent vancomycin using
a standard time-kill kinetics assay ranging from 0.5 to 4.0×
MIC. As presented in Fig. 1, time-kill kinetic experiments
dosing with 0.5 and 1.0× MIC suppressed more than 90% of
bacterial growth within 8 h, revealing a considerable bacteri-
cidal effect. When higher concentrations of 11k (2.0 and 4.0×
MIC) were tested, the time-kill kinetic experiments unambig-
uously disclosed that 11k exhibited a very rapid in vitro bacte-
ricidal activity (killing >90% MRSA cells within 2.0 and 0.5 h
respectively) and a complete bactericidal effect with a sharp
reduction in CFU of 99.9% after 2 and 4 h of exposure, re-
spectively. Meanwhile, vancomycin only achieved a 99.9% re-
duction at the same concentration in more than 8 h. In this
respect, RDSB analogue 11k exhibited a distinctly superior
advantage over vancomycin because of its rapid time-kill ef-
fect. This experimental evidence indicated that the RDSB ana-
logue 11k mainly exerted bactericidal activity at a dose of
more than 2-fold the MIC with rapid efficiency.
1.0
2.0
>32
>32
>32
>32
>32
S. dysenteriae (CMCC 51252) G−
C. albicans (ATCC 10231)
Fungi >32
aureus (CMCC 26003), B. cereus (ATCC 10876), P. acnes (ATCC
6919), E. faecalis (ATCC 29212), and S. epidermidis (ATCC
12228) with MIC values ranging from 0.125 to 0.5 μg mL−1,
demonstrating superior potency to that of the control drug
vancomycin (Table 2) and a widely diverse antibacterial spec-
trum. On the other hand, when the Gram-positive bacteria B.
polymyxa (GIM 1.467) and B. polymyxa (ATCC 842), the
Gram-negative bacteria E. coli (ATCC 8739), S. typhi (CMCC
44102) and S. dysenteriae (CMCC 51252), and fungus C.
albicans (ATCC 10231) were tested, 11k showed no noticeable
biological effects (MIC >32 μg mL−1); this collectively indi-
cated that 11k is a promising selective antibacterial agent. B.
polymyxa represents an exception to the general rule that
Gram-positive bacteria have a low content of PE compared
with Gram-negative bacteria,25 and the difference in the lipid
composition of the bacterial membranes between specific
and non-specific bacteria will be helpful for finding molecu-
lar targets of 11k in future work.
2.2.4. Toxicity analysis of potent RDSB analogue 11k
against mammalian cells. To further evaluate the selectivity
of antibacterial versus antiproliferative activities, the most ac-
tive RDSB analogue 11k was next subjected to a preliminary
cytotoxicity profiling. Hence, compound 11k was evaluated
against the RAW 264.7 cell line and the cancer cell lines
MCF-7 and SF-268 to determine the potential toxic effect
in vitro after 24 h of exposure. As depicted in Table 3, the
IC50 of 11k against RAW cells was 5.6 0.42 μg mL−1, which
was more than 20-fold higher than the MIC value (0.25 μg
mL−1) obtained against MRSA. Compound 11k exhibited in-
significant toxicity to RAW 264.7 cells when 0.25 μg mL−1 11k
was used to kill MRSA cells. The IC50 of 11k against cancer
cells MCF-7 and SF-268 was 9.98 1.08 and 10.15 0.98 μg
2.3. Bactericidal mechanism study
2.3.1. The effect on the MRSA membrane probed by
DiSC3(5) and SYTOX Green assay. In order to understand the
anti-bacterial mode of action for RDSB analogue 11k at the
morphological level and shed light on the mechanism under-
lying the antimicrobial activity against the clinical MRSA iso-
late (JCSC 4788), a mechanistic investigation using fluores-
cent probes was initiated combining the well-known
fluorescence agents DiSC3(5) and SYTOX Green through fluo-
rescence imaging technology.16,29 As shown in Fig. 2A, after
treating the MRSA (JCSC 4788) cells with serial
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2018

View Article Online
MedChemComm
Research Article
Fig. 1 Time-kill kinetics of MRSA (JCSC 4788). A) Vancomycin, MIC value (μg mL−1): vancomycin = 1.0; B) 11k, MIC value (μg mL−1): 11k = 0.25.
Fig. 2 Hyperpolarization of the bacterial cytoplasmic membrane. A) The fluorescence of the dye DiSC3(5) within a MRSA (JCSC 4788) bacterial
suspension was monitored with a spectrophotometer. B) The use of SYTOX Green demonstrated that treatment with over 1× MIC RDSB 11k for 3 h
induced membrane disruption of MRSA (JCSC 4788). In contrast, no increase in fluorescence was observed when vancomycin was added. Melittin
was used as a positive control. Error bars represent the standard deviation, n = 3. C) Compound 11k compromised the inner membrane of MRSA
and caused significant increases in fluorescence emission due to SYTOX Green binding to nucleic acids. D) In contrast, no significant fluorescence
emission was observed, when MRSA was treated with vancomycin.
concentrations of 11k (0.5× MIC, 1× MIC, 2× MIC, 4× MIC, 8×
MIC), the depolarization of the bacterial membrane potential
was intermediately induced within 1 min at 0.5–2× MIC, be-
cause an obvious increase of fluorescence was observed.
However, a significant decrease in the fluorescence intensity
was observed after more than 2 min at 1–2× MIC and in the
whole process at 4–8× MIC, suggesting that the favorable bac-
tericidal action of 11k might be attributed to a decrease of
the membrane potential due to its hyperpolarization.
Moreover, the SYTOX Green assay indicated that RDSB an-
alogue 11k showed an excellent ability to disrupt and
permeabilize the bacterial inner membrane, since an obvious
increase in fluorescence intensity was detected when the
MRSA suspension was treated even with 1× MIC for only 3 h.
In comparison, the positive control melittin also caused an
unambiguous increase in fluorescence in the whole concen-
tration range from 0.25 to 8 μg mL−1 after 3 h, whereas the
negative control vancomycin failed (Fig. 2B). Moreover, this
This journal is © The Royal Society of Chemistry 2018
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
aforementioned conclusion was further confirmed by the ob-
servation of the staining of 11k-treated bacteria using fluores-
cence microscopy. As a result, a large number of stained bac-
teria were observed in a culture incubated with 11k, clearly
clarifying that it compromised the living bacterial cytoplas-
mic membrane. Hence, these informative results collectively
pointed to a rational conclusion that the RDSB analogue 11k
resulted in Gram-positive bacterial death by disrupting their
membranes.
2.3.2. The visualization of cells by SEM. The morphologi-
cal changes of the MRSA (JCSC 4788) cells after treatment
with RDSB analogue 11k were then examined using scanning
electron microscopy (SEM), and the result was in good agree-
ment with that of the fluorescence probes, wherein RDSB an-
alogue 11k successfully disrupted the cytoplasmic membrane
structure of MRSA (JCSC 4788) cells following treatment with
4× MIC of 11k for 3 h. As shown in Fig. 3, the bacterial mem-
brane exhibited extensive damage, and almost all of the
tested MRSA cells were obviously distorted with intracellular
components released. However, the morphology of the
untreated cells remained intact with smooth spheres. The
findings of rapid structural effects and apparent membrane
changes also agreed well with the results of the time-kill as-
says. Based on the results of the membrane hyperpolarization
and membrane disruption assays, in conjunction with the
time-killing study (killing >99.9% after 4 h), it can be con-
cluded that RDSB analogue 11k acts rapidly as a bactericidal
antibiotic to induce cell lysis by causing membrane hyperpo-
larization and disruption.
important role in minimizing the development of drug-
resistance because of its rapid bactericidal action through
membrane disruption. As expected, when a multipassage re-
sistance study was conducted with serial passaging in the
presence of sub-MIC levels of antimicrobial drug 11k, no de-
tectable resistance was observed even after 20 serial passages
(Fig. 4). However, in the equivalent case of ofloxacin, the MIC
value began to increase after 5 passages and had increased
16 times after 20 passages. Therefore, this result strongly
demonstrated that RDSB analogue 11k resulted in significant
suppression of MRSA drug-resistance activity and was suit-
able for further clinical applications.
2.5. The in vivo efficacy of RDSB analogue 11k in MRSA-
infected mice
Motivated by its potent bactericidal activity and attractive se-
lectivity profiles, RDSB analogue 11k was then evaluated in a
mouse skin infection model against MRSA (JCSC 4788) to as-
sess its in vivo antibacterial efficacy. Compared to untreated
controls, a single dose of RDSB analogue 11k or vancomycin
was observed to significantly prevent skin ulcer formation
(Fig. 5A).
In addition, RDSB analogue 11k treated animals showed
significantly lower incidence of infection-induced morbidity,
as evaluated by weight loss (Fig. 5B) and wound size
(Fig. 5C). These initial investigations clearly proved that
RDSB analogue 11k had significant therapeutic efficacy to
suppress the virulence of MRSA in vivo.
3. Conclusion
2.4. Multistep drug-resistance assay
In summary, we report the synthesis and anti-MRSA activity
of a pool of twenty-two synthetic natural product-based RDSB
analogues with enhanced structural diversity, synthesized
through a new strategy of modifying the pendant acyl moiety.
The SAR study against the MRSA (JCSC 4788) strain clearly
suggested that the presence of an optical hydrophobic acyl
tail in the phloroglucinol scaffold was a prerequisite for the
anti-bacterial activity of the RDSB analogues. The reasons un-
derlying the aforementioned observation might be attributed
The emerging drug-resistance has been considered as one of
the most intractable problems for the treatment of infectious
diseases; however, a membrane-targeting agent would be very
unlikely to lead to resistance, since the membrane-targeted
mechanism would require the bacteria to change their entire
membrane chemistry, which would almost certainly lead to
the cells being unviable. With this scenario, the membrane-
targeting action of RDSB analogue 11k is expected to play an
Fig. 3 A) SEM of MRSA treated with 11k. B) SEM of untreated MRSA.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2018

View Article Online
MedChemComm
Research Article
bacterial anti-MRSA activity and many other admirable fea-
tures, including breadth of the antibacterial spectrum, rapid-
ity of bactericidal action, and excellent membrane selectivity.
Moreover, the investigation of its biophysical and morpholog-
ical effects revealed that 11k exerted a bactericidal action on
MRSA strains by membrane hyperpolarization to induce
membrane disruption, which represents a very favorable bac-
tericidal mechanism, rendering it a very promising candidate
for further extensive medicinal chemistry investigations with
an aim towards antibiotic drug discovery with novel chemical
scaffolds and mechanisms of action.
Fig. 4 Resistance acquisition during serial passaging in the presence
of sub-MIC levels of antimicrobial drugs. Ofloxacin was used as a posi-
tive control.
4. Experimental section
4.1. Materials and experimental apparatus
All of the reactions were carried out under nitrogen atmo-
sphere with dry solvents under anhydrous conditions, unless
otherwise noted. Reagents were purchased at high commer-
cial quality and used without further purification. Thin-layer
chromatography (TLC) was conducted with 0.25 mm
Tsingdao silica gel plates (60 F-254) and visualized by expo-
sure to UV light (254 nm) or staining with potassium per-
manganate. Silica gel (ZCX-II, 200–300 mesh) used for flash
to these substituents increasing the ability of the compounds
to penetrate the bacterial cell wall and distort the MRSA cell
membrane. Among them, RDSB analogue 11k was identified
as a very promising lead compound with significant anti-
bacterial activity against a wide panel of MRSA strains with
similar activity to the reference antibiotic vancomycin. Most
importantly, compound 11k possessed profound in vivo anti-
Fig. 5 Evaluation of RDSB analogue 11k in a mouse model of skin infection. A) BALB/c mice were intradermally challenged with an inoculum
mixture containing 1 × 108 CFU of MRSA (JCSC 4788) along with 40 μg of the vehicle control (DMSO) or 40 μg of 11k or vancomycin.
Representative images of the resulting cutaneous injuries sustained in the antibiotic and vehicle control mice are shown for 3 days and 8 days
post-infection (scale in cm), respectively. B) A single 40 μg dose of 11k or vancomycin profoundly attenuated dermatopathology following cutane-
ous MRSA (JCSC 4788) challenge. C) A single 40 μg dose of 11k or vancomycin reduced morbidity as measured by animal weight. Error bars repre-
sent the standard deviation, n = 6.
This journal is © The Royal Society of Chemistry 2018
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
column chromatography was purchased from Qing Dao Hai
CDCl3): δ 1.33 (s, 6H), 1.42 (s, 6H), 2.57 (s, 3H); 13C NMR
(125 MHz, CDCl3): δ 23.8, 24.3, 27.4, 52.0, 56.7, 109.4, 196.7,
199.1, 201.7, 210.0.
1
Yang Chemical Industry Co. of China. H NMR and 13C NMR
spectra were recorded on a Brüker Advance 500 instrument
(1H: 500 MHz, 13C: 125 MHz). Chemical shifts were reported
in parts per million relative to CDCl3 (1H NMR; 7.27 ppm,
13C NMR; 77.00 ppm). The following abbreviations were used
to explain the multiplicities: s = singlet, d = doublet, t = trip-
let, q = quartet, m = multiplet, br = broad. Mass spectromet-
ric data were obtained using an ABI-Q Star Elite high resolu-
tion mass spectrometer. Dichloroethane (DCE) was distilled
from calcium hydride. Yields refer to chromatographically
purified products unless otherwise stated.
4.2.4. Synthesis of syncarpic acid 7. A flame-dried 50 mL
flask was charged with 6 N HCl (30 mL) and 6 (3.17 g, 14.2
mmol), and the reaction mixture was stirred vigorously under
reflux for 24 h. The mixture was cooled to room temperature
and extracted with EtOAc (5 × 50 mL). The combined organic
phases were washed with brine (30 mL), dried over Na2SO4,
and filtered. Removal of the solvent by rotary evaporation and
purification by flash column chromatography (silica gel, hex-
ane : EtOAc = 1 : 1) afforded 7 (2.17 g, 11.4 mmol, 80% yield).
1H NMR (500 MHz, CDCl3): ketone: δ 1.31 (s, 12H), 3.61 (s, 2H);
enol: δ 1.40 (s, 12H), 5.74 (br d, J = 2.3 Hz, 1H), 8.00 (br s, 1H);
13C NMR (125 MHz, CDCl3): ketone: δ 21.8, 50.2, 59.1, 204.3,
208.9; enol: δ 24.5, 51.2, 59.1, 101.7, 191.9, 204.3, 212.6.
4.2. Chemical synthesis
4.2.1. General procedure for the synthesis of compounds
5d–5k. Aluminum trichloride (4 g, 40 mmol) was slowly and
carefully added to a solution of phloroglucinol 3 (1.26 g, 10
mmol) in CH2ClCH2Cl/PhNO2 (1 : 1, 50 mL) at 0 °C. After stir-
ring at this temperature for 10 min under nitrogen atmo-
sphere, acid chloride 4 (12 mmol) was added. Then, the ice
bath was removed, and the mixture was stirred at 80 °C for 3 h.
The crude reaction mixture was cooled to room temperature
and quenched with water (100 mL). The mixture was extracted
with EtOAc (5 × 100 mL), washed with brine, and concentrated
under vacuum. The crude product was purified by flash chro-
matography (silica gel, hexane : EtOAc = 2 : 1) to provide the rel-
ative products 5d–5k with yields ranging from 60–80%. Their
spectra data were referenced to the reported relative references.
4.2.2. General procedure for the synthesis of intermediate
compounds 5l–5v. To a solution of phloroglucinol 4 (630 mg,
5.0 mmol) in acid chloride 5 (6.0 mmol), was added methane-
sulfonic acid (1.45 g, 15 mmol). After stirring at room tem-
perature for 10 min under nitrogen atmosphere, the mixture
was stirred at 60–80 °C for another 1 h. Then, the crude mix-
ture was cooled to room temperature and quenched with wa-
ter (20 mL). The mixture was extracted with EtOAc (5 × 25
mL), washed with brine, and concentrated under vacuum.
The crude product was purified by flash chromatography (sil-
ica gel, hexane : EtOAc = 2 : 1) to provide compounds 5l–5v
with yields ranging from 25–80%. Their spectra data were
referenced to the reported relative references.
4.2.5. Synthesis of 2,2,4,4-tetramethyl-6-(isopentylidene)
cyclohexane-1,3,5-trione 9. Syncarpic acid 7 (182 mg, 1.0
mmol) was dissolved in CH2Cl2 (8 mL), and isovaleraldehyde
8 (172 mg, 2.0 mmol) and proline (12 mg, 0.1 mmol) were
added. The resulting mixture was stirred for 30 min at room
temperature, then purified by 3 cm long flash chromatography
(silica gel, CH2Cl2) to afford the desired product 9 (250 mg,
100% yield) as a colorless oil. 1H NMR (500 MHz, CDCl3): δ
0.95 (d, J = 6.7 Hz, 6H), 1.30 (s, 6H), 1.31 (s, 6H), 1.89 (m, J =
6.7 Hz, 1H), 2.59 (dd, J = 3.0, 3.0 Hz, 2H), 7.51 (dd, J = 3.0 Hz,
1H); 13C NMR (125 MHz, CDCl3): δ 21.9, 22.3, 22.6, 28.7, 38.9,
57.9, 58.6, 133.1, 159.1, 196.4, 199.5, 208.8. 42.6.
4.2.6. General procedure for the synthesis of acyl-
phloroglucinol analogues 10. Sodium hydride (80 mg, 2.0
mmol, 60% in mineral oil) was carefully added to a solution
of acylphloroglucinol 5 (1.0 mmol) in THF (10 mL), and then
the unsaturated triketone 9 (125 mg, 0.5 mmol) in THF (4
mL) was slowly added. The resulting mixture was stirred for
0.5 h at room temperature, and then quenched with 1 N HCl
(5 mL) and extracted with EtOAc (3 × 10 mL), washed with
brine, and concentrated under vacuum. The crude product
was purified by flash chromatography (silica gel, hexane :
EtOAc = 10 : 1 to 2 : 1) to provide derivatives of the general
structure 10 with yields ranging from 30% to 80%.
4.2.7. General procedure for the synthesis of RDSB ana-
logues 11. Acylphloroglucinol analogues 10 (0.5 mmol) were
added to a solvent of 5 mL toluene and heated to reflux, and
PTSA (0.05 mmol) was then added. The reaction mixture was
further stirred at that temperature for about 1 h until all of
the starting materials disappeared by TLC detection. After
cooling to room temperature, the mixture was directly puri-
fied through a short flash chromatography (silica gel; hex-
ane : EtOAc = 5 : 1) to afford the desired products 11a–11v as
slightly brown solids. The NMR spectra of products 11a–11v
can be seen in the ESI.†
4.2.3.
Synthesis
of
4-acyl-5-hydroxy-2,2,6,6-tetra-
methylcyclohex-4-ene-1,3-dione 6. Sodium methoxide (7.18 g,
133 mmol) was slowly dissolved in anhydrous methanol (60
mL) at 0 °C. To this clear solution, acylphloroglucinol 5c
(2.75 g, 16.5 mmol) was added carefully and the mixture was
stirred for 10 min under nitrogen atmosphere. After that,
methyl iodide (14.2 mL, 228 mmol) was slowly added. The ice
bath was then removed, and the mixture was stirred at room
temperature for 24 h. The crude mixture was quenched with
2 N HCl (60 mL) and extracted with CHCl3 (5 × 60 mL),
washed with brine, and concentrated under vacuum. The
crude product was purified by flash chromatography (silica
gel, hexane : EtOAc = 5 : 1) to provide 6 (3.17 g, 14.2 mmol,
86% yield) as a yellow rod-like crystal. 1H NMR (500 MHz,
4.3. Animals
Six-week-old female BALB/c mice from Guangdong Provincial
Laboratory Animal Public Service Center were housed at 23
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2018

View Article Online
MedChemComm
Research Article
1 °C with a 12 h light–dark cycle with free access to standard
pellet chow and water and allowed to acclimatize to the labo-
ratory environment for 7 days in propylene cages. The animal
studies were conducted according to the guidelines
established by the National Institutes of Health for the care
and use of laboratory animals, and the recommendations of
the Animal Care and Use policies at the South China Agricul-
tural University (Guangzhou, China). Procedures concerning
animal treatments were reviewed and approved by the Animal
Use Ethics Committee of South China Agricultural University
(No. SCAU2017-C003).
4.6. Mode of antimicrobial action of 11l
4.6.1. Cytoplasmic membrane depolarization analysis. The
membrane potential of MRSA (JCSC 4788) with 11k was in-
vestigated using a modified method.32 Briefly, exponential
phase cultures of MRSA (JCSC 4788) were washed and ad-
justed to an optical density at 620 nm (OD620) of 0.1 with a
buffer solution (5 mM HEPES at pH 7). 0.4 μM 3,3′-dipropyl-
thiacarbocyanine [DiSC3(5)] as well as 100 mM KCl were
added to the cell suspensions and incubated for 20 min at 37
°C with agitation. The desired concentration of 11k was
added, and the change in fluorescence was monitored at an
excitation wavelength of 660 nm and at an emission wave-
length of 670 nm.
4.4. Bacteria strains and cells
4.6.2. SYTOX Green assay. This assay was performed
according to the reported method.33 Briefly, exponential
phase cultures of MRSA (JCSC 4788) were washed and ad-
justed to an optical density at 620 nm (OD620) of 0.4 with a
0.85% NaCl buffer. The bacterial suspension was incubated
with serial concentrations of RDSB for a desired time, and
then treated with 3 μM of SYTOX Green for 20 min. The fluo-
rescence reading was monitored at an excitation wavelength
of 504 nm and at an emission wavelength of 523 nm.
Melittin, a cell lytic factor, was used as a positive control.
4.6.3. Visualization of bacterial membrane permeability.
The MRSA cell suspension (OD620 0.4) was treated with the
desired concentrations of 11k and vancomycin for 3 h, and
then incubated with 3 μM of SYTOX Green. Finally, the sus-
pension was fixed on glass slides and then observed by a
fluorescence microscope (ZEISS Model Axioplan 2IE) with an
excitation wavelength of 485 nm.
4.6.4. Scanning electron microscopy. For observation
using scanning electron microscopy, exponential-phase bacte-
ria were treated with the compounds at 8× MIC for 4 h at 37
°C. Then, the cells were washed twice, suspended in PBS,
and prefixed in 0.1 M phosphate buffer (pH 7.2) containing
2% glutaraldehyde and 2.5% paraformaldehyde overnight. Af-
ter washing 6 times with 0.1 M phosphate buffer, the sam-
ples were post-fixed in 1% osmium tetroxide for 2 h. After
washing another three times with 0.1 M phosphate buffer,
the samples were dehydrated through a graded ethanol series
and subjected to freeze-drying (JFD-310, JEOL, Japan). After-
wards, the samples were mounted on stubs and coated with
gold in a sputter coater (JFC-1600, JEOL, Japan), and then ex-
amined and photographed under a scanning electron micro-
scope (JSM-6360LV, JEOL, Japan).
S. aureus (ATCC 29213), S. aureus (ATCC 6538), S. aureus
(CMCC 26003), B. cereus (ATCC 10876), P. acnes (ATCC 6919),
E. faecalis (ATCC 29212), S. epidermidis (ATCC 12228), B. poly-
myxa (GIM 1.467), B. polymyxa (ATCC 842), E. coli (ATCC
8739), S. typhi (CMCC 44102), S. dysenteriae (CMCC 51252),
and C. albicans (ATCC 10231) were obtained from Guangdong
Institute of Microbiology (Guangzhou, China). Vancomycin-
resistant Enterococcus faecium (VRE, No. 151458137) isolated
from the abdominal cavity of patients was donated by the
first affiliated hospital of Sun Yat-Sen University, Guangzhou,
China. The SCCmec-carrying clinical isolates of MRSA (JCSC
4788, JCSC 2172, JCSC 3063, JCSC 4469, JCSC 4744, NCTC
10442,
N 315,3 85/2082) were kindly provided by K.
Hiramatsu and T. Ito from Juntendo University (Tokyo, Ja-
pan). RAW 264.7 cells were obtained from Kunming Institute
of Zoology, Chinese Academy of Sciences (Kunming, China).
MCF-7 and SF-268 cancer cells were obtained from the Cell
Bank of the Chinese Academy of Sciences (Shanghai, China).
4.5. Activity evaluation
4.5.1. Minimal inhibitory concentration (MIC) assays. MIC
determinations were performed by Microbroth dilution in
Mueller–Hinton broth medium (MHB) according to CLSI30
guidelines. Furthermore, P. acnes was cultured in the MGC
Anaero Pack. Resazurin was added as a visible indicator
according to the established method.31
4.5.2. The selectivity over cytotoxic activity study. The se-
lectivity over cytotoxic activity study was performed using the
MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bro-
mide] assay. 11k was applied at various concentrations (rang-
ing from 1.5625 to 100 μg mL−1) and the control cells were
treated with DMSO at the highest concentration used in the
test wells (0.5%).
4.5.3. Time-killing assays. Time-kill assays were carried
out based on the guidelines of CLSI.30 The overnight culture
of cells (MRSA JCSC 4788) was adjusted in a 0.85% NaCl
buffer to obtain a bacterial suspension with 106 to 107 CFU
mL−1. The inoculums were treated with various concentra-
tions (0.5×, 1×, 2×, and 4× MIC) of 11k and vancomycin at 35
°C. Aliquots were removed at 0.5 h, 2 h, 4 h, 8 h, and 24 h to
measure viable plate counts. Similar results were obtained by
two independent time-kill tests.
4.7. Resistance study
Resistance was defined as an increase of 4-fold more than
the initial MIC.34 For resistance research by sequential pas-
saging, MRSA (JCSC 4788) cells incubated in the exponential
phase at 37 °C were passaged into fresh MHB containing 11k
at a sub-inhibitory concentration, and then re-passaged at 24
h intervals until at least 20 passages were completed. The
MIC was determined as described above.
This journal is © The Royal Society of Chemistry 2018
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
4.8. Mouse skin infected model
11 D. Hardej, C. R. Ashby Jr and N. S. Khadtare, et al., Eur. J.
Med. Chem., 2010, 45, 5827–5832.
12 F. R. DeLeo, M. Otto and B. N. Kreiswirth, et al., Lancet,
2010, 375, 1557–1568.
13 A. Anandan, G. L. Evans and M. L. O'Mara, et al., Proc. Natl.
Acad. Sci. U. S. A., 2017, 114, 2218–2223.
14 H. X. Liu, W. M. Zhang and Z. F. Xu, et al., RSC Adv.,
2016, 6, 25882–25886.
15 H. X. Liu, H. B. Tan and S. X. Qiu, et al., J. Asian Nat. Prod.
Res., 2016, 18, 535–541.
16 J. J. Koh, S. M. Lin and T. T. Aung, et al., J. Med. Chem.,
2015, 58, 739–752.
17 H. B. Tan, H. X. Liu and L. Y. Zhao, et al., Eur. J. Med.
Chem., 2017, 125, 492–499.
18 S. Dachriyanus, M. V. Sargent and B. W. Skelton, et al., Aust.
J. Chem., 2002, 55, 229–232.
0.1 mL of MRSA (JCSC 4788) cells (1 × 108 CFU per mouse)
and either 11k (40 μg diluted in DMSO) or vancomycin (40 μg
diluted in DMSO) were injected intradermally into the ab-
dominal skin of matched BALB/c mice (female, 18–20 g per
mouse). The negative control mice were infected with DMSO
as a vehicle. The body weight and lesion size of each mouse
were determined once per day for 11 days after infection.
4.9. Data analysis
All assays were carried out in at least triplicate, and all of the
results were found to be reproducible. The cytotoxicity assay,
DiSC3(5) assay, SYTOX Green assay, and in vivo antimicrobial
activity results were expressed in terms of the mean stan-
dard deviation (SD).
19 S. Leejae, L. Hasap and S. P. Voravuthikunchai, J. Med.
Microbiol., 2013, 62, 421–428.
20 S. Leejae, P. W. Taylor and S. P. Voravuthikunchai, J. Med.
Microbiol., 2013, 62, 78–85.
Conflicts of interest
The authors declare no conflict of interest.
21 M. Morkunas, L. Dube, F. Götz and M. E. Maier,
Tetrahedron, 2013, 69, 8559–8563.
22 H. X. Liu, K. Chen and Y. Yuan, et al., Org. Biomol. Chem.,
2016, 14, 7354–7360.
23 I. R. Green, F. E. Tocoli and S. H. Lee, et al., Med. Chem.,
2007, 15, 6236–6241.
24 The biophysical processes mainly referred to that the
hydrophilic head moiety binds with an intermolecular
hydrogen bond like a ‘hook’ attaching itself to a hydrophilic
portion of the membrane, after which the hydrophobic tail
portion of the molecule is then able to enter into the
membrane lipid bilayer (ref. 23).
Acknowledgements
We thank the National Natural Science Foundation of China
(NSFC) (No. 81773602, 31600271, 81502949), the Natural Sci-
ence Foundation of Guangdong Province (2015A030310482,
2016A030313149, 2016A010105015), the Strategic Resources
Service Network Program on Plant Genetic Resources Innova-
tion of the Chinese Academy of Sciences (No. ZSZC-005), Pearl
River Science and Technology New Star Fund of Guangzhou
(201605120849569), and the Frontier Science Key Program of
Chinese Academy of Sciences (No. QYZDB-SSW-SMC018) for
financial support.
25 R. M. Epand, S. Rotem and A. Mor, et al., J. Am. Chem. Soc.,
2008, 130, 14346–14352.
26 S. R. Singh, A. E. Bacon and D. C. Young, et al., Antimicrob.
Agents Chemother., 2009, 53, 4495–4497.
References
1 D. M. Brogan and E. Mossialos, Global. Health, 2016, 12,
1–7.
27 P. Vergidis, M. S. Rouse and G. Euba, et al., Antimicrob.
Agents Chemother., 2011, 55, 1182–1186.
2 R. M. Klevens, M. A. Morrison and J. Nadle, et al., JAMA,
2007, 98, 1763–1771.
28 I. H. Eissa, H. Mohammad and O. A. Qassem, et al., Eur. J.
Med. Chem., 2017, 130, 73–85.
3 G. Taubes, Science, 2008, 321, 356–361.
29 J. J. Koh, S. X. Qiu and H. X. Zou, et al., Biochim. Biophys.
Acta, 2013, 1828, 834–844.
30 CLSI, Performance standards for antimicrobial susceptibility
testing, twenty-first informational supplement CLSI document,
2011, M100–S21.
31 A. J. Drummond, Recent Res. Dev. Phytochem., 2000, 4,
143–152.
32 M. Wu, E. Maier and R. Benz, et al., Biochemistry, 1999, 38,
7235–7242.
4 Centers for Disease Control and Prevention, Antibiotic
Resistance Threats in the United States, 2013.
5 H. F. Chambers, N. Engl. J. Med., 2005, 352, 1485–1487.
6 G. J. Moran, A. Krishnadasan and R. J. Gorwitz, et al.,
N. Engl. J. Med., 2006, 355, 666–674.
7 B. W. Frazee, J. Lynn and E. D. Charlebois, et al., Ann.
Emerg. Med., 2005, 45, 311–320.
8 I. M. Gould, Int. J. Antimicrob. Agents, 2009, 34, S2–S5.
9 L. L. Han, L. K. McDougal and R. J. Gorwitz, et al., J. Clin.
Microbiol., 2007, 45, 1350–1352.
33 R. Rathinakumar, W. F. Walkenhorst and W. C. Wimley,
J. Am. Chem. Soc., 2009, 131, 7609–7617.
10 R. Goldburg, S. Roach and D. Wallinga, et al., Science,
2008, 321, 1294.
34 D. J. Farrell, M. Robbins and W. Rhys-Williams, et al.,
Antimicrob. Agents Chemother., 2011, 55, 1177–1181.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2018