Red Light-Triggered Intracellular Carbon Monoxide Release Enables Selective Eradication of MRSA Infection

Article abstract of DOI:10.1002/anie.202104024

Carbon monoxide (CO) is an important gaseous signaling molecule. The use of CO-releasing molecules such as metal carbonyls enables the elucidation of the pleiotropic functions of CO. Although metal carbonyls show a broad-spectrum antimicrobial activity, i

Full text of DOI:10.1002/anie.202104024

Angewandte
Research Articles
Chemie
How to cite:
International Edition: doi.org/10.1002/anie.202104024
German Edition: doi.org/10.1002/ange.202104024
Antibacterial Agents
Hot Paper
Red Light-Triggered Intracellular Carbon Monoxide Release Enables
Selective Eradication of MRSA Infection
Jian Cheng, Guihai Gan, Zhiqiang Shen, Lei Gao, Guoying Zhang, and Jinming Hu*
Abstract: Carbon monoxide (CO) is an important gaseous
signaling molecule. The use of CO-releasing molecules such as
metal carbonyls enables the elucidation of the pleiotropic
functions of CO. Although metal carbonyls show a broad-
spectrum antimicrobial activity, it remains unclear whether the
bactericidal property originates from the transition metals or
the released CO. Here, we develop nonmetallic CO-releasing
micelles via a photooxygenation mechanism of 3-hydroxyfla-
vone derivatives, enabling CO release under red light irradi-
ation (e.g., 650 nm). Unlike metal carbonyls that non-specif-
ically internalize into both Gram-positive and Gram-negative
bacteria, the nonmetallic micelles are selectively taken up by S.
aureus instead of E. coli cells, exerting a selective bactericidal
effect. Further, we demonstrate that the CO-releasing micelles
can cure methicillin-resistant S. aureus (MRSA)-infected
wounds, simultaneously eradicating MRSA pathogens and
accelerating wound healing.
that it was the transition metals rather than the released CO
exerting the antibacterial effect. Indeed, a recent study
concluded that it was the thiol-reactive Ru2+ ions instead of
CO release that underlined the antibacterial property.[7]
Therefore, we surmised that it would be necessary to develop
metal-free CO-releasing platforms, which may be beneficial
to exclude the influence of metal ions and to clarify the
antibacterial capacity of CO.
In the context of nonmetallic CORMs, boranocarbon-
ates,[8] Diels–Alder reaction-based CO donors,[9] and several
photoresponsive CORMs (photoCORMs) have been success-
fully developed.[10] Among them, photoCORMs enabled
localized CO delivery by taking advantage of the spatiotem-
poral precision of light stimulus. Typical nonmetallic photo-
CORMs include 3-hydroxyflavone (3-HF),[11] cyclic dike-
tone,[12] xanthene-9-carboxylic acid,[13] and meso-carboxyl
BODIPY derivatives,[14] exhibiting triggered CO release
under specific light irradiation (Scheme 1a). Remarkably,
although metal-free photoCORMs have revealed great
potentials in anti-inflammatory and anticancer applica-
tions,[9b,11c] the antibacterial performance of nonmetallic
photoCORMs have far less been explored. Since bacterial
pathogens were inherently sensitive to ultraviolet (UV)
irradiation, it would be necessary to activate photoCORMs
under long-wavelength illumination with decreased photo-
toxicity. Unfortunately, many of previous photoCORMs were
Introduction
Apart from being an air pollutant, carbon monoxide (CO)
is also an endogenous gaseous singling molecule in living
organisms,[1] exerting physiological functions such as anti-
inflammatory, anti-apoptotic, and anti-cancer effects.[2] Car-
bon monoxide-releasing molecules (CORMs) such as metal
carbonyls are developed for safe and convenient delivery of
CO, and these CO-releasing prodrugs render it possible to
investigate and understand the physiological functions of
CO.[3] Although metal carbonyls provide a robust tool to
locally deliver CO and show therapeutic benefits similar to
that of CO gas, the use of metal carbonyls to evaluate the
antibacterial activity of CO leads to ambiguous conclusions.
Specifically, previous studies revealed that the most exten-
sively used CO-releasing metal carbonyls such as CORM-2
(tricarbonyldichlororuthenium(II) dimer) and CORM-3 (tri-
carbonylchloro(glycinato)ruthenium (II)) showed a broad-
spectrum antibacterial activity.[4] However, a saturated solu-
tion of CO gas had a negligible effect on inhibiting bacterial
growth.[5] Giving that no CO release was detected in
biological fluids of CORM-3,[6] these results likely implied
[*] J. Cheng, G. Gan, Z. Shen, L. Gao, Prof. Dr. G. Zhang, Prof. Dr. J. Hu
CAS Key Laboratory of Soft Matter Chemistry
Hefei National Laboratory for Physical Science at the Microscale
Department of Polymer Science and Engineering
University of Science and Technology of China
Hefei 230026, Anhui (China)
Scheme 1. a) Representative metal-free photoCORMs with typical acti-
vation wavelengths and CO yields. b) Illustration of the fabrication of
metal-free CO-releasing micelles by integrating 3-HF derivatives and
tetraphenylporphyrin (TPP) photosensitizer into the micelle cores,
enabling CO release under red light irradiation via a photooxygenation
mechanism, exerting a selective bactericidal effect toward Gram-
positive bacteria.
E-mail: jmhu@ustc.edu.cn
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202104024.
&&&&
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
primarily activated by detrimental UV or near-UV light with
poor tissue penetration,[15] and it remained a grand challenge
to develop photoCORMs capable of releasing CO under red
or near-infrared light irradiation.[16] Considering the facile
preparation, quantitative CO yield, and low toxicity of the
photolyzed products of 3-HF-based CO releasers, we envi-
sioned that 3-HF derivative might be an ideal candidate for
metal-free CORMs to evaluate the antibacterial capacity of
CO. Nevertheless, it was impractical to directly activate 3-HF
moieties under red light irradiation owing to the weak
absorbance. It was worth noting that the CO release from 3-
HF derivatives involved the oxygenation of excited 3-HF
derivatives by ground oxygen (3O2).[11d] We hypothesized it
was possible to oxidize 3-HF derivatives through a photo-
oxygenation mechanism using more oxidative singlet oxygen
(1O2) in the presence of a specific photosensitizer with long-
wavelength absorbance, thereby enabling CO release via an
indirect approach under red light irradiation.[17]
In this work, we report a metal-free CO-releasing plat-
form by integrating 3-HF derivatives and tetraphenylpor-
phyrin (TPP) moieties into the cores of micellar nano-
particles, serving as the CO donor and photosensitizer,
respectively. The formation of micellar nanoparticles not
only increased the water-dispersity but also locally concen-
trated 3-HF derivatives and the photosensitizer. When
exposed to 650 nm light irradiation, the excited TPP photo-
sensitizer converted 3O2 to 1O2 that spontaneously oxidized 3-
HF derivatives, resulting in CO release (Scheme 1b). The
CO-releasing micelles were selectively internalized by Gram-
positive Staphylococcus aureus (S. aureus) instead of Gram-
negative Escherichia coli (E. coli) bacteria, and the red light-
triggered intracellular CO release exerted an excellent
bactericidal effect against S. aureus. Moreover, in vivo studies
revealed that the CO-releasing micelles could efficiently
eradicate methicillin-resistant S. aureus (MRSA) bacteria and
accelerate MRSA-infected wounds in a full-thickness skin
wound model.
Scheme 2. Chemical structures of HF, TPP-HF, TPP-CHF, and TPP-Flav
diblock copolymers used in this study.
known as inhibitors of radicals, the HFM and FlavM
monomer containing a phenolic group at the 3-position could
be directly polymerized through RAFT polymerization.
With the synthesized block copolymers in hand, we first
investigated their self-assembly behavior in aqueous solu-
tions. TPP-HF, TPP-CHF, and HF block copolymers self-
assembled into micellar nanoparticles with hydrodynamic
diameters, hDhi, of ca. 20–60 nm (Figure S12). Upon exposure
to mild red light irradiation (650 nm, 26 mWcmꢀ2), no
significant changes in hDhi were observed for all three
micelles, whereas a decrease in scattering intensities was
observed for TPP-HF micelles (Figure 1a–c, S12). To under-
stand this phenomenon, we monitored the UV/Vis absorb-
ance spectra of all three micelles under 650 nm light
irradiation. We found that a continuous decrease centered
at 356 nm was only observed for TPP-HF micelles (Fig-
ure 1d), whereas TPP-CHF micelles with caged 3-HF moi-
eties and HF micelles without the labeling of TPP photo-
sensitizers did not reveal noticeable changes in UV/Vis
spectra (Figure S13). The apparent photolysis constant, kobs
,
of TPP-HF micelles was determined to be 0.08 minꢀ1 under
650 nm light irradiation (Figure S13d and Table S2). Remark-
ably, when TPP-HF micelles were exposed to 365 nm light
irradiation, the kobs value was increased to 0.151 minꢀ1
(Figure S14), which was likely ascribed to the higher absorb-
ance at 365 nm (Figure 1d). However, the kobs of HF micelles
without TPP photosensitizer dropped to 0.034 minꢀ1 under
identical 365 nm light irradiation (Figure S14 and Table S2),
demonstrating that the photolysis of 3-HF moieties proceed-
ed more efficiently in the presence of TPP photosensitizer
under 650 nm light irradiation.
Results and Discussion
Starting from 4-hydroxybenzaldehyde, both 3-HF-based
monomers, N-(2-(4-(3-hydroxy-4-oxo-4H-chromen-2-yl)phe-
noxy) ethyl)methacrylamide (HFM) and N-(2-(4-(3-hydroxy-
4-oxo-4H-benzo[g]chromen-2-yl)phenoxy)ethyl)methacryla-
mide (FlavM), were successfully synthesized using a similar
procedure. Moreover, the phenolic group of HFM could be
further modified with the formation of caged HF monomer
(CHFM) in the presence of benzyl bromide (Supporting
Information, Scheme S1). Also, we synthesized TPP photo-
sensitizer-based monomer (TPPM; Scheme S2) and HF, TPP-
HF, TPP-CHF, and TPP-Flav block copolymers through
reversible addition-fragmentation chain transfer (RAFT)
polymerization (Scheme S3).[18] All the precursors, targeted
monomers, and block copolymers were thoroughly charac-
terized (Figures S1–S11). The chemical structures and the
structural parameters of the as-synthesized diblock copoly-
mers are shown in Scheme 2 and Table S1, respectively.
Remarkably, although phenol-containing compounds were
Using the allyl-Flu (AFCO) probe,[19] red light-mediated
CO release from TPP-HF micelles was confirmed by the
fluorescence increase (Figure 1e). Moreover, the CO release
process was monitored by a portable CO detector (Drꢀger
Pac6500), revealing that a higher CO yield under 650 nm
irradiation than that of 365 nm irradiation (Figure 1 f).
Specifically, the CO release amounts were calculated to be
1.26 mmol and 1.06 mmol under 650 and 365 nm irradiation for
90 min, corresponding to ca. 69% and ca. 58% of the 3-HF
moieties, respectively (Figure 1 f). However, no CO release
was detected for TPP-CHF and HF micelles under identical
red light irradiation (Figure 1 f), in good agreement with the
negligible changes in UV/Vis spectra (Figure 1d and S13).
Taken together, although 3-HF moieties showed negligible
absorbance at 650 nm, the incorporation of TPP photosensi-
tizer and 3-HF moieties into the cores of micellar nano-
particles enabled the photooxygenation of 3-HF moieties,
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 1. a),b) TEM images and c) intensity-average hydrodynamic diameter distributions of TPP-HF micelles with 650 nm irradiation for 0–
120 min. d) UV/Vis spectra of TPP-HF micelles (0.1 gLꢀ1) under 650 nm irradiation. e) Evolution of fluorescence emission spectra (lex =490 nm)
of TPP-HF micelles with allyl-Flu probe (AFCO; 5 mM) and PdCl2 (5 mM) under 650 nm irradiation. f) CO release profiles of TPP-HF, TPP-CHF,
and HF micelles (0.1 gLꢀ1) in PBS under varying irradiation conditions (650 nm, 26 mWcmꢀ2 or 365 nm, 6.8 mWcmꢀ2). g) Fluorescence intensity
of SOSG at 533 nm in the presence of TPP-HF, TPP-HF, and HF micelles (0.1 gLꢀ1) with 650 nm irradiation (26 mWcmꢀ2). h) CO release profiles
of DMF solution of HF (0.2 mgmLꢀ1) in the presence of pre-irradiated PEP (1.0 mgmLꢀ1) under dark condition. Pre-treated PEP was irradiated
with 650 nm light for 60 min in the presence of TPPM (0.2 mgmLꢀ1). i) CO release profiles of TPP-Flav micelles (0.1 gLꢀ1) in PBS under 650 nm
irradiation. In all cases, the irradiation intensity was 26 mWcmꢀ2
.
resulting in CO release under red light irradiation. It should
be mentioned that the control experiments revealed that the
coexistence of free phenolic group and TPP photosensitizer
was indispensable to enable the 3-HF derivatives to be
activated by red light irradiation because neither TPP-CHF
with caged 3-HF moieties nor HF micelles without TPP
photosensitizer could release CO under otherwise identical
conditions.
ure 1g). These results corroborated the involvement of
oxygen in the photolysis process. To further support this
claim, we synthesized PYD diblock copolymer capable of
1
capturing and releasing O2. The conversion of pyridone to
endoperoxide moieties under 650 nm light irradiation in the
presence of TPP photosensitizer was monitored by UV/Vis
1
spectra. The formation of O2-releasing endoperoxide-con-
taining PEP diblock copolymers was evidenced by the
absorbance decrease at 302 nm (Figure S17).[20] Upon incu-
bation of HF micelles with PEP block copolymers, sponta-
neous CO release was detected again even under dark
conditions (Figure 1h). In addition, we found that TPP-Flav
micelles with extended 3-HF moieties underwent red light-
triggered CO release as well, suggesting a general procedure
for the activation of 3-HF derivatives (Figure 1i). Also, we
analyzed the photolyzed product of HFM monomer in the
presence of TPP photosensitizer under 650 nm light irradi-
ation, revealing the formation of 2-(benzoyloxy)benzoic acid
derivatives by HRMS (Figure S18), which was identical to the
photolyzed product under UV light irradiation.[11e] Building
on the above results, we surmised that, under 650 nm light
irradiation, the excited TPP photosensitizer produced 1O2
To clarify the CO release mechanism under red light
irradiation, we deoxygenated the TPP-HF micelles by three
freeze-pump-thaw cycles and exposed the degassed micelles
to the same light source. There were no remarkable changes
in the UV/Vis spectra (Figure S15), implying the involvement
of oxygen in the photolysis process. Moreover, the generation
1
of O2 under 650 nm light irradiation was observed for both
1
TPP-HF and TPP-CHF micelles, as probed by O2-specific
Singlet Oxygen Sensor Green (SOSG). The fluorescence
intensity of SOSG was stronger for TPP-CHF than that of
TPP-HF micelles with a similar TPP concentration, which was
tentatively ascribed to the consumption of 1O2 by 3-HF
moieties (Figure 1g and S16). By contrast, no 1O2 was
detected for HF micelles without TPP photosensitizer (Fig-
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
within micellar nanoparticles, which in turn oxidized the 3-HF
after 10 min irradiation and only a mild decrease (ca. 20%) in
moieties in close proximity within the cores, thereby trigger-
ing CO release under red light irradiation. Moreover, the
photooxidation process under red light irradiation (e.g.,
650 nm) gave rise to more efficient CO release in terms of
faster release kinetic and higher CO yield, which was quite
beneficial for potential biomedical applications with de-
creased phototoxicity and increased CO production.
cell viability after 30 min irradiation (Figure 2a). Although
1O2 has been extensively used for bacterial killing,[21] the
compromised bactericidal effect of TPP-CHF was likely due
to the short working distance of 1O2 (e.g., ca. 10–20 nm). The
generated 1O2 had to diffuse out from the inner cores of TPP-
CHF micelles (ca. 60 nm in diameter, Figure S12) before
taking any action on bacterial killing,[22] which may surpass
1
Next, we investigated whether the CO-releasing micelles
had an antimicrobial capacity. Although irradiation of Gram-
positive S. aureus bacteria with 650 nm (26 mWcmꢀ2) did not
affect the bacterial viability within 30 min (Figure 2a), we
found that the viability of S. aureus bacteria drastically
decreased (ca. 92%) after 10 min irradiation in the presence
of TPP-HF micelles, and no bacterial colonies were observed
after 20 min irradiation (Figure 2a). Notably, although HF
block copolymer could release CO under 365 nm irradiation
(Figure 1 f), we found that both S. aureus and E. coli cells
were sensitive to 365 nm irradiation even at a very low light
intensity (e.g., 2.2 mWcmꢀ2) for 20 min (Figure S19), which
was unfavorable for examining the antibacterial capacity of
CO due to the inherent phototoxicity. As such, the current
photooxidation strategy enabling CO release under 650 nm
irradiation was quite advantageous to evaluate the antimicro-
bial performance of CO due to decreased phototoxicity.
Notably, although TPP-CHF micelles produced cytotoxic
1O2 (Figure 1g), there was no appreciable bacterial death
the working distance of O2. Therefore, the micelle formula-
tion was advantageous not only for the photooxidation of 3-
HF moieties within micellar cores but also for minimizing the
1
interferences of O2. Notably, no antimicrobial capacity was
1
observed for HF micelles that released neither CO nor O2
under 650 nm light irradiation (Figure 1g). Moreover, TPP-
HF micelles showed a concentration-dependent bactericidal
performance under 650 nm light irradiation. Impressively, at
a micelle concentration of 0.025 gLꢀ1, ca. 98% bacteria were
killed after 30 min irradiation, showing an excellent antimi-
crobial performance of TPP-HF micelles (Figure 2b). By
sharp contrast, none of TPP-HF, TPP-CHF, or HF micelles
led to statistically significant decreases in bacterial viability of
E. coli after 650 nm irradiation for 30 min (Figure 2c,d).
These results demonstrated that the CO-releasing micelles
exerted a selective antimicrobial property toward S. aureus
bacteria. However, previous studies revealed a broad-spec-
trum bactericidal effect of metal carbonyls toward both
Gram-positive and Gram-negative bacteria.[4a,d]
To understand the selective bactericidal effect of TPP-HF
micelles toward S. aureus cells, we examined the bactericidal
performance of pre-irradiated TPP-HF micelles, revealing
bacterial killing capacity toward neither S. aureus nor E. coli
(Figure S20). This result likely suggested that photo-triggered
CO release was crucial for the antibacterial performance.
However, we found that the incubation of S. aureus and E.
coli cells in saturated CO solution (ca. 1 mM; much higher
than that of the release CO contents) for 30 min did not
induce bacterial death, regardless of 650 nm light irradiation
(Figure S21), in line with previous results.[5] The contradictory
results revealed that the localized delivery of CO produced
distinct antibacterial outcomes in comparison with direct CO
administration. A similar phenomenon was also observed for
the discrepant antimicrobial performance between CO gas
solution and CO-releasing metal carbonyls.[4b,f,g]
Previous results demonstrated that the antimicrobial
effects of metal carbonyls primarily originated from the
accumulated metal ions in cells rather than released CO.[4b,7]
We then evaluated the bacterial internalization performance
of CO-releasing micellar nanoparticles. Zeta-potential meas-
urements revealed that TPP-HF, TPP-CHF, and HF micelles
were negatively charged with zeta potentials of ꢀ32, ꢀ21, and
ꢀ26 mV, respectively (Figure S22a). Upon incubation of both
S. aureus and E. coli bacteria with micellar nanoparticles, no
significant changes in zeta potentials were observed either
(Figure S22b). This was in sharp contrast to cationic polymer-
based antibacterial agents that exerted an antibacterial
activity by disrupting bacterial membranes through electro-
static interaction between negatively charged bacterial mem-
branes and positively charged polymers.[23] Interestingly,
albeit negatively charged, we found that TPP-HF micelles
Figure 2. a) Irradiation time-dependent bacterial viability of S.aureus
after treatment with TPP-CHF, HF, and TPP-HF micelles (0.1 gLꢀ1
)
with 650 nm light irradiation (26 mWcmꢀ2) for 0–30 min, respectively.
##
p<0.01, in comparison with the non-irradiated group; ** p<0.01,
*** p<0.001, in comparison with the non-irradiated group; n.s., not
significant. b) Bacterial viability of S.aureus in the presence of varying
concentrations of TPP-HF micelles with or without 650 nm light
irradiation 30 min (26 mWcmꢀ2). *** p<0.001, **** p<0.0001, n. s.
not significant, in comparison with the PBS group with irradiation.
c) E. coli viability in the presence of TPP-CHF, HF, and TPP-HF
micelles (0.1 gLꢀ1) with or without 650 nm light irradiation
(26 mWcmꢀ2) for 30 min. n.s., not significant. d) Bacterial viability of
E. coli in the presence of varying concentrations of TPP-HF micelles
with or without 650 nm light irradiation 30 min (26 mWcmꢀ2). n.s.,
not significant.
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 3. CLSM images of a) S. aureus and b) E. coli bacteria incubated with TPP-HF (0.1 gLꢀ1) under varying conditions. The green channel was
excited at 488 nm and collected at 500–600 nm. The red channel was excited at 594 nm and collected at 630–700 nm. c) SEM images of S. aureus
and E. coli cells in the presence of TPP-HF, TPP-CHF, or HF micelles with or without 650 nm light irradiation for 30 min.
could be selectively taken up by S. aureus, as evidenced by the
appearance of red fluorescence of TPP moieties within S.
aureus cells. Moreover, red light-triggered CO release within
S. aureus cells could be detected by the CO-specific AFCO
probe as well (Figure 3a).[19] Nevertheless, TPP-HF micelles
cannot be internalized into E. coli cells and no intracellular
CO release was detected (Figure 3b). The different bacterial
uptake performance was likely due to the distinct membrane
structures of S. aureus and E. coli cells.[24] Scanning electron
microscopy (SEM) revealed no remarkable changes in cell
morphologies for both S. aureus and E. coli (Figure 3c),
demonstrating that the selective bacterial killing performance
was not ascribed to the disruption of cellular membranes, in
agreement with the negatively charged nature of micellar
nanoparticles. Therefore, we concluded that the localized CO
release within S. aureus bacteria played a crucial role in
bacterial killing, while the detailed antimicrobial mechanism
needs to be further elucidated. Remarkably, the discrimina-
tive antibacterial capacity is advantageous for inhibiting the
development of drug resistance in bacteria.[25]
Giving that CO-releasing TPP-HF micelles can selectively
kill S. aureus cells under red light irradiation, we further
investigated the CO-releasing micelles for the treatment of
methicillin-resistant S. aureus (MRSA) infection in a full-
thickness skin wound model by taking advantage of the
multiple roles of CO in terms of bacterial eradication and
accelerating wound healing.[3] MRSA serves as the main
reason for skin infections and it may cause pneumonia and
other severe diseases, representing a huge threat to human
health.[26] In vitro antibacterial assay revealed that TPP-HF
micelles (0.1 gLꢀ1) eradicated ca. 97% MRSA cells after
30 min irradiation, whereas neither TPP-CHF nor HF under
otherwise identical conditions gave rise to statistically sig-
nificant decreases in bacterial viability (Figure S23). There-
fore, CO-releasing TPP-HF micelles could be potentially used
for the treatment of MRSA infection.
Next, skin wounds (ca. 5 mm in diameter) were made by
a surgical blade and the wounds were inoculated with MRSA
cells.[27] After 6 h of infection, the infected wounds were then
treated with PBS (negative control), vancomycin (positive
control), and TPP-CHF and TPP-HF micelles with or without
650 nm irradiation (Figure 4a). HF micelles released neither
CO nor 1O2 were excluded for in vivo antibacterial evaluation
(Figure 2). The wounds were treated with 25 mL of micelles
following light irradiation (650 nm, 30 min) once per day and
the infected areas were continuously monitored within 9 days
post-treatment. Notably, the wounds of the TPP-HF micelle-
treated group with irradiation completely healed after 9 days,
whereas other groups including vancomycin-treated one
showed unhealed wounds (Figure 4b,c). Moreover, the bac-
terial amounts on days 1, 3, and 5 post-treatment were
checked by standard colony-forming unit (cfu) count, reveal-
ing the least bacterial burden for TPP-HF micelle-treated
group (Figures 4d,e and S24), in good agreement with the
faster wound healing performance (Figure 4b,c). In vitro
scratch assay further supported that the red light-triggered
CO release from TPP-HF micelles boosted the wound healing
(Figure S25), likely due to the inherent property of CO on
accelerating wound healing. Hematoxylin and eosin (H&E)-
and Sirius red-staining and CD31-immunostaining analysis
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Figure 4. a) Experimental outline of in vivo antibacterial assessment in an MRSA-infected wound healing model. b) Representative skin wound
images at 1, 3, 5, 7, 9 days and c) Quantitative analysis of the residual wounded areas receiving different treatments (PBS, TPP-HF(-hv), TPP-
HF(+hv), TPP-CHF(-hv), TPP-CHF(+hv), and vancomycin). *, p<0.05, **p<0.01, **** p<0.0001, compared with PBS group. d) Bacterial
colony-forming units obtained from different tissues of mice with varying treatments.***, p<0.001, ###, p<0.001, &&&&, p<0.0001, compared
with PBS groups on days 1, 3, 5, respectively. e) Photographs of bacterial colonies on the agar plates of the wounded tissues receiving PBS and
TPP-HF (+hv) treatments. f) Histologic analysis of the MRSA-infected wounds receiving different treatments. The scale bar is 250 mm.
revealed that TPP-HF micelles remarkably decreased the
infiltration of inflammatory cells, facilitating angiogenesis and
collagen deposition, showing better antibacterial and wound
healing performance than that of vancomycin and other
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Flavell, A. M. K. Choi, Nat. Med. 2000, 6, 422 – 428; e) S. Y. Tao,
J. Cheng, G. Su, D. Li, Z. Q. Shen, F. H. Tao, T. You, J. M. Hu,
Angew. Chem. Int. Ed. 2020, 59, 21864 – 21869; Angew. Chem.
2020, 132, 22048 – 22053; f) J. Cheng, B. Zheng, S. Cheng, G. Y.
Zhang, J. M. Hu, Chem. Sci. 2020, 11, 4499 – 4507; g) K. M.
Dillon, R. J. Carrazzone, J. B. Matson, K. Kashfi, Biochem.
Pharmacol. 2020, 176, 113931.
control groups (Figure 4 f and S26). Besides, cell viability test
against normal mammalian cells (using mouse fibroblast L929
cells as an example) revealed that none of TPP-HF, TPP-
CHF, and HF micelles showed obvious toxicity at a micelle
concentration of 0.1 gLꢀ1 (Figure S27). Moreover, hemolysis
assay demonstrated that the negatively charged TPP-HF,
TPP-CHF, and HF micelles did not lead to detectable
hemolysis either, regardless of 650 nm light irradiation (Fig-
ure S28). Therefore, the nonmetallic CO-releasing micelles
may provide an alternative strategy to combat MRSA
infection with low toxicity and hemolysis toward normal cells.
[3] R. Motterlini, L. E. Otterbein, Nat. Rev. Drug Discovery 2010, 9,
728-U724.
[4] a) L. S. Nobre, J. D. Seixas, C. C. Romao, L. M. Saraiva, Anti-
microb. Agents Chemother. 2007, 51, 4303 – 4307; b) S. McLean,
R. Begg, H. E. Jesse, B. E. Mann, G. Sanguinetti, R. K. Poole,
Antioxid. Redox Signaling 2013, 19, 1999 – 2012; c) M. Klinger-
Strobel, S. Glaser, O. Makarewicz, R. Wyrwa, J. Weisser, M. W.
Pletz, A. Schiller, Antimicrob. Agents Chemother. 2016, 60,
4037 – 4046; d) K. S. Davidge, G. Sanguinetti, C. H. Yee, A. G.
Cox, C. W. McLeod, C. E. Monk, B. E. Mann, R. Motterlini,
R. K. Poole, J. Biol. Chem. 2009, 284, 4516 – 4524; e) D. Nguyen,
T. K. Nguyen, S. A. Rice, C. Boyer, Biomacromolecules 2015, 16,
2776 – 2786; f) J. L. Wilson, H. E. Jesse, B. Hughes, V. Lund, K.
Naylor, K. S. Davidge, G. M. Cook, B. E. Mann, R. K. Poole,
Antioxid. Redox Signaling 2013, 19, 497 – 509; g) M. Desmard, R.
Foresti, D. Morin, M. Dagoussat, A. Berdeaux, E. Denamur,
S. H. Crook, B. E. Mann, D. Scapens, P. Montravers, J. Bocz-
kowski, R. Motterlini, Antioxid. Redox Signaling 2012, 16, 153 –
163; h) J. S. Ward, J. M. Lynam, J. Moir, I. J. S. Fairlamb, Chem.
Eur. J. 2014, 20, 15061 – 15068; i) C. S. Bang, R. Kruse, K.
Johansson, K. Persson, BMC Microbiol. 2016, 16, 64.
[5] L. K. Wareham, R. K. Poole, M. Tinajero-Trejo, J. Biol. Chem.
2015, 290, 18999 – 19007.
Conclusion
We have successfully developed a new strategy to activate
CO-releasing 3-HF moieties through a photooxygenation
mechanism, enabling CO release under red light irradiation.
This nonmetallic CO-releasing micelle eliminates the inter-
ference of transition metal ions, representing an ideal tool to
evaluate the antibacterial capacity of CO. Our results
suggested that the CO-releasing micelles could be selectively
taken up by S. aureus rather than E. coli cells, enabling
intracellular CO release under red light irradiation and
exerting a selective antimicrobial effect toward only S. aureus
bacteria. In vivo study revealed that the CO-releasing
micelles could be used for both eradicating MRSA pathogens
and accelerating MRSA-infected wound healing. This work
not only provides a new strategy to activate photoCORMs but
also sheds light on the crucial role of intracellular CO delivery
on antibacterial capacity.
[6] T. Santos-Silva, A. Mukhopadhyay, J. D. Seixas, G. J. L. Ber-
nardes, C. C. Romao, M. J. Romao, J. Am. Chem. Soc. 2011, 133,
1192 – 1195.
[7] H. M. Southam, T. W. Smith, R. L. Lyon, C. Liao, C. R. Trevitt,
L. A. Middlemiss, F. L. Cox, J. A. Chapman, S. F. El-Khamisy,
M. Hippler, M. P. Williamson, P. J. F. Henderson, R. K. Poole,
Redox Biol. 2018, 18, 114 – 123.
[8] R. Alberto, K. Ortner, N. Wheatley, R. Schibli, A. P. Schubiger,
J. Am. Chem. Soc. 2001, 123, 3135 – 3136.
Acknowledgements
[9] a) X. Y. Ji, K. Damera, Y. Q. Zheng, B. C. Yu, L. E. Otterbein,
B. H. Wang, J. Pharm. Sci. 2016, 105, 406 – 416; b) Y. Q. Zheng,
X. Y. Ji, B. C. Yu, K. L. Ji, D. Gallo, E. Csizmadia, M. Y. Zhu,
M. R. Choudhury, L. K. C. De La Cruz, V. Chittavong, Z. X.
Pan, Z. N. Yuan, L. E. Otterbein, B. H. Wang, Nat. Chem. 2018,
10, 787 – 794.
The financial supports of the National Key R&D Program of
China (2020YFA0710700), the National Natural Scientific
Foundation of China (NNSFC) Projects (51690150, 51690154,
52073270, 51722307, and 51973206), and the Fundamental
Research
Funds
for
the
Central
Universities
(WK2060190102) is gratefully acknowledged.
[10] a) N. Abeyrathna, K. Washington, C. Bashur, Y. Liao, Org.
ˇ
Biomol. Chem. 2017, 15, 8692 – 8699; b) L. Stackovꢂ, M. Russo,
ˇ
L. Muchovꢂ, V. Orel, L. Vꢁtek, P. Stacko, P. Klꢂn, Chem. Eur. J.
2020, 26, 13184 – 13190.
Conflict of interest
[11] a) L. S. Lazarus, A. D. Benninghoff, L. M. Berreau, Acc. Chem.
Res. 2020, 53, 2273 – 2285; b) Y. Li, Y. Z. Shu, M. W. Liang, X. L.
Xie, X. Y. Jiao, X. Wang, B. Tang, Angew. Chem. Int. Ed. 2018,
57, 12415 – 12419; Angew. Chem. 2018, 130, 12595 – 12599; c) M.
Popova, T. Soboleva, S. Ayad, A. D. Benninghoff, L. M. Berreau,
J. Am. Chem. Soc. 2018, 140, 9721 – 9729; d) S. N. Anderson,
J. M. Richards, H. J. Esquer, A. D. Benninghoff, A. M. Arif,
L. M. Berreau, ChemistryOpen 2015, 4, 590 – 594; e) T. Sobole-
va, H. J. Esquer, A. D. Benninghoff, L. M. Berreau, J. Am.
Chem. Soc. 2017, 139, 9435 – 9438.
The authors declare no conflict of interest.
Keywords: antibacterial agents · carbon monoxide · MRSA ·
photooxidation · wound healing
[1] A. Verma, D. J. Hirsch, C. E. Glatt, G. V. Ronnett, S. H. Snyder,
Science 1993, 259, 381 – 384.
[12] P. Peng, C. M. Wang, Z. Shi, V. K. Johns, L. Y. Ma, J. Oyer, A.
Copik, R. Igarashi, Y. Liao, Org. Biomol. Chem. 2013, 11, 6671 –
6674.
[13] L. A. P. Antony, T. Slanina, P. Sebej, T. Solomek, P. Klan, Org.
Lett. 2013, 15, 4552 – 4555.
[2] a) C. C. Rom¼o, W. A. Blꢀttler, J. D. Seixas, G. J. L. Bernardes,
Chem. Soc. Rev. 2012, 41, 3571 – 3583; b) S. Garcꢁa-Gallego,
G. J. L. Bernardes, Angew. Chem. Int. Ed. 2014, 53, 9712 – 9721;
Angew. Chem. 2014, 126, 9868 – 9877; c) U. Hasegawa, A. J.
van der Vlies, E. Simeoni, C. Wandrey, J. A. Hubbell, J. Am.
Chem. Soc. 2010, 132, 18273 – 18280; d) L. E. Otterbein, F. H.
Bach, J. Alam, M. Soares, H. T. Lu, M. Wysk, R. J. Davis, R. A.
[14] E. Palao, T. Slanina, L. Muchova, T. Solomek, L. Vitek, P. Klan,
J. Am. Chem. Soc. 2016, 138, 126 – 133.
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
[15] A. Lipovsky, Y. Nitzan, A. Gedanken, R. Lubart, Lasers Surg.
Y. Y. Yang, Nat. Chem. 2011, 3, 409 – 414; b) S. K. Samal, M.
Med. 2010, 42, 467 – 472.
[16] S. H. C. Askes, G. U. Reddy, R. Wyrwa, S. Bonnet, A. Schiller, J.
Am. Chem. Soc. 2017, 139, 15292 – 15295.
[17] a) S. L. Studer, W. E. Brewer, M. L. Martinez, P. T. Chou, J. Am.
Chem. Soc. 1989, 111, 7643 – 7644; b) M. Russo, P. Stacko, D.
Nachtigallova, P. Klan, J. Org. Chem. 2020, 85, 3527 – 3537.
[18] N. Corrigan, K. Jung, G. Moad, C. J. Hawker, K. Matyjaszewski,
C. Boyer, Prog. Polym. Sci. 2020, 111, 101311.
Dash, S. Van Vlierberghe, D. L. Kaplan, E. Chiellini, C. van
Blitterswijk, L. Moroni, P. Dubruel, Chem. Soc. Rev. 2012, 41,
7147 – 7194; c) T. K. Nguyen, S. J. Lam, K. K. K. Ho, N. Kumar,
G. G. Qiao, S. Egan, C. Boyer, E. H. H. Wong, ACS Infect. Dis.
2017, 3, 237 – 248.
[24] a) T. J. Beveridge, J. Bacteriol. 1999, 181, 4725 – 4733; b) H.
Strahl, J. Errington, Annu. Rev. Microbiol. 2017, 71, 519 – 538.
[25] a) M. Perros, Science 2015, 347, 1062 – 1064; b) R. J. Melander,
D. V. Zurawski, C. Melander, MedChemComm 2018, 9, 12 – 21.
[26] N. A. Turner, B. K. Sharma-Kuinkel, S. A. Maskarinec, E. M.
Eichenberger, P. P. Shah, M. Carugati, T. L. Holland, V. G.
Fowler, Nat. Rev. Microbiol. 2019, 17, 203 – 218.
[27] M. Zhou, Y. X. Qian, J. Y. Xie, W. J. Zhang, W. N. Jiang, X. M.
Xiao, S. Chen, C. Z. Dai, Z. H. Cong, Z. M. Ji, N. Shao, L. Q. Liu,
Y. Q. Wu, R. H. Liu, Angew. Chem. Int. Ed. 2020, 59, 6412 –
6419; Angew. Chem. 2020, 132, 6474 – 6481.
[19] S. M. Feng, D. D. Liu, W. Y. Feng, G. Q. Feng, Anal. Chem. 2017,
89, 3754 – 3760.
[20] a) J. H. Zou, J. W. Zhu, Z. Yang, L. Li, W. P. Fan, L. C. He, W.
Tang, L. M. Deng, J. Mu, Y. Y. Ma, Y. Y. Cheng, W. Huang, X. C.
Dong, X. Y. Chen, Angew. Chem. Int. Ed. 2020, 59, 8833 – 8838;
Angew. Chem. 2020, 132, 8918 – 8923; b) S. Benz, S. Notzli, J. S.
Siegel, D. Eberli, H. J. Jessen, J. Med. Chem. 2013, 56, 10171 –
10182.
[21] A. Hirao, S. Sato, D. Saitoh, N. Shinomiya, H. Ashida, M. Obara,
Photochem. Photobiol. 2010, 86, 426 – 430.
[22] J. Moan, K. Berg, Photochem. Photobiol. 1991, 53, 549 – 553.
[23] a) F. Nederberg, Y. Zhang, J. P. K. Tan, K. J. Xu, H. Y. Wang, C.
Yang, S. J. Gao, X. D. Guo, K. Fukushima, L. J. Li, J. L. Hedrick,
Manuscript received: March 22, 2021
Accepted manuscript online: April 7, 2021
Version of record online: && &&
,
&&&&
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Research Articles
Antibacterial Agents
Metal-free CO-releasing micelles can be
selectively taken up by Gram-positive
bacteria, exerting a narrow-spectrum
bactericidal activity toward only Gram-
positive bacteria under red light irradi-
ation. The CO-releasing micelles synerg-
istically eradicate MRSA pathogens and
accelerate MRSA-infected wound healing.
J. Cheng, G. Gan, Z. Shen, L. Gao,
G. Zhang, J. Hu*
&&&& — &&&&
Red Light-Triggered Intracellular Carbon
Monoxide Release Enables Selective
Eradication of MRSA Infection
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!