Exploiting a bacterial drug-resistance mechanism: A light-activated construct for the destruction of MRSA

Article abstract of DOI:10.1002/anie.200804804

(Figure Presented) A cunning and dangerous plan foiled! An enzyme-specific molecular construct exploits the overexpression of β-lactamase in several drug-resistant bacteria. Specific photodynamic toxicity was detected towards β-lactam-resistant methicilli

Full text of DOI:10.1002/anie.200804804

Communications
DOI: 10.1002/anie.200804804
Antimicrobial Photochemistry
Exploiting a Bacterial Drug-Resistance Mechanism: A Light-Activated
Construct for the Destruction of MRSA**
Xiang Zheng, Ulysses W. Sallum, Sarika Verma, Humra Athar, Conor L. Evans, and
Tayyaba Hasan*
The prevalence of bacterial drug resistance makes it imper-
ative to develop new targeted strategies that can be used
either as monotherapies or in conjunction with existing
antibiotic regimens. Photodynamic therapy (PDT) has such
potential. PDT is a photochemistry-based emerging technol-
ogy that relies on the wavelength-specific light activation of
certain nontoxic chemicals (photosensitizers, PSs) to form
active molecular species (AMSs) that are toxic to surrounding
biological targets.[1] The reported effectiveness of PDTagainst
pathogens in general, and methicillin-resistant Staphylococ-
cus aureus (MRSA) in particular,[2] makes it a potentially
powerful technology for the treatment of drug-resistant
infections. AMSs have multiple cellular targets, in contrast
to conventional antibiotics, such as b-lactams and amino-
glycocides,[3] which inhibit the activity of single enzymes. This
multifaceted nature of PDT action has the advantage that it
decreases the probability of generating PDT-resistant strains
of bacteria; however, this feature can also be a limitation,
owing to nonspecific PS accumulation, which results in
damage to healthy host tissue.[4]
The aim of this study was to exploit a bacterial drug-
resistance mechanism to activate the PS locally, only at the
site of infection, for a more specific PDT effect. The strategy
involves the synthesis of a construct which can be activated by
light and which recognizes a molecular target that is unique to
the bacterium of interest. The advantage of this approach is
that it enables much-enhanced selectivity, as the construct can
only be activated by light after interaction with the molecular
target. The construct can not be activated at any other site.
This enhanced selectivity could enable the use of PDT more
broadly for regional infections than is currently possible; at
present, PDT can only be used for highly localized infections.
This study focuses on the b-lactamase enzyme as the
molecular target. One way in which bacteria resist the
action of b-lactam antibiotics is through the production of
b-lactamase, which cleaves the b-lactam ring hydrolytically.[5]
In the case of cephalosporins, ring opening of the b-lactam is
accompanied by the release of the substituent at the 3’-
position. Zlokarnik et al. used this feature to design b-
lactamase reporter systems to detect gene-promoter activa-
tion in mammalian cells.[6,7]
In this study, we targeted the b-lactamase expressed by
MRSA. In designing the molecular construct (b-lactamase-
enzyme-activated photosensitizer, b-LEAP), we took advant-
age of the photophysical phenomenon known as quenching.
PSs can be quenched when in close proximity to each other.
As a result, the probability of a PS excited-state transition is
diminished, which leads to decreased fluorescence or AMS
formation. b-LEAP is designed in such a way that, upon
cleavage by b-lactamase (Scheme 1), the PS is released from
homodimeric, ground-state quenching to yield an enzyme-
specific, light-activated antimicrobial action. The b-lacta-
mases were ideal targets for this proof-of-principle study
owing to the prevalence of b-lactamase expression among
bacteria and its high enzymatic efficiency.[8,9]
The PS 5-(4’-carboxybutylamino)-9-diethylaminoben-
zo[a]phenothiazinium chloride (EtNBS-COOH), an EtNBS
derivative, was demonstrated previously to be a potent
antimicrobial agent.[10] In the current study, the free terminal
carboxy group of the PS was conjugated to a cephalosporin
derivative, 7-amino-3-chloromethyl-3-cephem-4-carboxylic
acid p-methoxybenzyl ester (ACLE), which contains a b-
lactam ring. Thus, ACLE was modified with two primary
amino groups and treated with EtNBS-COOH (Scheme 2).
The final product 3 (b-LEAP) was purified and characterized
by HPLC, mass spectrometry, and NMR spectroscopy (see
Figures 1 and 3 in the Supporting Information). b-LEAP
showed a nearly fivefold decrease in fluorescence emission
(excitation at 625 nm) relative to EtNBS-COOH (Figure 1).
This result indicates the quenching effect of the two PS
moieties in the molecule. There are two mechanisms of PS
quenching: 1) static (ground-state) quenching, such as fluo-
rophore homo- or heterodimerization, and 2) dynamic
(excited-state) quenching through Fꢀrster resonance energy
transfer (FRET). The distortion of the long-wavelength
absorption peak of b-LEAP (with an extra shoulder that is
blue-shifted from the maximum absorption peak by 30 nm)
with respect to that of EtNBS-COOH (see Figure 2 in the
Supporting Information) suggested a ground-state quenching
mechanism.[11] The short separation (ca. 2.4 nm) of the two PS
moieties assured a high quenching efficiency. Homodimeri-
[*] Dr. X. Zheng,[+] U. W. Sallum,[+] S. Verma, H. Athar, C. L. Evans,
Prof. T. Hasan
Wellman Center for Photomedicine
Harvard Medical School and Massachusetts General Hospital
40 Blossom Street, Boston, MA 02114 (USA)
Fax: (+1)617-726-8566
E-mail: thasan@partners.org
Homepage: http://www2.massgeneral.org/wellman/people/thasa-
n.asp
[+] X. Zheng and U. W. Sallum contributed equally to this work.
[**] We thank Otsuka Chemicals for their generous gift of ACLE and Dr.
Robert Moellering, Jr. for kindly providing the strains of MRSA. This
research was funded by the Department of Defense/Air Force Office
of Scientific Research (DOD/AFOSR) (grant number: FA9550-04-1-
0079). MRSA=methicillin-resistant Staphylococcus aureus.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804804.
2148
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 2148 –2151

Angewandte
Chemie
Michaelis constant (Km) and the
Kcat value at 50 units per milliliter of
B. cereus penicillinase were determined
to be 1.648 mm and 2.79 sÀ1, respectively,
by a global fitting procedure with the
program GraphPad Prism 5.0. The Kcat/
Km ratio of b-LEAP was comparable to
those reported for penicillins and greater
than those reported for cephalosporins,
and thus indicated efficient cleavage
(Table 1).
Four clinical blood isolates of MRSA
and a S. aureus (non-MRSA) control
strain were selected to assess b-LEAP
cleavage and PDT efficacy. The five
strains exhibited different levels of sus-
ceptibility to penicillin G, as determined
by a minimum inhibitory con-
Scheme 1. Mechanism for the cleavage of b-LEAP. The blue balls represent the inactive PSs in
the uncleaved construct, and the red balls represent the potentially phototoxic active PSs
following the b-lactamase-mediated cleavage of b-LEAP.
centration (MIC) assay (Fig-
ure 2a and Table 2).
Within the first 30 min of
incubation, the rates of b-
LEAP hydrolysis by the
three strains of MRSA were
relatively close (Figure 2c),
whereas the control strain,
29213, exhibited a rate of
hydrolysis no different from
that of the buffer alone. The
total amount of hydrolysis by
strain 9307 during the 3-h
incubation period was less
than that observed for strains
8150 and 8179. The suscepti-
bilities of all bacterial strains
for penicillin G and b-LEAP
PDT were determined. Bac-
teria were cultured overnight
in brain heart infusion
medium with penicillin G
(10 mgmLÀ1). Overnight cul-
tures were then diluted 1:1000
in
fresh
antibiotic-free
medium and cultured to mid
exponential phase. The bac-
teria were then diluted to a
final concentration of 5 ꢁ
Scheme 2. Synthesis of b-LEAP. DIPEA=N,N’-diisopropylethylamine, NMM=4-methylmorpholine,
HATU=O-(7-azabenzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate, TEA=triethylamine,
TFA=trifluoroacetic acid.
106 CFUmLÀ1
(CFU =
colony-forming unit) and
incubated with b-LEAP for
90 min in the dark to enable
zation (PS pair) quenching has the advantage of releasing two
active PSs after cleavage to yield twice the potential photo-
toxicity of a comparable heterodimer system.
The ability of Bacillus cereus penicillinase to cleave and
activate b-LEAP was assayed, and a concentration-depen-
dent increase in fluorescence emission as a function of time
was demonstrated for both the enzyme and the substrate
(Figure 2b, and Figure 4 in the Supporting Information). The
sufficient b-LEAP activation prior to exposure to the
appropriate fluence of laser light at 670 nm. The respective
levels of b-LEAP hydrolysis correlated with the observed
differences in the susceptibilities of the three strains to
penicillin G as well as PDT with b-LEAP (compare Table 2
and Figure 2). Strain 9307 was more susceptible to penicil-
lin G than strains 8150 and 8179 (Table 2, Figure 2a).
Interestingly, at a b-LEAP concentration of 2.5 mm, strain
Angew. Chem. Int. Ed. 2009, 48, 2148 –2151
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2149

Communications
Table 2: Susceptibility of strains of S. aureus to penicillin G and b-LEAP,
and relative b-lactamase activity.
Strain
MIC [mgmLÀ1 [a]
]
b-Lactamase
Survival with
activity [UmLÀ1 [b]
]
b-LEAP (2.5 mm)[c]
29213 ATCC
0.24Æ0.01
n.d.[d]
0.84
0.44Æ0.06
0.24Æ0.07
9307 clinical 5.92Æ1.0
MRSA
8150 clinical 15.7Æ1.2
MRSA
0.79
1.23
n.d.[d]
0.16Æ0.08
0.04Æ0.04
0.04Æ0.04
Figure 1. Physical characterization: a) Fluorescence quantum yield
8179 clinical 74.6Æ5.3
(F (fl)) and singlet-oxygen quantum yield (F (1O2)) of b-LEAP and
EtNBS-COOH; ethanol with 0.1% acetic acid was used as the solvent.
b) Fluorescence spectra of b-LEAP and EtNBS-COOH. The spectra
were recorded at equimolar concentrations; methanol was used as the
solvent. RFU=relative fluorescence units.
MRSA
8140 clinical n.d.[d]
MRSA
[a] MIC=minimum
inhibitory
concentration
of
penicillin G.
[b] 1 UmLÀ1 =b-lactamase activity of 1ꢀ109 bacteria per mL. [c] Fraction
of the bacteria that survived treatment with b-LEAP (2.5 mm). [d] n.d.=
not determined.
relationship between b-lactamase activity and bacterial
susceptibility to b-LEAP PDT. Taken together, these results
indicate the success of our approach. All MRSA strains were
inactivated effectively at a 5 mm concentration of b-LEAP
with illumination at 15 JcmÀ2 (Figure 2d); this result is
comparable to the observed efficacy of EtNBS-COOH (see
Figure 6 in the Supporting Information). The reference strain
exhibited a concentration-dependent inhibition of growth
upon incubation with b-LEAP that did not differ significantly
from the observed inhibition without PDT (Figure 2d, and
Figure 5 in the Supporting Information).
The phototoxicity of b-LEAP to human foreskin fibro-
blasts (HFF-1) was compared to that of EtNBS-COOH. b-
LEAP was found to be less phototoxic than the free PS to
HFF-1 cells (see Figure 7 in the Supporting Information).
Together, these findings indicated that b-LEAP could target
MRSA selectively and produce less damage to host tissue
than PDTwith free PS. Cocultures of MRSA and HFF-1 were
incubated with b-LEAP and examined by confocal micro-
scopy to compare their relative levels of b-LEAP fluores-
cence. Although there was some nonspecific uptake of b-
LEAP by the HFF-1 cells, the presence of MRSA8179, the
strain most resistant to penicillin G (Figure 2a and Table 2),
led to far greater fluorescence (Figure 3). The higher fluo-
rescence levels observed in MRSA (Figure 3, insets) demon-
strate quantitatively the specificity of b-LEAP in targeting the
drug-resistant strain.
PDT for entirely localized infections is currently in clinical
trials;[13] however, the concept of using PDT systemically for
regionally localized infections is new. The successful imple-
mentation of this strategy requires the combination of an
appropriate molecular target that is specific to the bacteria,
such as b-lactamase, with a robust chemical construct that
recognizes the target. A perceived limitation of this technol-
ogy could be light delivery to the site of infection. However,
with the advent of fiber optics, light delivery to complex
anatomical sites, although challenging, is feasible. PDT is an
FDA-approved treatment for lung cancer[14] and is in clinical
studies for systemic diseases, such as mesothelioma, dissemi-
nated ovarian cancer, and other intraperitoneal diseases.[15,16]
Figure 2. Specificity of b-LEAP for b-lactamase. a) Inhibition profiles
for selected strains of S. aureus with penicillin G. b) Nonlinear regres-
sion analysis of b-LEAP hydrolysis by B. cereus penicillinase. c) b-LEAP
hydrolysis by selected strains of S. aureus. d) Profiles for the inhibition
of selected strains of S. aureus by PDT with b-LEAP. (Key for a, c, and
d: purple 29213, green 9307, gold 8150, red 8179, blue 8140 (d only),
black no cells (c only)).
Table 1: Rate constants for the cleavage of b-LEAP and other b-lactams
by B. cereus penicillinase.
Substrate
K ]
cat/Km [mmÀ1 sÀ1
b-LEAP
1.69Æ0.15
18.7Æ1.5
phenethicillin[9]
l-propicillin[9]
13.2Æ0.7
d-propicillin[9]
15.4Æ0.4
carbenicillin[9]
2.84Æ0.2
phenoxymethylpenicillin[9]
benzylpenicillin[12]
cephaloridine[12]
cefuroxime[12]
38.4Æ1.0
0.427Æ0.06
0.0676Æ0.004
0.774Æ0.07
9307 was more resistant to PDT than strains 8150 and 8179
(Figure 2d). These findings demonstrate both an inverse
relationship between bacterial susceptibility to penicillin G
and bacterial susceptibility to b-LEAP PDT and a direct
2150
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 2148 –2151

Angewandte
Chemie
preseptic infections at sensitive anatomical sites. This strategy
is adaptable to other mechanisms of drug resistance.
Received: October 1, 2008
Revised: December 4, 2008
Published online: February 10, 2009
Keywords: drug resistance · lactams · photochemistry ·
.
photodynamic therapy · photosensitizers
[1] T. Hasan, B. Ortel, N. Solban, B. W. Pogue in Cancer Medicine 7,
7th ed. (Eds.: D. W. Kufe, E. Frei, J. F. Holland, R. R. Weich-
selbaum, R. E. Pollock, R. C. Bast, W. K. Hong, W. N. Hait), BC
Decker, Ontario, 2006, p. 537.
[2] M. R. Hamblin, T. Hasan, Photochem. Photobiol. Sci. 2004, 3,
436 – 450.
[3] H. M. Tang, M. R. Hamblin, C. M. Yow, J. Infect. Chemother.
2007, 13, 87 – 91.
[4] S. Verma, G. M. Watt, Z. Mai, T. Hasan, Photochem. Photobiol.
2007, 83, 996 – 1005.
[5] K. Bush, G. A. Jacoby, A. A. Medeiros, Antimicrob. Agents
Chemother. 1995, 39, 1211 – 1233.
[6] G. Zlokarnik, P. A. Negulescu, T. E. Knapp, L. Mere, N. Burres,
L. Feng, M. Whitney, K. Roemer, R. Y. Tsien, Science 1998, 279,
84 – 88.
[7] B. Xing, A. Khanamiryan, J. Rao, J. Am. Chem. Soc. 2005, 127,
4158 – 4159.
[8] R. Munch, H. Wombacher, F. Korber, J. Clin. Chem. Clin.
Biochem. 1981, 19, 953 – 960.
[9] H. Christensen, M. T. Martin, S. G. Waley, Biochem. J. 1990, 266,
853 – 861.
[10] S. Verma, U. W. Sallum, H. Athar, L. Rosenblum, J. W. Foley, T.
Hasan, Photochem. Photobiol. 2009, 85, 111 – 118.
[11] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd ed.,
Springer, New York, 2006, p. 277.
[12] S. Bounaga, A. P. Laws, M. Galleni, M. I. Page, Biochem. J. 1998,
331(Pt 3), 703.
[13] Phase II antimicrobial PDT studies in chronic leg ulcers and
chronic diabetic foot ulcers (Photopharmica Ltd): http://
www.photopharmica.com/clinical development.htm.
[14] J. S. Friedberg, R. Mick, J. P. Stevenson, T. Zhu, T. M. Busch, D.
Shin, D. Smith, M. Culligan, A. Dimofte, E. Glatstein, S. M.
Hahn, J. Clin. Oncol. 2004, 22, 2192; FDA: US Food and Drug
Administration.
[15] P. Baas, L. Murrer, F. A. N. Zoetmulder, F. A. Stewart, H. B. Ris,
N. van Zandwijk, J. L. Peterse, E. J. T. Rutgers, Br. J. Cancer
1997, 76, 819.
Figure 3. b-LEAP fluorescence in cell cultures. Two-dimensional projec-
tions of 90 depth-resolved confocal slices separated by 0.4 mm:
a) coculture of HFF-1 cells and strain 8179; b) HFF-1 cells alone.
Insets are plots of the fluorescence intensity along the dotted lines in
the images. The insets show the significantly greater fluorescence
intensities for MRSA than for the HFF cells and thus indicate the local
cleavage of b-LEAP only at the site of the bacteria. I=intensity
(arbitrary units).
We anticipate that the strategy developed in the current
study will be able to be used in combination with standard
antibiotic treatment to destroy resistant and nonresistant
bacteria. This study demonstrates the successful exploitation
of an antibiotic-resistance mechanism for the activation of a
novel therapeutic photodynamic molecule. The significance
of this study lies in the broader applicability of PDT for
[16] S. M. Hahn, D. L. Fraker, R. Mick, J. Metz, T. M. Busch, D.
Smith, T. Zhu, C. Rodriguez, A. Dimofte, F. Spitz, M. Putt, S. C.
Rubin, C. Menon, H. W. Wang, D. Shin, A. Yodh, E. Glatstein,
Clin. Cancer Res. 2006, 12, 2517.
Angew. Chem. Int. Ed. 2009, 48, 2148 –2151
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2151