Photodegradation of lipopolysaccharides and the inhibition of macrophage activation by anthraquinone-boronic acid hybrids

Article abstract of DOI:10.1039/c2cc33559j

Target-selective photodegradation of 3-deoxy-d-manno-2-octulopyranosonic acid (KDO) was achieved without additives and under neutral conditions using a designed anthraquinone-boronic acid hybrid and long wavelength UV light irradiation. The hybrid can photodegrade lipopolysaccharides (LPS) and inhibit macrophage activation induced by LPS.

Full text of DOI:10.1039/c2cc33559j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 7595–7597
www.rsc.org/chemcomm
COMMUNICATION
Photodegradation of lipopolysaccharides and the inhibition of
macrophage activation by anthraquinone–boronic acid hybridsw
Daisuke Takahashi, Takuya Miura and Kazunobu Toshima*
Received 17th May 2012, Accepted 7th June 2012
DOI: 10.1039/c2cc33559j
Target-selective photodegradation of 3-deoxy-D-manno-2-octulo-
pyranosonic acid (KDO) was achieved without additives and
under neutral conditions using a designed anthraquinone–boronic
acid hybrid and long wavelength UV light irradiation. The hybrid
can photodegrade lipopolysaccharides (LPS) and inhibit macro-
phage activation induced by LPS.
shock that do not cause serious side effects. There is therefore an
urgent need for the development of novel therapeutic agents which
can selectively bind to LPS and neutralize bacterial endotoxins.
Herein we describe the molecular design, chemical synthesis, and
biological evaluation of novel light-activatable small molecules
which can selectively photodegrade LPS and inhibit macrophage
activation, without the need for additives and under neutral
conditions. To the best of our knowledge, this is the first demon-
strated example of the target-selective degradation of glycolipid
LPS by light switching under neutral conditions.
Carbohydrates in living cells play important roles in many
biological events, including bacterial recognition and infection.1
Consequently, the development of innovative methods for
selectively controlling the specific function of certain carbo-
hydrates has attracted much attention in chemistry, biology,
and medicine. Lipopolysaccharides (LPS) represent the main
outer membrane component of Gram-negative bacteria such
as E. coli (Fig. 1), and play a key role during severe Gram-
negative infection, sepsis, and sepsis shock.2 However, there
are few effective therapeutic agents for LPS-induced sepsis
In our previous studies, designed anthraquinone (AQ)–lectin
hybrids with high affinities for specific oligosaccharides were found
to selectively degrade target oligosaccharides upon irradiation with
long wavelength UV light and without the need for further
additives.3,4 In addition, designed AQ–5 or fullerene–boronic acid
hybrids6 were found to selectively bind to and effectively degrade
the target D-galactofuranosides (Galfs) under neutral conditions
not only upon UV irradiation, but also upon visible light irradia-
tion. Furthermore, upon photo-irradiation, AQ–boronic acid
hybrids showed bactericidal activity against Mycobacterium bovis
BCG, which contains Galfs in the cell wall. On the basis of our
preliminary findings,7 we set out to design and synthesize
target-selective photodegrading agents for LPS.
We first targeted the structure of the KDO (3-deoxy-D-manno-
2-octulopyranosonic acid) residue that is covalently linked to
lipid A, a membrane anchoring element and the toxic principal of
LPS. KDO is a unique and common structural component of
LPS and is not found in mammals. Furthermore, KDO possesses
an acyclic 1,2-diol on the exocyclic side chain. Arylboronic acid
was chosen as the recognition moiety. It is already known that
phenylboronic acid can bind to 1,2- or 1,3-diols through
reversible boronate formation under physiological conditions.8
Thus, we expected that a hybrid molecule consisting of a
boronic acid, which would preferentially bind to an acyclic diol
on KDO, and a photo-degrading AQ derivative, could be used
for target-selective photodegradation of LPS.
To investigate our hypothesis, we first designed AQ–boronic
acid hybrids 2 and 3, which have different length spacers between
the AQ moiety and the boronic acid moiety (Fig. 2). After
chemical synthesis of the hybrids 2 and 3 (see Scheme S1 in the
ESIw), the binding affinities of the hybrids to different glycosides,
including KDOa–OMe (4), Neu5Aca–OMe (5), Galb–OMe (6),
Glca–OMe (7) and Mana–OMe (8), were examined using a
quantitative three-component alizarin red S (ARS) assay9 under
Fig. 1 Chemical structure of lipopolysaccharides (LPS) (1) from E. coli.
Department of Applied Chemistry, Faculty of Science and Technology,
Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522,
Japan. E-mail: toshima@applc.keio.ac.jp; Fax: +81 45-566-1576
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c2cc33559j
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7595–7597 7595

View Article Online
Fig. 3 Each glycoside (1.0 mM) was incubated with AQ–boronic acid
hybrid 2 or 3 (1.0 mM) in 10% DMF/0.1 M phosphate buffer (100 mL,
pH 7.4) at 25 1C for 2 h under irradiation with a UV lamp (365 nm,
100 W) placed 10 cm from the sample. The reaction mixture was
analysed by HPLC (Mightysil RP-18 GP 5 mm, 4.6 Â 150 mm); 40 1C;
detection by UV (210 nm or 254 nm) after appropriate modifications
of the resulting products.
Fig. 2 Chemical structures of AQ–boronic acid hybrids and several
glycosides.
neutral (buffered pH 7.4) aqueous conditions. The results are
summarized in Table 1. Hybrid molecules 2 and 3 bound with
good affinity to 4 and 5, which possess an acyclic 1,2-diol on the
exocyclic side chain, and not to 1,2- or 1,3 diols on other
glycosides. In addition, 2, which possesses a short spacer,
exhibited very similar Ka values towards 4 and 5. On the other
hand, the Ka value obtained for hybrid 3, which possesses a long
spacer, for binding to 4, was almost three times greater than that
of 3 binding to 5. These results suggest that our designed hybrid
molecule 3 selectively binds to the target glycoside 4 rather than
to other glycosides. Although the reason for this specificity
remains unclear, we believe that not only the interaction between
the boronic acid in 3 and an acyclic 1,2-diol in KDO, but also
other interactions, such as CH–p interaction between the AQ
moiety in 3 and the hydrogen atoms on a pyran ring of KDO, are
important factors for the binding.
of the hybrids for the glycosides. In addition, the degradation
activities of 2 for 4 and 5, which have moderate affinities for 2,
were also found to be low. In sharp contrast, when 3 was
exposed to 4 under photo-irradiation, significant degradation
took place. Moreover, the degradation of 4 by 3 was found to
be much more effective than that of 5 by 3. These results
suggest that the binding affinity of the hybrid 3 to glycoside,
and the distance between the protons in the glycoside and the
photo-activated AQ moiety are influential factors in the
target-selective photodegradation of glycosides by 3.
Next, we examined the photo-induced degradation of LPS
(1) (5.0 mg mLÀ1, 2.15 mM, 1.0 equiv.) by hybrid 2 or 3
at concentrations of 215 (100 equiv.), 64.5 (30 equiv.), 21.5
(10 equiv.) and 6.45 mM (3 equiv.) in 10% DMF/0.1 M
phosphate buffer (pH 7.4). The progress of the photodegradation
reaction was monitored by sodium dodecyl sulfate polyacrylamide
gel electrophoresis (SDS-PAGE). The results are shown in Fig. 4.
When 2 was exposed to LPS (1) and photo-irradiated, no
change in the SDS-PAGE profile was observed (Fig. 4a). On
the other hand, when LPS (1) was treated with 3 at a
concentration of 215 mM with photo-irradiation, the SDS-
PAGE band corresponding to LPS (1) disappeared (lane 4 in
Fig. 4b). These results indicate that our designed hybrid 3
photodegraded not only KDO but also lipid A in the target
LPS (1) upon irradiation with UV light in the absence of any
Next, we examined the photoinduced glycoside-degradation
abilities of 2 and 3 (1.0 mM) against glycosides 4–8 (1.0 mM)
in 10% DMF/0.1 M phosphate buffer (pH 7.4) at 25 1C for
2 h, using a long-wavelength UV lamp (365 nm, 100 W) placed
10 cm from the sample. The progress of the photodegradation
reaction was monitored by HPLC/UV (215 nm or 254 nm)
analysis after appropriate modifications of the resulting products
(see ESIw). The percent degradation was calculated based on the
peak area corresponding to each modified glycoside; the results
are shown in Fig. 3. When neither hybrid was added (control),
more than 95% of each modified glycoside was detected by
HPLC analysis. When hybrid 2 or 3 was exposed to glycosides
6–8, less than 10% degradation took place, owing to low affinity
Table 1 Association constants (Ka) for the hybrid molecules 2 and 3
with several glycosides
Ka [MÀ1
a
]
Hybrid 2
Hybrid 3
Entry
Glycoside
Fig. 4 Photodegradation of LPS (1) by hybrids 2 and 3. 1 (5.0 mg mLÀ1
,
1
2
3
4
5
KDOa–OMe (4)
Neu5Aca–OMe (5)
Galb–OMe (6)
Glca–OMe (7)
54
48
10
12
11
99
34
9
12
9
2.15 mM, 1.0 equiv.) was incubated with (a) 2 or (b) 3 in 10% DMF/
0.1 M phosphate buffer (pH 7.4) at 25 1C for 2 h under irradiation with a
UV lamp (365 nm, 100 W) placed 10 cm from the sample. The reaction
mixture was analysed using tricine–SDS-PAGE. Lane 1, 1 alone; lane 2, 1
following UV irradiation; lane 3, 1 + 2 or 3 without UV irradiation;
lanes 4–7, 1 + 2 or 3 (equiv. to 1; 100, 30, 10 and 3 equiv., respectively)
following UV irradiation.
Mana–OMe (8)
a
The Ka values were determined by an ARS assay in 10% DMF/
0.1 M phosphate buffer (pH 7.4) and are the average of at least two
reproducible measurements; see ESI.
c
7596 Chem. Commun., 2012, 48, 7595–7597
This journal is The Royal Society of Chemistry 2012

View Article Online
western blotting (Fig. 5b). It was confirmed that pre-incubation of
LPS with 3 upon photo-irradiation did not induce iNOS
expression in the cells (lane 5 in Fig. 5b). These results clearly
indicate that photodegradation of LPS by AQ–boronic acid
hybrid 3 upon photo-irradiation occurred effectively and
inhibited NO production and iNOS expression by LPS-induced
macrophage activation.
In conclusion, we have developed a new chemical agent, an
AQ–boronic acid hybrid, which can selectively bind to and
photo-degrade target glycoside KDO upon irradiation with
UV light, without the need for additives and under neutral
conditions. In addition, the hybrid molecule can photodegrade
target glycolipid LPS. Furthermore, it was revealed that
photo-degradation of LPS by the hybrid inhibits NO production
and iNOS expression by LPS-induced macrophage activation.
The results presented here will contribute to the molecular design
of novel artificial carbohydrate photodegradation agents, which
should find wide application in chemistry, biology, and medicine.
The development of more specific and tighter binding hybrid
molecules for LPS is now under investigation in our laboratory.
The authors wish to thank Prof. Dr K. Umezawa, Faculty
of Science and Technology, Keio University, for his valuable
suggestions. This research was supported in part by the
MEXT-Supported Program for the Strategic Research Foundation
at Private Universities, 2012–2016, Scientific Research (B)
(No. 20310140 and 23310153), and Scientific Research on
Innovative Areas ‘‘Chemical Biology of Natural Products’’,
from the Ministry of Education, Culture, Sports, Science and
Technology of Japan (MEXT).
Fig. 5 LPS-induced NO production and iNOS expression in RAW
264.7 cells. LPS (1) (10 mg mLÀ1, 4.30 mM, 1.0 equiv.) was pre-incubated
with 3 in 10% DMF/H2O at 25 1C for 2 h with or without photo-
irradiation using a UV lamp (365 nm, 100 W) placed 10 cm from the
sample. The resulting mixture was incubated with RAW 264.7 cells,
which were seeded into 96-well plates, at 37 1C for 24 h. NO production
was measured by the Griess assay (column). Cell viability was measured
by the MTT assay (circle). Lane 1, control (without treatment); lane 2, 1
without photo-irradiation; lane 3, 1 with photo-irradiation; lane 4, 1 + 3
(1000 equiv.) without photo-irradiation; lanes 5–9, 1 + 3 (equiv. to 1;
1000, 300, 100, 30 and 10 equiv., respectively) with photo-irradiation. (b)
iNOS expression was assessed by western blotting. Lane 1, control
(without treatment); lane 2, 1 without photo-irradiation; lane 3, 1 with
photo-irradiation; lane 4, 1 + 3 (1000 equiv.) without photo-irradiation;
lanes 5–7, 1 + 3 (equiv. to 1; 1000, 300, and 100 equiv., respectively) with
photo-irradiation.
additives at neutral pH. However, an excess amount of 3 to 1 was
required for photodegradation due to the low concentration of 1
used in the reaction.
Notes and references
Next, we conducted EPR studies10 using 3 and DMPO with
or without UV irradiation in order to confirm the generation of
ROS, a reactive species for oligosaccharide photodegradation.3–7
Photo-irradiation of 3 in the presence of DMPO gave the DMPO-
hydroxy radical spin adduct DMPO–ꢀOH. Furthermore, no peaks
corresponding to DMPO–ꢀOH were detected either when 3 was
photo-irradiated or when photo-irradiation was conducted in the
absence of 3 (see Fig. S1 in the ESIw).
Finally, the macrophage activation ability of photodegraded
LPS by AQ–boronic acid hybrid 3 was evaluated based on
LPS-induced nitric oxide (NO) production in macrophage
RAW 264.7 cells using the Griess assay.11 In addition, cell
viability was determined by the MTT assay. These results are
summarized in Fig. 5a. When the cells were treated with only
LPS (0.1 mg mLÀ1), NO production was significantly induced
(lanes 1–3 in Fig. 5a). In addition, it was found that
pre-incubation of LPS with 3 without photo-irradiation resulted
in neither cytotoxicity nor inhibition of NO production (lane 4 in
Fig. 5a). In contrast, pre-incubation of LPS with 3 upon photo-
irradiation inhibited NO production in a concentration-dependent
manner (lanes 5–9 in Fig. 5a). Next, to reveal whether these
changes in NO production were caused by changes in iNOS
expression, we assessed iNOS expression in RAW 264.7 cells by
1 Carbohydrate in Chemistry & Biology, ed. B. Ernst, G. W. Hart and
P. Sinay, Wiley-VCH, Weinheim, 2000; Comprehensive Glycoscience,
ed. J. P. Kamerling, G.-J. Boons, Y. C. Lee, A. Suzuki, N. Taniguchi
and A. G. J. Voragen, Elsevier, Oxford, 2007.
2 E. T. Rietschel and H. Brade, Sci. Am., 1992, 267, 54; E. T. Rietschel,
T. Kirikae, F. U. Schade, U. Mamat, G. Schmidt, H. Loppnow,
A. J. Ulmer, U. Zahringer, U. Seydel, F. Dipadova, M. Schreier and
H. Brade, FASEB J., 1994, 8, 217.
3 M. Ishii, S. Matsumura and K. Toshima, Angew. Chem., Int. Ed.,
2007, 46, 8396.
4 Y. Imai, S. Hirono, H. Matsuba, T. Suzuki, Y. Kobayashi, H.
Kawagishi, D. Takahashi and K. Toshima, Chem.–Asian J., 2012, 7, 97.
5 D. Takahashi, S. Hirono, C. Hayashi, M. Igarashi, Y. Nishimura
and K. Toshima, Angew. Chem., Int. Ed., 2010, 49, 10096.
6 D. Takahashi, S. Hirono and K. Toshima, Chem. Commun., 2011,
47, 11712.
7 D. Takahashi and K. Toshima, Chem. Commun., 2012, 48, 4397.
8 J. P. Lorand and J. O. Edwards, J. Org. Chem., 1959, 24, 769;
T. D. James, Boronic acids in Organic Synthesis and Chemical
Biology, Wiley-VCH, Weinheim, 2005, p. 441.
9 G. Springsteen and B. Wang, Chem. Commun., 2001, 1608.
10 J. E. Wertz and J. R. Bolton, Electron Spin Resonance, McGraw-
Hill, New York, 1972; H. M. Swartz, J. R. Bolton and D. C. Borg,
Biological Application of Electron Spin Resonance, Wiley-Inter-
science, New York, 1972.
11 M. Takeiri, M. Tachibana, A. Kaneda, A. Ito, Y. Ishikawa,
S. Nishiyama, R. Goto, K. Yamashita, S. Shibasaki, G. Hirokata,
M. Ozaki, S. Todo and K. Umezawa, Inflammation Res., 2011,
60, 879.
¨
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7595–7597 7597