Discovery of selective 2,4-diaminopyrimidine-based photoaffinity probes for glyoxalase i

Article abstract of DOI:10.1039/c3md00286a

Glyoxalase I (GLO-1) plays a critical role in the detoxification of 2-oxoaldehydes and is highly expressed in cancer cells. Through photo-affinity labelling and affinity pull-down approaches, a series of 2,4-diaminopyrimidine compounds were discovered to

Full text of DOI:10.1039/c3md00286a

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
Discovery of selective 2,4-diaminopyrimidine-
based photoaffinity probes for glyoxalase I†
Cite this: Med. Chem. Commun., 2014,
5, 352
Yiqing Zhou,a Tianlin Guo,a Xitao Li,a Yi Dong,a Paul Galatsis,b Douglas S. Johnsonb
a
*
and Zhengying Pan
Glyoxalase I (GLO-1) plays a critical role in the detoxification of 2-oxoaldehydes and is highly expressed in
cancer cells. Through photo-affinity labelling and affinity pull-down approaches, a series of 2,4-
diaminopyrimidine compounds were discovered to selectively bind to GLO-1 in cells. These compounds
show potent inhibition of GLO-1 enzyme activity and prevent proliferation of cancer cells. The cell
permeable and “clickable” photoaffinity probe L1-Bpyne presented here could be a valuable tool for
profiling GLO-1 in live cells.
Received 30th September 2013
Accepted 20th December 2013
DOI: 10.1039/c3md00286a
www.rsc.org/medchemcomm
Despite advancements in the development of GLO-1 inhibitors,
a selective GLO-1 labelling probe has not been reported.
Introduction
The 2,4-diaminopyrimidine scaffold has been widely used in
pharmaceutical agents, including the dihydrofolate reductase
inhibitor pyrimethamine, and the antibiotics iclaprim and
trimethoprim.11 Recently, 2,4-diaminopyrimidine also served as
a common template to generate inhibitors for a number of
kinases, including c-Met,12 MK2,13 ALK,14 and SYK.15 Since 2011,
several research groups have developed potent LRRK2 inhibi-
tors based on the 2,4-diaminopyrimidine scaffold.16–18 Of these
compounds, CZC-25146 was found to attenuate Parkinson's
disease-related toxicity in human neurons (Fig. 1A). While the in
vitro selectivity of this type of kinase inhibitor can be assessed
using various techniques,19–21 the selectivity of these
compounds in more complex biological systems is still
unknown.22 To examine the selectivity of this type of inhibitors
The glyoxalase system, composed of glyoxalase I (GLO-1, EC
4.1.1.5) and glyoxalase II (GLO-2, EC 3.1.2.6), is an important
and ubiquitous detoxication component of cellular metabo-
lism in most cells.1 Methylglyoxal (MG), a by-product in glycol-
ysis, is cytotoxic due to its covalent binding to DNA and proteins
in cells.1,2 Newly produced MG reacts with glutathione (GSH), an
abundant nucleophile in cells, to form hemithioacetal. GLO-1
catalyses the isomerisation of hemithioacetal into S-D-lactoyl-
glutathione. Subsequently GLO-2 catalyses the conversion of
lactoylglutathione into lactic acid and GSH via hydrolysis.3 The
glyoxalase system is highly expressed in cancer and inamma-
tory cells to rapidly remove the cytotoxic MG.4
Overexpression of GLO-1 has been associated with multi-
drug resistance in cancer chemotherapy.5 Much effort has been
made to develop inhibitors of GLO-1 as an anti-cancer therapy.
Most of the early GLO-1 inhibitors are GSH derivatives, but this
class of compounds has poor pharmacokinetic properties, as
GSH remains charged under physiological conditions.6 Addi-
tionally, GSH is a substrate or cofactor of many other enzymes;
thus, these inhibitors show poor selectivity. In recent years,
many non-GSH-based small molecule inhibitors, such as
methyl gerfelin, delphinidin, curcumin and zopolrestat, have
been reported to have inhibitory effects against GLO-1.7–10
aKey laboratory of Chemical Genomics, School of Chemical Biology and Biotechnology,
Peking University, Xili University Town, PKU Campus, Shenzhen, China, 518055.
E-mail: panzy@pkusz.edu.cn; Tel: +86-755-26033072
bNeuroscience Medicinal Chemistry and Chemical Biology, Pzer Worldwide Research
and Development, Cambridge, MA 02139, USA
† Electronic supplementary information (ESI) available: Detailed synthetic
procedures, 1H and 13C NMR spectra, HPLC chromatograms for compounds L1,
L2, L3, L1-Biotin and L1-Bpyne; detailed procedures for biological assays; mass
spectroscopy data for GLO-1. See DOI: 10.1039/c3md00286a
Fig. 1 Chemical structures of (A) the 2,4-diaminopyrimidine-based
kinase inhibitor CZC-25146; (B) the photoaffinity probes L1-Biotin and
L1-Bpyne; and (C) the analogues of the 2,4-diaminopyrimidine series
compounds used in this study.
352 | Med. Chem. Commun., 2014, 5, 352–357
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
in mammalian cells, we synthesised two photoaffinity probes, streptavidin–sepharose beads, then separated by SDS-PAGE and
L1-Biotin and L1-Bpyne, to detect these compounds cellular visualised by silver staining (Fig. 2B). In both experiments, the
targets (Fig. 1B). As a result, GLO-1 was identied as a major L1-Biotin probe bound specically to two protein bands at
target of these 2,4-diaminopyrimidine compounds in live cells. approximately 28 kDa and 55 kDa. The two corresponding
Several synthetic analogues of CZC-25146 were further tested in bands in silver staining gels were excised, destained, digested
in vitro enzyme activity and cell proliferation assays (Fig. 1C). by trypsin and analysed by MALDI-TOF mass spectrometry.
These compounds inhibited the enzyme activity of GLO-1 and Both bands were identied as human glyoxalase I (see ESI†). As
cancer cell proliferation, and the potency of these molecules the literature indicates, GLO-1 exists in both monomeric and
depended on the glucose concentrations in culture media. dimeric forms in cells;2,23 the higher molecular weight band
Finally, these compounds were shown to cause MG accumula- could represent the GLO-1 dimer, and the lower molecular
tion in live cells and induce cell apoptosis.
weight band could represent the monomer. The appearance of
the dimeric form band at 55 kDa in the SDS-PAGE experiments
might be caused by our probe molecules and cellular modi-
cations in HEK293T cells. These results were further conrmed
by western blotting using glyoxalase I antibody (Fig. 2C). Finally,
recombinant human GLO-1 protein was also successfully
labelled by L1-Biotin (Fig. 2D) in a separate experiment.
Competitive binding is clearly observed using the structurally
similar compound L1, suggesting that GLO-1 was specically
labelled by the probe L1-Biotin.
Further labelling experiments were performed to evaluate
the detection range of the probe L1-Biotin and the effect of UV
irradiation time. HEK293T lysates were incubated with different
concentrations of L1-Biotin; 0.5 to 1 mM of probe was sufficient
to label GLO-1, whereas higher concentrations of the probe
increased non-specic labelling (Fig. 3A). At a steady probe
concentration at 0.5 mM, cell lysates were incubated with
L1-Biotin and irradiated with UV light for various periods of
time. The band intensity increased with the UV irradiation time
and became saturated aer approximately 1 hour (Fig. 3B).
Aer GLO-1 was identied as a potential target in cell lysates,
we attempted to conrm it in live cells. However, the bio-
tinylated photoaffinity probe L1-Biotin was unable to bind to
proteins in live cells (data not shown), likely due to the low
trans-membrane permeability of the biotin tag. We then
designed the probe L1-Bpyne for labelling studies of GLO-1 in
live cells (Fig. 1B). Bio-orthogonal conjugation reactions have
been successfully employed to identify covalent targets of small
molecules in a variety of biological systems. The copper(I) cat-
alysed [3 + 2] azide–alkyne cycloaddition (CuAAC) reaction is
Results and discussion
First, we examined the potential targets of 2,4-diaminopyr-
imidine compounds in cells and designed a biotinylated pho-
toaffinity probe L1-Biotin to label and pull-down its binding
proteins in the cellular proteome (Fig. 1B). A benzophenone
group was introduced to covalently link the probe and its
potential targets upon photo irradiation. Human embryo
kidney 293T (HEK293T) cell lysate was pre-treated with strep-
tavidin–sepharose beads to reduce endogenously biotinylated
proteins. The cells were then incubated with 0.1 or 1 mM
L1-Biotin with and without 50 mM compound L1 overnight at 4
ꢀC. Aer photo-crosslinking by irradiation at 365 nm for 1 hour
on ice, the L1-Biotin binding proteins were separated by SDS-
PAGE and detected by streptavidin–HRP (Fig. 2A). Alternatively,
biotinylated proteins in lysates, both endogenous and those
bound to the L1-Biotin probe, were rst enriched with
Fig. 2 (A) HEK293T cell lysate labelling by L1-Biotin with and without
competitor L1; (B) silver staining of L1-Biotin binding proteins enriched Fig. 3 (A) HEK293T cell lysates treated with different concentrations
by streptavidin–sepharose beads; (C) western blot validation with of probe L1-Biotin; (B) HEK293T cell lysates treated with 0.5 mM of
GLO-1 antibody; (D) recombinant glyoxalse-1 protein labelling by L1-Biotin and irradiated by long-wave UV light for different lengths of
L1-Biotin.
time.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 352–357 | 353

View Article Online
MedChemComm
Concise Article
one of the most commonly used methods.24 Thus, in the new approximately 50 nM (Table 1), which suggests that the
L1-Bpyne probe, an alkyne tag replaced the biotin tag in benzophenone moiety and the alkyne tag did not signicantly
L1-Biotin to facilitate cellular studies.
alter the binding preference of these 2,4-diaminopyrimide
We then attempted to label GLO-1 using this new probe. For compounds towards GLO-1.
in vitro labelling, HEK293T lysates were incubated with 1 mM
Proliferation assays were performed to examine the effects of
L1-Bpyne with and without excess compound L1. Aer UV these compounds in cells. The primary cellular function of GLO-
irradiation and CuAAC conjugation with tetramethylrhod- 1 is MG detoxication, and the MG concentration has been
amine–azide (TAMRA–N3), the samples were analysed by SDS- reported to be elevated in high glucose conditions.25,26 There-
PAGE and detected by in-gel uorescence scan. For in situ fore, three types of cancer cells were cultured in Dulbecco's
labelling, live HEK293T cells were incubated with 1 mM of Modied Eagle Medium (DMEM) with two concentrations of
L1-Bpyne with and without excess compound L1. The cells were glucose at 5 mM (approximately the normal in vivo glucose level)
ꢀ
irradiated by UV for 1 hour at 37 C and washed with ice-cold and 25 mM (high glucose level). Cell viability was assessed using
PBS buffer solution to remove unbound probes. Following a CellTiter-Glo® luminescent kit aer incubating for 48 hours
centrifugation and homogenisation by lysis buffer, the samples with various concentrations of the tested compounds. These
were subjected to the CuAAC reaction with the same conditions compounds had GI50 values at low to mid micromolar concen-
as those used for the in vitro labelling. Both monomeric and trations (Table 1). In accordance with the data of the enzymatic
dimeric forms of GLO-1 were selectively labelled by the probe assay, the L1-Bpyne probe was also the most potent inhibitor in
L1-Bpyne in vitro and in situ. The labelling showed specic the cell proliferation assays. Between two human lung cancer
binding competition by compound L1 (Fig. 4A). Encouraged by cell lines, NCI-H522 cell, which was reported to express high
the results of in situ labelling, the intracellular CuAAC reaction intrinsic GLO-1 activity, is more susceptible to the GLO-1 inhi-
with TAMRA–N3 was performed to show the cellular localisation bition; A549 cell, which was reported to express relatively low
of L1-Bpyne targets. HeLa cells were cultured in normal culture intrinsic GLO-1 activity, showed about 5-fold less sensitivity
medium and treated with 20 mM L1-Bpyne for 2 hours with or towards L1-Bpyne.27 HeLa cell, a human cervical cancer cell line,
without 100 mM compound L1 as the competitor. Following UV could also be potently inhibited by this series of inhibitors. The
irradiation, the cells were xed, permeabilised, and subjected to glucose contents in the culture media had low to moderate
CuAAC reaction with TAMRA–N3 (Fig. 4B). Staining of the cells effects on the inhibition of tumor cell proliferation by this series
gave a strong signal as GLO-1 is a relatively abundant enzyme in of compounds. As expected, high concentrations of glucose
cytosol. The labelling was signicantly reduced following increased cellular sensitivity to GLO-1 inhibition. The GI50
competitive binding by excess compound L1.
Because GLO-1 appeared as a major cellular target of 2,4- glucose medium compared to high glucose medium.
diaminopyrimidine probes, we next investigated whether these Because GLO-1 inhibition causes methylglyoxal accumula-
values of compounds L1–L3 increased up to 2.6-fold in normal
types of compounds could inhibit the catalytic activity of GLO-1 tion followed by random DNA and protein labelling, we further
in vitro. A spectrophotometric method was employed by moni- investigated whether our compounds could induce an increase
toring the increase in UV absorbance at 240 nm from the of the cellular MG concentration. HeLa cells were grown in
formation of S-D-lactoylglutathione at 25 ꢀC.25 The L1-Bpyne normal or high glucose media for 3 days followed by treatment
probe was the most potent inhibitor with an IC50 value of with DMSO or 5 mM of L1-Bpyne for 24 hours. The cells were
26 nM, while other analogues L1–L3 showed IC50 values of harvested, washed with PBS buffer, resuspended in distilled
water and boiled. The MG content in the supernatant was
detected using the same spectrophotometric method as that in
the aforementioned enzyme activity assay. The MG concentra-
tions were slightly elevated in the high glucose medium with
untreated cells but were almost doubled in the L1-Bpyne treated
cells (Fig. 5A). This suggests that L1-Bpyne inhibited GLO-1
activity in HeLa cells. Cellular MG accumulation commenced
upon GLO-1 inhibition.
Finally, we tested whether GLO-1 inhibition would cause cell
apoptosis. HeLa cells were treated with various concentrations
of L1-Bpyne for 24 hours, the degradation of the apoptosis
marker protein caspase-3 clearly indicated cell apoptosis
(Fig. 5B). Moreover, aer HeLa cells were treated with 5 mM of
L1-Bpyne for 24 hours, the cells were xed with para-
formaldehyde and stained with 40,6-diamidino-2-phenylindole
(DAPI), the cell nuclei also appeared fragmented (Fig. 5C).
MG is a highly reactive a-oxoaldehyde that reacts predomi-
nantly with the lysine or arginine residues of proteins and the
Fig. 4 (A) In vitro labelling of HEK293T cell lysates and in situ labelling
guanyl residues in DNA and RNA. It may cause abnormal
enzyme activity and defects during transcription.28 Because
of live HEK293T cells by L1-Bpyne (1 mM); (B) in-cell CuAAC reaction
revealed the cellular distribution of L1-Bpyne.
354 | Med. Chem. Commun., 2014, 5, 352–357
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Table 1 Inhibition of GLO-1 enzyme activity and cancer cells proliferation by 2,4-diaminopyrimidine compoundsa
GLO-1
(nM)
NCI-H522-HG
(mM)
NCI-H522-NG
(mM)
Compound
HeLa-HG (mM)
HeLa-NG (mM)
A549-HG (mM)
A549-NG (mM)
L1-Bpyne
26.4
51.3
45.8
53.9
2.9
16.5
7.3
2.7
40.2
8.3
4.6
37.0
7.3
5.7
61.9
14.0
61.0
13.3
>50
26.9
>50
15.5
>50
37.4
>50
L1
L2
L3
12.2
25.0
23.2
a
Values are averages of at least two independent runs. HG: high glucose level; NG: normal glucose level.
the cell proliferation assay could result from multiple factors
including physiochemical properties of compounds and the
network linking GLO-1 activity to cell death. A combination of
GLO-1 inhibitors with compounds targeting multiple pathways
may be an attractive approach to achieve signicant anti-cancer
effects in the clinic.
Chemistry
The syntheses of compounds L1–L3, L1-Biotin and L1-Bpyne are
depicted in Scheme 1. Each was readily prepared using litera-
ture procedures with slight modications. The common
precursor 1 was coupled with substituted anilines by an SNAr
reaction to provide L1, L2, and L3 in good yields. L1-Biotin was
prepared from the intermediate L1-BpNH2 and biotin using
standard coupling conditions for amide bond formation.
Compounds 1 and 2 underwent a Hartwig–Buchwald amination
reaction to furnish L1-Bpyne in good yield. Detailed synthetic
procedures and compound characterisation data are provided
in the ESI.†
Fig. 5 (A) Intracellular MG concentration upon L1-Bpyne treatment;
(B) HeLa cells were treated with different concentrations of L1-Bpyne
and the intact caspase-3 levels were evaluated by western-blot; (C)
HeLa cells were treated with 5 mM of L1-Bpyne for 24 hours and
stained by DAPI.
tumour tissue has higher glycolysis activity than normal
tissue29,30 and the glyoxalase system is highly activated in cancer
Conclusions
cells,31,32 GLO-1 inhibitors could be selectively cytotoxic towards In summary, we have discovered two photoaffinity probes,
cancer cells. Our compounds did show inhibition of prolifera- L1-Biotin and L1-Bpyne, which specically label GLO-1 in
tion of several human cancer cells, thus indicating their cells. The biotinylated probe can specically label and
potential application to cancer therapy. The relative large pull-down GLO-1 in vitro, while the alkyne probe is able to
potency shi of these compounds from the enzymatic assay to passively penetrate living cells and label GLO-1 in situ.
Scheme 1 Synthesis of GLO-1 inhibitors. Reagents and conditions: (a) substituted anilines, concentrated HCl, n-BuOH, 120 ꢀC, 16–18 hours; (b)
HATU, Hunig's base, DMF, 0 ꢀC to rt, overnight; (c) 1, Pd2(dba)3, DPBP, K2CO3, nBuOH, 100 ꢀC, 4 hours.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 352–357 | 355

View Article Online
MedChemComm
Concise Article
Furthermore, all compounds potently inhibited the enzyme 11 R. Nelson and A. Rosowsky, Antimicrob. Agents Chemother.,
activity of GLO-1 with IC50 values in the range of 26 to 54 nM. 2001, 45, 3293–3303.
These compounds prevented cancer cell proliferation, 12 A. Zhao, X. Gao, Y. Wang, J. Ai, Y. Wang, Y. Chen, M. Geng
increased intracellular levels of methylglyoxal, and induced and A. Zhang, Bioorg. Med. Chem., 2011, 19, 3906–3918.
cell apoptosis. Interestingly, higher glucose content in the 13 M. Argiriadi, A. Ericsson, C. Harris, D. Banach, D. Borhani,
cell medium increased the cytotoxicity of these compounds.
Taken together, this series of 2,4-diaminopyrimidine
compounds directly interacts with GLO-1 in cells. Moreover,
L1-Bpyne is both a potent inhibitor and a selective probe of
GLO-1, and could be a powerful tool in studies of this
D. Calderwood, M. Demers, J. Dimauro, R. Dixon,
J. Hardman, S. Kwak, B. Li, J. Mankovich, D. Marcotte,
K. Mullen, B. Ni, M. Pietras, R. Sadhukhan, S. Sousa,
M. Tomlinson, L. Wang, T. Xiang and R. Talanian, Bioorg.
Med. Chem. Lett., 2010, 20, 330–333.
enzyme. With this cell permeable probe, GLO-1 was 14 C. Zicsak, J. Theroff, L. Aimone, T. Angeles, M. Albom,
successfully labelled both in vitro and in situ. This probe
could also be utilised to label GLO-1 in tissues of animal
models, though the long-wave UV light used in the photo-
M. Cheng, E. Mesaros, G. Ott, M. Quail, T. Underiner,
W. Wan and B. Dorsey, Bioorg. Med. Chem. Lett., 2011, 21,
3877–3880.
crosslinking step could be harmful to surrounding tissues 15 N. Powell, J. Hoffman, F. Ciske, M. Kaufman, J. Kohrt,
and limits the application of this probe in whole animal
studies. Further research is underway to understand the
binding mode of this type of probe, including the exact
J. Quin, D. Sheehan, A. Delaney, S. Baxi, C. Catana,
P. McConnell, J. Ohren, L. Perrin and J. Edmunds, Bioorg.
Med. Chem. Lett., 2013, 23, 1046–1050.
binding site and specic binding interactions with GLO-1, 16 N. Ramsden, J. Perrin, Z. Ren, B. Lee, N. Zinn, V. Dawson,
and its potential use in combination with other anti-cancer
agents.
D. Tam, M. Bova, M. Lang, G. Drewes, M. Bantscheff,
F. Bard, T. Dawson and C. Hopf, ACS Chem. Biol., 2011, 6,
1021–1028.
17 H. Chen, B. Chan, J. Drummond, A. Estrada, J. Gunzner-
Toste, X. Liu, Y. Liu, J. Moffat, D. Shore, Z. Sweeney,
T. Tran, S. Wang, G. Zhao, H. Zhu and D. Burdick, J. Med.
Chem., 2012, 14, 5536–5545.
18 H. Choi, J. Zhang, X. Deng, J. Hatcher, M. Patricelli,
Z. Zhao, D. Alessi and N. Gray, ACS Med. Chem. Lett.,
2012, 3, 658–662.
19 J. Bain, L. Plater, M. Elliott, N. Shpiro, C. J. Hastie,
H. McLauchlan, I. Klevernic, J. S. Arthur, D. R. Alessi and
P. Cohen, Biochem. J., 2007, 408, 297–315.
Acknowledgements
The following nancial support to ZP is gratefully acknowl-
edged: 2013CB910704 (973 Program); 21142005 (Sino-NSF);
CXB201005260059, AJC201005270281A and Peacock Program-
KQTD201103 (Shenzhen Science and Technology Innovation
Commission), 20100001120030 (MOE). Pzer, Inc. also
provided partial nancial support for this project.
20 M. W. Karaman, S. Herrgard, D. K. Treiber, P. Gallant,
C. E. Atteridge, B. T. Campbell, K. W. Chan, P. Ciceri,
M. I. Davis, P. T. Edeen, R. Faraoni, M. Floyd, J. P. Hunt,
D. J. Lockhart, Z. V. Milanov, M. J. Morrison, G. Pallares,
H. K. Patel, S. Pritchard, L. M. Wodicka and
P. P. Zarrinkar, Nat. Biotechnol., 2008, 26, 127–132.
21 M. P. Patricelli, T. K. Nomanbhoy, J. Wu, H. Brown,
D. Zhou, J. Zhang, S. Jagannathan, A. Aban, E. Okerberg,
C. Herring, B. Nordin, H. Weissig, Q. Yang, J. D. Lee,
N. S. Gray and J. W. Kozarich, Chem. Biol., 2011, 18, 699–
710.
Notes and references
1 P. Thornalley, Biochem. J., 1990, 269, 1–11.
2 P. Thornalley, Biochem. Soc. Trans., 2003, 31, 1372–1377.
3 F. Himo and Per E. M. Siegbahn, J. Am. Chem. Soc., 2001, 123,
10280–10289.
4 R. Vince and S. Daluge, J. Med. Chem., 1971, 14, 35–37.
5 D. Creighton, Z. Zheng, R. Holewinski, D. Hamilton and
J. Eiseman, Biochem. Soc. Trans., 2003, 31, 1378–
1382.
6 R. Takasawa, S. Takahashi, K. Saeki, S. Sunaga,
A. Yoshimoria and S. Tanumaa, Bioorg. Med. Chem. Lett., 22 M. Schenone, V. Dancik, B. K. Wagner and P. A. Clemons,
2008, 16, 3969–3975.
Nat. Chem. Biol., 2013, 9, 232–240.
7 M. Kawatani, H. Okumura, K. Honda, N. Kanoh, M. Muroi, 23 A. Cameron, B. Olin, M. Ridderstrom, B. Mannervik and
N. Dohmae, M. Takami, M. Kitagawa, Y. Futamura, T. Jones, EMBO J., 1997, 16, 3386–3395.
M. Imoto and H. Osada, Proc. Natl. Acad. Sci. U. S. A., 2008, 24 A. Speers and B. Cravatt, Proling enzyme activities in vivo
¨
105, 11691–11696.
using click chemistry methods, Chem. Biol., 2004, 11, 535–
8 R. Takasawa, K. Saeki, A. Tao, A. Yoshimori, H. Uchiro,
546.
M. Fujiwara and S. Tanuma, Bioorg. Med. Chem. Lett., 2010, 25 P. Thornalley, Biochem. J., 1988, 254, 751–755.
18, 7029–7033.
26 M. Shinohara, P. Thornalley, I. Giardino, P. Beisswenger,
S. Thorpe, J. Onorato and M. Brownlee, J. Clin. Invest.,
1998, 101, 1142–1147.
9 M. Liu, M. Yuan, M. Luo, X. Bu, H. Luo and X. Hu, Biophys.
Chem., 2010, 147, 28–34.
10 J. Zhai, H. Zhang, L. Zhang, Y. Zhao, S. Chen, Y. Chen, 27 H. Sakamoto, T. Mashima, S. Sato, Y. Hashimoto,
X. Peng, Q. Li, M. Yuan and X. Hu, ChemMedChem, 2013,
8, 1462–1464.
T. Yamori and T. Tsuruo, Clin. Cancer Res., 2001, 7,
2513–2518.
356 | Med. Chem. Commun., 2014, 5, 352–357
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
28 N. Shangari, W. Bruce, R. Poon and P. O'Brien, Biochem. Soc. 31 A. Rulli, L. Carli, R. Romani, T. Baroni, E. Giovannini,
Trans., 2003, 31, 1390–1393.
29 O. Warburg, Science, 1956, 123, 309–314.
G. Rosi and V. Talesa, Breast Cancer Res. Treat., 2001, 66,
67–72.
30 R. Gatenby and R. Gillies, Nat. Rev. Cancer, 2004, 4, 891– 32 C. Antognelli, F. Baldracchini, V. Talesa, E. Costantini,
899.
A. Zucchi and E. Mearini, Cancer J., 2006, 12, 222–228.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 352–357 | 357