Download A new family of covalent inhibitors block nucleotide binding to the

Biochem. J. (2012) 448, 67–72 (Printed in Great Britain)
67
doi:10.1042/BJ20121014
A new family of covalent inhibitors block nucleotide binding to the active
site of pyruvate kinase
Hugh P. MORGAN*, Martin J. WALSH†, Elizabeth A. BLACKBURN*, Martin A. WEAR*, Matthew B. BOXER†, Min SHEN†,
Henrike VEITH†, Iain W. MCNAE*, Matthew W. NOWICKI*, Paul A. M. MICHELS‡, Douglas S. AULD†,
Linda A. FOTHERGILL-GILMORE* and Malcolm D. WALKINSHAW*1
*Centre for Translational and Chemical Biology, School of Biological Sciences, University of Edinburgh, Michael Swann Building, The King’s Buildings, Mayfield Road, Edinburgh EH9
3JR, U.K., †NIH Chemical Genomics Center, NIH Center for Translational Therapeutics, National Human Genome Research Institute, National Institutes of Health, 9800 Medical Center
Drive, Rockville, MD 20850, U.S.A., and ‡Research Unit for Tropical Diseases, de Duve Institute and Laboratory of Biochemistry, Université catholique de Louvain, Avenue Hippocrate
74, B-1200 Brussels, Belgium
PYK (pyruvate kinase) plays a central role in the metabolism
of many organisms and cell types, but the elucidation of the
details of its function in a systems biology context has been
hampered by the lack of specific high-affinity small-molecule
inhibitors. High-throughput screening has been used to identify a
family of saccharin derivatives which inhibit LmPYK (Leishmania
mexicana PYK) activity in a time- (and dose-) dependent manner,
a characteristic of irreversible inhibition. The crystal structure
of DBS {4-[(1,1-dioxo-1,2-benzothiazol-3-yl)sulfanyl]benzoic
acid} complexed with LmPYK shows that the saccharin moiety
reacts with an active-site lysine residue (Lys335 ), forming a
covalent bond and sterically hindering the binding of ADP/ATP.
Mutation of the lysine residue to an arginine residue eliminated
the effect of the inhibitor molecule, providing confirmation of the
proposed inhibitor mechanism. This lysine residue is conserved
in the active sites of the four human PYK isoenzymes, which were
also found to be irreversibly inhibited by DBS. X-ray structures
of PYK isoforms show structural differences at the DBS-binding
pocket, and this covalent inhibitor of PYK provides a chemical
scaffold for the design of new families of potentially isoformspecific irreversible inhibitors.
INTRODUCTION
of the human M2PYK isoenzyme and oncogenesis [3], and
this isoenzyme is found in all tumours studied to date [3].
The effector-regulated HsM2PYK can facilitate a build-up of
phosphometabolites which are required for the cancer cell to
proliferate. A number of potent activators of HsM2PYK have
been identified with AC50 values around 30 nM [10]; however,
the only examples of HsM2PYK inhibitors bind relatively
weakly with IC50 values of 10–20 μM [11]. RNAi (RNA
interference) knockdown of PYK and other enzymes in the
glycolytic pathway in trypanosomatids has facilitated a systems
biology approach to elucidate the roles played by these enzymes
[12]. A complementary approach to regulate PYK activity by
small-molecule compounds has been hindered by the lack of
appropriate chemical tools. One of the few compounds currently
available is the polysulfonated drug suramin, one of the earliest
synthetic drugs used to treat human African trypanosomiasis.
It is a promiscuous binder with a complex pharmacology and
poorly understood mode-of-action. However, it has been shown
to inhibit seven out of the ten enzymes in the glycolytic
pathway of Trypanosoma brucei [13,14]. A crystal structure
of a complex of LmPYK with suramin shows that it acts as
an ATP/ADP mimic and binds competitively with the ADP
substrate [15]. Suramin also inhibits all four human isoforms
of PYK with K i values between 1 and 20 μM [15]. In addition,
affinity labelling of rabbit-muscle PYK has been achieved by
covalent modification of active-site residues using nucleotide
PYK (pyruvate kinase) catalyses the last step in glycolysis to
produce ATP and pyruvate and, in most organisms studied, PYKs
have similar homotetrameric architectures with each monomer
composed of four domains (Figure 1a). Four human tissue-specific
PYK isoenzymes have been described: HsRPYK (erythrocyte),
HsLPYK (liver), HsM1PYK (muscle) and HsM2PYK (embryonic
or tumour). The M1 isoform is constitutively active, whereas the
other three are allosterically regulated by the effector molecule
F16BP (fructose 1,6-bisphosphate) [1]. Trypanosomatid PYKs
are distinguished by their use of the chemically distinct molecule
F26BP (fructose 2,6-bisphosphate) as the effector, and recently
the detailed allosteric mechanism for PYK of the pathogenic
protist Leishmania mexicana (LmPYK) has been elucidated [2].
PYK has been implicated as playing a central role in a
number of proliferative and infectious diseases, and the discovery
of isoenzyme-specific inhibitors or activators of PYK could be of
potential interest in the elucidation of the aetiology of cancer
[3] and of metabolic diseases such as diabetes and obesity [4], as
well as infectious diseases caused by bacteria [5], trypanosomatid
parasites [6] and the malaria parasites Plasmodium spp. [7].
For example, PYK deficiency in erythrocytes results in nonspherocytic haemolytic anaemia, and over 130 mutations in
HsRPYK have been identified which contribute to the disease
[8,9]. There is also a strong link between the up-regulation
Key words: covalent inhibitor, Leishmania mexicana, lysine
covalent modification, nucleotide binding, pyruvate kinase,
saccharin analogue.
Abbreviations used: DBS, 4-[(1,1-dioxo-1,2-benzothiazol-3-yl)sulfanyl]benzoic acid; DTT, dithiothreitol; F26BP, fructose 2,6-bisphosphate; LmPYK,
Leishmania mexicana pyruvate kinase; PEG, poly(ethylene glycol); PEP, phosphoenolpyruvate; PTS, 1,3,6,8-pyrenetetrasulfonic acid; PYK, pyruvate
kinase; qHTS, quantitative high-throughput screening; RMS, root mean square; RNAi, RNA interference; TEA, triethanolamine; TLS, Translation–Libration–
Screw.
1
To whom correspondence should be addressed (email [email protected]).
The atomic co-ordinates of the LmPYK–DBS structure have been deposited in the PDB under code 3SRK.
c The Authors Journal compilation c 2012 Biochemical Society
68
Figure 1
H. P. Morgan and others
Proposed reaction mechanism of DBS
(a) The four-domain structure of Lm PYK–OX/DBS is indicated by different colours: green, short N-terminal domain; purple, domain A; blue, domain B; and yellow, domain C. The large (a-a) and
small (c-c) subunit interfaces are indicated by broken lines. The black box designates the position of the active site of one subunit, and is shown in close-up in (b). (b) The difference (F o − F c
map shown in green at a resolution of 3.5 Å, contoured to 2.5 σ ) electron density observed in all four active sites is associated with Lys335 in a novel orientation (blue). (c) Two-dimensional
structures of NCGC00188411, NCGC00186526 and NCGC00059857 compared with the structure of the synthesized analogue NCGC00188636 (DBS) that displayed enhanced stability and solubility.
(d) The proposed reaction mechanism for the covalent modification of Lys335 . (e) Time-dependent inhibition of Lm PYK by pre-incubation with 50 μM DBS under variable conditions. Curve A,
Lm PYK pre-incubated with 0.4 mM PEP and 50 μM DBS. Curve B, Lm PYK pre-incubated with 0.4 mM PEP (no inhibitor). Curve C, Lm PYK pre-incubated with 0.4 mM PEP, 4 μM F26BP and
50 μM DBS. Curve D, Lm PYKK335R pre-incubated with 0.4 mM PEP and 50 μM DBS.
analogues [16,17]. The only other known general PYK inhibitor
is the substrate analogue oxalate, which exhibits poor specificity
and binds with relatively weak affinity (K i = 220 μM) [18].
Selective inhibitors of PYK are needed as biochemical tools
for studying the glycolytic pathway and as potential leads for
drug development. In the present paper we report the discovery
of a novel covalent PYK inhibitor DBS {4-[(1,1-dioxo-1,2benzothiazol-3-yl)sulfanyl]benzoic acid} (Figure 1c).
EXPERIMENTAL
Expression and purification of wild-type and K335R mutant forms of
LmPYK
Chemically competent Escherichia coli Rosetta 2* (DE3)pLysS
cells (Merck, catalogue number 71403) were transformed with
c The Authors Journal compilation c 2012 Biochemical Society
either the wild-type or mutated plasmid (see Supplementary Material at http://www.BiochemJ.org/bj/448/bj4480067add.htm).
Both wild-type and K335R mutant forms of LmPYK (UniProtID,
Q27686) were expressed and purified as described previously
[15].
Synthesis and characterization of covalent inhibitors
A series of saccharin derivatives identified as inhibitors of
LmPYK by qHTS (quantitative high-throughput screening) was
further elaborated by de novo chemical synthesis, purification
and characterization. The procedures for the synthesis and
purification of compounds NCG00186526, NCGC00059857,
NCGC00188411 and NCGC00188636 (Figure 1c) and their
characterization are described in detail in the Supplementary
Covalent inhibition of pyruvate kinase
data. One of these analogues, DBS (NCGC00188636), displayed
improved stability and solubility profiles relative to the original
screening hit (NCGC00186526) and was therefore used for the
experiments described in the present paper.
PYK inhibitor assay
The following reagents were added to a 50 ml Falcon tube
(equivalent to 11×1 ml assays): 8.58 ml of assay mix {1×
assay buffer [50 mM TEA (triethanolamine), pH 7.2, 100 mM
potassium chloride, 3 mM magnesium chloride and 10 %
glycerol], 0.2 mM NADH (Roche, catalogue number 128023),
3.2 units/ml lactate dehydrogenase (Sigma, catalogue number
61309), 1.6 units/ml LmPYK, 0.4 mM PEP (phosphoenolpyruvate) (Sigma, catalogue number 79430) and 2.20 ml of
250 μM inhibitor solution (made up with 1× assay buffer from
a 170 mM stock in 100 % DMSO, final concentration of 50 μM,
added last to the reaction mix)}. The control reaction mix was
made up in an identical manner except 1 × assay buffer was used
in place of the inhibitor solution. Both the control and inhibitor
reaction mixtures were incubated throughout the experiment in
a 25 ◦ C water bath (before the addition of inhibitor which was
also incubated at 25 ◦ C). To 990 μl of the reaction mix, 10 μl
of 20 mM ADP [final concentration = 0.2 mM ADP (Sigma,
catalogue number A4386); made up with 1× assay buffer] was
added to start the reaction. The mixture was gently agitated and
the decrease in absorbance at 340 nm was measured for 2 min
(using Lambda Bio). The process was repeated every 20 min over
200 min for both the control and inhibitor. The initial rate was
then calculated using UV kinlab. The rate for each inhibitor assay
was expressed as a percentage of the control assay.
Preparation of inhibitor-modified Lm PYK
The DBS inhibitor (stock = 172 mM in 100 % DMSO) was added
to 200 μl of LmPYK [10 mg/ml: 184 μM in 20 mM TEA buffer
(pH 7.2) and 10 % glycerol] to a final concentration of 9 mM
(maintaining a similar molar ratio of inhibitor to protein as used
in the kinetic assay). The sample was then incubated overnight
at 4 ◦ C. DTT (dithiothreitol) was added to a final concentration
of 1 mM, and the LmPYK–DBS inhibitor mix was incubated at
room temperature (20 ◦ C) for 15 min. The DTT and leaving group
were removed by repeated dilutions [using 20 mM TEA buffer
(pH 7.2) and 10 % glycerol] and by concentrating the sample
in a Vivaspin column (molecular mass cut-off = 100 kDa). The
sample was concentrated to 12 mg/ml.
Table 1
69
Data collection and refinement statistics
Values in parentheses are for the highest-resolution shell.
Measurement
Data collection
Space group
Cell dimensions
a , b , c (Å)
Solvent content (%)
Wavelength (Å)
Resolution (Å)
R sym
I /σ (I )
Completeness (%)
Redundancy
Refinement
Resolution (Å)
Number of reflections
R work /R free
Average protein B -factor (Å2 )
Number of residues
RMS deviations
Bond lengths (Å)
Bond angles (◦ )
I222
122.4, 130.2, 166.5
60.00
0.98
60.85–2.65 (2.79–2.65)
0.09 (0.64)
8.3 (1.7)
98.0 (98.8)
2.8 (2.8)
60.85–2.65
35 936
22.3/26.6
31.4
894
0.01
0.90
buffer (pH 7.2), 50 mM magnesium chloride, 100 mM potassium
chloride and 25 % glycerol, which eliminated the appearance of
ice rings. Intensity data were collected (ϕ scans were 2o over 180o )
at the Diamond synchrotron radiation facility in Oxfordshire,
U.K. on beamline IO3 from a single crystal cryocooled in liquid
nitrogen. A single crystal gave data to a resolution of 2.65 Å
(1 Å = 0.1 nm) at 100 K.
Structure determination and analysis of model geometry
The LmPYK–DBS structure was solved and refined using a
method described previously [2], yielding R/Rfree values of
21.9/27.35. A further round of TLS (Translation–Libration–
Screw) restrained refinement (four optimal TLS groups were
determined using a TLSMD procedure [19]) yielded final R/Rfree
values of 22.3/26.6. The geometry of the model was assessed
using MolProbity [20]. Although electron density was well
defined for Thr296 (a key active-site residue), it exhibits geometry
outside the Ramachandran plot here and in many PYK structures.
This is primarily due to a restricted geometry, which facilitates
interactions with active-site ligands. Atomic co-ordinates and the
experimental structure factors have been deposited in the PDB
with the code 3SRK.
Crystallization and data collection
Samples of inhibitor-modified LmPYK (prepared as described
above) were diluted to 10 mg/ml using a buffer containing
20 mM TEA (pH 7.2) and PTS (1,3,6,8-pyrenetetrasulfonic
acid, final concentration 1 mM). Single crystals of inhibitormodified LmPYK complexed with PTS were obtained at 4 ◦ C
by vapour diffusion using the hanging-drop technique. The
drops were formed by mixing 1.5 μl of protein solution
with 1.5 μl of a well solution, composed of 12–16 % PEG
[poly(ethylene glycol)] 8000, 20 mM TEA buffer (pH 7.2),
50 mM magnesium chloride, 100 mM potassium chloride and
10 % glycerol. The drops were equilibrated against a reservoir
filled with 0.5 ml of well solution. Crystals grew to maximum
dimensions (1.0 mm×0.2 mm×0.1 mm) after 24–48 h. Before
data collection, crystals were equilibrated for 14 h over a
well solution composed of 14–18 % PEG 8000, 20 mM TEA
RESULTS AND DISCUSSION
High-throughput screening identified a series of saccharin-based
inhibitors
There were 292740 compounds in the NIH Molecular LibrariesSmall Molecule Repository tested in the primary screen for
the wild-type LmPYK (PubChem AID 1721). The screen was
performed at seven compound concentrations using qHTS [21,22]
and identified 1087 high-quality concentration–response curves,
corresponding to a hit rate of 0.4 % of the library. One of
the top actives from this series was the saccharin derivative
NCGC00186526, with an IC50 of 10 μM. The oxo linkage in this
compound was labile, and the molecule was found to hydrolyse to
saccharin and the corresponding phenol (Figure 1c). Stable sulfur
(NCGC00188411) and nitrogen (NCGC00059857) analogues
c The Authors Journal compilation c 2012 Biochemical Society
70
Figure 2
H. P. Morgan and others
DBS reacts with Lys335 at the active site of Lm PYK
All ligands and interacting amino acid residues are shown as sticks, and waters are shown as red spheres. (a) The superimpositions of the ATP and the modified Lys335 indicate that the ATP/ADP
binding may be sterically hindered. (b) The proposed position of Lys335 prior to covalent modification is shown by purple sticks. Initially the sulfur dioxide group of the DBS molecule (pink sticks) binds
to the active site, occupying a similar position to the sulfono group of suramin (Lm PYK–suramin structure [15], see Supplementary Figure S4 at http://www.BiochemJ.org/bj/448/bj4480067add.htm).
The network of interactions (broken red line, interactions <3 Å; broken yellow lines, interactions 3–4 Å; broken green lines, co-ordination of the inorganic cations) hold DBS in place, providing an
ideal reactive geometry for the reaction with Lys335 to occur. (c) Final refined position of covalently modified Lys335 as observed in the crystal structure. (d) Overlay of the X-ray structures of Lm PYK
(the present study) and the X-ray structure of Hs M2PYK showing differences in the amino acid side chains in three regions (R1, R2 and R3) around the modified Lys335 (yellow) that could enable the
design of isoenzyme-specific families of inhibitors.
were prepared (Figure 1c) and tested in the LmPYK activity assay.
Only NCGC00188411 showed inhibitory activity (IC50 = 5 μM).
At this point it was hypothesized that covalent modification of
either cysteine or lysine in the enzyme, as well as the leaving
group ability of the resultant phenol, thiophenol and aniline
explained the trend in activity [23]. The sulfur analogue DBS
(NCGC00188636, Figures 1c and 1d) was used in subsequent
experiments.
crystallization solution (see Table 1 for data collection and
refinement statistics). Electron density corresponding to the
covalent addition of the saccharin moiety to Lys335 is clearly
visible in all active sites (Figure 1b). The modified Lys335 residue is
located at the adenine-binding site and blocks ADP/ATP binding
(Figure 2a). Electron density was carefully examined around all
other lysine residues in the structure, but no evidence for their
covalent modification was observed.
Covalent modification of Lm PYK by DBS is confirmed by X-ray
crystal structure analysis
Inhibition of Lm PYK by DBS is time-dependent
LmPYK crystals grown in the presence of 2 mM oxalate
and 2.8 mM DBS (LmPYK–OX/DBS) were anisotropic, and
diffracted poorly to approximately 4 Å. Despite the relatively low
resolution, difference (F o − F c ) electron density was observed
near Lys335 in all active sites (Figure 1b) suggesting that Lys335
was covalently modified by the saccharin moiety (Figure 1d).
Improved-quality crystals diffracting to 2.65 Å were obtained
using a purification protocol of DBS-modified LmPYK in which
DMSO was removed by dilution and PTS was added to the
c The Authors Journal compilation c 2012 Biochemical Society
An inhibition assay was used to examine the covalent reaction
further, whereby LmPYK activity was monitored over time in
the presence of 50 μM DBS (Figure 1e). Maximal inhibition
of ∼ 80 % was achieved after ∼ 250 min (Figure 1e, curve A),
although LmPYK inhibition never reached 100 % inhibition even
after 10 h (after prolonged incubation periods at 25 ◦ C both the
wild-type and K335R LmPYK mutant began to aggregate). The
small amount of remaining activity could possibly be due to
weak binding of ADP to the DBS-modified active site. The
X-ray structure of the modified enzyme suggests that the saccharin
Covalent inhibition of pyruvate kinase
group covalently bound to Lys335 with its flexible side chain could
adopt conformations that would still allow ADP access to the
active site (Figure 2a), albeit with reduced affinity. In terms of
potential antiparasitic activity it is relevant to note that incomplete
depletion (approximately 75 %) of the intracellular concentration
of PYK by RNAi is sufficient to cause cell death in the pathogenic
bloodstream form of Trypanosoma brucei [24].
71
IC50 values of 8 and 16.3 μM respectively (see Supplementary
Figure S4 at http://www.BiochemJ.org/bj/448/bj4480067add.
htm). These values compare with an IC50 value of DBS for
LmPYK of 2.9 μM. Modelled poses of the pre-cleavage DBSbinding pocket highlight sequence differences between the
trypanosomatid and human enzymes (Figure 2d) and it is likely
that such differences in the saccharin-binding pocket provide
an opportunity for the design of more potent species-specific
inhibitors against either trypanosomatid or human PYK isoforms.
The K335R mutation confirms the covalent inhibitory mechanism
To test whether inhibition stems from the covalent modification
of Lys335 and not modification of other lysine residues in PYK,
we expressed and purified the K335R mutant of LmPYK. The
wild-type and K335R mutant of LmPYK enzymes exhibited
similar activity and kinetic parameters (Supplementary Table S1 at
http://www.BiochemJ.org/bj/448/bj4480067add.htm). However,
on addition of DBS and under identical assay conditions to that
of wild-type LmPYK, the K335R mutant exhibited essentially no
change in activity over time (Figure 1e, curve D).
Evidence of selectivity of DBS for Lys335 is suggested by
the inability of DBS to inhibit rabbit lactate dehydrogenase (a
coupling enzyme) through covalent modification of a similar
active-site lysine residue, Lys56 . This residue is similar in both
location (also found on the rim of the active site cleft) and
interaction (interacting with the ribose hydroxy group of the
nucleoside group of NAD) to Lys335 of LmPYK (Supplementary
Figure S2 at http://www.BiochemJ.org/bj/448/bj4480067add.
htm). A lysine residue (Lys531 ) also exists in the active site of firefly
luciferase (PDB code 2D1T). Both of these coupling enzymes
provide good controls to suggest that DBS displays selectivity
for binding Lys335 . The X-ray structural results discussed in the
following section provide a rationale for this specificity.
Mechanism of covalent modification by DBS is suggested by the
structure of Lm PYK–suramin
A series of phenyl sulfonated dye-like molecules, including the
trypanocidal drug suramin, has been shown to bind in a nearidentical position within the active site of LmPYK [15]. The
LmPYK–DBS monomer was superimposed on to the LmPYK–
suramin structure, with excellent alignment of the protein
backbones [RMS (root mean square) fit = 0.5 Å]. Modelling a fit
of the sulfonamide group of the unreacted DBS molecule on to the
sulfone group in the suramin complex perfectly positions Lys335
for nucleophilic attack on C-3 of the saccharin ring to release the
sulphide moiety (Figures 1d and 2). The requirement for DBS to
dock in such a specific pose could explain its specificity for Lys335
over other lysine residues in the structure. The X-ray structure,
however, suggests that once the covalent bond has formed, the
modified lysine residue adopts a different pose. Comparisons of
the relevant X-ray structures show that the sulfone groups
of suramin and of the saccharin moiety of DBS covalently attached
to Lys335 are 4.4 Å apart (Figure 2b and Supplementary Figure S3
at http://www.BiochemJ.org/bj/448/bj4480067add.htm).
DBS is a covalent inhibitor of both human and trypanosomatid PYKs
Lys335 is relatively well conserved among different PYK species
and it is of interest that naturally occurring mutations in
HsRPYK (equivalent residue Lys410 ) to either glutamic acid [9] or
aspartic acid [25] result in non-spherocytic haemolytic anaemia.
DBS was found to inhibit both HsRPYK (the human PYK
isoenzyme present in erythrocytes) and HsM2PYK (the human
PYK isoenzyme present in embryonic and tumour cells) with
AUTHOR CONTRIBUTION
Hugh Morgan carried out kinetic assays, X-ray structure determination and prepared the
paper. Martin Walsh performed the synthetic chemistry. Martin Wear, Matthew Nowicki
and Elizabeth Blackburn performed protein purification and biophysical characterization.
Matthew Boxer, Henrike Veith and Min Shen performed high-throughput screening and data
analysis. Paul Michels, Douglas Auld, Linda Fothergill-Gilmore and Malcolm Walkinshaw
performed data analysis and prepared the paper.
ACKNOWLEDGEMENTS
We thank Paul Shinn, Danielle VanLeer, Thomas Daniel, Christopher LeClair and James
Bougie for assistance with compound management and purification. We would also like
to thank the staff at the Diamond synchrotron radiation facility in Oxfordshire, U.K.
FUNDING
This work was supported, in part, by the Molecular Libraries Initiative of the National
Institutes of Health Roadmap for Medical Research and the Intramural Research Program
of the National Human Genome Research Institute, National Institutes of Health. Additional
funding was from the Medical Research Council, and the European Commission through
its INCO-DEV programme. The Centre for Translational and Chemical Biology and
the Edinburgh Protein Production Facility are funded by the Wellcome Trust and the
Biotechnology and Biological Sciences Research Council.
REFERENCES
1 Fothergill-Gilmore, L. A. and Michels, P. A. (1993) Evolution of glycolysis. Prog. Biophys.
Mol. Biol. 59, 105–235
2 Morgan, H. P., McNae, I. W., Nowicki, M. W., Hannaert, V., Michels, P. A.,
Fothergill-Gilmore, L. A. and Walkinshaw, M. D. (2010) Allosteric mechanism of pyruvate
kinase from Leishmania mexicana uses a rock and lock model. J. Biol. Chem. 285,
12892–12898
3 Christofk, H. R., Vander Heiden, M. G., Harris, M. H., Ramanathan, A., Gerszten, R. E.,
Wei, R., Fleming, M. D., Schreiber, S. L. and Cantley, L. C. (2008) The M2 splice isoform
of pyruvate kinase is important for cancer metabolism and tumour growth. Nature 452,
230–233
4 Vander Heiden, M. G., Cantley, L. C. and Thompson, C. B. (2009) Understanding the
Warburg Effect: the metabolic requirements of cell proliferation. Science 324, 1029
5 Zoraghi, R., Worrall, L., See, R. H., Strangman, W., Popplewell, W. L., Gong, H., Samaai,
T., Swayze, R. D., Kaur, S., Vuckovic, M. et al. (2011) Methicillin-resistant Staphylococcus
aureus (MRSA) pyruvate kinase as a target for bis-indole alkaloids with antibacterial
activities. J. Biol. Chem. 286, 44716–44725
6 Nowicki, M. W., Tulloch, L. B., Worralll, L., McNae, I. W., Hannaert, V., Michels, P. A. M.,
Fothergill-Gilmore, L. A., Walkinshaw, M. D. and Turner, N. J. (2008) Design, synthesis
and trypanocidal activity of lead compounds based on inhibitors of parasite glycolysis.
Bioorg. Med. Chem. 16, 5050–5061
7 Ayi, K., Min-Oo, G., Serghides, L., Crockett, M., Kirby-Allen, M., Quirt, I., Gros, P. and
Kain, K. C. (2008) Pyruvate kinase deficiency and malaria. N. Engl. J. Med. 358,
1805–1810
8 Zanella, A., Bianchi, P. and Fermo, E. (2007) Pyruvate kinase deficiency. Haematologica
92, 721–723
9 Zanella, A., Fermo, E., Bianchi, P. and Valentini, G. (2005) Red cell pyruvate kinase
deficiency: molecular and clinical aspects. Br. J. Haematol. 130, 11–25
10 Jiang, J., Boxer, M. B., Heiden, M. G. V., Shen, M., Skoumbourdis, A. P., Southall, N.,
Veith, H., Leister, W., Austin, C. P. and Park, H. W. (2010) Evaluation of thieno [3,2-b]
pyrrole [3,2-d] pyridazinones as activators of the tumor cell specific M2 isoform of
pyruvate kinase. Bioorg. Med. Chem. Lett. 20, 3387–3393
c The Authors Journal compilation c 2012 Biochemical Society
72
H. P. Morgan and others
11 Vander Heiden, M. G., Christofk, H. R., Schuman, E., Subtelny, A. O., Sharfi, H., Harlow,
E. E., Xian, J. and Cantley, L. C. (2009) Identification of small molecule inhibitors of
pyruvate kinase M2. Biochem. Pharmacol. 79, 1118–1124
12 Verlinde, C., Hannaert, V., Blonski, C., Willson, M., Périé, J. J., Fothergill-Gilmore, L. A.,
Opperdoes, F. R., Gelb, M. H., Hol, W. G. J. and Michels, P. A. M. (2001) Glycolysis as a
target for the design of new anti-trypanosome drugs. Drug Resistance Updates 4,
50–65
13 Willson, M., Callens, M., Kuntz, D. A., Périé, J. J. and Opperdoes, F. R. (1993) Synthesis
and activity of inhibitors highly specific for the glycolytic enzymes from Trypanosoma
brucei . Mol. Biochem. Parasitol. 59, 201–210
14 Albert, M. A., Haanstra, J. R., Hannaert, V., Van Roy, J., Opperdoes, F. R., Bakker,
B. M. and Michels, P. A. M. (2005) Experimental and in silico analyses of glycolytic
flux control in bloodstream form Trypanosoma brucei . J. Biol. Chem. 280,
28306–28315
15 Morgan, H. P., McNae, I. W., Nowicki, M. W., Zhong, W., Michels, P. A. M., Auld, D. S.,
Fothergill-Gilmore, L. A. and Walkinshaw, M. D. (2011) The trypanocidal drug suramin
and other trypan blue mimetics are inhibitors of pyruvate kinases and bind to the
adenosine site. J. Biol. Chem. 286, 31232–31240
16 DeCamp, D. L. and Colman, R. F. (1986) Identification of tyrosine and lysine peptides
labeled by 5 -p-fluorosulfonylbenzoyl adenosine in the active site of pyruvate kinase.
J. Biol. Chem. 261, 4499–4503
17 Vollmer, S. H., Walner, M. B., Tarbell, K. V. and Colman, R. F. (1994) Guanosine
5 -O-[S-(4-bromo-2,3-dioxobutyl)]thiophosphate and adenosine 5 -O-[S-(4-bromo2,3-dioxobutyl)]thiophosphate. New nucleotide affinity labels which react with rabbit
muscle pyruvate kinase. J. Biol. Chem. 269, 8082–8090
Received 25 June 2012/27 July 2012; accepted 20 August 2012
Published as BJ Immediate Publication 20 August 2012, doi:10.1042/BJ20121014
c The Authors Journal compilation c 2012 Biochemical Society
18 Dombrauckas, J. D., Santarsiero, B. D. and Mesecar, A. D. (2005) Structural basis for
tumor pyruvate kinase M2 allosteric regulation and catalysis. Biochemistry 44,
9417–9429
19 Painter, J. and Merritt, E. A. (2006) Optimal description of a protein structure in terms of
multiple groups undergoing TLS motion. Acta Crystallogr. Sect. D Biol. Crystallogr. 62,
439–450
20 Davis, I. W., Leaver-Fay, A., Chen, V. B., Block, J. N., Kapral, G. J., Wang, X., Murray,
L. W., Arendall, 3rd, W. B., Snoeyink, J., Richardson, J. S. and Richardson, D. C. (2007)
MolProbity: all-atom contacts and structure validation for proteins and nucleic acids.
Nucl. Acids Res. 35, W375–W383
21 Inglese, J., Auld, D. S., Jadhav, A., Johnson, R. L., Simeonov, A., Yasgar, A., Zheng, W.
and Austin, C. P. (2006) Quantitative high-throughput screening: a titration-based
approach that efficiently identifies biological activities in large chemical libraries. Proc.
Nat. Acad. Sci. U.S.A. 103, 11473–11478
22 Shukla, S. J., Sakamuru, S., Huang, R. L., Moeller, T. A., Shinn, P., VanLeer, D., Auld,
D. S., Austin, C. P. and Xia, M. H. (2011) Identification of clinically used drugs that
activate pregnane X receptors. Drug Metab. Disp. 39, 151–159
23 Carey, F. A. and Sundberg, R. J. (2007) Advanced Organic Chemistry: Structure and
Mechanisms. Springer Verlag
24 Bakker, B. M., Michels, P. A. M., Opperdoes, F. R. and Westerhoff, H. V. (1997) Glycolysis
in bloodstream form Trypanosoma brucei can be understood in terms of the kinetics of
the glycolytic enzymes. J. Biol. Chem. 272, 3207–3215
25 Pendergrass, D. C., Williams, R., Blair, J. B. and Fenton, A. W. (2006) Mining for
allosteric information: natural mutations and positional sequence conservation in
pyruvate kinase. IUBMB Life 58, 31–38
Biochem. J. (2012) 448, 67–72 (Printed in Great Britain)
doi:10.1042/BJ20121014
SUPPLEMENTARY ONLINE DATA
A new family of covalent inhibitors block nucleotide binding to the active
site of pyruvate kinase
Hugh P. MORGAN*, Martin J. WALSH†, Elizabeth A. BLACKBURN*, Martin A. WEAR*, Matthew B. BOXER†, Min SHEN†,
Henrike VEITH†, Iain W. MCNAE*, Matthew W. NOWICKI*, Paul A. M. MICHELS‡, Douglas S. AULD†,
Linda A. FOTHERGILL-GILMORE* and Malcolm D. WALKINSHAW*1
*Centre for Translational and Chemical Biology, School of Biological Sciences, University of Edinburgh, Michael Swann Building, The King’s Buildings, Mayfield Road, Edinburgh EH9
3JR, U.K., †NIH Chemical Genomics Center, NIH Center for Translational Therapeutics, National Human Genome Research Institute, National Institutes of Health, 9800 Medical Center
Drive, Rockville, MD 20850, U.S.A., and ‡Research Unit for Tropical Diseases, de Duve Institute and Laboratory of Biochemistry, Université catholique de Louvain, Avenue Hippocrate
74, B-1200 Brussels, Belgium
EXPERIMENTAL
Site-directed mutagenesis and characterization
Site-directed mutagenesis of the LmPYK gene was performed on
plasmid pET3a-LmPYK [1]. For the Lys335 to arginine mutation,
two complementary oligonucleotides containing the desired
mutation were synthesized (forward primer, 5 -GCTGTCTGGTGAGACGGCGCGAGGCAAGTATCCGAATGAGG-3 , and
reverse primer, 5 -CCTCATTCGGATACTTGCCTCGCGCCGTCTCACCAGACAGC-3 , mutated codons are shown in italics).
The total volume of the amplification mixture was 50 μl
containing 50 ng of plasmid, 125 ng of each primer, 200 μM each
of the four dNTPs and 2.5 units of Pfu polymerase. PCR was
performed using the following programme: first 30 s at 95 ◦ C;
16 cycles of 30 s at 95 ◦ C, 1 min 55 ◦ C and 10 min 68 ◦ C. Then,
10 units of the DpnI restriction enzyme was added directly to each
amplification reaction. The reaction mixtures were incubated at
37 ◦ C for 2 h to digest the parental DNA, and used to transform
E. coli DH5α cells (Invitrogen, catalogue number 18263-012).
The presence of the mutations and the absence of other changes
in the gene were verified by sequencing.
General chemistry methods
All air- or moisture-sensitive reactions were performed under positive pressure of nitrogen with oven-dried glassware. Anhydrous
solvents such as toluene, DMF (N,N-dimethylformamide), dioxane and triethylamine were obtained from Sigma–Aldrich. Preparative purification was performed on a Waters semi-preparative
HPLC system. The column used was a Phenomenex Luna C18
(5 μm, 30 mm×75 mm) at a flow rate of 45 ml/min. The mobile
phase consisted of acetonitrile and water [each containing 0.1 %
TFA (trifluoroacetic acid)]. A gradient of 10–50 % acetonitrile
over 8 min was used during the purification. Fraction collection
was triggered by UV detection (220 nm). Analytical analysis
of compound stability in assay buffer and towards Ac-N-LysOMe, as well as purity determinations were performed on an
Agilent LC (liquid chromatography)/MS (Agilent Technologies).
The LC/MS method was as follows. A 7 min gradient of
4–100 % acetonitrile (containing 0.025 % TFA) in water
(containing 0.05 % TFA) was used with an 8 min run time
at a flow rate of 1 ml/min. A Phenomenex Luna C18 column
(3 μm, 3 mm×75 mm) was used at a temperature of 50 ◦ C.
Mass determinations were performed using an Agilent 6130
mass spectrometer with electrospray ionization in the positive
mode. 1 H-NMR spectra were recorded on Varian 400 MHz
spectrometers. Chemical shifts are reported in p.p.m. with
residual chloroform (7.27 p.p.m.) as an internal standard for
deuterated chloroform solutions, or residual non-deuterated
DMSO (2.50 p.p.m.) as an internal standard for [2 H6 ]DMSO
solutions. All of the analogues for assay have a purity greater
than 95 % on the basis of both analytical methods. HRMS
(high-resolution MS) was recorded on an Agilent 6210 time-offlight LC/MS system. Confirmation of molecular formulae was
accomplished using electrospray ionization in the positive mode
with the Agilent Masshunter software (version B.02).
General procedure for the synthesis of suicide inhibitors
The above chloride 1 was prepared according to the procedure
described by Kotake and co-workers [2]. To a solution of
saccharin (20.0 g, 109 mmol, 1.0 equiv.) and DMF (1.65 ml)
in dioxane (108 ml) was added thionyl chloride (11.95 ml,
164 mmol, 1.5 equiv.). The reaction vessel was fitted with a
condenser and was heated to reflux for 48 h. After cooling,
the reaction was concentrated in vacuo and the resulting oily
solid was recrystallized from hot toluene to give the chloride
1 (14.02 g, 64 %) after filtration. To a suspension of 1 (30 mg,
0.149 mmol, 1.0 equiv.) in toluene (0.595 ml) was sequentially
added triethylamine (21 μl, 0.149 mmol, 1.0 equiv.) and an
appropriate nucleophile (0.149 mmol, 1.0 equiv.). After stirring
for 1 h, the reaction was concentrated, reconstituted in DMSO and
purified via reverse-phase preparative HPLC to afford the targeted
saccharin-derived suicide inhibitors.
NMR data for NCGC00186526 was as follows: 1 H-NMR
(400 MHz, [2 H6 ]DMSO) δ p.p.m. 8.22 (1 H, d, J 5.5 Hz), 8.17
(1 H, d, J 7.0 Hz), 7.92–8.08 (2 H, m), 7.23–7.35 (2 H, m),
7.14 (1 H, d, J 7.4 Hz), 2.33 (3 H, br. s.), 2.15 (3 H, s); LC/MS
1
To whom correspondence should be addressed (email [email protected]).
The atomic co-ordinates of the LmPYK–DBS structure have been deposited in the PDB under code 3SRK.
c The Authors Journal compilation c 2012 Biochemical Society
H. P. Morgan and others
data for NCGC00186526 was as follows: RT (retention time)
(min) = 6.101; HRMS data for NCGC00186526 was as follows:
(M + H) + = 288.0695 (calculated for C15 H14 NO3 S, 288.0689).
NMR data for NCGC00059857 was as follows: 1 H-NMR (400
MHz, [2 H6 ]DMSO) δ p.p.m. 10.77 (1 H, s), 8.37 (1 H, d, J 7.0
Hz), 8.04 (1 H, d, J 7.8 Hz), 7.84–7.95 (2 H, m), 7.18–7.29 (2 H,
m), 7.13 (1 H, d, J 7.8 Hz), 2.33 (3 H, s), 2.21 (3 H, s); LC/MS data
for NCGC00059857 was as follows: RT (min) = 5.128; HRMS
data for NCGC00059857 was as follows: (M + H) + = 287.0855
(calculated for C15 H15 N2 O2 S, 287.0849).
NMR data for NCGC00188411 was as follows: 1 H-NMR
(400 MHz, [2 H6 ]DMSO) δ p.p.m. 8.18 (1 H, d, J 7.0 Hz), 8.10
(1 H, d, J 7.4 Hz), 7.90 – 8.01 (2 H, m), 7.51 (1 H, s), 7.30
– 7.43 (2 H, m), 2.35 (3 H, s), 2.33 (3 H, s); LC/MS data
for NCGC00188411 was as follows: RT (min) = 6.394; HRMS
data for NCGC00188411 was as follows: (M + H) + = 304.0463
(calculated for C15 H14 NO2 S2 , 304.0460).
Figure S1
Stabilities of suicide inhibitor analogues
The percentage remaining is calculated by using the equation [AUC of SM]/[AUC of SM + AUC of
product], where AUC is the area under the curve as monitored at 254 nm using the LC/MS method
described in the Experimental section, SM is the starting material peak, and product correlated
with saccharin for (a) and lysine adduct for (b). (a) Stability of NCGC00188411, NCGC00059857
and NCGC00 186 526 in assay buffer. The ether-containing analogue (NCGC00186526) shows
hydrolysis to the saccharin and the phenol, whereas the sulfide and aniline derivatives are
completely stable. (b) Stability of NCGC00188411, NCGC00059857 and NCGC00186526
towards five equivalents of N α-acetyl-L-lysine methyl ester in assay buffer. The ether-containing
analogue (NCGC00186526) shows rapid formation of the lysine adduct with concomitant release
of the phenol, whereas the sulfide (NCGC00188411) analogue shows the same adduct being
formed with concomitant release of the thiophenol, but at a lower rate. The aniline derivative
(NCGC00059857) is completely stable and shows no formation of the lysine adduct.
(area under the curve) of the SM (starting material) and product
(saccharin). No other by-products or side reactions were observed
by LC/MS in all experiments. The percentage remaining was
determined using the following equation:
NMR data for NCGC00188636 was as follows: 1 H-NMR (400
MHz, [2 H6 ]DMSO) δ p.p.m. 13.49 (1 H, br. s.), 8.18 (1 H, d,
J 7.0 Hz), 8.11 (1 H, d, J 7.0 Hz), 8.06 (1 H, d, J 7.4 Hz),
7.94 (3 H, dd, J 16.0, 7.0 Hz), 7.68–7.81 (2 H, m); LC/MS data
for NCGC00188636 was as follows: RT (min) = 4.795; HRMS
data for NCGC00188636 was as follows: (M + H) + = 320.0044
(calculated for C14 H10 NO4 S2 , 320.0046).
General procedure for analysis of stability of suicide inhibitors in
assay buffer
To aqueous buffer (135 μl of 100 mM TEA buffer, pH 7.5)
in an LC/MS vial was added the suicide inhibitor (15 μl of a
10 mM DMSO stock solution). The vial was capped and the
mixture was vortex-mixed for 2 min. Hydrolysis of the suicide
inhibitor was monitored by injecting 3 μl of the solution on to
the Agilent LC/MS system and tracking, at 254 nm, the AUC
c The Authors Journal compilation c 2012 Biochemical Society
Percentage remaining = [AUC of SM]/[AUC of SM
+ AUC of product]
Subsequent injections were made at intervals denoted in Figure
S1(a).
General procedure for analysis of stability of suicide inhibitors to
lysine
To aqueous buffer (60 μl of 100 mM TEA, pH 7.5) in an
LC/MS vial was sequentially added a solution of Nα-acetyl-Llysine methyl ester in TEA buffer (75 μl of a 10 mM solution)
and a DMSO solution of suicide inhibitor (15 μl of a 10 mM
DMSO stock solution). The vial was capped and the mixture was
Covalent inhibition of pyruvate kinase
Figure S2
Similar lysine residues found in both rabbit lactate dehydrogenase and Lm PYK
56
Lys observed in the active site of rabbit lactate dehydrogenase (PDB code 3H3F) (a) fulfils a similar function to the active-site Lys335 observed in Lm PYK (b) which is covalently modified by DBS.
Figure S3
DBS and suramin bind in a similar position
Model of Lm PYK–DBS (modelled in its initial binding position) superimposed on to the Lm PYK–suramin complex [6]. Notice that the sulfonamide group of DBS (pink sticks) and the sulfone group
in suramin (yellow sticks) occupy a similar position.
vortex-mixed for 2 min. Covalent modification of the suicide
inhibitor was monitored by injecting 3 μl of the solution on to the
Agilent LC/MS system and tracking, at 254 nm, the AUC (area
under the curve) of the SM (starting material) and product (lysine
adduct). No other by-products or side reactions were observed
by LC/MS in all experiments. The percentage remaining was
determined using the following equation:
Percentage remaining = [AUC of SM]/[AUC of SM
+AUC of product]
Subsequent injections were made at intervals denoted in Figure
S1(b).
Screening and hit optimization
There were 292 740 compounds in the NIH Molecular Libraries
Small-Molecule Repository tested in the primary screen for
the wild-type LmPYK (PubChem AID 1721). The screen was
performed at seven compound concentrations using qHTS [3,4]
and identified 1087 high-quality concentration–response curves,
corresponding to a hit rate of 0.4 % of the library. A series of
saccharin derivatives were identified after hits were filtered on the
basis of promiscuity within internal screens run at the NCGC (NIH
Chemical Genomics Center) and on assay artifacts (e.g. luciferase
enzyme inhibitors). One of the top actives from this series was
NCGC00186526, with an IC50 of 10 μM. It was quickly realized
that the oxo linkage in this compound was labile, and the molecule
was found to hydrolyse to saccharin and the corresponding
phenol in the assay buffer at room temperature (supporting
Figure S1a). In an effort to modulate this instability, both
the sulfur (NCGC00188411) and nitrogen (NCGC00059857)
analogues were prepared, and their stability was evaluated in
the assay buffer. Both molecules showed complete stability
up to 48 h with no observable hydrolysis as monitored by
LC/MS analysis (Figure S1a). These two new analogues were
subsequently tested in the LmPYK activity assay and only
NCGC00188411 showed inhibitory activity (IC50 = 5 μM). At
this point it was hypothesized that covalent modification of
either cysteine or lysine residues in the enzyme, as well as
the leaving group ability of the resultant phenol, thiophenol
and aniline, may explain the trend in activity [5]. As such,
NCGC00188411, NCGC00059857 and MLS000713929 were
tested for their stability towards five equivalents of Nα-acetylL-lysine methyl ester in aqueous buffer using LC/MS analysis.
Stability was assessed by tracking the disappearance of the parent
compound peak and the appearance of the lysine adduct peak over
a 48 h period (Figure S1b). These experiments clearly indicated
that both the oxo- and sulfur-bridged compounds could participate
c The Authors Journal compilation c 2012 Biochemical Society
H. P. Morgan and others
REFERENCES
Figure S4
1 Morgan, H. P., McNae, I. W., Nowicki, M. W., Hannaert, V., Michels, P. A. M.,
Fothergill-Gilmore, L. A. and Walkinshaw, M. D. (2010) Allosteric mechanism of pyruvate
kinase from Leishmania mexicana uses a rock and lock model. J. Biol. Chem. 285,
12892–12898
2 Ahmed, A., Taniguchi, N., Fukuda, H., Kinoshita, H., Inomata, K. and Kotake, H. (1984) A
new and effective synthetic method for the preparation of the esters, peptides, and lactones
using 3-(5-nitro-2-oxo-1,2-dihydro-1-pyridyl)-1,2-benzisothiazole 1,1-dioxide: synthesis
of ( + / − )-(E)-8-dodecen-11-olide, Recifeiolide. Bull. Chem. Soc. Japan 57, 781–786
3 Inglese, J., Auld, D. S., Jadhav, A., Johnson, R. L., Simeonov, A., Yasgar, A., Zheng, W. and
Austin, C. P. (2006) Quantitative high-throughput screening: a titration-based approach
that efficiently identifies biological activities in large chemical libraries. Proc. Nat. Acad.
Sci. U.S.A. 103, 11473–11478
4 Shukla, S. J., Huang, R., Austin, C. P. and Xia, M. (2010) The future of toxicity testing: a
focus on in vitro methods using a quantitative high-throughput screening platform. Drug
Discovery Today 15, 997–1007
5 Carey, F. A. and Sundberg, R. J. (2007) Advanced Organic Chemistry, Part A. Springer,
New York
6 Morgan, H. P., McNae, I. W., Nowicki, M. W., Zhong, W., Michels, P. A. M., Auld, D. S.,
Fothergill-Gilmore, L. A. and Walkinshaw, M. D. (2011) The trypanocidal drug suramin and
other trypan blue mimetics are inhibitors of pyruvate kinases and bind to the adenosine
site. J. Biol. Chem. 286, 31232–31240
DBS inhibition of Lm PYK and human M2 PYK
Concentration–response curves observed for the titration of DBS against Lm PYK and human
M2PYK activity.
Table S1
Kinetic parameters for wild-type and the K335R mutant of Lm PYK
Ligand
Effector
Kinetic parameter
Wild-type
K335R
ADP*
PEP†
None
None
K m (mM)
S 0.5 (mM)
n
S 0.5 (mM)
n
0.12 +
− 0.02
1.06 +
− 0.06
1.5
0.11 +
− 0.03
1.0
0.17 +
− 0.10*
0.96 +
− 0.14
1.7
0.04 +
− 0.02†
∼ 1.0§
F26BP‡
*The ADP titration gives a typical hyperbolic rate plot but at higher concentrations of ADP
(>2 mM) the rate begins to decrease slowly.
†The PEP titration in the presence of F26BP did not give a typical hyperbolic rate plot. At
higher concentrations of PEP (>1 mM) the rate begins to increase slowly.
‡F26BP at 10 μM.
§Velocity increases linearly at higher substrate concentrations.
in covalent modification, whereas the nitrogen analogue remained
intact. To further investigate the covalent modification event and
to aid in crystallization studies, DBS (NCGC00188636), a more
soluble sulfur analogue was synthesized and used for subsequent
experiments.
Received 25 June 2012/27 July 2012; accepted 20 August 2012
Published as BJ Immediate Publication 20 August 2012, doi:10.1042/BJ20121014
c The Authors Journal compilation c 2012 Biochemical Society