Selective inhibition of Trypanosoma cruzi GAPDH by "bi-substrate" analogues

Article abstract of DOI:10.1039/b504703j

A new series of "bi-substrate" analogues have been synthesized as potential inhibitors of the glyceraldehyde-3-phosphate dehydrogenase and one lead compound has been identified that inhibits the enzyme from Trypanosoma cruzi with good affinity and very high (50-fold) specificity. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b504703j

View Article Online / Journal Homepage / Table of Contents for this issue
C O M M U N I C A T I O N
Selective inhibition of Trypanosoma cruzi GAPDH by
“bi-substrate” analogues
Sylvain Ladame,*a,b Re´gis Faure´,b Colette Denier,b Faouzi Lakhdar-Ghazalb and
Miche`le Willsonb
a University Chemical Laboratory, University of Cambridge, Cambridge, CB2 1EW, UK.
E-mail: sl324@cam.ac.uk; Fax: (+44) (0)1223336913; Tel: (+44) (0)1223762933
b Groupe de Chimie Organique Biologique, LSPCMIB, Universite´ Paul Sabatier, 31062,
Toulouse, France
Received 5th April 2005, Accepted 29th April 2005
First published as an Advance Article on the web 9th May 2005
A new series of “bi-substrate” analogues have been syn-
thesized as potential inhibitors of the glyceraldehyde-3-
phosphate dehydrogenase and one lead compound has been
identified that inhibits the enzyme from Trypanosoma cruzi
with good affinity and very high (50-fold) specificity.
Trypanosomes are flagellated protozoa responsible for seri-
ous parasitic diseases that have been classified by the World
Health Organization as tropical diseases of major importance.1
Trypanosoma cruzi in particular is the causative agent of
Chagas’ disease in South America and is highly dependent on
its host’s glycolysis in infectious stages for its energy supply.
Thus, glycolytic enzymes have been considered as a promising
target for the design of new drugs against parasitic trypanoso-
matid protozoa.2 Among all glycolytic enzymes, the glyco-
somal glyceraldehyde-3-phosphate dehydrogenase (gGAPDH)
was chosen as our target of interest since (i) its inhibition was
reported to significantly reduce the ATP supply of the parasite3
and (ii) its specific inhibition by adenosine analogues was shown
to kill bloodstream-form Trypanosoma brucei amastigotes
within a few minutes without affecting the growth of fibroblasts.4
GAPDH catalyses the reversible oxidative phosphorylation of
D-glyceraldehyde-3-phosphate (GAP) into 1,3-bis-phospho-D-
glyceric acid (1,3-diPG) in the presence of NAD+ as cofactor and
inorganic phosphate. Two highly conserved cationic sites have
been identified in GAPDH active site from the co-crystallisation
of two sulfate (or phosphate) ions within the protein. These two
sites correspond to the interaction sites of the C-3 phosphate
of GAP (called Ps) and of inorganic phosphate (called Pi).
The Ps site plays a crucial role in the catalytic mechanism
since it is occupied by the C-3 phosphate of GAP during the
phosphorylation step. The crystal structure revealed that the
cofactor NAD+ participates in the stabilisation of the phosphate
anion at the Ps site via an electrostatic interaction involving the
3ꢀ-hydroxyl of the ribose of the nicotinamide scaffold.5
Scheme 1
Herein, we describe the design and synthesis of a family
of original bi-substrate (GAP + NAD+) analogues and their
detailed inhibition properties tested against the GAPDHs of
both mammal and parasite. We report the selective inhibition
of Trypanosoma cruzi GAPDH with competitive phosphonate
inhibitors that were designed to target simultaneously GAP and
NAD+ binding sites.
First generation molecules were synthesized that consist of a
phosphonate group (mimic of the C-3 phosphate of GAP) asso-
ciated to a truncated NAD+ (limited to its ribose–nicotinamide
moiety) through linkers of different length (Scheme 1). While
more than 200 NADH analogues have been already described
and studied for their ability to work as hydride donor in various
dehydrogenases, most of these cofactor modifications have been
carried out on the adenine moiety. Only a few ribosyl-b-
nicotinamides have been synthesized by displacing an anomeric
halide with nicotinamide.6 In order to selectively phosphory-
late the ribofuranose at the 2ꢀ position, 3,5-di-O-benzoyl-D-
ribofuranosyl chloride (2) was used as starting material. Initially
prepared by selective chlorination (HClg, CCl4) of the 2,3,5-
tri-O-benzoyl-a-D-ribofuranose (1), the ribofuranosyl chloride
quantitatively and stereoselectively reacts with nicotinamide,
leading exclusively to the b-anomer of the resulting 3-carbamoyl-
1-ribofuranosyl-pyridinium chloride 3.7 On this modified ribose,
two phosphonates were introduced by formation of an ester
linkage with the 2ꢀ hydroxyl. Due to the high unstability of the
ribonicotinamide under alkaline conditions, this esterification
was carried out under neutral conditions by coupling the
appropriate carboxylic acids, activated with the DCC–DMAP
couple, on the 2ꢀ hydroxyl. By following such a strategy, diethyl-
2-phosphonoacetic acid and diethyl-4-phosphonocrotonic acid
were coupled, thus leading to compounds 4 and 5 with 75 and
65% yields, respectively. The cleavage of the diethoxyphosphoric
2 0 7 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 0 7 0 – 2 0 7 2
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
©

View Article Online
protecting groups by a standard and specific procedure (Me3SiBr
and then H2O) led to the expected phosphonic acids 6† and
7‡. Both 3ꢀ and 5ꢀ benzoyl protecting groups were kept for all
inactivation studies in order to prevent an acyl migration of the
phosphonic chain towards the 3ꢀ position and to replace the
pyrophosphate at the 5ꢀ position as in the entire NAD+.
Second generation inhibitors were synthesized in which the
ribose ring was replaced by either an aliphatic spacer or a
more conformationally constrained aromatic ring. These new
inhibitors (Scheme 2) were synthesized from the dibromoalka-
nes and dibromomethylbenzenes by an Arbuzov condensation
followed by addition of nicotinamide on the remaining halide.
In a last step, the deprotection of the diethoxy phosphoric esters
by a standard procedure (Me3SiBr and then H2O) led to the
expected phosphonic acids. In these simplified “phosphono-
nicotinamides”, the phosphonyl and the nicotinamide moities
are joined either by flexible and aliphatic spacers (compounds
8 and 9) or by constrained and hydrophobic aromatic linkers
(compounds 10 and 11).
The inhibitory activities of all “bi-substrate” analogues were
tested against the GAPDHs from Trypanosoma cruzi and rabbit
muscle as previously reported by us by spectrophotometrically
monitoring the reduction of NAD+ into NADH at 340 nm,
using the substrate GAP and cofactor (NAD+) at saturating
concentrations of 0.8 and 2 mM, respectively. The inhibitory
activities were measured after preincubation of the enzyme with
the ligand for 5 min followed by addition of the substrates
and cofactor. The reaction was then monitored by following
the absorbance change of NADH at 340 nm using a SAFAS
spectrophotometer equipped with a kinetic accessory unit.
Initial reaction rates were calculated from the slopes of the
curves recorded during the first 3 min of the reaction. The
concentration of inhibitor required for 50% inhibition (IC50) was
calculated from the percentage of remaining enzyme activity by
comparison with an inhibitor free control experiment and based
on measurements at five different concentrations of inhibitor.8
The inhibition pattern and inhibition constants (Ki) were
subsequently determined from Lineweaver–Burk plots. The
inhibitions with respect to GAP and then NAD+ were studied
using at least five different concentrations of substrate/cofactor
and at least three concentrations of ligand. The inhibition results
are summarized in Table 1. Initial experiment carried out using
these molecules in replacement of the natural cofactor NAD+
showed no enzyme activity, even at very high concentration of
ligand. This confirmed that these molecules are not recognised
as an enzyme pseudo-substrate and do not interfere with the
enzyme activity measurements.
Scheme 2
molecules based on a similar scaffold but functionalized at
position 2ꢀ of the sugar with a phosphonate group (compounds
6 and 7) were shown to inhibit Trypanosoma cruzi GAPDH
with an IC50 value of 5 and 75 lM, respectively. Even more
interesting is the fact that these molecules appeared to be specific
for the parasite’s enzyme and exhibit a very moderate inhibitory
effect of the mammal’s enzyme (IC50 values ca. mM). Detailed
inactivation kinetics studies were then carried out which showed
that these inhibitors are competitive with respect to the GAP,
The non-phosphorylated intermediate 3 was also tested as a
potential inhibitor of GAPDH but appeared to have no effect
on the enzyme activity at a 3 mM concentration. However,
Table 1 Inhibitory effect of “bi-substrate” analogues with respect to either GAP or NAD+ against GAPDHs from Trypanosoma cruzi (Tc) and
rabbit muscle (Rm)
Compound
IC50/lM
Inhibition of GAP/lM
Inhibition of NAD+/lM
a
a
a
3
Rm
Tc
—
—
—
—
—
—
a
a
a
6
Rm
Tc
Rm
Tc
Rm
Tc
Rm
Tc
Rm
Tc
Rm
Tc
1300
5
650
75
Ki = 200 10 (competitive)
Ki = 4 1 (competitive)
Ki = 180 15 (competitive)
Ki = 6 1 (competitive)
NDb
Ki = 900 50 (mixed)
Ki = 10 1 (non competitive)
Ki = 40 2 (mixed)
Ki = 33 2 (mixed)
NDb
7
8
1200
1000
NDb
NDb
a
a
a
9
—
—
—
1100
850
750
1800
2000
Ki = 1050 50 (competitive)
Ki = 950 40 (non competitive)
10
11
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
a No inhibition observed at a ligand concentration of 3 mM. b Not determined.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 0 7 0 – 2 0 7 2
2 0 7 1

View Article Online
thus suggesting an interaction of the phosphonate group at the
binding site of the C3 phosphate of GAP. However, these com-
pounds also show a mixed-type or non-competitive inhibition
with respect to the cofactor, suggesting that they do not fully
occupy the NAD+ binding site or do not completely prevent its
binding. This is particularly remarkable for compound 6, the
most potent inhibitor of this series, which was non-competitive
with regard to NAD+ for T. cruzi GAPDH, indicating that the
ligand binds to the enzyme–cofactor binary complex and to the
free enzyme with the same affinity. This could be a consequence
of the favourable interaction of one or both benzoyl protecting
groups in a hydrophobic pocket within the active site preventing
the inhibitor ribose nicotinamide from binding to the cofactor
natural binding site. These results demonstrate that, although
the phosphonate groups of compounds 6 and 7 probably have the
same binding sites in T. cruzi and Rm GAPDH (i.e. the so-called
Ps site), the ribose–nicotinamide moiety interacts specifically
with T. cruzi GAPDH, thus leading to a highly specific (50-
fold) inhibition of the parasite’s enzyme. Adenosine analogues
were previously reported to selectively inhibit the GAPDH of
Trypanosomatidae by targeting a hydrophobic cleft around the
adenosyl binding site, far from the catalytic cysteine.4,9 Here,
we demonstrate that it is possible to inhibit the enzyme from
parasite with a comparable specificity by targeting the enzyme
catalytic site. Co-crystallisation experiments of these inhibitors
with Trypanosoma cruzi GAPDH are in progress to elucidate
their mode of binding.
T. cruzi GAPDH. Compound 6 represents a good lead for the
design of new potential trypanocide drugs.
Acknowledgements
We acknowledge Dr M.S. Castilho for providing us with the
GAPDH from Trypansoma cruzi and Cancer Research UK for
financial support.
Notes and references
† 3-Carbamoyl-1-(2-(2-oxoethanephosphonate)-3,5-di-O-benzoyl-D-ri-
bofuranosyl) pyridinium chloride (6). dH (CD3OD) 9.61 (s, 1H, H2),
9.36 (d, 1H, H4), 9.02 (d, 1H, H6), 8.24 (dd, 1H, H5), 7.25–8.13 (m, 10H,
ꢀ
ꢀ
ꢀ
Ph), 6.82 (d, 1H, H1 ), 5.90 (m, 2H, H5 ), 5.15 (m, 1H, H4 ), 4.91 (m,
ꢀ
ꢀ
2H, H2 and H3 ), 2.96 (ddd, 2H, CH2–P); dC (CD3OD) 168.2 (d, JC–P
= 6.8 Hz), 167.5, 166.8, 164.8, 147.7, 144.1, 142.1, 136.1, 135.2, 134.9,
129.9–131.1, 98.9, 85.3, 78.0, 72.5, 64.8, 37.0 (d, JC–P = 123.0 Hz); dP
(CD3OD) 14.8; MS (ES) m/z 585 ([M+]). Purity was > 99% as assessed
by HPLC (CH3CN–H2O).
‡ 3-Carbamoyl-1-(4-(4-2-enedihydroxyphosphonobutane)-3,5-di-O-ben-
zoyl-D-ribofuranosyl) pyridinium chloride (7). dH (CD3OD) 9.67 (s, 1H,
H2), 9.40 (d, 1H, H4), 9.05 (d, 1H, H6), 8.28 (dd, 1H, H5), 7.38–8.09
ꢀ
(m, 10H, Ph), 7.12 (m, 1H), 6.85 (d, 1H, H1 ), 6.12 (dd, 1H), 5.91 (m,
ꢀ
ꢀ
ꢀ
ꢀ
2H, H5 ), 5.13 (m, 1H, H4 ), 4.55 (m, 2H, H2 and H3 ), 2.76 (dd, 2H,
CH2–P); dC (CD3OD) 167.6 (d, JC–P = 6.7 Hz), 166.8, 166.2, 164.9,
147.0, 143.6, 142.4, 136.0, 135.2, 134.9, 127.5–131.8, 123.8 (d, JC–P
=
13.1 Hz), 99.4, 84.5, 76.5, 71.7, 64.5, 33.5 (d, JC–P = 132.4 Hz); dP
(CD3OD) 20.4; MS (ES) m/z 611 ([M+]). Purity was > 99% as assessed
by HPLC (CH3CN–H2O).
In an attempt to simplify the general scaffold of these
inhibitors and investigate the influence of the ribose ring on
both affinity and selectivity of inhibition, second generation
ligands were synthesized in which the ribose had been removed
and substituted by either an aliphatic or aromatic spacer. The
inhibition results obtained with compounds 8–11 are summa-
rized in Table 1. As for compounds 6 and 7, these new ligands
appeared to be very poor inhibitors of Rm GAPDH (IC50 values
higher than 0.8 mM). Unfortunately, equally poor inhibition
effects were observed with these molecules when tested against
T. cruzi GAPDH. These results demonstrate the importance of
the ribose motif for the recognition of these inhibitors by T. cruzi
GAPDH.
In summary, we have reported the synthesis of an original
family of GAPDH inhibitors that inhibit the enzyme of Try-
panosoma cruzi with good affinity (Ki value up to 4 lM) but
also very high specificity (up to 50-fold). Detailed inactivation
studies suggested that (i) the phosphonate group interacted with
the GAP C3-phosphate binding site, and (ii) the ribose moiety is
necessary but not sufficient for inducing a specific inhibition of
1 W. H. O. The World Health Report 2000: Health Systems, Improving
Performance, 2000.
2 C. L. Verlinde, V. Hannaert, C. Blonski, M. Willson, J. J. Perie, L. A.
Fothergill-Gilmore, F. R. Opperdoes, M. H. Gelb, W. G. Hol and P. A.
Michels, Drug Resist. Updates, 2001, 1, 50.
3 B. M. Bakker, P. A. M. Michels, F. R. Opperdoes and H. V. Westerhoff,
J. Biol. Chem., 1999, 274, 14551.
4 J. C. Bressi, C. L. M. J. Verlinde, A. M. Aronov, M. Le Shaw, S. S.
Shin, L. N. Nguyen, S. Suresh, F. S. Buckner, W. C. VanVoorhis, I. D.
Kuntz, W. G. J. Hol and M. H. Gelb, J. Med. Chem., 2001, 44, 2080.
5 T. Skarzynski, P. C. E. Moody and A. J. Wonacott, J. Mol. Biol.,
1987, 193, 171; D. H. F. Souza, R. C. Garrat, A. P. U. Arau`jo, B. G.
Guimara˜es, W. D. P. Jesus, P. A. M. Michels, V. Hannaert and G. Oliva,
FEBS Lett., 1998, 424, 131; S. Ladame, M. S. Castilho, C. H. T. P.
Silva, C. Denier, V. Hannaert, J. Perie, G. Oliva and M. Willson,
Eur. J. Biochem., 2003, 270, 4574.
6 R. K. Ness and H. G. Fetcher, J. Am. Chem. Soc., 1956, 78, 4710.
7 M. Jarman and W. C. J. Ross, J. Chem. Soc., 1969, 199.
8 O. Misset, O. J. M. Bos and F. R. Opperdoes, Eur. J. Biochem., 1986,
157, 441.
9 A. M. Aronov, C. L. Verlinde, W. G. Hol and M. H. Gelb, J. Med.
Chem., 1998, 43, 4790; S. Suresh, J. C. Bressi, K. J. Kennedy, C. L.
Verlinde, M. H. Gelb and W. G. Hol, J. Mol. Biol., 2001, 309, 423.
2 0 7 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 0 7 0 – 2 0 7 2