Substrate analogs for the investigation of deoxyxylulose 5-phosphate reductoisomerase inhibition: Synthesis and evaluation

Article abstract of DOI:10.1016/j.bmcl.2004.08.023

Deoxyxylulose 5-phosphate (DXP) analogs were synthesized and evaluated as alternative substrates and inhibitors of recombinant Synechocystis PCC6803 DXP reductoisomerase (DXR; EC 1.1.1.267). Five of the compounds tested (1,2-dideoxy-d-threo-3-hexulose 6-phosphate, 1-deoxy-l-ribulose 5-phosphate, 2S,3R-dihydroxybutyramide 4-phosphate, 4S-hydroxypentan-2-one 5-phosphate, and 3S-hydroxypentan-2-one 5-phosphate) acted as relatively weak competitive inhibitors when compared to fosmidomycin. A sixth compound, 3R,4S-dihydroxy-5- oxohexylphosphonic acid, served as an alternate substrate, as has recently been reported for the same compound with Escherichia coli DXR.

Full text of DOI:10.1016/j.bmcl.2004.08.023

Bioorganic & Medicinal Chemistry Letters 14 (2004) 5309–5312
Substrate analogs for the investigation of deoxyxylulose
5-phosphate reductoisomerase inhibition: synthesis and evaluation
Chanokporn Phaosiri and Philip J. Proteau*
College of Pharmacy, Pharmacy Building. Rm. 203, Oregon State University, Corvallis, OR 97331-3507, USA
Received 7 July 2004; revised 10 August 2004; accepted 10 August 2004
Available online 9 September 2004
Abstract—Deoxyxylulose 5-phosphate (DXP) analogs were synthesized and evaluated as alternative substrates and inhibitors of
recombinant Synechocystis PCC6803 DXP reductoisomerase (DXR; EC 1.1.1.267). Five of the compounds tested (1,2-dideoxy-
D-threo-3-hexulose 6-phosphate, 1-deoxy-L-ribulose 5-phosphate, 2S,3R-dihydroxybutyramide 4-phosphate, 4S-hydroxypentan-2-
one 5-phosphate, and 3S-hydroxypentan-2-one 5-phosphate) acted as relatively weak competitive inhibitors when compared to
fosmidomycin. A sixth compound, 3R,4S-dihydroxy-5-oxohexylphosphonic acid, served as an alternate substrate, as has recently
been reported for the same compound with Escherichia coli DXR.
Ó 2004 Elsevier Ltd. All rights reserved.
DXR, Mg2+
NADPH
The discovery of the methylerythritol phosphate (MEP)
pathway as an alternate pathway for isoprenoid biosyn-
thesis in some organisms including most bacteria, plants,
and the malarial parasite Plasmodium falciparum, has
stimulated extensive research in this area.1,2 Studies have
revealed the potential of finding novel antibacterials,
antimalarials, and herbicides from enzyme inhibitors of
this pathway.3 The natural product fosmidomycin has
effective antimalarial activity in mice4 and is currently
undergoing clinical trials in humans in Africa and Thai-
land.5 Fosmidomycin (1) inhibits the second enzyme in
the MEP pathway, deoxyxylulose 5-phosphate reducto-
isomerase (DXR, EC 1.1.1.267),6 which mediates the
conversion of 1-deoxy-D-xylulose 5-phosphate (DXP,
2) into 2C-methyl-D-erythritol 4-phosphate (MEP, 3)
(Fig. 1).
OH
OH
H3C
O
2-
2-
OPO3
OPO3
X
CH3 OH
OH OH
DXP, 2
MEP, 3
OH
N
2-
O
PO3
Fosmidomycin, 1
H
Figure 1. Conversion of DXP to MEP mediated by DXR.
DXP that were considered as potential sites for designing
analogs of DXP were the methyl ketone group, the
hydroxylated backbone, and the polar phosphate head
group. All DXP analogs were synthesized and tested
against recombinant DXR from Synechocystis
PCC6803, which was used for previous studies in our
laboratory.11,12 A trivial name has been provided for
each analog that readily distinguishes the structural
change.
Although fosmidomycin and its analogs are potent
inhibitors of DXR,6 analogs of the DXR substrate,
DXP, should also be examined in order to more fully
understand the nature of DXP binding and DXR inhibi-
tion. Analysis of these compounds as substrates/inhibi-
tors for DXR will provide data that complements the
recent X-ray crystal structures of DXR7–10 and may help
in the design of putative DXR inhibitors. The features of
The compound, 1,2-dideoxy-L-threo-3-hexulose 6-phos-
phate (1-Me-DXP, 8) was designed to provide extra steric
bulk at the methyl ketone moiety of DXP. It was synthe-
sized by modifying the synthesis of DXP by Blagg and
Poulter.13 The key differences in the synthetic procedure
were use of EtMgBr instead of MeMgBr to install the
ethyl group and phosphorylation with diphenyl chloro-
phosphate (Scheme 1). The phosphate protecting groups
Keywords: Deoxyxylulose 5-phosphate; Reductoisomerase; Methyl-
erythritol phosphate pathway; Isoprenoids; Synechocystis.
*
Corresponding author. Tel.: +1 541 737 5776; fax: +1 541 737
3999; e-mail: phil.proteau@oregonstate.edu
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.08.023

5310
C. Phaosiri, P. J. Proteau / Bioorg. Med. Chem. Lett. 14 (2004) 5309–5312
OH
O
1) EtMgBr,THF, 0 ˚C
(74%; 2 Steps)
1) NaH, TIPSCl
THF (88%)
O
O
O
O
2) TPAP/NMO
CH2Cl2 (83%)
2) Swern Ox.
OH
OTIPS
4
5
O
O
O
1) HOCH2CH2OH,
p-TsOH, benzene (47%)
O
O
O
O
2) TBAF, THF (82%)
OTIPS
OH
6
7
O
1) PO(OPh)2Cl, DMAP,
pyr., THF (78%)
HO
2) PtO4/H2, EtOH
3) HCl (0.75 M, 5 h, rt)
(70%; 2 steps)
HO
OPO(OH)2
8
Scheme 1. Synthesis of 1-Methyl-DXP (8).
were removed by catalytic hydrogenolysis with Pt(IV)
oxide and acidic hydrolysis with aqueous HCl cleaved
the acetonide and dioxolane protecting groups to
provide 8.14 This analog has also been prepared enzy-
matically in low yield from a-oxobutyrate and glycer-
aldehyde 3-phosphate.15
mine in methanol directly afforded the methyl ester 10.
A reaction between the ester 10 and aqueous ammonia
gave amide 11. Finally, catalytic hydrogenolysis and
acid hydrolysis of 11 were performed to yield DXP
carboxamide (12) (12% overall yield, eight steps from
diisopropyl-D-tartrate).
Replacement of the methyl ketone moiety with an amide
should provide a compound with similar steric require-
ments to DXP, but with poorer electrophilicity at the
carbonyl moiety compared to the substrate DXP. There-
fore, this compound should not be a substrate for DXR
and may act as a DXR inhibitor. Scheme 2 shows the
procedure for the synthesis of 2S,3R-dihydroxybutyr-
amide 4-phosphate (DXP-carboxamide, 12). Monopho-
sphorylation of (ꢀ)-2,3-O-isopropylidene-D-threitol (4;
obtained in two steps from diisopropyl-D-tartrate) with
tetrabenzyl pyrophosphate provided 9. Swern oxidation
of 9 and oxidation of the resulting aldehyde with bro-
The compound 1-deoxy-L-ribulose 5-phosphate (4-epi-
DXP, 18) has the opposite stereochemistry at the C-4
position relative to DXP and will give information
about how crucial the stereochemistry of the hydroxyl
functional group is for binding to DXR and catalysis.
The synthesis of 4-epi-DXP began with L-arabinose
(13, Scheme 3). Selective isopropylidene formation and
oxidative cleavage of 13 generated a diastereomeric
mixture of 14.16 A Wittig reaction of 14 with methyltri-
phenylphosphonium bromide/potassium tert-butoxide
yielded the alkene 15. Phosphorylation of 15 was accom-
plished using tetrabenzyl pyrophosphate (TBPP).
Dihydroxylation and periodate cleavage provided the
aldehyde 16 in 62% overall yield for three steps. A single
diastereomer was obtained from the methyl Grignard
addition to the aldehyde 16, albeit in low yield. Oxida-
tion of the secondary alcohol using TPAP and NMO
provided ketone 17. Finally, 4-epi-DXP (18, 5% overall
yield, 10 steps from L-arabinose) was obtained by cata-
lytic hydrogenolysis and acid hydrolysis of 17.
OH
OH
1) Swern Ox.
O
O
O
O
NaH, TBPP/THF
(50%)
2) Br2
NaHCO3
(58%)
/MeOH,
OH
OPO(OBn)2
4
9
O
O
The three remaining compounds that were synthesized
have also been prepared and tested against the E. coli
DXR.17,18 The 3S-hydroxypentan-2-one 5-phosphate
(4-deoxy-DXP, 19) and 4S-hydroxypentan-2-one 5-
phosphate (3-deoxy-DXP, 20) compounds were de-
signed to test the importance of the hydroxyl groups
at the C3 and C4 positions for binding and for catalysis.
The phosphonate analog of DXP, 3R,4S-dihydroxy-5-
oxohexylphosphonic acid (DX-phosphonate, 21), was
prepared to determine the effect of changing from a
phosphate to a phosphonate functional group. Our syn-
thetic approaches to 3-deoxy-DXP and 4-deoxy-DXP
were similar to the routes previously published.17 The
O
O
O
O
NH3/H2O, MeOH
(68%)
OCH3
NH2
OPO(OBn)2
OPO(OBn)2
11
10
O
1) Pd/C, H2, EtOH
HO
HO
12
NH2
2) Dowex 50x8 H+,
MeOH, 12 h, rt
(77%)
OPO(OH)2
Scheme 2. Synthesis of DXP-carboxamide (12).

C. Phaosiri, P. J. Proteau / Bioorg. Med. Chem. Lett. 14 (2004) 5309–5312
5311
OH
O
1) 2,2-dimethoxypropane,
p-TsOH.H2O/DMF
CH3PPh3Br,
KOtBu, THF
HO
HO
O
O
O
OH
2)NaIO4/Na2CO3
(68% for 2 steps)
(74%)
OH
13
14
O
1) MeMgCl/THF
(37%)
1) NaH, TBPP/THF
2) OsO4/NMO
O
O
O
O
H
2) TPAP/NMO
(61%)
3) NaIO4
(62% for 3 steps)
OH
15
OPO(OBn)2
16
O
O
1) Pd/C, H2, EtOH
2) Dowex 50x8, H+
O
HO
HO
CH3
CH3
O
MeOH, 12 h, rt (90%)
OPO(OBn)2
17
OPO(OH)2
18
Scheme 3. Synthesis of 4-epi-DXP (18).
synthesis of the DX-phosphonate and analysis as an
alternate substrate for E. coli DXR were recently re-
ported.18 Our synthetic approach to DX-phosphonate
also started with a tartrate derivative, but involved a dif-
ferent series of steps to provide the final phosphonate
analog. The details of the syntheses of all three of these
compounds are provided elsewhere.19
The five DXP analogs that were not alternate substrates
were tested as potential inhibitors. All five phosphate
analogs were found to be competitive inhibitors of
DXR with respect to DXP (Fig. 2). The inhibition by
all of the compounds is rather weak, especially when
compared to the slow, tight-binding inhibitor fosmido-
mycin (Ki = 21nM).21
All six of the DXP analogs were tested as alternate sub-
strates for the Synechocystis DXR.20 All five phosphate
analogs of DXP showed no sign of being an alternate
substrate for DXR when tested at concentrations up
to 30mM. However, DX-phosphonate (21) did show
activity as an alternate substrate. This compound has
a Km = 690lM, which is fourfold higher than the
Km(DXP) (170lM).12 This is the same general result ob-
served for this phosphonate analog with the E. coli
DXR, although the Km with the Synechocystis DXR is
significantly higher than that measured with the E. coli
DXR.18 The Km value of DX-phosphonate (120lM)
with E. coli DXR was also fourfold higher than the re-
ported Km(DXP) (30lM). The change from a phosphate
group to a phosphonate leads to a decrease in binding,
but because this structural change is distant from the
portion of the molecule undergoing transformation,
turnover is still possible.
The 1-Me-DXP analog, which has an ethyl ketone
group rather than the methyl ketone moiety of DXP,
has a Ki about three and a half-fold higher than the
Km(DXP) The competitive nature of the inhibition indi-
cates that 1-Me-DXP binds at the same site as DXP,
although less tightly, but the added steric bulk of the
ethyl group prevents successful turnover of this com-
pound. As the addition of the methyl group does not
alter the electronic characteristics of the molecule, the
lack of turnover may be due to improper alignment of
the substrate in the active site or a steric interaction aris-
ing upon rearrangement.
Although the amide group of DXP carboxamide and the
methyl ketone of DXP are isosteric, the amide is much
less electrophilic than the ketone, which likely explains
the lack of turnover of this compound. The Ki value is
about half of the Km for DXP, suggesting slightly more
OH
OH
O
OH
O
O
2-
2-
2-
OPO3
OPO3
OPO3
OH
CH3 OH
NH2 OH
H3C
1-Me-DXP (8)
Ki = 630 µM
4-epi-DXP (18)
Ki = 180 µM
DXP carboxamide (12)
Ki = 90 µM
OH
OH
2-
O
O
PO3
O
2-
2-
OPO3
OPO3
CH3
CH3 OH
CH3 OH
4-Deoxy-DXP (19)
Ki = 30 µM
DX-phosphonate (21)
Km = 690 µM
3-Deoxy-DXP (20)
Ki = 150 µM
Figure 2. DXP analogs and inhibition (Ki) or Michaelis (Km) constants. The alternate substrate DX-phosphonate is indicated by a box.

5312
C. Phaosiri, P. J. Proteau / Bioorg. Med. Chem. Lett. 14 (2004) 5309–5312
5. Wiesner, J.; Borrmann, S.; Jomaa, H. Parasitol. Res. 2003,
90(Suppl. 2), S71.
6. Kuzuyama, T.; Shimizu, T.; Takahashi, S.; Seto, H.
Tetrahedron Lett. 1998, 39, 7913.
favorable interactions for the amide derivative, possibly
through additional hydrogen bonds.
None of the three analogs that have changes in the
hydroxylation pattern of DXP were alternate substrates,
indicating the importance of the presence and stereo-
chemistry of both hydroxyls for catalysis. The Ki for
4-epi-DXP and the Km for DXP are similar, suggesting
that the stereochemistry of the hydroxyl at C4 of the
substrate is not critical for binding, but it is essential
for catalysis. The absence of the C4 hydroxyl in 4-
deoxy-DXP reinforces the idea that the C4 hydroxyl is
not critical for binding because the Ki for this compound
(30lM) is about sixfold lower than the Km(DXP). The Ki
for 3-deoxy-DXP is similar to the Km(DXP) indicating
that the C3 hydroxyl also is not critical for binding,
but essential for catalysis. When tested with the E. coli
DXR, 3-deoxy-DXP and 4-deoxy-DXP were found to
be mixed type inhibitors, with Ki values of 800 and
120lM, respectively. Our results differ slightly from
the E. coli results in that both compounds were compet-
itive inhibitors for the Synechocystis DXR, rather than
mixed type inhibitors. The Ki values were also lower,
although in both cases, the Ki value for 4-deoxy-DXP
was 5–6-fold lower than for 3-deoxy-DXP. These results
may reflect slight differences in the active sites of the
E. coli and Synechocystis enzymes.
7. Reuter, K.; Sanderbrand, S.; Jomaa, H.; Wiesner, J.;
Steinbrecher, I.; Beck, E.; Hintz, M.; Klebe, G.; Stubbs,
M. T. J. Biol. Chem. 2002, 277, 5378.
8. Yajima, S.; Nonaka, T.; Kuzuyama, T.; Seto, H.; Ohsawa,
K. J. Biochem. 2002, 131, 313.
9. Steinbacher, S.; Kaiser, J.; Eisenreich, W.; Huber, R.;
Bacher, A.; Rohdich, F. J. Biol. Chem. 2003, 278,
18401.
10. Ricagno, S.; Grolle, S.; Bringer-Meyer, S.; Sahm, H.;
Lindqvist, Y.; Schneider, G. Biochim. Biophys. Acta 2004,
1698, 37.
11. Proteau, P. J.; Woo, Y.-H.; Williamson, R. T.; Phaosiri,
C. Org. Lett. 1999, 1, 921.
12. Yin, X.; Proteau, P. J. Biochim. Biophys. Acta 2003, 1652,
75.
13. Blagg, B. S. J.; Poulter, C. D. J. Org. Chem. 1999, 64,
1508.
14. All compounds were characterized by 1H, 13C, and 31P
NMR as well as HRMS. 1-Me-DXP, 8, Na+ salt:
25
½aꢁD ꢀ 1:7 (c 3.0, H2O); 1H NMR (300MHz, D2O): d
4.53 (d, J = 1.8Hz, 1H, H4), 4.33 (dt, J = 1.8, 6.5Hz, 1H,
H5), 3.88 (dd, J = 7.1, 6.7Hz, 2H, H6), 2.71 (m, 2H, H2a,
H2b), 1.07 (t, J = 7.2Hz, 3H, H1); 13C NMR (75MHz,
D2O): d 216.98 (C3), 76.78 (C4), 70.96 (d, J = 6.5Hz, C5),
65.56 (d, J = 3.8Hz, C6), 32.31 (C2), 7.13 (C1); 31P NMR
(121MHz, D2O):
d
1.05; HRMS-FAB calcd for
C6H13O7P: 228.0399; found 228.0392. DXP-Carboxamide,
25
1
12, Na+ salt: ½aꢁD ꢀ 26 (c 2.0, H2O); H NMR (300MHz,
D2O): d 4.28 (d, J = 2.1Hz, 1H, H2), 4.19 (dt, J = 2.0,
6.8Hz, 1H, H3), 4.00 (m, 2H, H4); 13C NMR (75MHz,
D2O): d 178.75 (C1), 71.80 (C2), 71.34 (d, J = 8.3Hz, C3),
67.18 (d, J = 4.0Hz, C4); 31P NMR (121MHz, D2O): d
0.62; HRMS-FAB calcd for C4H9NO7P: 214.0117; found
The biochemical analysis of these DXP analogs has pro-
vided further information about the interactions of the
substrate DXP with DXR. The importance of the hydr-
oxyl groups has been clearly demonstrated, as well as
the steric limitations near the ketone group of DXR.
These data should be helpful, when used in concert with
the current X-ray crystal structures of the E. coli7–9 and
Zymomonas10 DXR, for designing potent inhibitors of
this important enzyme.
25
214.0129. 4-epi-DXP, 18, Na+ salt: ½aꢁD ꢀ 4 (c 0.7, H2O);
1H NMR (300MHz, D2O): d 4.26 (d, J = 5.0Hz, 1H, H3),
4.09 (dt, J = 5.3, 5.6Hz, 1H, H4), 3.90 (dd, J = 6.0, 5.8Hz,
2H, H5), 2.18 (s, 3H, H1); 13C NMR (75MHz, D2O): d
213.93 (C2), 78.31 (C3), 71.50 (d, J = 8.8Hz, C4), 66.57 (d,
J = 5.3Hz, C5), 27.84 (C1); 31P NMR (121MHz, D2O): d
0.54; HRMS-FAB calcd for C5H10O7P: 213.0164; found
213.0167.
Acknowledgements
The authors would like to thank Dr. Xihou Yin for
assistance in the preparation of the recombinant DXR
for kinetic studies. Support for this work was provided
by a grant from the National Institutes of Health
(RO1 AI42258A). This publication was made possible
in part by grant number P30 ES00210 from the National
Institute of Environmental Health Sciences, NIH.
15. Schurmann, M.; Schurmann, M.; Sprenger, G. A. J. Mol.
¨
Catal. B: Enzym. 2002, 19–20, 247.
¨
16. Thompson, D. K.; Hubert, C. N.; Wightman, R. H.
Tetrahedron 1993, 49, 3827.
17. Hoeffler, J. F.; Tritsch, D.; Grosdemange-Billiard, C.;
Rohmer, M. Eur. J. Biochem. 2002, 269, 4446.
18. Meyer, O.; Grosdemange-Billiard, C.; Tritsch, D.; Roh-
mer, M. Org. Biomol. Chem. 2003, 1, 4367.
19. Phaosiri, C. Ph.D. Dissertation, Oregon State University,
2004.
References and notes
20. The recombinant enzyme used in these experiments was
the N-terminal hexahistidine-tagged DXR.12 General
assay conditions were as previously described.12 For the
determination of the Km for DX-phosphonate, seven
concentrations of substrate were used in triplicate assays.
Inhibition assays were also performed in triplicate, with
seven different concentrations of substrate assayed at five
different concentrations of inhibitor.
1. Rodriguez-Concepcion, M.; Boronat, A. Plant Physiol.
2002, 130, 1079.
2. Kuzuyama, T. Biosci. Biotechnol. Biochem. 2002, 66, 1619.
3. Testa, C. A.; Brown, M. J. Curr. Pharm. Biotechnol. 2003,
4, 248.
4. Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek, B.;
Weidemeyer, C.; Hintz, M.; Turbachova, I.; Eberl, M.;
¨
21. Koppisch, A. T.; Fox, D. T.; Blagg, B. S. J.; Poulter, C. D.
Biochemistry 2002, 41, 236.
Zeidler, J.; Lichtenthaler, H. K.; Soldati, D.; Beck, E.
Science 1999, 285, 1573.