Study of 1-deoxy-D-xylulose-5-phosphate reductoisomerase: Synthesis and evaluation of fluorinated substrate analogues

Article abstract of DOI:10.1021/ol048459b

(Chemical Equation Presented) 1-Deoxy-D-xylulose-5-phosphate (DXP) reductoisomerase is a NADPH-dependent enzyme catalyzing the conversion of DXP to methyl-D-erythritol 4-phosphate (MEP). In this study, each of the hydroxyl groups in DXP and one of its C-1

Full text of DOI:10.1021/ol048459b

ORGANIC
LETTERS
2004
Vol. 6, No. 20
3625-3628
Study of 1-Deoxy- -xylulose-5-phosphate
D
Reductoisomerase: Synthesis and
Evaluation of Fluorinated Substrate
Analogues
Alexander Wong,† Jeffrey W. Munos, Vidusha Devasthali,‡
Kenneth A. Johnson, and Hung-wen Liu*
DiVision of Medicinal Chemistry, College of Pharmacy and Department of Chemistry
and Biochemistry, UniVersity of Texas, Austin, Texas 78712
h.w.liu@mail.utexas.edu
Received August 3, 2004
ABSTRACT
1-Deoxy-D-xylulose-5-phosphate (DXP) reductoisomerase is a NADPH-dependent enzyme catalyzing the conversion of DXP to methyl-D-erythritol
4-phosphate (MEP). In this study, each of the hydroxyl groups in DXP and one of its C-1 hydrogen atoms, were separately replaced with a
fluorine atom and the effect of the substitution on the catalytic turnover was examined. It was found that the 1-fluoro-DXP is a poor substrate,
while both 3- and 4-fluoro-DXP behave as noncompetitive inhibitors.
Terpenoids are a family of secondary metabolites that are
widely distributed in nature and are rich in biological
activities.1,2 The terpenoid building block is a five-carbon
unit known as isoprene, which has long been established to
be derived from acetate via the mevalonate pathway.3
However, a new isoprene biosynthetic pathway has recently
been discovered where the isoprenoid unit is formed from
pyruvate and glyceraldehyde-3-phosphate.4-6 It has been
shown that 2-C-1-deoxy-D-xylulose-5-phosphate (DXP, 1)
is the immediate product resulting from condensation of these
two precursors,7,8 and conversion of DXP to methyl-D-
erythritol 4-phosphate (MEP, 2) is the first pathway-specific
transformation en route to the basic isoprenoid unit.9 Since
this MEP pathway appears to operate in eubacteria, archea-
bacteria, algae, and plastids of plants, but not in humans,
enzymes involved in this pathway could be excellent targets
† Current address: Cerilliant Corp., 811 Paloma Drive Suite A, Round
Rock, TX 78664.
‡ Current address: Department of Pathology, Howard Hughes Medical
Institute, Stanford, CA 94304.
(4) Eisenreich, W.; Schwarz, M.; Cartayrade, A.; Arigoni, D.; Zenk, M.
H.; Bacher, A. Chem. Biol. 1998, 5, R221-R233.
(5) Rohmer, M. Nat. Prod. Rep. 1999, 16, 565-574.
(6) Kuzuyama, T.; Seto, H. Nat. Prod. Rep. 2003, 20, 171-183.
(7) Sprenger, G. A.; Scho¨rken, U.; Wiegert, T.; Grolle, S.; deGraaf, A.
A.; Taylor, S. V.; Begley, T. P.; Bringer-Meyer, S.; Sahm, H. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 12857-12862.
(8) Lois, L. M.; Campos, N.; Putra, S. R.; Danielsen, K.; Rohmer, M.;
Boronat, A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 2105-2110.
(9) Takahashi, S.; Kuzuyama, T.; Watanabe, H.; Seto, H. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 9879-9884.
(1) Qureshi, N.; Porter, J. W. In Biosythesis of Isoprenoid Compounds;
Porter, J. W., Spurgeon, S. L., Eds.; Wiley: New York, 1981; Vol. 1, pp
47-94.
(2) Cane, D. E. In ComprehesiVe Chemistry of Natural Products; Barton,
D., Nakanishi, K., Meth-Cohn, O., Eds.; Pergamon: New York, 1999; Vol.
2, pp 1-14.
(3) Bochar, D. A.; Freisen, J. A.; Stauffacher, C. V.; Bodwell, V. W. In
ComprehesiVe Chemistry of Natural Products; Barton, D., Nakanishi, K.,
Meth-Cohn, O., Eds.; Pergamon: New York, 1999; Vol. 2, pp 15-44.
10.1021/ol048459b CCC: $27.50
© 2004 American Chemical Society
Published on Web 09/10/2004

DXR from E. coli19 as well as a complex of DXR with
NADPH have been determined.20 Despite the advances made
so far in our understanding of the catalytic properties of
DXR, it remains elusive whether catalysis is better described
by route A or route B. In an attempt to gain more insight
into the mechanism of this intriguing enzyme, we prepared
three fluorinated substrate analogues 5-7 and examined their
competence as substrates or inhibitors upon incubation with
DXR.
Scheme 1
for the development of new antibiotics.4-6 In fact, fosmido-
mycin, which inhibits the conversion of 1 to 2,10 has been
shown to be effective for the treatment of malaria.11,12
The enzyme catalyzing the conversion of DXP (1) to MEP
(2), 1-deoxy-D-xylulose-5-phosphate reducto-isomerase (DXR,
also known as MEP synthase), has been isolated from several
sources. It is a NADPH-dependent catalyst, which also
requires a divalent metal ion for activity.9,13-15 Two possible
mechanisms for the conversion of DXP (1) to MEP (2) have
been proposed. As shown in Scheme 1, the first mechanism
involves an R-ketol rearrangement to yield methylerythrose
phosphate (3) as an intermediate, followed by the reduction
of the aldehyde group in 3 by NADPH to drive the
equilibrium to completion (route A). A mechanism involving
a retroaldolization/aldolization rearrangement is another
alternative (route B). This route yields first a bimolecular
intermediate (4), which is condensed to form 3. Subsequently,
3 is reduced by NADPH to generate the product 2.
The catalysis by the Escherichia coli enzyme has been
shown to proceed via an ordered mechanism in which
NADPH binds before the substrate, DXP, and NADP+ is
released after the discharge of MEP.14,15 Interestingly, a
random mechanism has recently been determined for the
Mycobacterium tuberculosis enzyme.16 However, due to the
significant differences in Km of the substrates in the forward
and reverse reaction, a preferred order of binding exists in
the latter case rendering its mechanism effectively similar
to that of the E. coli enzyme. Early studies have also
established that the C-1 pro-S hydrogen in MEP (2) is derived
from H-3 of DXP (1) and the hydride transfer from NADPH
is pro-S specific.17,18 In addition, the crystal structures of
If the hydroxyl substituent(s) of DXP indeed play(s) a
crucial role in DXR catalysis, its replacement with a
chemically inert fluorine will effect the catalytic turnover.
We therefore replaced each of these hydroxyl groups with
an inert fluorine atom and examined the effect of the
substitution on the catalytic turnover. In route A, deproto-
nation of the C-3 hydroxyl group followed by a 1,2-migration
to yield an aldehyde intermediate is the key step in the
mechanism. Since the C-4 hydroxyl group is not directly
involved in catalysis, compound 7 is likely to be a substrate,
while compound 6 would not be a substrate if the reaction
proceeds via an R-ketol rearrangement. The catalytic rate
for 7 might be retarded, however, due to the increased
electronegativity of fluorine. In the case of a retroaldolization
mechanism, while the C-4 hydroxyl group is involved in the
first step of the reaction, the C-3 hydroxyl group participates
in the aldolization step such that a new bond forms between
carbons derived from C-2 and C-4 of DXP to yield the
aldehyde intermediate. Thus, it is expected that if the reaction
proceeds via route B, both compounds 6 and 7 will not be
turned over by DXR. In contrast, compound 5 should always
be recognized by DXR as a substrate, since C-1 is not directly
involved in catalysis. However, the electron-withdrawing
nature of the fluorine substituent at C-1 will render the 2-keto
group more electrophilic and thus facilitate the R-ketol
rearrangement. Although a similar argument holds for the
retroaldol bond cleavage step in route B, stabilization of the
resulting enolate anion by the fluorine substituent at C-1 may
compromise its capability as a nucleophile and slow the
second half of the reaction. The effect on the overall catalysis
will depend on which step is more rate limiting.
(10) Kuzuyama, T.; Shimizu, T.; Takahashi, S.; Seto, H. Tetrahedron
Lett. 1998, 39, 7913-7916.
(11) Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek, B.; Weidemeyer,
C.; Hintz, M.; Tu¨rbachova, I.; Eberl, M.; Zeidler, J.; Lichtenthaler, H. K.;
Soldati, D.; Beck, E. Science 1999, 285, 1573-1576.
(12) Lell, B.; Ruangweerayut, R.; Wiesner, J.; Missinou, M. A.;
Schindler, A.; Baranek, T.; Hintz, M.; Hutchinson, D.; Jomaa, H.; Kremsner,
P. G. Antimicrob. Agents Chemother. 2003, 47, 735-738.
(13) Kuzuyama, T.; Takahashi, S.; Takagi, M.; Seto, H. J. Biol. Chem.
2000, 275, 19928-19932.
To test these predictions, we developed methodology to
prepare this series of molecules.21 The reaction sequence used
to synthesize the 1-fluoro analogue 5 from 822 is shown in
Scheme 2. Fluorination at C-5 followed by hydrolysis,
benzylation, and reductive ring opening of 10 led to the key
intermediate 11, which upon benzoylation and Swern oxida-
tion gave 13 as the product. Subsequent ketalization and de-
(14) Koppisch, A. T.; Fox, D. T.; Blagg, B. S. J.; Poulter, C. D.
Biochemistry 2002, 41, 236-243.
(15) Hoeffler, J.-F.; Tritsch, D.; Grosdemange-Billiard, C.; Rohmer, M.
Eur. J. Biochem. 2002, 269, 4446-4457.
(19) Reuter, K.; Sanderbrand, S.; Jomaa, H.; Wiesner, J.; Steinbrecher,
I.; Beck, E.; Hintz, M.; Klebe, G.; Stubbs, M. T. J. Biol. Chem. 2002, 277,
5378-5384.
(16) Argyrou, A.; Blanchard, J. S. Biochemistry 2004, 43, 4375-4384.
(17) Proteau, P. J.; Woo, Y.-H.; Williamson, T.; Phaosiri, R. C. Org.
Lett. 1999, 1, 921-923.
(20) Yajima, S.; Nonaka, T.; Kuzuyama, T.; Seto, H.; Ohsawa, K. J.
Biochem. 2002, 131, 313-317.
(18) Radykewicz, T.; Rohdich, F.; Wungsintaweekul, J.; Herz, S.; Kis,
K.; Eisenreich, W.; Bacher, A.; Zenk, M. H.; Arigoni, D. FEBS Lett. 2000,
465, 157-160.
(21) For experimental details, see the Supporting Information.
(22) Pakulski, Z.; Zamojski, A. Tetrahedron 1995, 51, 871-908.
3626
Org. Lett., Vol. 6, No. 20, 2004

Scheme 2
Scheme 3
esterification afforded 15. This compound was then subjected
to treatment of trimethyl phosphite, 2,6-lutidine and tellurium
tetrachloride to phosphorylate the 5-OH group.23 The final
conversion was achieved by successive treatment of 16 with
trimethylsilyl bromide and palladium on carbon. The ketal
group was hydrolyzed during the hydrogenolysis reaction,
presumably due to either the residual acid from the TMS--
Br treatment or the slightly acidic nature of the Pd-C
catalyst. The desired product 5 was purified using cellulose
chromatography.
The 3- and 4-fluoro analogues, 6 and 7, were synthesized
using a similar strategy except for the initial steps. As shown
in Scheme 3, synthesis of 6 from 1724,25 involved selective
protection/deprotection of the C-2 and C-5 hydroxyl groups
followed by C-5 deoxygenation to give 21. The same
intermediate bearing a fluorine substituent at C-2 (34) was
prepared, as shown in Scheme 4, from 2926 via a sequence
of C-5 deoxygenation, methanolysis, and fluorination reac-
tions.27 Since the product of methanolysis of 31, was obtained
as a mixture of â/R isomers in a 6:1 ratio, acetylation to
give 32 was necessary to facilitate their separation by silica
gel chromatography. Subsequent transformations of 21 and
34 to 6 and 7, respectively, followed the procedures used to
make compound 5.
enzyme activity as a function of inhibitor concentration at
different substrate concentrations shows that 6 behaves as a
noncompetitive inhibitor with respect to DXP with a Ki value
of 444 µM (Figure 1 of the Supporting Information). Similar
results were also noted for 7 with a Ki of 733 µM. A possible
Scheme 4
When compounds 6 and 7 (0.5-5 mM) were incubated
in the presence of DXP (1, 50-100 µM), Mg2+, and NADPH
with heterologously expressed DXR from E. coli,21 neither
one was turned over by DXR. Instead, both compounds were
found to be inhibitors. A Dixon plot28 of the reciprocal
(23) Blagg, B. S. J.; Poulter, C. D. J. Org. Chem. 1999, 64, 1508-1511.
(24) Wright, J. A.; Taylor, N. F. Carbohydr. Res. 1967, 3, 333-339.
(25) Thome, M. A.; Giudicelli, M. B.; Picq, D.; Anker, D. J. Carbohydr.
Chem. 1991, 10, 923-926.
(26) van Straten, N. C. R.; van der Marel, G. A.; van Boom, J. H.
Tetrahedron 1997, 53, 6509-6522.
(27) Su, T.-L.; Klein, R. S.; Fox, J. J. J. Org. Chem. 1981, 46, 1790-
1792.
Org. Lett., Vol. 6, No. 20, 2004
3627

of 6 (or 7) is much higher than that of DXP in the incubation
mixture, the inhibitor (instead of substrate) would dominate
the second binding event. Because the inability of DXR to
process 6 and 7 is most likely a consequence of having
chemically inert functional groups at site of action preventing
turnover to the product, the above results may be considered
as preliminary evidence implying that the retroaldolization/
aldolization mechanism (route B) is operative.
Scheme 5
On the basis of the above findings, we anticipated that
the 1-fluoro substrate analogue 5 would bind favorably to
DXR. Incubation of DXR with 5 showed that this compound
is, in fact, a substrate for DXR, with a Km value of 100 µM
and kcat value of 4.5 s-1, albeit only about 13% (kcat/Km)
effective when compared to DXP, which has a Km value of
61 µM and a kcat value of 21.3 s-1 under the same assay
conditions.21 The slightly elevated Km suggests that the
fluorine substitution at C-1 likely affects the electronic as
well as the steric properties of the substrate. However, the
reduction in catalytic efficiency is mainly the result of
decreased kcat of 5, which is also consistent with a mechanism
based on route B.
explanation of the noncompetitive inhibition pattern observed
is that 6 and 7, both as DXP mimics, are capable of binding
two distinct forms of the DXR enzyme.29
As illustrated in Scheme 5, the first encounter between
DXR and 6 (or 7) is with the E‚NADPH complex, taking
place before DXP (1) binds. The second encounter is with
the E‚NADP+ complex and occurs after MEP (2) is released
but prior to the departure of NADP+. The first binding
interaction will affect the slope of the double reciprocal plot
due to the reversible binding of 6 (or 7) to the same enzyme
form as for DXP. The second binding interaction will affect
the intercept of the double reciprocal plot due to the binding
of 6 (or 7) and DXP to two different enzyme forms
(E‚NADP+ for 6/7 and E‚NADPH for DXP) that are
separated by an irreversible step.30 When the Ki values for
these two interactions are comparable, the inhibition pattern
will appear as noncompetitive.
While further experiments to determine the inhibition
kinetics are necessary to fully corroborate the proposed
mechanism, the above inhibition model is favored in view
of the close structural resemblance of 6 and 7 to DXP, which
are expected to bind to DXR at the same site. More
importantly, the fact that substrate inhibition is observed at
high concentrations of DXP (>500 µM), which may be
attributed to the formation of an E‚NADP+‚DXP ternary
complex, also supports the proposed model. It should be
noted that in the inhibition studies, since the concentration
Clearly, both hydroxyl groups at C-3 and C-4 of DXP are
crucial for catalysis. A similar conclusion was also reached
in a recent report in which the corresponding 3- and 4-deoxy-
DXP were found to be mixed-type inhibitors for DXR.15 It
is possible that these hydroxyl groups are sites for divalent
metal chelation, which play an essential role facilitating the
proposed R-ketol rearrangement. However, our findings may
more appropriately be interpreted in favor of the retroal-
dolization/aldolization mechanism for the isomerization
catalyzed by DXR, yet it is not possible at this time to fully
distinguish it from the R-ketol rearrangement. Undoubtedly,
more experiments are needed to unravel the details of this
intriguing enzymatic reaction. This knowledge is essential
for the development of novel therapeutic agents directed
against the reactions involved in the formation of isoprenoids
in many pathogens.
Acknowledgment. We gratefully acknowledge financial
support provided by the National Institutes of Health
(GM40541 and GM54346 to H.-w.L. and GM44613 to
K.A.J.) and the Welch Foundation (F-1511 to H.-w.L.).
(28) Segel, I. H. In Enzyme Kinetics BehaVior and Analysis of Rapid
Equilibrium and Steady-State Enzyme Systems; Wiley-Interscience: New
York, 1975; pp 100-160.
(29) An alternative but less favored interpretation of these data is that 6
and 7 cannot directly compete with DXP (1) for the same binding site.
This phenomenon could be a result of their inability to chelate the divalent
metal ion to form a complex, which may be essential for efficient binding
to the active site of DXR. Instead, 6 and 7, in the presence of DXP, may
bind DXR at a site away from the active site perhaps causing inhibition
through an allosteric effect.
Supporting Information Available: Experimental de-
tails, kinetic data, and NMR and mass spectral data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(30) Cleland, W. W. Kinetics and Mechanisms. In The Enzymes; Boyer,
P. D., Ed.; Academic Press: New York, 1970; Vol. 2, pp 1-65.
OL048459B
3628
Org. Lett., Vol. 6, No. 20, 2004