Synthesis and activity of two trifluorinated analogues of 1-deoxy-d-xylulose 5-phosphate

Article abstract of DOI:10.1016/j.tetlet.2006.11.091

An improved synthesis of 1,1,1-trifluoro-1-deoxy-d-xylulose 5-phosphate and an access to the reduced diastereomer mixture analogues 1,1,1-trifluoro-1-deoxy-d-xylitol 5-phosphate and 1,1,1-trifluoro-1-deoxy-d-lyxitol 5-phosphate are described. Inhibitor ac

Full text of DOI:10.1016/j.tetlet.2006.11.091

Tetrahedron Letters 48 (2007) 711–714
Synthesis and activity of two trifluorinated analogues
of 1-deoxy-D-xylulose 5-phosphate
Odile Meyer, Catherine Grosdemange-Billiard, Denis Tritsch and Michel Rohmer*
Universite´ Louis Pasteur – CNRS (UMR 7177 LC3), Institut Le Bel, 4 rue Blaise Pascal, 67070 Strasbourg Cedex, France
Received 21 September 2006; revised 13 November 2006; accepted 14 November 2006
Available online 8 December 2006
Abstract—An improved synthesis of 1,1,1-trifluoro-1-deoxy-D-xylulose 5-phosphate and an access to the reduced diastereomer mix-
ture analogues 1,1,1-trifluoro-1-deoxy-D-xylitol 5-phosphate and 1,1,1-trifluoro-1-deoxy-D-lyxitol 5-phosphate are described. Inhib-
itor activity of all compounds on the MEP pathway for isoprenoid biosynthesis was evaluated.
Ó 2006 Elsevier Ltd. All rights reserved.
In animals, fungi, plant cytoplasm, archaebacteria and
some eubacteria, isopentenyl diphosphate and dimethyl-
allyl diphosphate, the precursors of isoprenoids, are syn-
thesized through the mevalonate pathway,1 while they
derive from the more recently discovered methylerythri-
tol phosphate (MEP) pathway in most bacteria, in the
chloroplasts of all phototrophic organisms and in some
unicellular eukaryotes, including Plasmodium falcipa-
rum, the parasite responsible for malaria.2 As the inhibi-
tion of the biosynthesis of essential isoprenoids is lethal,
each enzyme of the MEP pathway is a potential target
for antibacterial or antiparasitic drugs.3 We focused
our attention on the second enzyme of the MEP path-
way, the DXP reductoisomerase (DXR), which mediates
the conversion of 1-deoxy-D-xylulose 5-phosphate
FR900098 inhibits the DXR of P. falciparum, and ester
prodrugs thereof showed an improved in vivo antimi-
crobial activity against Plasmodium vinckei in mice.8
Fosmidomycin or its prodrugs have also been shown
to present some positive effects for the treatment of uri-
nary infections9 and simple malaria cases10 in humans,
indicating that DXR is a valuable antimicrobial target.
Compounds containing fluorine are useful tools for the
elucidation of enzyme mechanisms and often used for
the design of enzyme inhibitors. In this context, Bouvet
and O’Hagan synthesized the 1-fluoro and 1,1-difluoro
analogues of DX as the potential inhibitors of enzymes
involved in the DX and DXP metabolism, but these
compounds did not show any antibacterial properties
against E. coli and Staphylococcus aureus,11 perhaps be-
cause of the absence of an efficient phosphorylating sys-
tem. To circumvent this problem, monofluorinated
DXP analogues have been synthesized and tested on
the DXR from E. coli.12 The C-1 fluorinated analogue
was found to behave as a substrate, whereas the C-3
(DXP)
1 into 2-C-methyl-D-erythritol 4-phosphate
(MEP) 2 (Fig. 1).4 This enzyme is a promising target
to inhibit microbial growth. Indeed, fosmidomycin
blocks the DXR in Escherichia coli,5 Zymomonas mobi-
lis6 and P. falciparum3a and leads to the inhibition of the
MEP pathway in higher plants.7 Its structural analogue
OH
HO CH3
O
HO CH3
OPO(OH)2
OPO(OH)2
OPO(OH)2
CH3 OH
O
OH
OH OH
NADP+
NADPH
2
1
Figure 1. Reaction catalyzed by the deoxyxylulose phosphate reductoisomerase (DXR) in the MEP pathway: 1-deoxy-D-xylulose 5-phosphate 1, 2-C-
methyl-D-erythritol 4-phosphate 2.
Keywords: 1-Deoxy-D-xylulose 5-phosphate reductoisomerase; Isoprenoid biosynthesis: 1,1,1-trifluoro-1-deoxy-D-xylulose 5-phosphate.
*
Corresponding author. Tel.: +33 3 90 24 13 46; fax: +33 3 90 24 13 45; e-mail: mirohmer@chimie.u-strasbg.fr
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.11.091

712
O. Meyer et al. / Tetrahedron Letters 48 (2007) 711–714
OBn OBn
OBn
i
iv
iii
HO
R1O
HO
OH
OR2
OPO(OBn)2
OBn
OBn
OBn
3
4
5
6
R1 = TBDMS, R2 = H
R1 = TBDMS, R2 = PO(OBn)2
ii
OBn
OBn
OH
v
vi
O
F3C
F3C
OPO(OBn)2
OPO(OBn)2
OPO(OH)2
OBn
OH OBn
OH OH
7
8
9
vii
OBn
OBn
OH
vi
F3C
F3C
F3C
OPO(OBn)2
OPO(OBn)2
OPO(OH)2
O
OBn
HO OH
HO OH
OBn
OH
10
10'
11
Figure 2. Synthesis of 1,1,1-trifluoro-1-deoxy-D-xylulose 5-phosphate 11 and the diastereomer mixture of 1,1,1-trifluoro-1-deoxy-D-xylitol 5-
phosphate/1,1,1-trifluoro-1-deoxy-D-lyxitol 5-phosphate 9. (i) TBDMSCl, NaH, THF (87%); (ii) (BnO)2PNEt2, tetrazole, mCPBA, CH2Cl2 (85%);
˚
(iii) Bu4NF, THF, (87%); (iv) PCC, CH2Cl2, 3 A molecular sieves (82%); (v) (a) TMSCF3, CsF, (b) TBAF, THF (72%); (vi) H2, 10% Pd/C, MeOH/
H2O (9:1) (quantitative); (vii) Dess–Martin, DMPI, CH2Cl2 (54%).
O
O
O
O
O
O
O
OMe
OMe
OR
OPO(OBn)2
CF3
i
vi
iii
11
OMe
O
O
O
13
14
R = H
R = PO(OBn)2
12
15
ii
Figure 3. Improved synthesis of 1,1,1-trifluoro-1-deoxy-D-xylulose 5-phosphate 11. (i) NaBH4, MeOH, 88% (conversion: 55%); (ii) (BnO)2PNEt2,
tetrazole, mCPBA, CH2Cl2 (81%); (iii) TMSCF3, CsF, TBAF, THF (70%); (iv) H2, 10% Pd/C, MeOH/H2O (9:1) (quantitative).
or C-4 fluorinated derivatives were weak non-competi-
tive inhibitors. In the search of novel drugs acting on
the MEP pathway, DXP analogues containing a trifluo-
romethyl group were also synthesized. A synthesis and
the evaluation of 1,1,1-trifluoro-1-deoxy-D-xylulose 5-
phosphate 11 were recently reported by Fox and Poul-
ter.13 In this contribution, an improved synthesis of
the latter compound and an access to the reduced diaste-
reomer mixture analogues 9, 1,1,1-trifluoro-1-deoxy-D-
xylitol 5-phosphate and 1,1,1-trifluoro-1-deoxy-D-lyxitol
5-phosphate, are described (Figs. 2 and 3).
alcohol groups was protected as silylether using sodium
hydride and t-butyldimethylsilyl chloride to afford 4
(87% yield). The remaining hydroxyl group was phos-
phorylated by the phosphoramidite method to give 5
in an 85% yield. Removal of the O-silyl group with
tetrabutylammonium fluoride led to compound 6 in an
87% yield from 5, corresponding to a 65% global yield
from threitol derivative 3. Due to the sensitivity of the
phosphate group in basic conditions, the Swern oxida-
tion could not be used, and aldehyde 7 was obtained
by the oxidation of 6 with PCC (82% yield). The key
step of this route, the introduction of the trifluoromethyl
group, was performed with (trifluoromethyl)trimethyl-
silane (Ruppert’s reagent). This nucleophilic trifluoro-
methylating agent leads to trifluoromethyl alcohols from
aldehydes and ketones.15 A modified procedure using
cesium fluoride as the catalyst was employed. Aldehyde
7 was mixed with Ruppert’s reagent and cesium fluoride.
When the starting material was totally consumed, the
addition of tetrabutylammonium fluoride allowed the
formation of the secondary alcohol 8 in a 72% yield.16
Removal of the benzyl protecting groups by catalytic
hydrogenolysis in the presence of Pd/C provided quan-
The trifluorinated compounds 9 and 11 were synthesized
from the commercially available (2R,3R)-2,3-O,O-di-
benzyl-threitol 3 following the reaction sequence outlined
in Figure 2. The first step of the synthesis is the phos-
phorylation of a single hydroxyl group of the starting
threitol derivative 3. Direct phosphorylation of com-
pound 3 with dibenzylphosphorochloridate was unsatis-
factory since it resulted in a mixture of the desired
monophosphate 6 (40% yield), the corresponding 1,4-
bisphosphate and the starting material.14 To avoid the
formation of the bisphosphate, one of the two primary

O. Meyer et al. / Tetrahedron Letters 48 (2007) 711–714
713
titatively the fluorinated derivative 9 as a 1:2 mixture of
two diastereomers in six steps and a 38% global yield.17
Separation of the diastereomer mixture by silica gel
chromatography was not possible, neither for the pro-
tected alcohol 8, nor for the free compound 9. Dess–
Martin conditions18 oxidized alcohol 8 into intermediate
10/100 in a rather modest yield (54%). Other oxidation
reagents (PCC, TPAP) did not increase the yield of
methyl)trimethylsilane on an ester rather than on an
aldehyde is of a general scope and may be applied to
the synthesis of other 1,1,1-trifluoro-1-deoxyketose
derivatives.
1,1,1-Trifluoro-DXP 11 was not a substrate of the E. coli
DXR since no oxidation of NADPH was observed. The
inhibitory activity of trifluoro-DXP 11 and of diastereo-
mer mixture 9 on the DXR was also investigated.22
1,1,1-Trifluoro-DXP 11 revealed as a poor inhibitor of
the E. coli DXR with a 1.7 mM IC50, quite similar to
that reported by Fox and Poulter (2 mM),13 and to be
compared to the Km found for DXP (97 lM) or MEP
(158 lM).4c The hydration of the carbonyl group may
prevent the binding to the active site of the enzyme.
Compound 9 was found to be more efficient on DXR
with a 0.6 mM IC50. It was therefore investigated in
more details and revealed as a reversible non-competi-
tive inhibitor with a 360 lM Ki value for the 2:1 diaste-
reomer mixture resulting from the synthetic scheme.
None of compounds 9 or 11 inhibited the growth of
E. coli in test on Petri dishes by the agar diffusion meth-
od on filter paper discs (6 mm diameter). Even the high-
est amount tested (100 lg/disc) did not prevent bacterial
growth. Despite their low activities, weak inhibitors may
be used for screening of inhibitors in pharmaceutical re-
search or for RX studies on enzymes. They are allowed
to bind to the enzyme, and compound libraries may be
screened to identify those, which displace the weak
binders.
this
reaction.
NMR
spectra
showed
that
trifluoromethylketone 10 was, as expected,19 only in
the hydrate form 100. The final deprotection by catalytic
hydrogenolysis in the presence of Pd/C gave quantita-
tively and without epimerization at C-3 the enantiomeri-
cally pure trifluorinated analogue 11 of DXP in seven
steps and a 21% overall yield. Its spectroscopic data
were identical with those reported for the same com-
pound by Fox and Poulter.13
The synthesis of the latter compound 11 was improved
by a shorter reaction sequence starting from the com-
mercially available dimethyl 2,3-O,O-benzylidene-D-tar-
trate 12 (Fig. 3). The phosphate group was introduced
on the primary alcohol 13 obtained in an 88% yield by
treating tartrate diester 12 with NaBH4 in methanol.11
The presence of a participating neighbouring group
facilitated the ester reduction.20 Such conditions did
not give a complete conversion of the starting material,
but compound 13 was easily isolated by silica gel flash
chromatography, and the starting material was recycled.
The primary alcohol 13 was phosphorylated using the
phosphoramidite method affording the phosphate deriv-
ative 14 in an 81% yield. The introduction of the fluori-
nated group was performed as described above and
directly provided trifluoromethylketone 15, only in the
hydrate form as judged from the 13C and 19F NMR
spectra, from methyl ester 14 in a 70% yield.16,21 Finally,
the enantiomerically pure trifluorinated DXP analogue
11 was released quantitatively and without epimeriza-
tion by catalytic hydrogenolysis of all benzyl protecting
groups in a global yield of 50% in four steps. Both triflu-
orinated analogues of DXP 9 and 11 could be stored
without decomposition at ꢀ18 °C for one year as lyo-
philized samples.
Acknowledgments
We thank Mr. R. Huber for the MS analysis. This inves-
tigation was supported by a grant to M.R. from the
‘Institut Universitaire de France’. O.M. was a recipient
`
of a grant from the ‘Ministere de la Jeunesse, de l’Edu-
cation Nationale et de la Recherche’.
References and notes
1. (a) Bloch, K. Steroids 1992, 57, 378–383; (b) Bochar, D.
A.; Friesen, J. A.; Stauffacher, C. V.; Rodwell, V. W. In
Comprehensive Natural Product Chemistry; Cane, D. E.,
Ed.; Isoprenoids Including Carotenoids and Steroids;
Pergamon: Oxford UK, 1999; Vol. 2, pp 15–44.
2. (a) Rohmer, M. Nat. Prod. Rep. 1999, 16, 565–574; (b)
Lichtenthaler, H. K. Annu. Rev. Plant Physiol. Plant Mol.
Biol. 1999, 50, 47–65; (c) Eisenreich, W.; Rohdich, F.;
Bacher, A. Trends Plant Sci. 2001, 6, 78–84.
Fox and Poulter recently described a synthesis of 11 in
seven steps in a 45% overall yield from the commercially
available 2,3-O-isopropylidene-D-threitol.13 In contrast
with our route towards 11, they first introduced the tri-
fluoromethyl group before the phosphate group. The
final removal of the isopropylidene 1,2-diol protecting
group after hydrogenolysis of the benzylidene phosphate
protections proved, however, to be difficult and required
5 days. In contrast, a single hydrogenolysis removes in
our method simultaneously and in one step all benzyl
protecting groups of the phosphate as well as of the
hydroxyls, does not require purification of the end prod-
uct and allows an easy scale-up of the synthesis. The
hydrogenolysis was performed on small quantities of
precursor 15 (up to 150 mg), which is a convenient stock
material for compound 11: it can be readily prepared in
a larger amount (i.e., 1 g), stored for several months at
ꢀ18 °C and hydrogenolyzed into 11 just before use.
Our synthesis involving the addition of (trifluoro-
3. (a) Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek,
B.; Weidemeyer, C.; Hintz, M.; Turbachova, I.; Eberl, M.;
¨
Zeidler, J.; Lichtenthaler, H. K.; Soldati, D.; Beck, E.
Science 1999, 285, 1573–1576; (b) Rohmer, M.; Grosde-
mange-Billiard, C.; Seemann, M.; Tritsch, D. Curr. Opin.
Invest. Drugs 2004, 5, 154–162.
4. (a) Kuzuyama, T.; Takahashi, S.; Takagi, M.; Seto, H.
J. Biol. Chem. 2000, 275, 19928–19932; (b) 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; (c) Hoeffler, J.-F.; Tritsch,
D.; Grosdemange-Billiard, C.; Rohmer, M. Europ. J.
Biochem. 2002, 269, 4446–4457.

714
O. Meyer et al. / Tetrahedron Letters 48 (2007) 711–714
5. Kuzuyama, T.; Shimizu, T.; Takahashi, S.; Seto, H.
(1 mL), and the solution was adjusted to pH 6–7 by the
addition of 1 M sodium hydroxide and lyophilized.
Vitreous solid, which decomposes at 88 °C. Rf = 0.28, (i-
propanol/water/ethyl acetate, 6/3/1). NMR spectroscopy.
Tetrahedron Lett. 1998, 39, 7913–7916.
6. Grolle, S.; Bringer-Meyer, S.; Sahm, H. FEMS Microbiol.
Lett. 2000, 191, 131–137.
7. Fellermeier, M.; Kis, K.; Sagner, S.; Maier, U.; Bacher,
A.; Zenk, M. Tetrahedron Lett. 1999, 40, 2743–2746.
8. (a) Reichenberg, A.; Wiesner, J.; Weidemeyer, C.; Dreis-
eidler, E.; Sanderbrand, S.; Altincicek, B.; Beck, E.;
Schlitzer, M.; Jomaa, H. Bioorg. Med. Chem. Lett. 2001,
11, 833–835; (b) Ortmann, R.; Wiesner, J.; Reichenberg,
A.; Henschker, D.; Beck, E.; Jomaa, H.; Schlitzer, M.
Bioorg. Med. Chem. Lett. 2003, 13, 305–314; (c) Ortmann,
R.; Wiesner, J.; Reichenberg, A.; Henschker, D.; Beck, E.;
Jomaa, H.; Schlitzer, M. Arch. Pharm. Chem. Life Sci.
2005, 338, 305–314.
9. (a) Kuemmerle, H. P.; Murakawa, T.; Sakamoto, H.;
Sato, N.; Konishi, T.; De Santis, F. Int. J. Clin. Pharma-
col. Ther. Toxicol. 1985, 23, 521–528; (b) Kuemmerle,
H. P.; Murakawa, T.; De Santis, F. Chemioterapia 1987, 6,
113–119.
Signals labelled with the superscript correspond to those
that are differentiated in the spectra of each diastereomer.
*
dH (300 MHz, D2O) 3.67 (2H, m, 5-H), 3.75 (1H, m, 3-H),
3.87 (1H, m, 4-H), 4.04 (1H, d, J2–3 = 6.6 Hz, 2-H); dC
(75 MHz, D2O) 65.6 (CH2, d, JC–P = 5.0 Hz, C-5), 65.8
(CH2, d, JC–P = 5.0 Hz, C-5*), 68.1 (CH, d, JC–P = 6.0 Hz,
C-4), 68.5 (CH, d, JC–P = 6.0 Hz, C-4*), 68.8 (CH, C-3),
68.9 (CH, C-3*), 71.0 (CH, q, JC–F = 10 Hz, C-2), 124.6
(quaternary C, q, JC–F = 282 Hz, C-1), 125.3 (quaternary
C, q, JC–F = 282 Hz, C-1*); dP (121.5 MHz, D2O) 2.0
(s); dF (282 MHz, D2O) ꢀ76.56 (d, JH–F = 6.2 Hz,
CF3), ꢀ75.51 (d, JH–F = 6.2 Hz, CFꢁ3). IR (KBr)
m
max/cmꢀ1: 3382, 1648, 1393, 1273, 1177, 1141, 1070,
925. MS (ESꢀ) m/z: 269 (MꢀH+).
18. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277–7287.
10. (a) Missinou, M. A.; Borrmann, S.; Schindler, A.; Issifou,
S.; Adegnika, A. A.; Matsiegui, P.-B.; Binder, R.; Lell, B.;
Wiesner, J.; Baranek, T.; Jomaa, H.; Kremsner, P. G.
Lancet 2002, 360, 1941–1942; (b) 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.
11. Bouvet, D.; O’Hagan, D. Tetrahedron 1999, 55, 10481–
10486.
12. Wong, A.; Munos, J. W.; Devasthali, V.; Johnson, K. A.;
Liu, H.-W. Org. Lett. 2004, 6, 3625–3628.
13. Fox, T.; Poulter, D. C. J. Org. Chem. 2005, 70, 1978–1985.
14. Taylor, S. V.; Vu, L. D.; Begley, T. P.; Scho¨rken, U.;
Grolle, S.; Sprenger, G. A.; Bringer-Meyer, S.; Sahm, H.
J. Org. Chem. 1998, 63, 2375–2377.
15. Prakash, G. K. S.; Krishnamurti, R.; Olah, G. A. J. Am.
Chem. Soc. 1989, 111, 393–395.
16. Singh, R. P.; Cao, G.; Kirchmeier, R. L.; Shreeve, J. M. J.
Org. Chem. 1999, 64, 2873–2876.
17. 1,1,1-Trifluoro-1-deoxy-D-xylitol 5-phosphate and 1,1,1-
trifluoro-1-deoxy-D-lyxitol 5-phosphate diastereomer mix-
ture 9. Compound 8 (120 mg, 0.19 mmol) was hydrogeno-
lyzed over 20% Pd/C (25 mg) in MeOH/H2O (9:1, 5 mL)
for 24 h at room temperature and atmospheric pressure.
After filtration on Celite and evaporation to dryness, the
2:1 mixture of diastereomers 9 was dissolved in water
19. Gelb, M. H.; Svaren, J. P.; Abeles, R. H. Biochemistry
1985, 24, 1813–1817.
20. (a) Dalla, V.; Catteau, J. P.; Pale, P. Tetrahedron Lett.
1999, 40, 5193–5196; (b) Nichols, B. W.; Safford, R. Chem.
Phys. Lipids 1973, 11, 222–227; (c) Giuminiani, A. G.;
Tubaro, F. J. Prakt. Chem. 1990, 332, 755–761.
21. Wiedemann, J.; Heiner, T.; Mloston, G.; Prakash, G. K.
S.; Olah, G. A. Angew. Chem., Int. Ed. 1998, 37, 820–821.
22. The His-tagged E. coli DXR was obtained as previously
reported.4c The enzymatic activity was determined in a
50 mM Tris–HCl, pH 7.5 buffer containing 3 mM MgCl2
and 2 mM DTT at 37 °C. The concentrations of NADPH
and DXP were 0.15 and 0.11 mM, respectively. The
reaction was initiated by adding His-tagged DXR (2 lg).
Initial rates were measured by following the decrease of
the absorbance at 340 nm due to the oxidation of NADPH
(Uvikon 933, Kontron Instruments). The influence of
compounds 9 and 11 on the enzymatic activity was studied
by adding them to the reaction medium at various
concentrations (0.1–3 mM). The IC50 was calculated from
a semi-log plot of enzyme residual activity as a function of
inhibitor concentration. The Ki of compound 9 was
calculated from a double reciprocal plot of enzymatic rate
versus DXP concentration (0.05–0.5 mM) in its absence
and in its presence at various concentrations (0.3–1 mM)
in the reaction medium.