Synthesis of a ss-ketophosphonate bioisostere of UDP-N- acetylglucosamine

Article abstract of DOI:10.1002/ejoc.200900399

A concise route to a new β-ketophosphonate analog of glycosyl nucleotides is described. Such a diphosphate bioisostere is a stable mimic of enzyme substrates involved, in peptidoglycan biosynthesis and will be the starting point for the development of new

Full text of DOI:10.1002/ejoc.200900399

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200900399
Synthesis of a β-Ketophosphonate Bioisostere of UDP-N-acetylglucosamine
Nicolas Auberger,[a] Christine Gravier-Pelletier,*[a] and Yves Le Merrer*[a]
Keywords: Antibiotics / C-glycosides / Nucleotides / Drug design / Inhibitors
A concise route to a new β-ketophosphonate analog of glyco-
syl nucleotides is described. Such a diphosphate bioisostere
is a stable mimic of enzyme substrates involved in peptidog-
lycan biosynthesis and will be the starting point for the de-
velopment of new potential antibiotics. The synthesis is car-
ried out by condensing a lithiomethylenephosphonate deriv-
ative on a methyl N-acetylglucosaminylacetate followed by
esterification with a uridine derivative.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
Introduction
The cell wall is a unique and essential feature for bacteria
and it is a key target for antibiotics development. Its bio-
synthesis[1] (Figure 1) takes place first at the cytoplasm
from fructose-6-phosphate, which is converted under the
successive action of three enzymes (GlmS, GlmM, and
GlmU) into UDP-N-acetylglucosamine (UDP-GlcNAc). Its
further transformation into UDP-N-acetylmuramic acid
pentapeptide (UDP-MurNAc-pentapeptide) is catalyzed by
the Mur enzymes (MurA, MurB, MurC, MurD, MurE, and
MurF). Then, at the plasma membrane two distinctive pro-
teins (MraY and MurG) subsequently synthesize polypre-
nyl-linked precursors (Lipids I and II) that carry one com-
plete cell wall subunit. Thereafter, a transport protein flips
lipid II across the membrane to deliver the cell wall subunit
to the polymerization enzymes, that is, the penicillin-bind-
ing proteins (PBPs). The latter catalyze both the polymeri-
zation of the lipid bearing monomer units of peptidoglycan
to make long polysaccharide chains and their cross linking
at the peptide unit. Owing to their high specificity and their
sole occurrence in bacteria, the enzymes implicated in pepti-
doglycan biosynthesis are potential targets of particular
interest for the search of novel antibacterial agents.
Furthermore, no counterparts have been identified in eu-
karyotes, and this renders these enzymes ideal targets for
new antibiotics. Several natural products with various
chemical structures have been identified as inhibitors[2] of
the enzymes involved in peptidoglycan biosynthesis, such as
fosfomycin for MurA, tunicamycins, muraymycins, murei-
domycins and liposidomycins for MraY, ramoplanin for
MurG, vancomycin for transglycosylases, and, for instance,
moenomycin or penicillins for PBPs.
Figure 1. Biosynthesis of the bacterial peptidoglycan.
Results and Discussion
In an ongoing program[3] aiming at the inhibition of new
targets for fighting antibiotics resistance,[4] the synthesis of
stable substrate analogs (Figure 2) for the enzymes involved
in this biosynthesis (MurA-G and MraY) is currently under
study.
[a] Laboratoire de Chimie et Biochimie Pharmacologiques et Tox-
icologiques, Université Paris Descartes, CNRS UMR 8601,
45 rue des Saints-Pères, 75006 Paris, France
Fax: +33-142868387
E-mail: christine.gravier-pelletier@parisdescartes.fr
yves.le-merrer@parisdescartes.fr
Eur. J. Org. Chem. 2009, 3323–3326
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3323

N. Auberger, C. Gravier-Pelletier, Y. Le Merrer
SHORT COMMUNICATION
philic opening of conveniently protected epoxide 3 by tri-
alkylphosphite followed by the oxidation of the resulting
secondary alcohol function. The second one relies on the
condensation of a dialkyl lithiomethylenephosphonate on
the ester function of conveniently protected methyl N-ace-
tylglucosaminylacetate 4. Then, esterification with uridine
should afford in both cases the targeted UDP-GlcNAc
mimic. Both epoxide and ester building blocks result from
the single orthogonally protected α-1C-allyl-N-acetylglu-
cosamine 1, readily obtained in four steps (50% overall
yield as an α/β mixture of anomers) from commercially
available N-acetylglucosamine.[7]
Figure 2. Enzyme substrates and structure of the targeted β-keto-
phosphonate bioisosteres.
Indeed, the central core of these substrates is based on
a pyrophosphate moiety linked on the one hand to an N-
acetylglucosamine derivative, and on the other hand to
either uridine or undecaprenol. Within the targeted analogs,
the oxygen atom at the anomeric position of the sugar is
replaced by a methylene group in order to enhance the sta-
bility of the resulting inhibitors towards hydrolysis.
Furthermore, a carbonyl group is introduced to replace the
phosphate at the sugar anomeric position, the latter being
the site[1d,5] of the enzymatic reaction catalyzed by MraY.
The resulting ketone is proposed as an alternative electro-
philic center for the corresponding phosphate. Finally, the
central oxygen atom of the pyrophosphate moiety is ex-
changed for a methylene group expected to prevent the re-
lease of UMP in the particular case of the reaction cata-
lyzed by MraY. One can assume that the resulting glycosyl
β-ketophosphonate bioisostere, stable to hydrolysis, should
have the required electronic properties to coordinate the
metallic cofactor of this reaction (Mg2+ or Mn2+). There-
fore, it should be a pertinent substrate mimic. Moreover,
such a pyrophosphate bioisostere can also serve as a rel-
evant stable surrogate of most substrates involved in various
steps of peptidoglycan biosynthesis.
Figure 4. Retrosynthetic analysis.
The preparation of these key building blocks (Scheme 1)
required the protection of the secondary alcohol function
of 1 as its tert-butyldimethylsilyl ether 2, which was isolated
by flash chromatography as a pure α anomer from the 24:1
α/β mixture.
A few different glycosyl nucleotide analogs have already
been described[6] as mimics of the pyrophosphate moiety
(Figure 3). However none of them displays a glycosyl β-
ketophosphonate skeleton.
Scheme 1. Preparation of the building blocks.
The subsequent m-CPBA oxidation of the double bond
of 2 led to 3 as a 3:2 mixture of epimers. Alternatively, the
ozonolysis of the alkene under basic conditions in the pres-
ence of methanol led to methyl ester 4.[8]
Figure 3. Pyrophosphate analogs of glycosyl nucleotides previously
described.
To reach these new analogs of UDP-GlcNAc and their
The synthesis of N-acetylglucosamine derivative 6
derivatives, the retrosynthetic analysis (Figure 4) relies on (Scheme 2) involved the nucleophilic opening of epoxide 3
two complementary strategies. The first one involves nucleo- by triethylphosphite in the presence of activated zinc chlo-
3324
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3323–3326

A β-Ketophosphonate Bioisostere of UDP-N-acetylglucosamine
ride under microwave irradiation at 60 °C[9] (CEM dis- efficiently achieved in the presence of quinuclidine[11] in re-
cover®) followed by Collins oxidation of resulting alcohol fluxing toluene to give monobenzyl β-ketophosphonate 8
5.
(Scheme 3).
We next turned to the introduction of the isopropylidene-
protected uridine, which was tentatively carried out under
Mitsunobu conditions[12] in the presence of triphenylphos-
phane and diisopropyl azodicarboxylate. However, the lat-
ter conditions were unsuccessful. Interestingly, esterification
of 8 with isopropylidene N-Boc uridine[13] under the above-
mentioned Mitsunobu conditions afforded the targeted pro-
tected glycosyl nucleotide analog 9 in 20% yield. Alterna-
tive conditions for uridine introduction were then studied
and we showed that coupling of isopropylidene N-Boc urid-
ine with β-ketophosphonate 8 could also be achieved in the
presence of a coupling agent (PyBOP, BOP, or HATU)[14]
and a base (diisopropylethylamine or triethylamine) in
moderate yield (30%).
Scheme 2. Formation of the β-ketomethylenephosphonate key
moieties.
Simultaneous acidic hydrolysis of the benzylidene and
isopropylidene ketals, the tert-butyldimethylsilyl ether, and
the tert-butylcarbamate groups of 9 afforded 10 (53%
yield). Finally, subsequent hydrogenolysis of benzyl phos-
phonate gave targeted compound 11 (84% yield).
In a complementary and more efficient way, compound
6 could also result from ester 4 through condensation of
diethyl lithiomethylenephosphonate at –78 °C. This lithio
derivative was readily generated at –78 °C by addition of
butyllithium to the commercially available diethyl methyl-
phosphonate. Furthermore, analogous dibenzyl β-ketopho-
sphonate 7 could be obtained from ester 4 under similar
conditions to those used for dibenzyl methylphosphonate,
which could be prepared from commercial dibenzyl H-
phosphonate by NaH treatment in the presence of methyl
iodide.[10] The monodeprotection of diethyl phosphonate 6
in the presence of sodium azide in DMF could not be con-
trolled, whereas that of dibenzyl phosphonate 7 could be
Conclusions
In conclusion, we describe a concise and straightforward
route to a new β-ketophosphonate mimic of nucleotide
sugars. It is a key intermediate towards substrate analogs
of enzymes involved in peptidoglycan biosynthesis and
identified as a target for the development of new antibiotics.
The methodology reported herein should serve as a major
tool for the further elaboration of a related series of inhibi-
tors. Indeed, the introduction of a peptidic chain at C3 of
the glycosyl part should afford substrate analogs of both
cytoplasmic enzymes and MraY, whereas that of polyiso-
prenol residues in place of uridine should afford substrate
analogs of MurG. Current work is in progress towards that
goal.
Acknowledgments
This work was supported by the European Community, EUR-IN-
TAFAR integrated project (LSHM-CT-2004-512138) and the
“Région Ile de France” for a doctoral grant to N.A. We wish to
thank Dr. A. Thellend for the close attention she paid to this work.
[1] a) J. van Heijenoort, Nat. Prod. Rep. 2001, 18, 503–519; b) H.
Barreteau, A. Kovac, A. Boniface, M. Sova, S. Gobec, D.
Blanot, FEMS Microbiol. Rev. 2008, 32, 168–207; c) R. S. Na-
rayan, M. S. VanNieuwenhze, Eur. J. Org. Chem. 2007, 1399–
1414; d) A. Bouhss, A. E. Trunkfield, T. D. H. Bugg, D. Men-
gin-Lecreulx, FEMS Microbiol. Rev. 2008, 32, 208–233.
[2] a) R. C. Goldman, D. Gange, Curr. Med. Chem. 2000, 7, 801–
820; b) A. H. Katz, C. E. Caufield, Curr. Pharm. Des. 2003, 9,
857–866.
[3] See for example: a) O. Monasson, M. Ginisty, G. Bertho, C.
Gravier-Pelletier, Y. Le Merrer, Tetrahedron Lett. 2007, 48,
8149–8152; b) A. Clouet, C. Gravier-Pelletier, B. Al-Dabbagh,
Scheme 3. Synthesis of a β-ketophosphonate bioisostere of UDP-
GlcNAc.
Eur. J. Org. Chem. 2009, 3323–3326
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3325

N. Auberger, C. Gravier-Pelletier, Y. Le Merrer
SHORT COMMUNICATION
A. Bouhss, Y. Le Merrer, Tetrahedron: Asymmetry 2008, 19,
397–400.
[4] a) C. Walsh, G. Wright, Chem. Rev. 2005, 105, 391–394 and all [7] V. R. Bouvet, R. N. Ben, J. Org. Chem. 2006, 71, 3619–3622.
ard, A. E. Christina, V. Desvergnes, O. R. Martin, J. Org.
Chem. 2006, 71, 7337–7345.
references of this volume dedicated to antibiotics resistance; b)
G. H. Talbot, J. Bradley, J. E. Edwards Jr, D. Gilbert, M.
Scheld, J. G. Bartlett, Clin. Infect. Dis. 2006, 42, 657–668.
[5] a) M. G. Heydanek Jr, F. C. Neuhaus, Biochemistry 1969, 8,
[8] J. A. Marshall, A. W. Garofalo, R. C. Sedrani, Synlett 1992,
643–645.
[9] A. R. Sardarian, Z. Shahsavari-Fard, Synth. Commun. 2007,
37, 289–295.
1474–1481; b) A. Bouhss, M. Crouvoisier, D. Blanot, D. Men- [10] L. Lebeau, C. Mioskowski, M. Saady, WO 96/00733, PCT/
gin-Lecreulx, J. Biol. Chem. 2004, 279, 29974–29980.
[6] a) M. M. Vaghefi, R. J. Bernacki, W. J. Hennen, R. K. Robins,
J. Med. Chem. 1987, 30, 1391–1399; b) R. Chang, T.-T. Vo,
N. S. Finney, Carbohydr. Res. 2006, 341, 1998–2004; c) J.
Hajduch, G. Nam, E. J. Kim, R. Fröhlich, J. A. Hanover, K. L.
Kirk, Carbohydr. Res. 2008, 343, 189–195; d) A. Khaled, O.
Piotrowska, K. Dominiak, C. Augé, Carbohydr. Res. 2008, 343,
167–178; e) A. Caravano, H. Dohi, P. Sinaÿ, S. P. Vincent,
Chem. Eur. J. 2006, 12, 3114–3123; f) A. Babic, S. Gobec, C.
Gravier-Pelletier, Y. Le Merrer, S. Pecar, Tetrahedron 2008, 64,
9093–9100; g) V. Liautard, V. Desvergnes, K. Itoh, H.-w. Liu,
O. R. Martin, J. Org. Chem. 2008, 73, 3103–3115; h) V. Liaut-
FR95/00871, 1995.
[11] M. Saady, L. Lebeau, C. Mioskowski, J. Org. Chem. 1995, 60,
2946–2947.
[12] a) M. Saady, L. Lebeau, C. Mioskowski, Synlett 1995, 643–
644; b) M. Saady, L. Lebeau, C. Mioskowski, Tetrahedron Lett.
1995, 36, 2239–2242.
[13] C. Dini, N. Drochon, J. C. Guillot, P. Mauvais, P. Walter, J.
Aszodi, Bioorg. Med. Chem. Lett. 2001, 11, 533–536.
[14] J.-M. Campagne, J. Coste, P. Jouin, J. Org. Chem. 1995, 60,
5214–5223.
Received: April 10, 2009
Published Online: May 27, 2009
3326
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3323–3326