Multi gram synthesis of UDP-N-acetylmuramic acid

Article abstract of DOI:10.1016/S0960-894X(99)00638-1

As part of an effort to discover novel antibacterial agents, a new and efficient synthesis was established in order to provide a large amount of UDP-N-acetylmuramic acid (UDP-MurNAc).

Full text of DOI:10.1016/S0960-894X(99)00638-1

Bioorganic & Medicinal Chemistry Letters 10 (2000) 143±145
Multi Gram Synthesis of UDP-N-Acetylmuramic Acid
C. Dini, a,* N. Drochon, a P. Ferrari b and J. Aszodi a
aMedicinal Chemistry Department, Hoechst Marion Roussel, 102 rte de Noisy, Romainville 93235, Cedex, France
bBiotechnology Department, Hoechst Marion Roussel, 102 rte de Noisy, Romainville 93235, Cedex, France
Received 9 August 1999; accepted 15 November 1999
AbstractÐAs part of an e€ort to discover novel antibacterial agents, a new and ecient synthesis was established in order to pro-
vide a large amount of UDP-N-acetylmuramic acid (UDP-MurNAc). # 2000 Elsevier Science Ltd. All rights reserved.
Resistance to antibiotics has become a major problem
for the treatment of nocosominal infections.1 One way
to overcome this phenomenon would be to discover new
chemical entitities acting on novel targets. Among the
enzymes of the peptidoglycan biosynthesis pathway,2
which is speci®c to bacteria3 and essential for bacteria
survival,4 MurC5±7 was identi®ed as a relevant target by
M.T. Bocquel et al.8 This prompted us to initiate a
programme for the discovery of speci®c inhibitors of
MurC, able to induce an antibacterial e€ect on whole
cells. This enzyme catalyses the amide bond formation
berween the carboxylic group of UDP-MurNAc, and
the amino function of l-alanine (Scheme 1).
results prompted us to establish an alternative and e-
cient access to a MurNac 1-P derivative.
Lactone 4 was identi®ed as a relevant key intermediate,
because of its ease of synthesis. It was prepared from
3,15 by hydrogeneolysis of the benzyl protecting group.
Phosphorylation in the anomeric positon was accom-
plished by using commercially available diphenyl
chlorophosphate in the presence of 4-pyrrolidinopyridine
(which was found to be the best acylating catlayst).
After work up and puri®cation on a short silica gel col-
umn (15 parts), only the expected a anomer was recov-
ered.16 It is worth noting that a good and reproducible
yield (43%) of 5ꢀ17 was only obtained by limiting the
amount of crude material applied to silica gel, corre-
sponding to the use of 10 g of 4. PtO2 catalysed hydro-
genolysis of 5ꢀ in THF for 48 h, gave the intermediate 6
which was immediately treated with 4 equivalents of
aqueous lithium hydroxide. The crude material was
puri®ed through a Dowex 50WX2 column (triethyl-
ammonium form) providing 7 in 72% yield for the two
steps. The ®nal coupling step was accomplished with an
improved Roseman et al.18 procedure: activated 4 A
molecular sieves and dry commercially available uridine
Large quantities of UDP-MurNac, which is not com-
mercially available, were needed for the programme
envisaged. Three approaches could be used to produce
this nucleotide substrate: (A) isolation from cell wall
precursors leads to limited amounts,9 (B) enzymatic
production using upstream enzymes of the peptidogly-
can biosynthesis, namely MurA and MurB (Scheme 2)
and the commercially available substrate of MurA,
UDP-GlcNAc.10,11 However, this necessitates large
quantities of both proteins, (C) chemical synthesis,
which is often delicate in these series, but gives the pro-
spect of the preparation of gram quantities. Of these,
the last option was retained.
50-monophosphomorpholidate
8
were successively
added to a solution of 7 in dry DMF. The mixture was
heated to 70 ꢀC for 10 h. The rather high reaction yield
(52%) with respect to values from literature is a result of
thorough drying of 8 at 100 ꢀC under vacuum for 18 h,
prior to use. Compound 8 is sold as a hydrate and this
coupling reaction is known to be water-sensitive.19 Under
these conditions, heating was possible, reducing the
reaction time from 1 week (rt) to 10 h (70 ꢀC). Puri®ca-
tion was simpli®ed by limiting the amount of 10 (product
of auto-condensation of 8 formed under these condi-
tions). Furthermore, a reduced amount of this expensive
Blanot et al. described in 1995 the ®rst full chemical
synthesis of UDP-MurNAc.12,13 Unfortunately, we
found the scale-up of the phosphorylation step (according
to MacDonald's procedure14), gave neither a reproductive
yield nor a satisfactory a/b ratio (Scheme 3). These
*Corresponding author. Fax: +33-1-49-91-50-87; e-mail: christophe.
dini@hmrag.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(99)00638-1

144
C. Dini et al. / Bioorg. Med. Chem. Lett. 10 (2000) 143±145
Scheme 1.
Scheme 2.
Scheme 3.
reagent 8 was required (1.5 eq. instead of a large excess
used previously). After usual work-up the UDP-Mur-
NAc was puri®ed by reverse phase and desalting
chromatography. Thus, the crude product containing
UDP-MurNAc was dissolved in 2 M ammonium formate
and then applied on a C18 Hyperprep (SHANDON)
column equilibrated with 50 mM ammonium formate at
pH 4.0. Pure fractions were pooled, concentrated by
distillation and then applied on a G25 Sephadex gel
(Pharmacia) column equilibrated with water. The main
fraction was lyophilised and the pure compound 9 was
isolated in 52% yield.

C. Dini et al. / Bioorg. Med. Chem. Lett. 10 (2000) 143±145
145
Society for Microbiology. Abs Genetics and Molecular Biology
1997, H-2, 285.
9. Flouret, B.; Mengin-Lecreulx, D.; van Heijenoort, J. Anal.
Biochem. 1981, 114, 59.
10. Benson, T. E.; Marquardt, J. L.; Marquardt, A. C.; Etz-
korn, F. A.; Walsh, C. T. Biochemistry 1993, 32, 2024.
11. Reddy, S. G.; Waddell, S. T.; Kuo, D. W.; Wuong, K. K.;
Pompliano, D. L. J. Am. Chem. Soc. 1999, 121, 1175.
12. Blanot, D.; Auger, G.; Ligier, D.; van Heijenoort, J. Car-
bohydr. Res. 1994, 252, 107.
Physical analyses are in accordance with data from the
literature.9±11 This new and ecient synthesis of UDP-
MurNAc was carried out on a multi gram scale and
more than 25 g of pure MurC substrate were prepared
this way.
References and Notes
1. Plattner, J. Annual Reports in Medicinal Chemistry 1997, 32,
111.
13. Heymann, H.; Turdiu, R.; Lee, B. K.; Barkulis, S. S. Bio-
chemistry 1968, 7, 1393.
2. Rogers, H. J.; Perkins, H. R.; Ward, J. B. Microbial Cell
Walls and Membranes. Chapman and Hall: London, 1980. pp
239.
14. MacDonald, D. L. J. Org. Chem. 1962, 27, 1107.
15. Osawa, T.; Sinay, P.; Halford, M.; Jeanloz, R. W. Bio-
chemistry 1969, 8, 3369.
3. Kandler, O. Cell wall structure and their phylogenis impli-
cations. Zbl. Bakt. Hyg., I Abt. Orig. 1982, C3, 149±160.
4. Bacterial Cell Wall; Ghuysen, J. M.; Hakenbeck, R. Eds.,
Elsevier, Oxford, 1994.
5. Ligier, D.; Masson, A.; Blanot, D.; van Heijnoort, J.; Par-
quet, C. Eur. J. Chem. 1995, 230, 80.
16. Sabesan, S.; Neira, S. Carbohydr. Res. 1992, 223, 169.
17. Data for 5ꢀ;: 1H NMR (CDCl3): 1.52 (d, 3H, CH3 lactoyl),
1.86 (s, 3H, Ac), 2.02 (s, 3H, Ac), 4.72 (q, 1H, CHCH3 lactoyl),
3.86 (dd, 1H, J=9 Hz, 10.5 Hz, H3 or H4), 4.03 (dm, 1H, H5),
4.13 (dd ,1H, J=3.5, 12.5 Hz, H6), 4.26 (dd, 1H, J=3.5, 12.5
Hz, H60), 4.44 (m, 2H, H2 + H3 or H4), 5.59 (d, 1H, J=9 Hz,
NHAc), 5.59 (dd, 1H, J=3 Hz, 6Hz, H1, a anomer), 7.15±7.44
(m, 10H, OPh). MS (ES): 572 (M+Na)+, 550 (M+H)+.
18. Roseman, S.; Distler, J. J.; Mo€att, J. G.; Khorana, H. G.
J. Am. Chem. Soc. 1961, 83, 659.
6. Falk, P. J.; Ervin, K. M.; Volk, K. S.; Ho, H. T. Biochem-
istry 1996, 35, 1417.
7. Gubler, M.; Appoldt, Y.; Keck, W. J. Bact. 1996, 178, 906.
8. Bocquel, M. T.; Fairley, M.; Mengin-Lecreulx, D.; Parquet,
C.; Realo, E.; Salah Bey, K.; Taburet, Y.; Harnois, M.; van
Heijenoort, J. Knock-out of the MurC gene coding for the l-
Ala ligase activity in E. coli: 97th General Meeting American
19. Kochetkov, N. K.; Budokovkii, E. I.; Shibaev, V. N.;
Lebedeva, K. S. Izv. Akad. Nauk. SSSR, Ser. Khim. 1969, 4,
897.