Synthesis of an orthogonally protected precursor to the glycan repeating unit of the bacterial cell wall.

Article abstract of DOI:10.1021/ol016692t

[reaction: see text]. A synthesis of stereochemically pure and orthogonally protected N-acetyl-(2-deoxy-2-aminoglucopyranosyl)-beta-[1,4]-N-acetylmuramyl (NAG-NAM) monopeptide (2), the glycopeptide core of lipid II and of the bacterial cell wall repeating

Full text of DOI:10.1021/ol016692t

ORGANIC
LETTERS
2001
Vol. 3, No. 22
3575-3577
Synthesis of an Orthogonally Protected
Precursor to the Glycan Repeating Unit
of the Bacterial Cell Wall
Shankar L. Saha,† Michael S. Van Nieuwenhze,* William J. Hornback,
James A. Aikins, and Larry C. Blaszczak*
Lilly Research Laboratories, A DiVision of Eli Lilly and Company,
Lilly Corporate Center, Indianapolis, Indiana 46285
MSV@lilly.com
Received September 4, 2001
ABSTRACT
A synthesis of stereochemically pure and orthogonally protected N-acetyl-(2-deoxy-2-aminoglucopyranosyl)-â-[1,4]-N-acetylmuramyl (NAG-
NAM) monopeptide (2), the glycopeptide core of lipid II and of the bacterial cell wall repeating unit, is reported.
Lipid II1 (1) is the ultimate monomeric intermediate of
bacterial cell wall (peptidoglycan) biosynthesis.2 To fuel our
synthetic efforts toward lipid II, we required an efficient and
dependable synthetic route to orthogonally protected di-
saccharide 2. Having successfully addressed this central issue,
we report here a practical synthesis of an orthogonally
protected version of the core NAG-NAM disaccharide 2; the
key starting material for our projected total synthesis of lipid
II.
is sufficiently effective in all regards for our purpose:
preparation of 2 in quantities to enable the first total synthesis
of lipid II.
Formation of the desired â-[1,4]-glycosidic bond requires
a 2-deoxy-2-aminoglucopyranose glycosyl donor with a
nitrogen substituent that will favor equatorial approach3 of
the very modestly nucleophilic C(4) hydroxyl group of a
muramic acid derivative.4 Several approaches to this for-
midable glycosidation problem have been recorded,5 but none
† Present Address: MediChem Research, Inc., 12305 S. New Avenue,
Lemont, IL 60439.
(1) This version of lipid II is common to most Gram-positive bacteria.
In Gram-negative bacteria, the lysine residue is replaced by meso-
diaminopimelic acid (see ref 2).
(2) For a comprehensive review of bacterial cell wall biosynthesis, see:
Ghuysen, J.-M.; Hackenbeck, R. Bacterial Cell Wall; Elsevier Biomedi-
cal: Amsterdam, 1994.
(3) Jeanloz, R. W. The Amino Sugars: The Chemistry and Biology of
Compounds Containing Amino Sugars; Academic Press: New York, 1969.
The overall synthetic challenge of lipid II required
protection and activation chemistries for utilization in several
nested applications: (1) the functional group activation
(4) The C(4) hydroxyl group is reported to be the least nucleophilic
hydroxyl group in glucopyranose-based acceptors. See, for example:
Toshima, K.; Tatsuta, K. Chem. ReV. 1993, 93, 1503.
10.1021/ol016692t CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/10/2001

scheme required for the glycosidation reaction needed to be
changed to a functional group protection scheme at the
disaccharide stage, (2) triple orthogonality of disaccharide
protecting groups (anomeric OH, peripheral OH, and car-
boxyl OH) was required to ensure staging flexibility in
exploratory transformations for the lipid II endgame, (3) a
protection scheme to preserve global deprotection by hy-
droxide ion as the final step to lipid II, and (4) minimizing
the number of functional group interchange reactions for
overall synthetic efficiency.
reactivity increase of trichloroethoxycarbonyl8 (Troc) over
phthaloyl, we chose to employ the former for protection of
the 2-amino substituent of our glycosyl donor.
Our synthesis began with a study to assess the reductive
opening of benzyl N-acetyl-4,6-benzylidinemuramic ester 6
as an efficient means for regioselective introduction of benzyl
protection/activation at the C(6) hydroxyl of the muramic
acid starting material (Scheme 2).
Our synthetic strategy (Scheme 1) dictated the use of
benzyl protection for the anomeric hydroxyl of glycosyl
Scheme 2. Reductive Opening of the 4,6-O-Benzylidene
Acetala
Scheme 1. Retrosynthetic Analysis of the Lipid II
Glycopeptide Core
a Reagents and conditions: (a) Mel, Cs2CO3, MeCN; (b) Et3SiH
(3 equiv), TFA (6 equiv), CH2Cl2, 0 °C, 5 h; (c) N-methylmor-
pholine, 2-chloro-4,6-dimethoxy-1,3,5-triazine, CH2Cl2, 2 h.
acceptor 4. Literature precedent5 reinforced by our own
experience suggested that the acceptor would also require
an electron-donating protective group at C(6) for activation
in the glycosidation reaction; benzyl was selected for stability
and ease of selective introduction. Of the several options
available for orthogonal protection of the C(3) lactate ester,
we selected phenylsulfonylethyl on the basis of its successful
performance in the synthesis of Park nucleotide.6
With respect to design of glycosyl donor 3, Wong and
co-workers7 recently measured relative rates of glycosidic
bond formation using various 2-amino-2-deoxyglucopyranose
donors (inter alia). Since they noted a 40-fold relative
When 69 was treated with trifluoroacetic acid and tri-
ethylsilane under the reported10 conditions, 76% of lactone
75i was isolated along with a small amount of the desired 8.
The simple expedient of installing L-alanine, the first amino
acid residue along the muramyl peptide chain, completely
suppressed acid-catalyzed cyclization of the reduction prod-
uct. Thus, subjecting 9 (Y ) HN-[L-Ala]-CO2CH2CH2SO2-
Ph) to the same reductive ring opening conditions afforded
the muramylmonopeptide ester 4 in 61% yield.
When glycosyl donor 311 and acceptor 4 were combined
under rigorously anhydrous Ko¨nigs-Knorr conditions (silver
triflate/dichloromethane), the desired â-linked disaccharide
10 was obtained in 74% yield after chromatographic
purification (Scheme 3).
(5) (a) Merser, C.; Sinay¨, P. Tetrahedron Lett. 1973, 13, 1029. (b) Durette,
P. L.; Meitzner, E. P.; Shen, T. Y. Carbohydr. Res. 1979, 77, C1. (c) Kiso,
M.; Kaneda, Y.; Shimizu, R.; Hasegawa, A. Carbohydr. Res. 1980, 83,
C8. (d) Kiso, M.; Kaneda, Y.; Shimizu, R.; Hasegawa, A. Carbohydr. Res.
1982, 104, 253. (e) Kusumoto, S.; Yamamoto, K.; Imoto, M.; Inage, M.;
Tsujimoto, M.; Kotani, S.; Shiba, T. Bull. Chem. Soc. Jpn. 1986, 59, 1411.
(f) Kusimoto, S.; Imoto, M.; Ogiku, T.; Shiba, T. Bull. Chem. Soc. Jpn.
1986, 59, 1419. (g) Farkas, J.; Ledvina, M.; Brokes, J.; Jezek, J.; Zajicek,
J.; Zaoral, M. Carbohydr. Res. 1987, 163, 63. (h) Kinzy, W.; Schmidt, R.
R. Liebigs Ann. Chem. 1987, 407. (i) Kantoci, D.; Keglevic, D.; Derome,
A. Carbohydr. Res. 1987, 162, 227. (j) Termin, A.; Schmidt, R. R. Liebigs
Ann. Chem. 1989, 789. (k) Ledvina, M.; Farkas, J.; Zajicek, J.; Jezek, J.;
Zaoral, M. Collect. Czech. Chem. Commun. 1989, 54, 2784. (l) Termin,
A.; Schmidt, R. R. Liebigs Ann. Chem. 1992, 527.
Finally, the activation scheme of the glycosidation stage
was exchanged for the protection scheme of the lipid II
endgame. Four protecting group interchange reactions were
accomplished in a single operation. Glycosidation product
(6) Hitchcock, S. A.; Eid, C. N.; Aikins, J. A.; Zia-Ebrahimi, M.;
Blaszczak, L. C. J. Am. Chem. Soc. 1998, 120, 1916.
(7) Zhang, Z.; Ollmann, I.; Ye, X.-S. Wischnat, R.; Baasov, T.; Wong,
C.-H. J. Am. Chem. Soc. 1999, 121, 734.
(8) Windholz, T. B.; Johnston, D. B. R. Tetrahedron Lett. 1967, 7, 2555.
3576
Org. Lett., Vol. 3, No. 22, 2001

dust was then added to the same reaction mixture to remove
the Troc group from the glucosamine nitrogen, which in turn
acetylated in situ. Removal of solvent, extractive workup,
flash chromatography, and crystallization provided the target
compound, benzyl N-acetyl-(2-deoxy-2-aminoglucopyrano-
syl)-â-[1,4]-N-acetyl-muramylmonopeptide phenylsulfonyl-
ethyl ester 2, in 67% yield.13
Scheme 3. Glycosidation and Protective Group Exchangea
The concise synthesis of the glycopeptide 2 outlined here
offers efficient access to multiple gram quantities of starting
material for our projected synthesis of lipid II. The conver-
gent approach, protecting group economy, and crystalline
nature of the intermediates permit the process to be scaled
to substantial volume. Moreover, the route provides a flexi-
bility that is well suited to support syntheses of N-acetyl-
(2-deoxy-2-aminoglucopyranosyl)-â-[1,4]-N-acetylmuramyl-
peptide analogues that are unavailable from nature.
Acknowledgment. The authors thank Dr. David A. Peake
and his colleagues for application of their expertise in mass
spectrometry to analysis of complex reaction mixtures. In
addition, we are indebted to the Physical Chemistry Depart-
ment for characterization of the carbohydrate intermediates.
a Reagents and conditions: (a) AgOTf (3 equiv), 4Å molecular
sieves (5.5 wt equiv), CH2Cl2, 25 °C, 20 h; (b) (i) anhydrous ZnCl2,
Ac2O/AcOH (2:1), 25 °C, 20 h, (ii) Zn0 dust (20 reducing equiv),
25 °C, 4 h.
Supporting Information Available: Experimental pro-
cedures and spectral data for compounds 2, 4, 9, and 10.
This material is available free of charge via the Internet at
http://pubs.acs.org.
10 was dissolved in acetic anhydride/acetic acid. Anhydrous
zinc chloride was added to remove the muramic C(6) benzyl
ether,12 revealing the alcohol, which acetylated in situ. Zinc
(9) For a preparation of 5, see: Jeanloz, R. W.; Walker, E.; Sinay¨, P.
Carbohydr. Res. 1968, 6, 184.
OL016692T
(10) DeNinno, M. P.; Etienne, J. B.; Duplantier, K. C. Tetrahedron Lett.
1995, 36, 669.
(11) Donor 3 may be prepared in three steps from commercially available
glucosamine. See: Imoto, M.; Yoshimura, H.; Shimoto, T.; Sakaguchi, N.;
Kusumoto, S.; Shiba, T. Bull. Chem. Soc. Jpn. 1987, 60, 2205.
(12) Yang, G.; Ding, X.; Kong, F. Tetrahedron Lett. 1997, 38, 6725.
(13) All compounds gave satisfactory combustion analysis (C, H, N).
All spectral data (HRMS, 1H NMR, 13C NMR, and IR) was consistent with
the assigned structures.
Org. Lett., Vol. 3, No. 22, 2001
3577