Synthesis and evaluation of a new fluorescent transglycosylase substrate: Lipid II-based molecule possessing a dansyl-C20 polyprenyl moiety

Article abstract of DOI:10.1021/ol100338v

(Figure Presented) The preparation of a novel fluorescent lipid II-based substrate for transglycosylases (TGases) is described. This substrate has characteristic structural features including a shorter lipid chain, a fluorophore tag at the end of the lipid chain rather than on the peptide chain, and no labeling with a radioactive atom. This fluorescent substrate is readily utilized In TGase activity assays to characterize TGases and also to evaluate the activities of TGase inhibitors.

Full text of DOI:10.1021/ol100338v

ORGANIC
LETTERS
2010
Vol. 12, No. 7
1608-1611
Synthesis and Evaluation of a New
Fluorescent Transglycosylase Substrate:
Lipid II-Based Molecule Possessing a
Dansyl-C20 Polyprenyl Moiety
Chen-Yu Liu,† Chih-Wei Guo,‡ Yi-Fan Chang,‡ Jen-Tsung Wang,† Hao-Wei Shih,‡
Yu-Fang Hsu,† Chia-Wei Chen,‡ Shao-Kang Chen,‡ Yen-Chih Wang,‡
Ting-Jen R. Cheng,‡ Che Ma,‡ Chi-Huey Wong,‡ Jim-Min Fang,*,†,‡ and
Wei-Chieh Cheng*,‡
Department of Chemistry, National Taiwan UniVersity, Taipei, Taiwan, and
The Genomics Research Center, Academia Sinica, Nankang, Taipei, Taiwan
wcheng@gate.sinica.edu.tw; jmfang@ntu.edu.tw
Received February 9, 2010
ABSTRACT
The preparation of a novel fluorescent lipid II-based substrate for transglycosylases (TGases) is described. This substrate has characteristic
structural features including a shorter lipid chain, a fluorophore tag at the end of the lipid chain rather than on the peptide chain, and no
labeling with a radioactive atom. This fluorescent substrate is readily utilized in TGase activity assays to characterize TGases and also to
evaluate the activities of TGase inhibitors.
Due to serious antibiotic drug resistance, such as VRE
(vancomycin resistant enterococcus) and MRSA (methicillin-
resistant Staphylococcus aureus), it is urgent for scientists
to search for new antimicrobial targets for the development
of effective antibiotics.1 In bacterial cell-wall biosynthesis,
transglycosylases (TGases), one class of glycosyl-trans-
ferases, catalyze the transfer of the sugar moiety from the
activated polymeric peptidoglycan (a glycosyl donor) to the
specific hydroxyl group (4-OH) of Lipid II (1, a glycosyl
acceptor) with concomitant release of an undecaprenyl
pyrophosphate moiety (Figure 1).2 These bacterial enzymes
are attractive antibiotic targets since their functions are
† Department of Chemistry, National Taiwan University.
‡ The Genomics Research Center, Academia Sinica.
Figure 1. Lipid II as the substrate in transglycosylation for bacterial
(1) (a) Mazel, D.; Dychinco, B.; Webb, V. A.; Davies, J. Science 1998,
280, 605–608. (b) Pfaller, M. A.; Jones, R. N.; Doern, G. V.; Kugler, K.
Antimicrob. Agents Chemother. 1998, 42, 1762–1770. (c) Chu, D. T. W.;
Plattner, J. J.; Katz, L. J. Med. Chem. 1996, 39, 3853–3874. (d) Kahne, D.;
Leimkuhler, C.; Lu, W.; Walsh, C. T. Chem. ReV. 2005, 105, 425–448.
(2) (a) Silver, L. L. Curr. Opin. Microbiol. 2003, 6, 431–438. (b) Walsh,
C. Nature 2000, 406, 775–781.
peptidoglycan formation.
essential, their structures are conserved, and they are located
on the external surface of bacterial membranes to allow easy
10.1021/ol100338v  2010 American Chemical Society
Published on Web 02/26/2010

access by inhibitors. Furthermore, the polysaccharide back-
bone of peptidoglycan remains conserved in methicillin-
resistant strains; thus, antibiotics targeting the transglyco-
sylation (TG) step may therefore be less liable to resistance.3
However, the study of transglycosylation for drug discovery
has been hampered by the difficulty in acquirement and
modification of the TGase substrate 1. Isolation of 1 from
bacterial sources is extremely difficult and tedious due to
its low natural abundance and its inherent structural com-
plexity.4 Fortunately, synthetic approaches toward 1 and its
analogues via chemical or chemo-enzymatic methods have
shed light on this study.5-7 For example, the Lipid II
analogue 2 bearing a fluorophore of the dansyl group at the
lysine residue of the pentapeptide chain has been prepared
(Figure 2).8 A chemo-enzymatic approach has culminated
betulaheptaprenyl lipid chain (C35) with defined double bond
configurations is a better TGase substrate than the natural Lipid
II with the C55 lipid chain.9 The transglycosylation products
usually contain mixtures of immature peptidoglycans with
varying degrees of polymerization that are not easily quantified.
The formed immature peptidoglycan can be subjected to
enzymatic cleavage by N-acetylmuramidase to release the
GlcNAc-MurNAc-pentapeptide molecules, the truncated Lipid
II without the moiety of undecaprenyl diphosphate.8,10
During the course of our research on the development of
new antibiotics that target TGase in bacteria,11,12 we explored
an efficient method for the preparation of polyprenyl alcohols
with varying chain lengths and double bond configurations
via solution- and solid-phase organic synthesis.13 We have
also prepared the fluorescent Lipid II (2)8 and its analogues
3-5 carrying a dansyl tag at the terminal ε-NH2 site of lysine
(see Supporting Information). In accord with our expecta-
tions, the dansyl-labeled Lipid II (2) and analogue 4, bearing
a betulaheptaprenyl chain, are active substrates for TGase.
Compound 3 bearing a benzene ring at the end of the lipid
chain is still an active substrate toward transglycosylation,
whereas compound 5 with a pentaprenyl chain is inert to
TGase. Inspired by this result, we proposed that compound
7 would be a possible substrate that could further improve
the TGase assay (Figure 3). Compound 7 incorporates a
Figure 2. Structures of Lipid II analogues 2-5 bearing a fluorescent
tag at the ε-NH2 site of lysine.
in the syntheses of radioactive Lipid II and several analogues
with all glucosamine carbons having 14C-labeling.9 The assay
systems for evaluation of the TGase activity have thus been
established with the assistance of the fluorescent or radioac-
tive Lipid II substrates.8,9
Walker and co-workers have shown, in their assay using
radioactive substrate, that a Lipid II analogue containing the
Figure 3. A schematic diagram for design of the fluorescent TGase
substrate 7 by mimicking a truncated Lipid II analogue 6 via an
imaginary transformation A.
(3) Ritter, T. K.; Wong, C.-H. Angew. Chem., Int. Ed. 2001, 40, 3508–
3533.
(4) (a) van Heijenoort, Y.; Gomez, M.; Derrien, M.; Ayala, J.; van
Heijenoort, J. J. Bacteriol. 1992, 174, 3549–3557. (b) Nakagawa, J.; Tamaki,
S.; Tomioka, S.; Matsuhashi, M. J. Biol. Chem. 1984, 259, 13937–13946.
(5) (a) VanNieuwenhze, M. S.; Mauldin, S. C.; Zia-Ebrahimi, M.;
Winger, B. E.; Hornback, W. J.; Saha, S. L.; Aikins, J. A.; Blaszczak, L. C.
J. Am. Chem. Soc. 2002, 124, 3656–3660. (b) Schwartz, B.; Markwalder,
dansyl group into the lipid chain for facile detection of the
progress of transglycosylation. The dansyl group can be
J. A.; Wang, Y. J. Am. Chem. Soc. 2001, 123, 11638–11643
.
(6) Lo, M.-C.; Men, H.; Branstrom, A.; Helm, J.; Yao, N.; Goldman,
(10) (a) Hsu, S.-T. D.; Breukink, E.; Tischenko, E.; Lutters, M. A. G.;
de Kruijff, B.; Kaptein, R.; Bonvin, A. M. J. J.; van Nuland, N. A. J. Nat.
Struct. Mol. Biol. 2004, 11, 963–967. (b) Baizman, E. R.; Branstrom, A. A.;
Longley, C. B.; Allanson, N.; Sofia, M. J.; Gange, D.; Goldman, R. C.
R.; Walker, S. J. Am. Chem. Soc. 2000, 122, 3540–3541
.
(7) Breukink, E.; van Heusden, H. E.; Vollmerhaus, P. J.; Swiezewska,
E.; Brunner, L.; Walker, S.; Heck, A. J. R.; de Kruijff, B. J. Biol. Chem.
2003, 278, 19898–19903
.
Microbiology 2000, 1463129-3140
.
(8) (a) Schwartz, B.; Markwalder, J. A.; Seitz, S. P.; Wang, Y.; Stein,
R. L. Biochemistry 2002, 41, 12552–12561. (b) Fraipont, C.; Sapunaric,
F.; Zervosen, A.; Auger, G.; Devreese, B.; Lioux, T.; Blanot, D.; Mengin-
Lecreulx, D.; Herdewijn, P.; Van Beeumen, J.; Frre, J.-M.; Nguyen-Distche,
M. Biochemistry 2006, 45, 4007–4013.
(11) Liu, H.; Wong, C.-H. Bioorg. Med. Chem. 2006, 14, 7187–7195
.
(12) Cheng, T.-J. R.; Sung, M.-D.; Liao, H.-Y.; Chang, Y.-F.; Chen,
C.-W.; Huang, C.-Y.; Chou, L.-Y.; Wu, Y.-D.; Chen, Y.-H.; Cheng, Y.-
S. E.; Wong, C.-H.; Ma, C.; Cheng, W.-C. Proc. Natl. Acad. Sci. U.S.A.
2008, 105, 431–436
.
(9) Ye, X.-Y.; Lo, M.-C.; Brunner, L.; Walker, D.; Kahne, D.; Walker,
S. J. Am. Chem. Soc. 2001, 123, 3155–3156.
(13) Chang, Y.-F.; Liu, C.-Y.; Guo, C.-W.; Wang, Y.-C.; Fang, J.-M.;
Cheng, W.-C. J. Org. Chem. 2008, 73, 7197–7203.
Org. Lett., Vol. 12, No. 7, 2010
1609

visualized as a mimic of the diprenyl unit, i.e. the lipid moiety
in 7 can be considered equivalent to the hexaprenyl chain in
6. Though the biological function of the polyprenyl chain in
Lipid II is not fully understood, it is generally thought to
promote penetration into the hydrophobic bacterial mem-
brane.10 Lipid II analogues having a proper fluorophore on
the lipid chain, instead of the peptide chain,8 might not
interfere with substrate-enzyme recognition. In comparison,
the conventional use of a Lipid II analogue with a dansyl
group on the peptide chain, e.g., 2, in TG assay could result
in many fluorescent elongation products that could compli-
cate the HPLC analysis. On the other hand, the fluorescent
moiety of lipid diphosphate, e.g., in 7, would not be
incorporated into the elongation products, so that the HPLC
analysis would be much simplified.
expected problem was encountered in the desulfonation of
20. After several unsuccessful attempts at using LiEt3BH/
PdCl2(dppp),15 SmI2/HMPA,16 Na/naphthalene, Li/naphtha-
lene,17 or Mg/HgCl2 as the reducing agents, we finally
18
carried out the desulfonation reaction with Na/EtOH at 0
°C to give a 66% yield of an inseparable mixture containing
1
21a and isomer 21b in a ratio of 7:3 as shown by the H
NMR analysis. Isomer 21b, presumably from double bond
migration of the allylic radical intermediate,13 was character-
ized by the C11-methyl group showing as a doublet at δ 0.94
ppm. The isomeric mixture was carried through the sequence
of O-THP deprotection, potassium phthalimide substitution,
and phosphorylation to give compound 23 (Scheme 2) The
We herein describe the synthesis of compound 7 and its
use in a functional assay for the TGase represented by a full-
length penicillin-binding protein (PBP). To circumvent the
problem in handling the acid labile R-glycosyl diphosphate
group,14 the diphosphate moiety in 7 was constructed at the
final stage by conjugation of a disaccharyl pentapeptide 8
with the dansyl-lipid phosphate 9 as illustrated in the
retrosynthetic route (Scheme 1).
Scheme 2.
Synthesis of the Probe Molecule 7a
Scheme 1
.
Retrosynthetic Analysis of 7 Bearing a Dansyl Label
in the Lipid Chain
a The structures of isomers 22b and 23b are omitted for clarity.
phthaloyl group in 23 was removed by hydrazine, and a
subsequent reaction with dansyl chloride under basic condi-
tions delivered 9. The coupling reaction of 8 and 9 in the
presence of 1,1′-carbonyldiimidazole (CDI),5 followed by
global deprotection under basic conditions and purification
by reverse-phase HPLC, afforded the desired probe
molecule 7.
The feasibility of 7 for a TGase substrate was demonstrated
by HPLC-based TGase assays. As we had anticipated, 7
Compound 8 was prepared by using the reported
procedure5 with slight modifications, and syntheses of the
intermediate compounds 11-18 are described in the
Supporting Information. For the preparation of 9, nerol
(19) was converted to a key intermediate 20 according to
our previously described procedures.13 However, an un-
(15) Grassi, D.; Lippuner, V.; Aebi, M.; Brunner, J.; Vasella, A. J. Am.
Chem. Soc. 1997, 119, 10992–10999.
(16) Keck, G. E.; Savin, K. A.; Weglarz, M. A. J. Org. Chem. 1995,
60, 3194–3204.
(17) Liu, H.-J.; Yip, J.; Shia, K.-S. Tetrahedron Lett. 1997, 38, 2253–
2256.
(14) Narayan, R. S.; VanNieuwenhze, M. S. Eur. J. Org. Chem. 2007,
1399–1414.
(18) Lee, J. W.; Son, H. J.; Jung, Y. E.; Lee, J. H. Synth. Commun.
1996, 26, 1499–1505.
1610
Org. Lett., Vol. 12, No. 7, 2010

was recognized by both TGases from C. difficile and E.
coli. The transglycosylation reaction catalyzed by C.
difficile TGase (3 µig, 0.3 mg/mL) proceeded more
efficiently, consuming ∼90% of 7 (7.5 × 10-4 µmol, 75
µM) in 4 h at 25 °C (Figure 4).
Figure 5. Relative TGase inhibition percentage of moenomycin A
(24) and its analogue 25 via HPLC analysis with substrate 7.
than 1 and 3 µM, respectively. In contrast, when concentra-
tions were below 0.1 µM 24 or 0.3 µM 25, no inhibitory
activities were observed. In the presence of 0.3 µM 24 or
25, the TG inhibition was 59% or 14%, corresponding to
76% or 50% of nonconsumed 7, respectively.
Figure 4. Measurement of the transglycosylation progress by HPLC.
In conclusion, we have designed a new fluorescent TG
substrate 7 by incorporating a dansyl group to mimic the
diprenyl moiety in the lipid chain. We have also explored a
convergent approach to synthesize this probe molecule.
Unlike the Lipid II analogue 2 bearing the dansyl group at
the peptide site, substrate 7 could be readily used in the
functional assay of TGases without an additional treatment
with N-acetylmurmidase. The assay system with substrate 7
was also able to screen TGase inhibitors, as exemplified by
distinguishing the inhibitor potency of moenomycin A versus
its analogue in this study. In principle, one can also modify
compound 7 to construct a high throughput TG probe by
installation of another chromophore at the peptide site as
the FRET donor or acceptor to the dansyl group at the lipid
site. We are currently investigating the feasibility of this
research approach.
The reactions (a-g) are at t ) 0, 0.25, 1, 2, 4, 8, and 16 h. (see
details in the Supporting Information).
We further demonstrated the use of the fluorescent
substrate 7 in the characterization of TGase inhibitors.
Moenomycin A (24) is a potent natural inhibitor against
TGases,19-21 though its poor bioavailability has limited its
development as an antibacterial drug. According to our
binding assay fluorescence anisotropy method,12 moenomy-
cin A and its analogue 25 (see the structures in the SI) exhibit
IC50 values of 0.36 and 2.10 µM against TGase, respectively.
In agreement with this result, the TGase inhibition assays
with 7 as the substrate also showed that the two inhibitors
could be easily distinguished at concentrations between 0.3
and 1 µM (Figure 5). The TGase activity was completely
inhibited when the concentrations of 24 and 25 were higher
Acknowledgment. We thank the National Science Council
and Academia Sinica for financial support.
(19) (a) Wasielewski, E. V.; Muschaweck, R.; Schutz, E. Antimicrob.
Agents Chemother. 1965, 743–748. (b) Scherkenbeck, J.; Hiltmann, A.;
Hobert, K.; Bankova, W.; Siegels, T.; Kaiser, M.; Mu¨ller, D.; Veith, H. J.;
Fehlhaber, H.-W.; Seibert, G.; Markus, A.; Limbert, M.; Huber, G.; Bo¨ttger,
D.; Sta¨rk, A.; Takahashi, S.; van Heijenoort, Y.; van Heijenoort, J.; Welzel,
Supporting Information Available: Experimental pro-
cedures, characterization of the synthetic compounds, and
HPLC analysis of transglycosylation. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
P. Tetrahedron 1993, 49, 3091–3100
(20) Ostash, B.; Walker, S. Curr. Opin. Chem. Biol. 2005, 9, 459–466
(21) Stembera, K.; Vogel, S.; Buchynskyy, A.; Ayala, J. A.; Wezel, P.
ChemBioChem 2002, 3, 559–565
.
.
.
OL100338V
Org. Lett., Vol. 12, No. 7, 2010
1611