Antibody-catalyzed cleavage of the D-Ala-D-Lac depsipeptide: an immunological approach to the problem of vancomycin resistance.

Article abstract of DOI:10.1016/S0960-894X(02)00047-1

Vancomycin resistance is currently a major healthcare problem. The development of a catalytic monoclonal antibody (mAb) that hydrolyzes the D-Ala-D-Lac depsipeptide provides a potentially novel antibiotic strategy. A phosphonate hapten design was used to program antibody catalysis. The characteristics of the hapten were shown to be important for obtaining a viable immune response and several catalytic mAbs that cleave a peptidoglycan model substrate. The best mAb afforded a >500-fold rate enhancement over background.

Full text of DOI:10.1016/S0960-894X(02)00047-1

Bioorganic & Medicinal Chemistry Letters 12 (2002) 861–864
Antibody-Catalyzed Cleavage of the D-Ala-D-Lac Depsipeptide:
An Immunological Approach to the Problem of Vancomycin
Resistance
Shigeki Isomura, Jon A. Ashley, Peter Wirsching* and Kim D. Janda*
The Scripps Research Institute and the Skaggs Institute for Chemical Biology, Department of Chemistry,
BCC-582, 10550 N. Torrey Pines Road, La Jolla, CA 92037, USA
Received 2 January 2002; accepted 2 January 2002
Abstract—Vancomycin resistance is currently a major healthcare problem. The development of a catalytic monoclonal antibody
(mAb) that hydrolyzes the d-Ala-d-Lac depsipeptide provides a potentially novel antibiotic strategy. A phosphonate hapten design
was used to program antibody catalysis. The characteristics of the hapten were shown to be important for obtaining a viable
immune response and several catalytic mAbs that cleave a peptidoglycan model substrate. The best mAb afforded a >500-fold rate
enhancement over background. # 2002 Published by Elsevier Science Ltd.
Vancomycin was widely instituted in the 1960s and has
arguably been the world’s most essential antibiotic over
the past three decades. During this time, the vancomy-
cin group of glycopeptides were propelled to the
forefront in combating Gram-positive streptococcal,
staphylococcal, and enterococcal infections that
acquired resistance to the penicillins and cephalospor-
ins, amongst others.1,2 However, since 1989, healthcare
environments have witnessed an escalation in vanco-
mycin-resistant enterococci (VRE) that has resulted in
high mortality rates, particularly in immunocompro-
mised individuals.3,4 The real and serious threat posed
by the prevalent and opportunistic VRE arises not only
from their own broad antibiotic resistance, but also
from the potential for transmission of resistance to
more virulent organisms, such as methicillin-resistant
Staphylococcus aureus (MRSA).5À7 Although yet to be
observed clinically, transfer of vancomycin resistance to
MRSA and other pathogenic bacteria could have
devastating consequences.
pentapeptide cell-wall peptidoglycan (PG) sub-
structures, which inhibits PG cross-linking and thus
leads to cell-wall defects and bacterial lysis (Scheme 1).
The predominant strategy used by VRE is replacement
of pentapeptide d-Ala-d-Ala termini with the d-Ala-d-
Lac depsipeptide, resulting in the loss of one hydrogen
bond with vancomycin, a 1000-fold decrease in affinity
and essentially useless antibiotic activity. Semisynthetic
glycopeptides1,11 and glycopeptide analogues12,13 that
address the resistance problem are in various phases of
development and, clearly, other alternative approaches
are warranted. Recently, screening of a small-molecule
library led to the design of prolinol derivatives that
cleaved d-Ala-d-Lac substrates and resensitized bac-
teria to vancomycin.14 Herein, we describe the founda-
tion for immunotherapy as a new method in the
treatment of VRE based on catalytic antibodies that
hydrolyze the d-Ala-d-Lac ester bond in PG precursors.
The phosphonate transition-state (TS) analogue strat-
egy has been paramount for obtaining antibody ester-
ases.15,16 To this end, hapten 1 (VCA) was prepared in
homochiral form as a peptidomimetic for the l-Lys-d-
Ala-d-Lac fragment starting from known 10 and using
literature modifications to obtain 11–13 (Fig. 1, Scheme
2).17À19 Notably, the linker design incorporated a pre-
activated carboxylic group that afforded regiospecific
immunoconjugation. The l-Lys residue was included to
improve the elicitation of an immune response, since
For this reason, the explosive emergence of VRE has led
to the intensive investigation of the molecular mechan-
isms of glycopeptide resistance.8À10 Vancomycin binds
noncovalently to the l-Lys-d-Ala-d-Ala portion of
*Corresponding author. Tel.: +1-858-784-2516; fax: +1-858-784-
2595; e-mail: kdjanda@scripps.edu
0960-894X/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.
PII: S0960-894X(02)00047-1

862
S. Isomura et al. / Bioorg. Med. Chem. Lett. 12 (2002) 861–864
previous studies in our laboratory suggested a smaller
hapten would not be sufficiently immunogenic. Immu-
nization of mice with 1 coupled to keyhole limpet
hemocyanin (KLH) as a carrier protein yielded a panel
of six monoclonal antibodies (mAbs). Indeed, the
unusually low number of stable clones obtained was
apparently due to a poorly immunogenic hapten which
afforded low titer values. Yet, when screened for cata-
lytic activity using 3, four of the six mAbs afforded a
moderate rate enhancement over background. The sub-
strate 3 is structurally congruent to the l-Lys-d-Ala-d-
Lac target, but fitted at the distal end of the requisite
linker with a chromophoric auxiliary that allows mon-
itoring of cleavage product 6 by HPLC (Fig. 1). The
mAb VCA33H3 demonstrated the best esterase activity
and was analyzed in further detail.
saturable mAb active site was the same as that elicited
by hapten 1. Surprisingly, the Ki value was much higher
than the tight binding (<10À6 M) generally observed in
Well-behaved
Michaelis–Menten
kinetics
were
observed, in which it was possible to approach a prac-
tical Vmax at nearly physical saturation with no evidence
of product inhibition. Upon mAb saturation with 3,
d-Ala-d-Lac cleavage was >500-fold that of the spon-
taneous rate (Table 1). The phosphonate 2 competitively
inhibited the reaction, which demonstrated that the
Scheme 1. Illustration of PG biochemistry in Gram-positive bacteria
and the vancomycin resistance problem. Glycopeptide precursors are
synthesized in the cytoplasm, anchored via a hydrophobic tail, and
transported outside the cell membrane for assembly (*=N-acetyl-
muramate, *=N-acetylglucosamine). In pathway A, disaccharide
subunits are appended to a growing chain by a transglycosylase. In
pathway B, a transpeptidase cross-links the N-terminus of a pentagly-
cine unit, attached to l-Lys in a prior step, with another peptide moi-
ety. Cross-linking between different poly(glycopeptides), in which
either d-Ala or d-Lac can serve as a leaving group, forms a stable PG
matrix. The precise architecture has been hypothesized.28 Vancomycin
sterically blocks transglycosylation and transpeptidation. Loss of one
hydrogen bond (X=*) results in drug resistance. A catalytic antibody
designed to cleave d-Ala-d-Lac could restore sensitivity to vancomycin.
Figure 1. Structures of compounds under discussion.

S. Isomura et al. / Bioorg. Med. Chem. Lett. 12 (2002) 861–864
863
Scheme 2. Synthesis of the transition-state analogue hapten: (a) (i) 30% HBr; (ii) di-t-butyldicarbonate; (b) DCC, benzyl alcohol; (c) (i) NaIO4;
(ii) 1-adamantanamine; (d) (i) aq HCl; (ii) benzyl d-lactate, BOP, DIPEA; (e) (i) TFA, CH2Cl2; (ii) Boc-Lys(Ac)-OH, HBTU, NMM; (f) (i) TFA, CH2Cl2;
(ii) Boc-b-Ala-OH, HBTU, NMM; (g) (i) TFA, CH2Cl2; (ii) N-hydroxysuccinimidyl hemiglutaryl chloride, DIPEA, CH2Cl2; (h) H2, Pd/C, MeOH.
Table 1. Kinetic data for VCA33H3 that cleaves the d-Ala-d-Lac
bond in model PG substrate 3a
diastereomer 4. This compound showed no detectable
activity with VCA33H3, indicating that the configura-
tion of the methyl group is critical to binding and/or
catalysis. The lactate methyl group, as programmed by
1, can help ‘lock’ the substrate 3 and the developing TS
in the correct orientation, affording perhaps 3.2 kcal/
mol in binding energy,25 some of which translates into
catalysis. Given the numerous conformational minima
that are available to flexible structures such as 1 and 3,
together with the absence of a dominant epitope (e.g.,
aromatic ring), makes the additivity of weak interac-
tions particularly important. Additional support for this
premise was furnished by peptide 5, devoid of the lysine
side chain, where the mAb did not catalyze formation of
7. Hence, it is clear that all stereoelectronic components
programmed by the hapten are essential for substrate
activity.
kcat
(hÀ1
Km
(mM)
kcat=Km
(MÀ1 minÀ1
kcat=kuncat
Ki
(mM)b
)
)
0.120
2.17
0.921
530
160
aDetermined in 100 mM MOPS, pH 7.4, 5% DMSO cosolvent at
23 ꢂC. The kobs (3)=2.25Â10À4 hÀ1
.
bThe value using compound 2.
catalytic antibody reactions. On the contrary, the bind-
ing of VCA33H3 to the conjugate VCA–KLH titered in
the nanomolar range similar to other haptens. It is likely
that antibody avidity and linker/protein recognition
effects20,21 are operative, which further underscores
the borderline immunogenic properties of VCA. Even
though weak immunogenicity of the TS analogue prob-
ably leads to reduced TS stabilization for hydrolysis,
additional physicochemical interactions invoked by the
hapten could fortuitously contribute to catalysis (e.g.,
hydrogen bonding, electrostatics, ‘bait–switch’ effects).20
In fact, the Km=Ki ꢀkcat=kuncat relationship, according
to a standard thermodynamic cycle based on TS
theory,22À24 would suggest a rate enhancement for
VCA33H3 of only ꢁ12.5, or ꢁ42-fold less than that
observed. Hence, it is possible that some aspect of general/
acid–base or nucleophilic chemistry, anchored by the core
structure that directs TS stabilization, helps promote
catalysis.
The d-Ala-d-Lac phenotype in VRE is generally
induced, although constitutive expression may
arise.3,8À10 Since a given fraction of PG in VRE still
relies on d-Ala-d-Ala precursors in the presence or
absence of vancomycin, we examined interaction of
VCA33H3 with 8 and 9. Hydrolysis or merely com-
plexation of d-Ala-d-Ala could also serve as immuno-
therapy. No hydrolytic activity using 8 was observed,
apparently from a lack of catalytic power and not
ground-state binding, as judged from 9, which acted as
a competitive inhibitor (Ki=13 mM). In principle, mAb
binding could mimic the action of vancomycin and
supplement catalytic action. Although the Ki value for 9 is
high, transpeptidases show very weak binding (Km ꢀ 40–
100 mM) to model substrates26 and so competition in
Evidence that the locale of the phosphonate group is
central to hapten recognition and to the energetics of
d-Ala-d-Lac hydrolysis came from the study of

864
S. Isomura et al. / Bioorg. Med. Chem. Lett. 12 (2002) 861–864
vivo is possible. New hapten designs should address this
issue, since some bacterial species, notably MRSA, can
evolve vancomycin resistance simply by increasing the
production of d-Ala-d-Ala termini.5,6
7. Levy, S. B. Sci. Am. 1998, 3, 46.
8. Walsh, C. T.; Fisher, S. L.; Park, I.-S.; Prahalad, M.; Wu,
Z. Chem. Biol. 1996, 3, 21.
9. Walsh, C. T. Science 1993, 261, 308.
10. Arthur, M.; Courvalin, P. Antimicrob. Agents Chemother.
1993, 37, 1563.
Bacteria have always demonstrated the ability to readily
acquire resistance to small-molecule drugs. Hence, there
is a need for alternative approaches to supplement
standard antibiotic treatments. Peptide and protein-
based structures that operate via catalysis present one
such option. In this regard, proteins up to 50kDa (e.g.,
scFv, 25 kDa; Fab, 50kDa) are good candidates to
penetrate the cell-wall fabric during normal growth
conditions.27 Studies are underway in our laboratory to
determine the bactericidal potential for VCA33H3 and
related mAbs, as well as associated strategies for using
immunological reagents as antibiotics.
11. Malabarba, A.; Nicas, T. I.; Ciabatti, R. Eur. J. Med.
Chem. 1997, 32, 459.
12. Ge, M.; Chen, Z.; Onishi, H. R.; Kohler, J.; Silver, L. L.;
Kerns, R.; Fukuzawa, S.; Thompson, C.; Kahne, D. Science
1999, 284, 507.
13. Xu, R.; Greiveldinger, G.; Marenus, L. E.; Cooper, A.;
Ellman, J. A. J. Am. Chem. Soc. 1999, 121, 4898.
14. Chiosis, G.; Boneca, I. G. Science 2001, 293, 1484.
15. Schultz, P. G.; Lerner, R. A. Science 1995, 269, 1835.
16. Janda, K. D.; Chen, Y.-C. J. Handb. Exp. Pharmacol.
1994, 113, 209.
17. Baylis, E. K.; Campbell, C. D.; Dingwall, J. G. J. Chem.
Soc., Perkin Trans. 1 1984, 2845.
18. Karanewsky, D. S.; Badia, M. C. Tetrahedron Lett. 1986,
27, 1751.
Acknowledgements
19. Campagne, J.-M.; Coste, J.; Jouin, P. Tetrahedron Lett.
1993, 34, 6743.
20. Matsushita, M.; Hoffman, T. Z.; Ashley, J. A.; Zhou, B.;
Wirsching, P.; Janda, K. D. Bioorg. Med. Chem. Lett. 2001,
11, 87.
21. Beekman, N. J. C. M.; Schaaper, W. M. M.; Langeveld,
J. P. M.; Boshuizen, R. S. J. Peptide Res. 2001, 58, 237.
22. Radzicka, A.; Wolfenden, R. Methods Enzymol. 1995,
249, 284.
This work was supported by funding from the Skaggs
Institute for Chemical Biology and a grant from the
National Institutes of Health (GM 43858) (K.D.J.).
References and Notes
23. Stewart, J. D.; Benkovic, S. J. Nature 1995, 375, 388.
24. Benkovic, S. J.; Napper, A. D.; Lerner, R. A. Proc. Natl.
Acad. Sci. U.S.A. 1988, 85, 5355.
1. Nicolaou, K. C.; Boddy, C. N. C.; Brase, S.; Winssinger, N.
Angew. Chem. Int. Ed. 1999, 38, 2096.
25. Fersht, A. R.; Shindler, J. S.; Tsui, W.-C. Biochemistry
1980, 19, 5520.
2. Williams, D. H.; Bardsley, B. Angew. Chem. Int. Ed. 1999,
38, 1172.
26. Kozarich, J. W.; Strominger, J. L. J. Biol. Chem. 1978,
253, 1272.
3. Cetinkaya, Y.; Falk, P.; Mayhall, C. G. Clin. Microbiol.
Rev. 2000, 13, 686.
27. Demchick, P.; Koch, A. L. J. Bacteriol. 1996, 178, 768.
28. Dmitriev, B. A.; Ehlers, S.; Rietschel, E. Th. Med. Micro-
biol. Immunol. 1999, 187, 173.
4. Perl, T. M. Am. J. Med. 1999, 106, 26S.
5. Livermore, D. M. Int. J. Antimicrob. Agents 2000, 16, S3.
6. Hiramatsu, K. Drug Resist. Updates 1998, 1, 135.