K.-W. Yang et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1085±1087
1087
Figure 3. (A) Proposed structure of tetrahedral intermediate of
cephalosporin hydrolysis by metallo-b-lactamase L1. (B) structure of
phosphonamidate 1.
that the positioning of a hydrogen in an inhibitor to
form a hydrogen bond to the enzyme can result in a 10-
fold increase in binding anity. The ability to control
the rotation about the P±C bond in 1 will likely result in
further increases in binding anity and potentially in a
clinically-useful inhibitor.
Acknowledgements
Figure 2. Slope and intercept replots of VanX inhibition by phospho-
namidate 1.
The authors thank Professor John Grunwell for reading
the manuscript and oering suggestions, also Eli Lilly
Co. for sharing the overexpression plasmid for VanX
with us. This work was supported by a grant (R29-
AI40052) from the National Institutes of Health and a
1998 Research Challenge grant to M.W.C. Lisa Chat-
wood acknowledges funding from Miami University's
Committee on Undergraduate Research and the
Howard Hughes Medical Institute.
to the phosphonate 3, and the increased binding anity
of the phosphonamidate is most likely due to the addi-
tional H-bond from the N±H of 1 to the enzyme. The
crystal structure of VanX demonstrated that the com-
parable N±H in substrate d-Ala-d-Ala is hydrogen
bonded to the backbone carbonyl of Tyr109.6
To extend the synthetic route in Scheme 1 to generate
additional potential inhibitors of VanX, chlorophos-
phonate 10 was coupled with methyl mercaptoacetate
and deprotected using standard procedures8,10,14 to yield
the phosphothioate 2. This compound was shown to be
stable for over a week at 4 ꢀC and at pH 7.0. Kinetic
studies demonstrated that the phosphothioate 2 did not
inhibit VanX at concentrations of 2 up to 1 mM, sug-
gesting an important role of the methyl group in sub-
strate/inhibitor binding. This result is in contrast to the
crystallographic studies that showed minimal contacts
between VanX and the C-terminal methyl on d-Ala-d-
Ala, phosphonate 3, and APMP.6 We cannot rule out
the possibility that the replacement of the methylene in
APMP, the amine in 1, and the oxygen in 3 with sulfur
explains the loss of inhibitory behavior by 2. Attempts
to synthesize a phosphothioate analogue containing a
C-terminal methyl substituent were unsuccessful.
References
1. Wu, Z.; Wright, G. D.; Walsh, C. T. Biochem. 1995, 34,
2455.
2. Walsh, C. T.; Fisher, S. L.; Park, I. S.; Prahalad, M.; Wu, Z.
Chem. Biol. 1996, 3, 21.
3. Reynolds, P. E.; Depardieu, F.; Dutka-Malen, S.; Arthur,
M.; Courvalin, P. Mol. Microbiol. 1994, 13, 1065.
4. Wu, Z.; Walsh, C. T. J. Am. Chem. Soc. 1996, 118, 1785.
5. Wu, Z.; Walsh, C. T. Proc. Nat. Acad. Sci. USA 1995, 92,
11603.
6. Bussiere, D. E.; Pratt, S. D.; Katz, L.; Severin, J. M.;
Holzman, T.; Park, C. H. Mol. Cell. 1998, 2, 75.
7. Kaplan, A. P.; Bartlett, P. A. Biochem. 1991, 30, 8165.
8. Bartlett, P. A.; Marlowe, C. K. Biochem. 1983, 22, 4618.
9. Hanson, J. E.; Kaplan, A. P.; Bartlett, P. A. Biochem. 1989,
28, 6294.
10. Jacobsen, N. E.; Bartlett, P. A. J. Am. Chem. Soc. 1981,
103, 654.
The phosphorus-containing dipeptide analogues 1, 2
and 3 were also tested as inhibitors of metallo-b-lacta-
mase L1 from Stenotrophomonas maltophilia because
these compounds have a similar structure as a part of
the putative tetrahedral intermediate formed during b-
lactam bond hydrolysis (Fig. 3). However, none of the
analogues inhibited L1 at concentrations up to 1 mM at
pH 7.0 using nitroce®n as the substrate.17
11. Tronrud, D. E.; Holden, H. M.; Matthews, B. W. Science
1987, 235, 571.
12. Bartlett, P. A.; Marlowe, C. K. Science 1987, 235, 569.
13. Oleksyszyn, J.; Subotkowska, L. Mastalerg, P. Synthesis
1979, 985.
14. Ellsworth, B. A.; Tom, N. J.; Bartlett, P. A. Chem. Biol.
1996, 3, 37.
15. Brandt, J. J.; Chatwood, L. L.; Yang, K. W.; Crowder, M.
W. Anal. Biochem. 1999, 272, 94.
16. Webb, J. L. Enzyme and Metabolic Inhibitors; Academic:
New York, 1963; Vol. 1.
17. Crowder, M. W.; Walsh, T. R.; Banovic, L.; Pettit, M.;
Spencer, J. Antimicrob. Agents Chemother. 1998, 42, 921.
The extremely narrow and selective active site of VanX
presents a major obstacle in designing small molecule
inhibitors. The results on phosphonamidate 1 suggest