view release on metacpan or  search on metacpan

t/data/good/10398587.utf8  view on Meta::CPAN

   superimposes with the aromatic ring of triclosan containing the
   2-hydroxyl group. In both cases the arrangement of the aromatic rings
   leads to extensive stacking interactions with the nicotinamide. The
   bridging oxygen of the triclosan is positioned almost in the center of
   the boron-containing diazaborine ring. The second half of the inhibitor
   is not involved in interactions with the NADH. In different
   2-hydroxydiphenyl ethers [[118]Rozwarski et al 1998], this half seems
   to be of minor importance for antimicrobial activity, since variations
   of the substituents at the second ring have little effect on the
   minimal inhibitory concentration (MIC). This finding is in very good
   agreement with this structure, since the first half of the inhibitor is
   buried in the binding pocket and is involved in protein and cofactor
   interactions whereas the second half is partially solvent exposed and
   forms only a few hydrophobic interactions with the protein.

   Recently, the IC[50] values for EnvM inhibition and MICs against E.
   coli containing wild-type and mutant envM genes were determined for six
   2-hydroxydiphenylethers [[119]Heath et al 1998]. The 2-hydroxyl group
   appeared to be crucial for antibacterial activity of the triclosan
   analogs, consistent with the structural data discussed above.
   Furthermore, it was shown that replacement of the bridging ether oxygen
   atom with sulfur atom abolished antibacterial activity of the triclosan
   analogs and raised the IC[50] 32-fold. This result is not surprising,
   since the bulkier sulfur atom will lead to an increase in drug-cofactor
   distance, thereby disrupting the hydrogen-bond network involving the
   triclosan 2-hydroxyl group.

   Three mutations in the gene encoding E. coli EnvM have been identified
   which lead to resistance against triclosan: Gly93Val, Met159Thr and
   Phe203Leu [[120]McMurry et al 1998]. All three residues are located in
   close proximity to the inhibitor and are involved in the formation of
   the cofactor binding site ([121]Figure 4). The MIC for the Gly93Val
   mutation exceeds the other two mutations by almost a factor of 10.
   Substitution of Gly93 with a valine residue could have two effects: the
   C^α of Gly93 is positioned towards the inhibitor and a bigger
   side-chain would lead to steric interference with the triclosan,
   thereby preventing binding of the inhibitor. In addition, the
   side-chain of Val93 would be in close proximity to the side-chain of
   Lys163. Movement of Lys163 will disrupt hydrogen bond interactions
   between this residue and the NADH ribose. In InhA, mutation of Lys165
   (residue equivalent to Lys163 in EnvM) lowers the affinity of the
   enzyme for NADH (P.J.T., unpublished results). Since NADH has to be
   bound for inhibition by diazaborine inhibitors as well as, possibly,
   for triclosan, this could also account for the increased resistance
   resulting from mutations close to this Lys. The second mutation,
   Met159Thr, leads to a 12-fold increase in the MIC for triclosan
   compared to the wild-type, but results in a decrease in the MIC for
   diazaborines. Triclosan forms several hydrophobic interactions with the
   side-chain of Met159, which are not possible with the shorter
   side-chain of Thr159, thus the drug might not be bound as strongly as
   in the wild-type protein. On the other hand, the mutation of Met159 to
   Thr seems to eliminate an unfavorable interaction between the C^ε atom
   of Met159 and one of the oxygen atoms of the diazaborine sulfonyl
   group, which may explain the increased sensitivity of this mutant to
   diazaborines. The third mutation, Phe203Leu, is located on the opposite
   side of the binding pocket relative to the other two mutations and
   leads to a sixfold increase in the MIC compared to the wild-type
   protein with a similar effect on the diazaborines. The side-chain of
   Phe203 seems to be important for the formation of the inner surface of
   the binding pocket and participates in hydrophobic interactions with
   both types of inhibitors. These interactions could be weakened in the
   mutant leading to a decreased binding affinity of either inhibitor.
   Very recently, mutations in InhA from M. smegmatisleading to resistance
   against triclosan have been identified [[122]McMurry et al 1999]. Two
   of the mutated residues, Met161 and Met103, superimpose or are in close
   proximity to Met159 in EnvM and might have the same effect as the
   Met159Thr mutation. The Ala124Val mutation in InhA is not located in
   the putative triclosan-binding site and its structural effect can not
   be explained with our current model.
     __________________________________________________________________

   [123]Full-size image (32K) - Opens new window [124]Full-size image
   (32K)

   Figure 4. Location of residues conferring resistance against triclosan.
   Stereo view of the cofactor/inhibitor binding pocket with the molecular
   surface of EnvM rendered transparent. Residues leading to triclosan
   resistance in E. coli are shown in all-bonds representation and are
   mapped in red onto the molecular surface. Phe203 forms part of the
   binding pocket near the 2-hydroxyl containing ring of triclosan, which
   can not be seen in this orientation. [125]Figure 4 has been generated
   with the program SPOCK [[126]Christopher 1998].
   [127]View Within Article
   The high-resolution structure of EnvM in complex with triclosan and
   NADH provides a framework for understanding the inhibitory mechanisms
   of triclosan in bacterial fatty acid biosynthesis. This structure
   suggests explanations for the decreased effectiveness of triclosan in
   certain E. coli strains harboring mutations in the envM gene and
   provides a model for structure based drug design of new triclosan
   derivatives lacking the toxic effects of the diazaborine inhibitors.

 §3§ Protein Data Bank accession numbers §3§

   The coordinates will be deposited in the Brookhaven Data Bank with
   accession code 1QSG, and can be requested by e-mail to:
   [128]kisker@pharm.sunysb.edu.

 §3§ Acknowledgements §3§

   This work was supported by a National Institute of Health Training
   Grant to M.J.S. and by an NIH grant to P.J.T. S.P. is a DOE/GAANN
   fellow. The NSLS in Brookhaven is supported by DOE and NIH and beamline
   X26C is supported in part by the SUNY Stony Brook Research Foundation.

 §3§ References §3§

   [129]Baldock et al 1996. C. Baldock, J.B. Rafferty, S.E. Sedelnikova,
   P.J. Baker, A.R. Stuitje, A.R. Slabas, T.R. Hawkes and D.W. Rice, A
   mechanism of drug action revealed by structural studies of enoyl
   reductase. Science 274 (1996), pp. 2107–2110. [130]Full Text via
   CrossRef | [131]View Record in Scopus | [132]Cited By in Scopus (99)

   [133]Baldock et al 1998a. C. Baldock, G.-J.D. Boer, J.B. Rafferty, A.R.
   Stuitje and D.W. Rice, Mechanism of action of diazaborines. Biochem.
   Pharm. 55 (1998), pp. 1541–1549. [134]View Record in Scopus |
   [135]Cited By in Scopus (30)

   [136]Baldock et al 1998b. C. Baldock, J.B. Rafferty, A.R. Stuitje, A.R.
   Slabas and D.W. Rice, The X-ray structure of Escherichia coli enoyl
   reductase with bound NAD^+ at 2.1 Ã… resolution. J. Mol. Biol. 284
   (1998), pp. 1529–1546. [137]Article | [138]PDF (3521 K) | [139]View