Active analog modelling of novel siderophores

Michael Meyer 1 and Wolfram Trowitzsch-Kienast 2




1Konrad-Zuse-Zentrum für Informationstechnik
Takustr. 7
14195 Berlin
Germany
E-mail: meyer@zib.de
2Technische Fachhochschule Berlin Fachbereich 3, Chemie und Biotechnik
Luxemburger Str. 10
D-13353 Berlin
Germany
E-mail: kienast@tfh-berlin.de






Siderophores are low molecular weight ligands produced by bacteria for iron uptake [Drechsel et al., 1997; Drechsel et al., 1998]. Therfore it is a promising strategy to use conjugates of siderophores and antibiotics as a trojan horse to increase the antibiotic concentration at the site of action. Siderophores are also used for the medical treatment of patients su ering from an excess of iron (thalassemiae). An example of natural siderophores is enterobactin consisting of three catechol groups for iron chelation, which are linked to a cyclic lactone skeleton of three serine residues. Starting from enterobactin we have desinged the novel siderophore Pyridinochelin with an active analog approach.

The complex structures of enterobactin and MECAM [Harris et al., 1979] with metals have been determined with ab initio methods at HF/6-31G(d) level. Subsequently the structures have been superimposed to build a pseudoreceptor using LUDI [Böhm, 1996]. Then catechoylamide groups have been selected as a pharmacophore necessary for binding of iron. In the nal 254 304 272 317 step the anchoring skeleton has been replaced by suitable molecular fragments satisfying sterical constraints and atom type properties. The iron transporters generated this way have been subjected to docking studies with the enterobactin receptor [Buchanan et al., 1999] using Autodock [Morris et al., 1998].

The computational studies indicate that the serine anchoring skeleton of en- terobactin can be replaced by thiazole or pyridine drivatives, like in Pyridinochelin. These fragments fit accurately to the geometry of the pharmacophore and have acceptor atoms like the natural analog. In contrast to previous synthetic siderophores derived from enterobactin, only two catechol groups are used for iron complexation. This is in agreement with computational and experimental studies showing that the third catechol group is not necessary for iron transport [Meyer et al., 1997; Ambrosi et al., 1998]. Furthermore, the enterobactin receptor shows no threefold symmetry requiring a corresponding symmetry of the siderophores.

Pyridinochelin has been synthesized and growth promotion experiments with a series of bacteria indicate that it enables a very efficient iron uptake.


REFERENCES