Synthesis and characterisation of the bacteriostatic effects of aminocarboxylate and aminophenolate ligands

Abstract

This thesis describes efforts towards the synthesis and evaluation of ligands that may be used in
place of EDTA, a ligand that is slow to biodegrade after use, in antibacterial formulations.

A number of synthetic strategies have been employed in the search for candidate ligands, such as substituting
the carboxylate groups of EDTA with amide groups to form systems bearing 1 to 4 amide groups, using ligands which are able to form extremely stable complexes with Fe3+ on account of their phenolic donor groups, and the design and synthesis of ligands incorporating motifs known to be biodegradable.

To gauge the bacteriostatic efficacy of the studied ligands, optical-density based methods of monitoring the growth of E. coli were used, allowing ranking of the ligands according to their inhibitory
power. Of the synthesised ligands, symmetrical EDTA bis-amides bearing strongly coordinating pen-
dent groups (e.g. carboxylates and pyridines) gave the greatest extent of inhibition against E. coli.
So as to understand the chemical reasons dictating the extent of inhibition of the studied ligands,
thermodynamic characterisation (via pH- potentiometry and spectroscopic measurements) was un-
dertaken. Comparison of the metal ion binding constants obtained for a selection of synthesised EDTA amides with those of other aminocarboxylate ligands, indicated that the amides almost always had smaller binding affinities for Ca2+ , Mg2+ , Fe3+, Mn2+, and Zn2+ , indicating that the strength of ligand binding to these ions is not the sole determinant of the efficacy for the EDTA amides.

Studies into the importance of association constants of a selection of known, commercially available aminocarboxylate ligands to Ca2+ , Mg2+ , Fe3+, Mn2+ , and Zn2+ indicate that a certain level of Mg2+ binding (at the experimental pH) is relevant to inhibition, but this is not the case for phenolic ligands if the Fe3+ binding is sufficiently high.

Assessment of a number of variables on the growth
inhibition of E. coli showed that growth inhibition was most closely correlated with parameters pertaining to Fe3+ binding. ICP-MS studies show that aminocarboxylate ligands can extensively deplete
manganese and iron from cells, which may not have been apparent from inspection of their Mn2+
binding constants. In constrast, the phenolic EHPG depleted iron almost exclusively.

Based on the above, rational ligand design was attempted and some suggestions for future work have been given.