Desferriferrithiocin

Nordesferriferrithiocin, NDFFTH(2), is a derivative of the siderophore desferriferrithiocin, DFFTH(2), in which the methyl group is substituted by a hydrogen atom. Both compounds show high oral activity as possible drugs for the treatment of iron overload. While DFFTH(2) is significantly toxic, NDFFTH(2) exhibits a lower toxicity and offers a much better therapeutic window than other orally active iron chelators. In this study, complexes of DFFTH(2) and NDFFTH(2) with various trivalent metals have been synthesized and characterized. Five isomers (the maximum possible) have been observed in the case of [Co(DFFT)(2)](-) in solution, as proved by (1)H-NMR measurements. Although normally labile, complexes of Al(3+) ([Al(DFFT)(2)](-)) have been separated by HPLC. In general, DFFTH(2) forms kinetically inert complexes whereas complexes of NDFFTH(2) tend to isomerize quickly in solution, as indicated by CD spectroscopy of separated HPLC fractions of [Cr(NDFFT)(2)](-). The most stable isomers of the aluminum complexes of both ligands have been characterized by X-ray crystallography; K[Al(DFFT)(2)] crystallizes from methanol/diethyl ether in the orthorhombic space group P2(1)2(1)2 with a = 11.238(3) Å, b = 31.719(11) Å, c = 7.684(2) Å, V = 2739.2(24) Å(3), and Z = 4. This isomer has the mer-(N,O-Lambda)(S,S) configuration, while K[Al(NDFFT)(2)] crystallizes from methanol/diethyl ether in the space group P6(1) (a = 21.269(8) Å, c = 9.643(3) Å, V = 3777.8(42) Å(3), Z = 6) and has the same coordination geometry. The solution thermodynamics of the Al(3+), Ga(3+), and Fe(3+) complexes have been studied by spectrophotometric titration. The stability constants (log K) are 23.6(1), 29.2(3), and 31.04(3), respectively, for the DFFTH(2) complexes and 22.0(1), 27.8(2), and 29.09(3), respectively, for the NDFFTH(2) complexes. Cyclic voltammograms of both iron complexes have been recorded in water at a carbon disk working electrode and in DMF at a graphite working electrode. The reduction waves measured in DMF indicate no reversibility whereas in water a quasi-reversible reduction is observed. The reduction potentials (E(1/2)'s) in water are -166 mV for [Fe(DFFT)(2)](-) and -97 mV for [Fe(NDFFT)(2)](-) versus NHE. These potentials are well in the range for biological reductants, which makes possible an in vivo reduction mechanism for the iron removal from the siderophore.

The microbial iron chelators desferriferrithiocin and desferricrocin as well as human lactoferrin were tested in vitro against Plasmodium falciparum. The microbial chelators inhibit the growth of P. falciparum in a dose dependent way. Parasite multiplication is stopped at 25-30 microM desferriferrithiocin, whereas 60-90 microM desferricrocin are needed to exhibit the same effect. After iron saturation, the microbial chelators are ineffective. Human lactoferrin (30 microM), both iron free and iron saturated, inhibits P. falciparum. A 3-day preincubation of host erythrocytes with iron free and iron saturated lactoferrin prior to infection enhances this effect, which is therefore attributed to lactoferrin bound iron. It has been suggested that the lactoferrin/iron complex generates oxygen free radicals, which may cause membrane damage of both erythrocyte and parasite. This process can be considered to lead to growth inhibition of the parasite.