FERREDOXIN-NADP+ REDUCTASE OF PLASMODIUM FALCIPARUM (PFFNR): PROTEIN ENGINEERING AND INHIBITION STUDIES

Malaria is recognised as one of the main health priorities by the World Health Organization. The most severe form of the disease, tropical malaria, is caused by Plasmodium falciparum. The diffusion of strains of this pathogen, resistant to traditional antimalarial treatments, urgently imposes the development of new therapies. P. falciparum belongs to the phylum Apicomplexa, which consists of unicellular, obligate intracellular parasites. These organisms possess an organelle of algal origin, the apicoplast, which has been shown to be required for pathogen survival and represents a known site of action of antimalarial compounds. In the apicoplast of P. falciparum a ferredoxin-NADP+ reductase (PfFNR) has been identified. This enzyme is directly involved in the electron transfer pathway from NADPH to LytB which catalyzes the last step of the mevalonate-independent isoprenoid biosynthesis. The crystallographic structure of PfFNR in complex with the substrate analogue 2’-P-AMP showed the presence of two basic residues, His286 and Lys249, in the NADP(H)-binding site, which are conserved within the Plasmodium genus but not in other plant-type homologues. Aim of this work has been to clarify the role of His286 and Lys249 in substrate binding and catalysis, and to identify and characterize PfFNR inhibitors. The study of His286 and Lys249 was carried out by site-directed mutagenesis. The replacement of His286 resulted in a multifaceted effect, which highly depended on the replacement made. Steady-state kinetics showed that the substitution with aliphatic residues, i.e. Ala and Leu, decreases both kcat/KmNADPH and kcat indicating that His286 plays a critical role not only in NADP(H) binding but also in catalysis. 2’-P-AMP inhibition studies and titrations with NADP+ also support this conclusions showing a decreasing in substrate affinity. Moreover rapid kinetic studies clearly demonstrated that the substitution His286Leu leads to a 3.4-fold decrease in the hydride-transfer (HT) rate between NADP(H) and FAD, with a destabilization of CT complexes between them. The His286Lys mutation resulted in the lowest kcat but surprisingly this enzyme form binds 2’-P-AMP and NADP+ with high affinity, indicating that this mutation destabilized the HT-competent conformation of the substrate NADP(H). Unexpectedly, steady-state kinetics and stopped-flow experiments showed that the mutation to Gln gave an enzyme more active than the wild-type, providing a better stabilization of the HT-competent conformation between nicotinamide and isoalloxazine ring. Thus, His286 has a role in modulating the enzyme affinity for NADP(H) and in the precise positioning of the nicotinamide ring in the active site. The replacement of Lys249 with an Ala resulted in an enzyme form with a 10-fold decrease in KmNADPH and Kd for NADP+ or 2’-P-AMP. In particular, as came out from the study of the NADH-K3Fe(CN)6 reductase reaction, this residue participates in substrate recognition by interacting with the 2’-phosphate of the pyridine dinucleotide. By the in silico screening of two virtual libraries I have identified a pool of compounds potentially active against PfFNR. Five compounds, i.e. I8, I19, I21, I24 and I27, showed an IC50 in the micromolar range and I deeply characterized their inhibition mechanism by steady-state kinetic studies, also evaluating their specificity. Although in the in silico screening the substrate-binding site of PfFNR was targeted for the selection of enzyme ligands, all the compounds found turned out to inhibit PfFNR by a mixed-type mechanism, and did not show the expected pure competitive behavior towards NADPH. Some of the inhibitors turned out to be very specific for Apicomplexan FNRs. Inhibition studies were carried out also to characterize the mechanism of ebselen, a compound selected within the experimental of a library of more than thousand diverse compounds. Steady-state kinetic studies and spectrofluorimetric analysis showed that ebselen is able to inhibit PfFNR with a two-phases mechanism consisting of a rapid enzyme inactivation followed by a slow FAD release. The inactivation process resulted to be irreversible without any formation of an initial reversible enzyme-inhibitor complex. Moreover the inactivation constant is higher at pH 8.2 rather than at pH 7 consistently with a covalent modification of PfFNR sulphydrylic groups. I also found that FAD release in PfFNR-C99A-C284S is markedly slowed down, suggesting a critical role of the active-site Cys284 in ebselen inhibition mechanism. Nevertheless the mild effect of mutation Cys284Ser on inactivation process suggests that FAD release and inactivation processes involve the modification of different Cys residues. The need of large amount of purified enzyme for in-depth functional and structural characterization and inhibition studies prompted us to find a new expression system for PfFNR, able to decrease the cost of protein purification. Thus, I developed an expression system for PfFNR-C99A based on the fusion with the yeast SUMO protein. The partial in vivo degradation of poly-His-SUMO-PfFNR-C99A prevented the obtainment of high expression levels. However, a procedure for the rapid and cheap isolation of the recombinant protein was developed, representing an attractive alternative to the previous protocol. The features of PfFNR highlighted by the studies here reported, point out that PfFNR could represent an attractive drug target, suitable for the development of novel antimalarial compounds.