A, structural changes of loop 113–119 occurring in the dimer (light blue model and yellow loop; Protein Data Bank code 2AAW) to tetramer (blue model and orange loop; Protein Data Bank code 1OKT) transition. Red spheres indicate the amino acids replaced in this study to obtain mutants A, B, and C. B, model of hemin·PfGST complex obtained by docking simulation using the crystal structure for Protein Data Bank code 1Q4J (15). Hemin is shown in red, loop 113–119 is in orange, and GSH is shown as yellow sticks.

GSH-induced tetramer/dimer transition. Inactive tetrameric wild-type PfGST (5 μm) obtained by 12 h of incubation in the absence of GSH was incubated with 1 mm GSH in 0.1 m potassium phosphate buffer, pH 6.5, at 25 °C. At fixed times, fluorescence anisotropy (r) (■) and enzymatic activity (▴) were monitored. As a control, changes of fluorescence anisotropy of mutant A (incubated for 12 h in the absence of GSH and then treated with 1 mm GSH) are also reported (▾). Mutants B and C (data not shown) display the same behavior as mutant A. The standard deviations of anisotropy values were calculated from nine replicate experiments.

Isothermic binding of hemin to wild-type and mutant PfGSTs as determined by fluorescence quenching. Hemin was added to PfGSTs (1 μm) in the presence or absence of GSH or GSSG in 0.1 m potassium phosphate buffer, pH 6.5, 25 °C. Quenching of the intrinsic fluorescence was measured as described under “Experimental Procedures,” and the data were analyzed by Equations 3 and 4. The continuous lines represent global fits to the high affinity and low affinity data, respectively. WT, wild type.

Tetramerization/inactivation of native and mutated PfGSTs in the absence of GSH. Active dimeric PfGSTs (with 10 mm GSH) were passed through a Sephadex G-25 column. After GSH removal, the enzymes were incubated at 25 °C at a concentration of ∼1 mg/ml. A, enzymatic activity, assayed at different incubation times (25 °C) after GSH removal. Activity was assayed as described under “Experimental Procedures.” B, size exclusion chromatography performed on wild-type PfGST and mutant A in the presence of 10 mm GSH (upper curves) and immediately after GSH removal by a Sephadex G-25 column (lower curves). Very similar results have been found with mutant B and mutant C (data not shown).

Effects of GSSG on PfGSTs. ■, active native PfGST (2 mg) was passed through a Sephadex G-25 column saturated with 1 mm GSSG in 10 mm potassium phosphate buffer, pH 7.2, at 25 °C. The specific activity of enzyme was measured at different times at 25 °C. ●, inactivated PfGST tetramer (2 mg) was incubated at 25 °C with 10 mm GSSG. Inset, Lineweaver-Burk plot of 1/v versus 1/[GSH] obtained by adding variable amounts of GSH (from 0.1 to 4 mm) to the native PfGST in the presence of fixed GSSG concentrations (♦, 0.53 mm; ▴, 0.23 mm; and ▾, 0.03 mm) in 1 ml of 0.1 m potassium phosphate buffer, pH 6.5 (25 °C).

PfGST is able to bind hemin only when the enzyme is in the active dimeric structure, i.e. in complex with GSH (15). Because of the fast tetramerization of the enzyme in the absence of GSH, it remains uncertain whether the hemin binding is prevented by steric hindrance caused by the tetrameric structure or by the absence of GSH in the G-site. The particular inertness of the mutated enzymes makes it possible to clarify this question. Binding of hemin has been visualized on the basis of the quenching of intrinsic fluorescence. The results, reported in Table 3 and Fig. 5, indicate that all mutants are able to bind hemin in the absence of GSH. However, the affinity is lower than that shown by the GSH·enzyme complex. It appears that all of these mutants display two different states characterized by two different affinities for hemin, one high affinity conformation, R-state (relaxed state), and one low affinity conformation, T-state (tensed state), determined by the presence or absence of GSH. The ability of loop 113–119 to adopt different conformations is possibly linked to this high-low affinity transition for hemin binding. Because Asn-112 is involved in the hemin stabilization through coordination to the iron atom (15), it is not surprising that any structural change of loop 113–119 may cause such relevant effects for the hemin binding. The model for hemin-associated PfGST reported in Fig. 1B (15) shows the close proximity of loop 113–119 with the bound hemin that also contacts the adjacent subunit. These contacts could represent the structural basis for cooperativity. Importantly, we also observed that the native enzyme is able to bind hemin also in the presence of GSSG. In this case the binding is noncooperative, and the dissociation constant indicates a low affinity conformation (Table 3 and Fig. 5). This behavior parallels what was observed in the presence of S-methylglutathione (15) and indicates that the free thiol group in the active site is important for the cooperative mechanism and to shift the enzyme toward a high affinity state. The ability of PfGST to bind hemin in the presence of GSSG may be of physiological relevance because this enzyme would protect the cell from hemin even under severe oxidative stress conditions. The model in Scheme 1 is consistent with all of the results described above.