The naphthothiophenquinone TNQ2, available from our previous work (Zani et al. 1997), was re-crystallised from methanol-ethyl acetate to more than 99.5% purity, as assessed by HPLC analysis. A stock solution at 100 mM was prepared in DMSO and appropriate dilutions were prepared in the assay buffer.

We initially checked that the incubation time of TNQ2 with TR does not affect its activity (data not shown) ensuring that the inhibition is not time dependent. Initial rates (v₀) were then measured after 5 min incubation before the addition of the trypanothione with vigorous mixing. The decrease in the absorbance at 340 nm was measured and initial rates determined using the first 10-20 sec of data, in such way that the correlation coefficient to a straight line was greater than 0.95. Non-linear regression of the raw data was performed using GraFit to calculate the kinetic parameters K_m and K_i (Table). The graphical results are presented in Fig. 1 as Lineweaver-Burk plots, showing a clear simple linear non-competitive profile with respect to both NADPH and the substrate trypanothione.

TNQ2 was also tested on hGR, the enzyme with a similar function in the mammalian host. At 100 µM the activity of this enzyme was reduced by only 15% (data not shown), indicating a much lower activity against the host enzyme and thus a good degree of selectivity.

In the case of a non-competitive inhibitor, the relative activity of TR can be calculated as follows:

where v_i and v₀ are the inhibited and uninhibited rates for trypanothione reductase, respectively, K_m is the Michaelis constant, [S] and [I] are concentrations of substrate and inhibitor, respectively and K_i the inhibitor constant. When v_i/v₀ = 0.1, it can be seen by rearranging the above equation that the concentration of inhibitor required for 90% inhibition ([I]₉₀) is equal to 9K_i. Similar equations can be derived for competitive and uncompetitive inhibition (see Fig. 2). Since intracellular levels of reduced trypanothione are ~1 mM in T. cruzi (Ariyanayagam & Fairlamb 1997) the behaviour of these types of reversible inhibitor is presented as a function of substrate concentration up to 0.5 mM trypanothione disulphide. Note that, unlike competitive and uncompetitive inhibition, the [I]₉₀ for non-competitive inhibition is independent of the substrate concentration. Thus, even though trypanothione disulphide levels would be expected to increase in cells due to the metabolic roles of trypanothione as an antioxidant and as an electron donor for reduction of ribonucleotides, the level of inhibition of trypanothione reductase would remain constant.

Data from our previous work (Zani et al. 1997) shows that TNQ2 inhibited growth of epimastigotes with an IC₅₀ of 14.3 µM. Assuming TNQ2 equilibrates across the membrane of the parasite, then, for [I] = 14.3 µM (50% inhibition of growth) and the K_i values given in the Table, TR would be inhibited by about 75%. In studies on conditional knockouts of trypanothione reductase in T. brucei, it was noted that growth completely ceased when enzyme levels fell to 95% of normal, but that lysis occurred only at even lower levels (Krieger et al. 2000) . Also, in L. donovani expressing a dominant-negative mutant form of trypanothione reductase, growth was unaffected by up to 85% reduction in activity, although sensitivity to oxidant stress was increased (Tovar et al. 1998). T. cruzi epimastigotes would appear to be slightly more sensitive than these organisms, bearing in mind the assumption made above about the intracellular levels of TNQ2 and assuming TNQ2 does not affect other cellular processes. This compound was only partially effective (27 ± 7% lysis) against the trypomastigote form of T. cruzi in blood at much higher concentrations (200 µM). This discrepancy could be due to factors such as: (a) the greatly reduced metabolic activity of T. cruzi at 4°C retarding the trypanocidal effect, (b) the reduced requirement for ribonucleotide reduction in non-dividing trypomastigote stages or (c) the difference in end point measurement (i.e. growth versus lysis). Since the most striking difference in the properties of TR and hGR is at their respective disulphide-binding sites, most compounds have been designed as competitive inhibitors. As discussed above, such inhibitors suffer from the disadvantage that their inhibitory effects are reversed by the accumulation of trypanothione disulphide. Irreversible inhibitors such as the antimonial drugs (Cunningham & Fairlamb 1995) or lunarine (Bond et al. 1999) or its analogues (Hamilton et al. 2003), or non-competitive inhibitors such as TNQ2 may offer a more fruitful approach. Work is in progress to identify the binding site for TNQ2 in order to increase potency and selectivity of this interesting lead.