In the anaerobic protozoon <i>Trichomonas vaginalis</i>, the oxidative decarboxylation of pyruvate is catalysed in a CoA-dependent reaction by pyruvate: ferredoxin oxidoreductase (PFOR). This enzyme has been identified as a potential target for the development of a relatively non-toxic anti-trichomonal agent. 1. <i>T. vaginalis</i> PFOR was localised in the hydrogenosomal membrane fraction, and could be solubilised by buffer of high ionic strength. A high salt concentration was required to prevent aggregation of PFOR. These results suggested that PFOR was either an extrinsic protein bound to the hydrogenosomal membrane or that the enzyme exists <i>in vivo</i> in the hydrogenosomal matrix in an aggregated state. PFOR was solubilised and purified to homogeneity, the most _ffective step being salting-out chromatography on Sepharose 4B. Low recoveries of active enzyme were caused by inactivation by oxygen and the irreversible loss of thiamin pyrophosphate (TPP). 2. PFOR is a dimeric enzyme of overall M<SUB>r</SUB>240000. The enzyme contains 0.5 mol of TPP per mol of dimer, and equivalent amounts of non-haem iron and acid-labile sulphur, consistent with the presence of two [4Fe-4S] centres per enzyme molecule. Flavin nucleotides and lipoic acid are absent. PFOR from <i>T. vaginalis</i> is therefore broadly similar to the 2-oxo acid:ferredoxin (flavodoxin) oxidoreductases purified from bacterial sources, and clearly different from the 2-oxo acid dehydrogenase multienzyme complexes which occur in aerobic organisms. 3. A steady-state kinetic analysis of purified PFOR demonstrated that the enzyme obeyed Bi Bi Ping Pong kinetics except at very high CoA concentrations, where substrate inhibition occurred. The inhibition produced by the product of acetyl-CoA, in the presence of saturating CoA, was competitive with respect to pyruvate. In the absence of CoA, stoichiometric amounts of pyruvate were decarboxylated by PFOR. These results suggest that decarboxylation, formation of the stable imtermediate and its reaction with CoA to form acetyl-CoA all take place at one active site. 4. Spectroscopic investigations using electron paramagnetic resonance indicated that the stable intermediate formed after pyruvate decarboxylation was a free-radical species. This substrate-based radical is proposed to arise by the transfer of a single electron from the initial decarboxylation product to a [4Fe-4S] centre. The free-radical signal was greatly diminished if the enzyme was subsequently incubated with CoA, suggesting that it represents a real catalytic intermediate. 5. <i>T. vaginalis</i> PFOR was inactivated by incubation with pyruvate alone, a reaction the enzyme has in common with the <i>E. coli</i> pyruvate dehydrogenase (PDH) complex and yeast pyruvate decarboxylase, suggesting similarities between these enzymes at least in the initial formation of the decarboxylated intermediate, presumed to be the enamine of hydroxyethyl-TPP. The conjugated 2-oxo acid, (E)-4-(-chorophenyl)-2-oxo-3-butenoic acid, was an irreversible inhibitor of <i>T. vaginalis</i> PFOR and yeast pyruvate decarboxylase, a result taken to reflect the initial formation of an enamine intermediate in each case. 6. 3-hydroxypyruvate was a potent irreversible inhibitor of <i>T. vaginalis</i> PFOR. The observation that 3-hydroxypyruvate was also an alternative substrate for pyruvate in the overall reaction suggested that it might be acting as a mechanism-based inactivator. 3-hydroxypyruvate was ineffective against <i>E. coli</i> PDH complex suggesting an interesting difference in active site geometry that might be exploited for potential drug design.