School of Biosciences (Microbiology, Main Building), Cardiff University of Wales, PO Box 915,
Cardiff, CF1 3TL, Wales UK
*ML Laboratories PLC, 60 London Road, St. Albans, AL1 1NG, UK

†Corresponding author (Fax, 44 1222 874305; Email, lloydd@cardiff.ac.uk).

The amitochondriate sexually-transmitted human parasitic protozoan Trichomonas vaginalis (Bushby strain) grown anaerobically on complex medium containing cysteine and ascorbic acid consumed O₂ avidly (6·9 µM
min^–1 per 10⁶ organisms) with an apparent K_m value of 5·1 µM O₂ : O₂ uptake was inhibited by O₂ > 120 µM. Spectrophotometric assays in the presence of microperoxidase (419–407 nm) indicated that H₂O₂ was produced and that inhibition by high O₂ concentrations was again evident. Hydrogenosomes oxidizing pyruvate in the presence of ADP and succinate showed similar patterns of O₂ consumption, H₂O₂ production (33·5 pmol min^–1 per mg protein), and O₂ inhibition. Cytosolic NADH oxidase gave no detectable H₂O₂, whereas the cytosolic NADPH oxidase produced H₂O₂ at a rate (43 pmol min^–1 per mg protein) greater than that of hydrogenosomes. These results are discussed in relation to the oxidative stress experienced by the pathogen in its natural habitat.

An understanding of the oxidative metabolism of the amitochondriate flagellated parasitic protozoan, Trichomonas vaginalis, an extremely common cause of human genito–urinary infections, is paramount in devising strategies for chemotherapy (Milne et al 1978; Lloyd and Petersen 1985). The redox-active organelles of trichomonads, the hydrogenosomes (Lindmark and Müller 1973) are responsible for the oxidation of pyruvate (Cerkasov et al 1978). Under anaerobic conditions hydrogenase is active and protons serve as the terminal electron acceptor for a chain of flavin and iron sulphur carriers (Lindmark and Müller 1973; Chapman et al 1986). When exposed to low O₂ concentrations (e.g., 20 µM O₂) the carbon fluxes through the alternative branching pathways from pyruvate are adjusted (Ellis et al 1992), hydrogenase activity becomes inhibited (Lloyd and Kristensen 1985), and O₂ serves as terminal electron acceptor (Lloyd et al 1983). NADH and NADPH oxidation also occurs in hydrogenosome-enriched fractions (Lindmark and Müller 1973; Lloyd et al 1983) with a variety of electron acceptors including O₂. T. vaginalis also possesses a highly active cytosolic NADH oxidase (Tanabe 1979; Brown et al 1996) and an NADPH: acceptor oxidoreductase (EC 1.6.99.1) which can use O₂ as electron acceptor (Linstead 1983; Linstead and Bradley 1987). As the vagina is not a strictly anaerobic habitat, but rather may have regions where O₂ can occasionally reach almost 60 µM (Wagner and Levin 1978), it is important to investigate O₂ scavenging as a mechanism of protection against O₂ toxicity (Ellis et al 1994) in this microaerophilic protozoan (Paget and Lloyd 1990). In this report we have determined the oxygen affinities of the major oxidases in the Bushby strain, and shown that H₂O₂ is a minor product, both in the hydrogenosomal and cytosolic compartments.

2. Methods

2.1 Growth of the organisms and preparation of subcellular fractions

T. vaginalis (Bushby strain) was grown with a logarithmic phase doubling time of 3·6 h in modified Diamond’s medium as described previously (Chapman et al 1986). Disruption was with a motor-driven Potter–Elvejhem homogenizer with a Teflon pestle rotating at about 800 rev min^–1 in a buffer containing 1 mM EDTA (Cerkasov et al 1978). The hydrogenosomal-enriched particulate fraction was obtained by centrifugation at 2·500 g_av for 15 min. The cytosolic fraction was prepared by centrifuging cell-free extracts at 94,000 g_av for 1 h.

2.2 Dependence of O₂ uptake on O₂ concentration

This was determined in the specially-constructed open O₂ electrode system (Lloyd and Kristensen 1985; Yarlett
et al 1986) with windows for simultaneous spectrophotometric measurements. Argon (< 3 ppm O₂) was used to obtain anaerobiosis of the liquid in the vessel. One hundred per cent O₂ air and a standard gas mixture of 5% O₂ in N₂ was also used (Air Products, Cardiff, UK). Various O₂ tensions were obtained by the use of a digital gas mixer to provide O₂ dilutions in 5% steps. Apparent K_m O₂ values were obtained at low O₂ concentrations
(0–50 µM). The evolving gas mixture (150 ml/min) was humidified by passage through water-soaked cotton wool before passage over the liquid vortex in the vessel. O₂ concentrations were calculated from percentage dilutions of stock gas mixtures.

Calibration for dissolved O₂ and determination of t_1/2 (half times for equilibration) were obtained using standard gas mixtures in the absence of biological material. Values for the O₂ concentration in the liquid phase (T_L) in the presence of respiring cells or subcellular fractions at a series of fixed values of O₂ concentration in the gas
phase (T_G) then allowed calculation of respiration rates (from V_r = k(T_G – T_L) where V_r = respiration rate and k = log_e 2/t_1/2).

2.3 Production of H₂O₂ by T. vaginalis cells and subcellular fractions

Microperoxidase (MPII, Sigma) was used (3·0 µg/ml final concn.) for the detection of H₂O₂. The absorbance maxima of the free enzyme and the peroxide complex are at 407 nm and 419 nm respectively. The rate of H₂O₂ production was determined by the rate of change of absorption at 419–407 nm in the dual wavelength mode (molar absorption coefficient 78000 M^–1 cm^–1).

3. Results

Figure 1a shows a double reciprocal (Lineweaver–Burk) plot of the glucose-supported dependence of O₂ consumption rates on O₂ tension by intact T. vaginalis cells. In a series of experiments at low O₂ concentrations (0–50 µM), the apparent K_m value was determined to be 5·12 ± 1·35 µM (SD from 3 experiments). The respiration of whole cells was inhibited by high oxygen concentrations (> 120 µM O₂). The extrapolated value for V_max of O₂ consumption was found to be 6·9 µM O₂/min/10⁶ cells. H₂O₂ production by intact cells revealed a similar pattern of inhibition by high O₂ concentrations in the liquid phase (figure 1b). Thus both O₂ consumption rate and rate of H₂O₂ production increased as the O₂ concentration in solution was raised to a value of about 150 µM, above which, the cellular O₂ uptake and rate of H₂O₂ production decreased. The ratio of H₂O₂ produced to O₂ consumed increased as the O₂ concentration increased up to 250 µM (figure 1c).

A similar pattern of O₂ uptake, H₂O₂ production, and inhibition was observed for hydrogenosomes oxidizing pyruvate in the presence of ADP and succinate, as for intact cells. Figure 2a shows the Lineweaver–Burk plot of hydrogenosomal O₂ consumption at low O₂ concentrations with an apparent K_m for O₂ of 11·2 µM. A value of 11·0 ± 3·2 µM (SD from 3 experiments) was obtained in a series of experiments. Inhibition of hydrogenosomal O₂ consumption occurred at an oxygen concentration of 110 µM, with an extrapolated V_max of 79·6 µM O₂/min/mg protein. The rate of production of H₂O₂ (figure 2b) increased up to an O₂ tension of 30 µM to a rate of 33·5 pmol/min/mg protein, above this O₂-concentration, production rapidly fell, to a rate of 11 pmol/min/mg protein. At 30 µM O₂, most of the O₂ consumed was converted to H₂O₂, but at high O₂ concentrations, the ratio of H₂O₂ production to O₂ consumption progressively decreased. Almost 14·4% of the O₂ consumed was converted to H₂O₂ at an O₂ concentration of 150 µM compared with 65% at a O₂ concentration of 30 µM (figure 2c).

The double reciprocal plot obtained for O₂ consumption rate against O₂ concentration by cytosolic NADPH oxidase (NADPH: flavin oxidoreductase) is shown in figure 3a. The apparent K_m for O₂ was determined after a series of experiments to be 27·2 ± 5·1 µM (SD from 3 experiments). The V_max was determined to be 33·1 µM O₂/min/mg protein. Cytosolic NADPH oxidase was shown to produce H₂O₂ (figure 3b). The rate of H₂O₂ produced (43·5 pmol/min/mg protein) was higher for the NADPH oxidase system than the hydrogenosomal O₂ uptake system. This production was decreased at an O₂ concentration greater than 90 µM: the ratio of H₂O₂ production to O₂ consumption at low O₂ concentrations was twice that of whole cells (figure 3c).

No increase in the rate of H₂O₂ production was observed when superoxide dismutase (SOD) (20 EU/ml) was added to the reaction vessel containing intact cells or subcellular fractions. The addition of catalase (5 EU/ml) prevented any further H₂O₂ production; i.e., no increase in absorbance of the microperoxidase complex was observed at 419–407 nm after its addition.No H₂O₂ could be detected when the cytosolic fraction was incubated aerobically with NADH.

4. Discussion

The microaerophilic parasite, T. vaginalis (Paget and Lloyd 1990) is finely tuned so as to be able to utilize the very low (< 0·25 µM) levels of dissolved O₂ that accelerate hydrogenosomal acetate production (Ellis et al 1992) and hence substrate-level phosphorylation.

Presumably low levels of O₂ occur in microenvironments at the mucosal surface of the vagina. Increased blood flow can lead to more oxygenated conditions (Wagner and Levin 1978) and there may be other situations where elevated ambient O₂ can increase oxidative stress in the protozoon. We have previously shown that inhibition of O₂ consumption occurs when a threshold of O₂ concentration is exceeded, and that the critical level of dissolved O₂ varies from one strain of T. vaginalis to another (Yarlett et al 1986; Ellis et al 1994). One of these studies (Ellis et al 1994) confirmed earlier reports of the presence of superoxide dismutase activity in this (Linstead and Bradley 1988) and other anaerobic flagellates (Lindmark and Müller 1974). Other O₂-defense mechanisms include high levels of intracellular thiols especially methanethiol, propanethiol and H₂S (Ellis et al 1994). Therefore the sensitivity of T. vaginalis to O₂ at above physiological levels comes about from the production of H₂O₂, with the consequent likelihood of the generation of other reactive oxygen species in an organism where no catalase can be detected (Honigberg 1978).

The present work shows correlation of O₂ inhibition of O₂ uptake with H₂O₂ production both at the whole cell level as well as in cytosolic and hydrogenosomal-enriched fractions. Similar dependencies of both O₂ consumption and H₂O₂ generation rates suggests that this product of univalent reduction of the terminal electron acceptor arises at the cytosolic NADPH oxidase as well as at hydrogenosomal oxidases responsible for accepting electrons from pyruvate and NADPH.

The apparent K_m for O₂ of an organism is an indication of its ability to scavenge oxygen from its surroundings both for useful and protective functions. Its measured (apparent) value in intact cells depends upon (i) the nature of the terminal oxidase(s) (especially its affinity for O₂), and (ii) in large cells, the existence of O₂ gradients between the plasma membrane and sites of O₂ reduction. The ability of intact cells of T. vaginalis to remove O₂ from their surroundings is similar to that determined for the cattle parasite Tritrichomonas foetus KV1. Similar apparent K_m values for O₂ have been obtained for the hydrogenosome-containing anaerobic rumen holotrich Dasytricha ruminantium, and also for the free-living mitochondriate protozoan, Tetrahymena pyriformis (Lloyd et al 1982).

The apparent K_m value for O₂ obtained for the isolated hydrogenosomal or cytosolic fractions indicate that these subcellular preparations do not have such high affinities for O₂ as those exhibited by whole cells. The higher values of K_m for O₂ of both fractions may reflect either (i) the impairment of the activities of the fractions
during processing and prolonged exposure to O₂ during the experiments, or (ii) loss of normal interaction between the hydrogenosome and the cytoplasm. The V_max values for O₂ consumption by hydrogenosomal or cytoplasmic NADPH oxidase obtained in the open system do not agree exactly with the values obtained from the rates of O₂ uptake recorded immediately after subcellular fractionation using the closed oxygen electrode. This is possible due to the extended times (about 2 h) in which the subcellular fractions are subjected to varying O₂ tensions in the open system; this exposure to O₂ may partially inactivate the enzyme system(s) acting as terminal oxidases.

From the results presented here, it is not possible to distinguish between reduced flavin (Massey et al 1971), an iron–sulphur centre (Misra and Fridovitch 1971), or superoxide as the source of hydrogenosomal or
cytoplasmic H₂O₂ production. The decrease in the rate of H₂O₂ production observed at high O₂ concentrations can have several alternative explanations: (i) at high O₂ concentrations, the terminal oxidase is inhibited; this results in decreased H₂O₂ production; (ii) H₂O₂ itself becomes inhibitory to the terminal oxidase at high concentrations, i.e., product inhibition of the oxygen-utilizing systems of either the oxidase itself or an intermediate redox carrier by H₂O₂ may occur; (iii) oxidative damage to the electron transport systems occurs as a result of H₂O₂ production. The primary reactant responsible for damage to the terminal oxidase systems may thus be oxygen itself, a product of its partial reduction such as H₂O₂, or a free radical such as superoxide or hydroxyl species. Because T. vaginalis contains no catalase, partial reduction of oxygen must result in a build-up of H₂O₂ capable of leading to irrepairable damage to sensitive cellular components including redox components of the electron transport chain.

Inhibition of H₂O₂ production by intact cells, hydrogenosomes and NADPH oxidase occurred at different oxygen concentrations. H₂O₂ production in hydrogenosomes was inhibited at a lower oxygen concentration than the overall oxygen consumption; this could indicate that H₂O₂ is not only produced by the terminal oxidase but also by another electron transfer component in the hydrogenosome. If superoxide radicals are produced by reduction of oxygen either at the terminal oxidase or elsewhere in the respiratory chain, then exogenous SOD added to the respiring system might not produce any effect if endogenous SOD activities are high. Indeed, addition of SOD to the reaction vessel did not increase the rate of change of absorbance when measuring H₂O₂ production by formation of the microperoxidase-H₂O₂ complex using intact cells or with either of the subcellular O₂ consuming systems.

AC held a postgraduate studentship provided by the Science and Engineering Research Council.

References

MS received 8 October 1998; accepted 20 July 1999