Catalases are NAD(P)H-dependent tellurite reductases

Abstract

Reactive oxygen species damage intracellular targets and are implicated in cancer, genetic disease, mutagenesis, and aging. Catalases are among the key enzymatic defenses against one of the most physiologically abundant reactive oxygen species, hydrogen peroxide. The well-studied, heme-dependent catalases accelerate the rate of the dismutation of peroxide to molecular oxygen and water with near kinetic perfection. Many catalases also bind the cofactors NADPH and NADH tenaciously, but, surprisingly, NAD(P)H is not required for their dismutase activity. Although NAD(P)H protects bovine catalase against oxidative damage by its peroxide substrate, the catalytic role of the nicotinamide cofactor in the function of this enzyme has remained a biochemical mystery to date. Anions formed by heavy metal oxides are among the most highly reactive, natural oxidizing agents. Here, we show that a natural isolate of Staphylococcus epidermidis resistant to tellurite detoxifies this anion thanks to a novel activity of its catalase, and that a subset of both bacterial and mammalian catalases carry out the NAD(P)H-dependent reduction of soluble tellurite ion (TeO(3)(2-)) to the less toxic, insoluble metal, tellurium (Te(o)), in vitro. An Escherichia coli mutant defective in the KatG catalase/peroxidase is sensitive to tellurite, and expression of the S. epidermidis catalase gene in a heterologous E. coli host confers increased resistance to tellurite as well as to hydrogen peroxide in vivo, arguing that S. epidermidis catalase provides a physiological line of defense against both of these strong oxidizing agents. Kinetic studies reveal that bovine catalase reduces tellurite with a low Michaelis-Menten constant, a result suggesting that tellurite is among the natural substrates of this enzyme. The reduction of tellurite by bovine catalase occurs at the expense of producing the highly reactive superoxide radical.

Figure 1. The tellurite reductase activity in crude extracts prepared from S. epidermidis CH is dependent on NADH as cofactor.

The tellurite reductase activity present in crude extracts of S. epidermidis CH cells was followed over time as the increase in absorbance at 500 nm due to the conversion of tellurite to tellurium in the presence (▪) and absence (•) of NADH under standard assay conditions. The values shown represent the means of three independent determinations made with crude extracts containing 1.2 mg/ml total protein.

Figure 2. The product of tellurite reduction by catalase is elemental tellurium.

Absorption spectra of products formed upon the reduction of K₂TeO₃ in vitro by (A) 2-mercaptoethanol, (B) a crude extract prepared from S. epidermidis CH cells prior to chromatographic enrichment, and (C) purified bovine liver catalase, were resolved by Induced Coupled Plasma-Optical Emission spectroscopy . All products show an absorption maximum at 214.281 nm, the peak wavelength of the Te° standard (triangle in panel A). We note that the scale used to measure absorbance in the crude extract differs by a factor of ten from those used to measure absorbance in chemically prepared tellurium (A) and in the product of tellurite reduction by bovine liver catalase (C). This is because the crude extract includes a plethora of components with absorption maxima at or near this wavelength.

Figure 3. Catalases have tellurite reductase activity in situ.

Assays for the H₂O₂ dismutase and TeO₃ ²⁻ reductase activities of catalase in situ were carried out by resolving proteins on polyacrylamide gels, incubating gel strips in the presence of substrate, and developing gel strips to reveal the presence of specific products of each reaction (Woodbury et al. 1971; Gregory and Fridovich 1974). For dismutase assays, 25 µg of protein were loaded in each gel lane; 250 µg of protein was used for tellurite reductase assays. (A) The results in Fig. 3A show that bovine liver catalase (a tetramer of 255 kDa) migrates as a single band on native gels (lane 2) with an apparent molecular mass similar to that of a 244 kD standard (lane 1), and has both hydrogen peroxide dismutase (lane 3) and tellurite reductase (lane 4) activities. (B) The activities of catalase present in extracts made from E. coli Top10 cells carrying the plasmid vector pBAD display lower H₂O₂ dismutase and TeO₃ ²⁻ reductase activities (lanes 7 and 9) than does an otherwise isogenic strain with plasmid pCAT, which expresses the S. epidermidis CH katA (catalase) gene (lanes 8 and 10); these activities most likely correspond to the predominant tetrameric form of catalase. In crude extracts of S. epidermidis, two different bands of each activity are observed (lanes 5 and 6).

Figure 4. Figure 4. Expression of the S. epidermidis katA gene in E. coli confers increased resistance to both tellurite and hydrogen peroxide.

(A) Aliquots of exponentially growing cultures of E. coli strain JWK3914 (katG) carrying plasmids pBAD (a and c) or pCAT (b and d) were spread onto the surface of plates with LB medium, 15.5 µl K₂TeO₃ (4 mM)(a and b) or hydrogen peroxide (3%) (c and d) were spotted onto the centers of the bacterial lawns, and cells were grown to reveal zones of inhibition (Fuentes et al. 2005). (B) The mean areas of growth inhibition zones (cm²) were determined from three independent experiments; thin bars represent the standard deviations.

Figure 5. The reduction of tellurite by bovine liver catalase produces superoxide anion and obeys simple Michaelis-Menten kinetics.

In these experiments, we measure the rate of evolution of superoxide radical as the product of the tellurite reductase reaction mediated by catalase, by the ability of superoxide to reduce the tetrazolium derivative WST-1. (A) The rates of change in absorbance at 438 nm in standard reactions with 5 mM K₂TeO₃ were measured. Reactions with tellurite (◂;complete) show a higher rate of WST-1 reduction than do reactions without substrate (•; - K₂TeO₃). Without substrate, WST-1 is reduced at a lower, background rate by NADPH. When superoxide dismutase is added to the reaction (▴; +SOD), only the lower rate of reduction is observed; in contrast, the addition of β-amylase (▪; +AMY) has little effect on the rate of reduction, demonstrating that the most abundant reductant of WST-1 in these assays is superoxide radical. (B) The reciprocals of the initial velocities of tellurite reduction, measured as the changes in absorbance at 438 nm min⁻¹ due to the coupled reduction of WST-1, plotted versus the reciprocals of the concentrations of the substrate (S), potassium tellurite, (mM) for reactions carried out under standard conditions (see Materials and Methods for details).