CO₂, not , facilitates oxidations by Cu,Zn superoxide dismutase plus H₂O₂

Abstract

The Cu,Zn superoxide dismutase catalyzes -dependent oxidations by H₂O₂. This activity has been shown to depend on the creation of a bound oxidant at the Cu(II) by interactions with H₂O₂. The bound oxidant was then thought to oxidize to , which diffuses into the bulk solution and there oxidizes diverse substrates. We now find that CO₂ rather than facilitates the peroxidations catalyzed by Cu,Zn superoxide dismutase. This fact was shown by a lag in the rate of peroxidation of NADPH when was added last and by a burst in the rate when aqueous CO₂ was added last. Both the lag and the burst were eliminated by carbonic anhydrase.

dramatically increases the rate of oxidation of diverse substrates by Cu,Zn superoxide dismutase (SOD1) plus H₂O₂, while only modestly increasing the intrinsically slow inactivation of this enzyme by H₂O₂ (1–4). The following reactions have been advanced as part of a scheme to explain the behavior of this system (1).

Thus, the cuprous enzyme produced in reaction i is oxidized in reaction ii by a second H₂O₂ to yield a strong bound oxidant, which can be written as Cu(I)O, Cu(II)OH, or Cu(III). This bound oxidant can either oxidize an adjacent histidine residue and in so doing inactivate the enzyme, or it can oxidize to the carbonate radical that is free to diffuse into the bulk solution and cause further oxidations. Before leaving the active site, the carbonate radical can oxidize residues in the ligand field of the copper and thus cause inactivation.

was assumed to be the species in the carbonate buffer that facilitated the oxidations observed. This was the case because, at neutral to mildly alkaline pH, it is the major species. It also seemed reasonable that an active site evolved to deal with the monovalent anion would preferentially react with other monovalent anions. Nevertheless, the experiments described below establish that at pH 7.4, it is CO₂ rather than that profoundly facilitates the oxidation of NADPH by SOD1 plus H₂O₂.

Materials and Methods

SOD1 was from Chemie Grünenthal (Aachen, Germany); H₂O₂, from Mallinckrodt; and NaHCO₃, from Fisher; whereas NADPH and carbonic anhydrase were from Sigma. NADPH at 0.1 mM was reacted with 0.1 or 0.2 mg/ml SOD1 and 10 mM H₂O₂ in 100 mM potassium phosphate at pH 7.4. When was used to facilitate the oxidation of NADPH (followed at 340 nm), it was added to 20 mM from a 1.0 M stock of NaHCO₃ in water. When CO₂ was used, it was taken from ice-cold water saturated with CO₂ gas at 1.0 atmosphere (atm) (1 atm = 101.3 kPa) that would contain 76 mM CO₂. Carbonic anhydrase when used was present at 0.1 mg/ml. Additional experimental details are presented in the legends to Figs. 1 and 2.

Fig. 1.

requires dehydration before facilitating. Shown is the oxidation of NADPH by SOD1 plus H₂O₂. The final reaction mixtures contained 0.1 mM NADPH, 0.1 mg/ml SOD1, 20 mM , and 10 mM H₂O₂, plus 100 mM sodium phosphate at pH 7.4 and at 23°C. Line 1, SOD or H₂O₂ added as the last component at the arrow; line 2, NaHCO₃ added last at the arrow; line 3, as in line 2, except that 0.1 mg/ml carbonic anhydrase was present in the reaction mixture. Note the slow oxidation before the addition of .

Fig. 2.

CO₂ added as such supports the peroxidation activity of SOD1. Reaction conditions shown are essentially as in Fig. 1, except that water at 0°C saturated with CO₂ gas was the last component added in place of . Line 1, CO₂ added at the arrow to reaction mixture as in Fig. 1; line 2, CO₂ added at the arrow to a reaction mixture as in Fig. 1 but containing 0.2 mg/ml SOD1; line 3, as in line 2 but with 0.1 mg/ml carbonic anhydrase in the reaction mixture.

Results

Line 1 in Fig. 1 demonstrates an immediately linear rate of NADPH oxidation that was seen when the HCO₃-dependent oxidation by SOD1 plus H₂O₂ reaction was started by adding either SOD1 or H₂O₂. However, when NaHCO₃ was the last component added, there was a distinct lag before the linear rate was achieved, shown by line 2. Because NaHCO₃ was added from an alkaline stock solution into a reaction mixture buffered at pH 7.4, it seemed possible that this lag was due to the perceptibly slow dehydration of H₂CO₃ to CO₂. That this was the case was shown by the effect of carbonic anhydrase (line 3), which eliminated this lag. These results indicated that CO₂ rather than was the essential reactant.

In that case, it could be predicted that starting the reaction with CO₂-saturated water would yield an initially fast reaction that would slow to a linear rate as the hydration and ionization to reached an equilibrium. Moreover, carbonic anhydrase should eliminate the burst seen when CO₂ was added last, because it had eliminated the lag when was added last. Fig. 2 demonstrates that these predictions were correct. Thus line 1 was obtained when 150 μl of CO₂ solution was added to a 3-ml reaction mixture otherwise identical to that used in Fig. 1. The very evident initial burst was made even more pronounced when the concentration of SOD1 was doubled (line 2). Finally, carbonic anhydrase eliminated the initial burst (line 3). It should be emphasized that the initial CO₂ concentration in the reaction mixture would have been 3.8 mM, substantially lower than the concentration of (20 mM) used in Fig. 1, yet the initial rate was greater than that seen with .

Discussion

These results establish that, at pH 7.4, the peroxidase activity of SOD1 depends on CO₂ rather than on . The methods used to demonstrate this were virtually identical to those Cooper et al. (5) used to show that CO₂ is the substrate for phosphoenol pyruvate carboxy kinase and carboxy transphosphorylase. In the case of the peroxidase activity of SOD1, we can envision two routes for the oxidation of CO₂ to by the bound oxidant. Thus,

and then

Elam et al. (6) have proposed still another route to ; thus,

However, the rate of formation of from CO₂ plus is slow (7), yet we saw the same initial rate of oxidation of NADPH when the and H₂O₂ were preincubated before the addition of SOD1 as we did when and SOD1 were preincubated before the addition of H₂O₂, rendering the mechanism defined by reactions viii and ix less likely.

CO₂ is a linear molecule that exists partially as a zwitterion, i.e., –O—CO⁺ (8). This structure is also supported by its considerable solubility in water (9), which results in 76 mM CO₂ in water equilibrated with 1-atmosphere CO₂ at 0°C compared with only 2.18 mM for O₂ under the same conditions. Perhaps the entry of CO₂ into the active site of SOD1 is facilitated by this zwitterionic form.

Carbonic anhydrase contains Zn(II) at its active site, and SOD1 contains Zn(II) joined to Cu(II) by the imidazolate moiety of a histidine residue. Here, it is possible that the first point of interaction of CO₂ with SOD1 is at the Zn(II) rather than at the Cu center. The proximity of the Zn and Cu would allow CO₂ interacting with the Zn to be oxidized by the Cu(II)—OH. While considering carbonic anhydrases, it is reasonable to ask whether this ubiquitous family of enzymes might play a defensive role by catalyzing the hydration of CO₂, which can be oxidized to , in contrast to , which cannot. In support of this supposition is the report by Götz et al. (10) that deletion of the putative carbonic anhydrase from Saccharomyces cerevisiae caused an imposed oxygen sensitivity. Although it is certainly CO₂ that is oxidized to by SOD1 plus H₂O₂ at pH 7.4, there may be entities in cells that can similarly oxidize . In fact, it cannot be excluded that, at high pH, SOD1 plus H₂O₂ may be able to oxidize or .

The results and discussion presented above preclude some hypotheses presented earlier, including the notions that facilitates oxidations by SOD1 plus H₂O₂ by creating a binding site for H₂O₂ (4) and that specifically binds in the solvent access channel of SOD1 and, while so bound, becomes oxidized, at which time the resulting protrudes sufficiently to oxidize substrates in the bulk solution (6).