Kinetic study by pulse radiolysis of the lactate dehydrogenase-catalyzed chain oxidation of nicotinamide adenine dinucleotide by HO2 and O2-RADICALS.

The lactate dehydrogenase-catalyzed chain oxidation of NADH (LDH-NADH) by the superoxide radicals, HO2 and O2, has been studied with pulse radiolysis in the pH range between 4.5 and 9.0. The rate constants for the oxidation of the LDH-NADH by HO2 and O2 determined at 23 degrees are 1.2 times 10-6 M(-1) s(-1) and 3.6 times 10-4 M(-1) s(-1), respectively. The latter represents an activation of over 1000-fold by the enzyme. A chain reaction mechanism consistent with the results from these kinetic studies has been proposed.

(LDH-NADH) by the superoxide radicals, HOz and 0~~, has been studied with pulse radiolysis in the pH range between 4.5 and 9.0. The rate constants for the oxidation of the LDH-NADH by HOz and 02-determined at 23" are 1.2 X IO6 I@ s-l and 3.6 X lo4 M-' s-l, respectively. The latter represents an activation of over lOOO-fold by the enzyme.
A chain reaction mechanism consistent with the results from these kinetic studies has been proposed.
Earlier reports (1, 2) have shown that lactate dehydrogenase (EC 1.1.1.27) catalyzes the superoxide radical-induced chain oxidation of reduced nicotinamide adenine dinucleotide in the presence of molecular oxygen. The superoxide radical necessary for the induction of the chain reaction can be generated either by 6oCo y rays or ,by the xanthine-Oz-xanthine oxidase system (3).
The purpose of the present investigation is to determine the rates of interaction between the lactate dehydrogenase-bound NADH and the superoxide radicals (which refer to 0~~ and HO2 in equilibrium).
Because of the fast reaction rates, the technique of pulse radiolysis was chosen to generate superoxide radicals in an oxygenated sodium formate solution.
It has been established that although formate protects lactate dehydrogenase against radiation damages (4)) it does not affect the catalytic activity of the enzyme on NADH oxidation by pyruvate (1). The molecular oxygen plays a multiple role in this system; it serves as a precursor for the HO2 and 02-radicals, it protects the enzyme against reductive attack by the hydrated electron and atomic hydrogen (Reactions 1 and 2), and it also acts as a chain carrier (Reaction 9).
Based on the results from various studies (4-6), the over-all The numerical values in parentheses in Equation I represent the G values, the number of molecules formed or transformed per 100 e. v. of energy dissipated in the system.
Most conveniently, G values are computed from calibrations with the ferrous dosimeter, G(Fe3+) = 15.5 (7). In the above mechanism, the G value for the total number of superoxide radicals formed is G(H02 + 02) = 6.05.
In the presence of high formate concentrations (0.1 M and higher), this G value is slightly higher on account of the spur scavenging.
In the present investigation, the experimentally determined G(H02 + 01) = 6.45. With this G value, a ferrous dosimeter calibration curve, and the monitored energy input per pulse, one can calculate the concentration of superoxide radicals produced in the irradiated sample (5  The reaction mixture contained 50 mM of phosphate, 0.25 mM 01,O.l M sodium formate, and varying amounts of NADH and LDH-NADH (pH 8.1; 23"). Aliquots of 4 ml of this solution were irradiated with a single lops pulse of 1.9 m.e.v. electrons which generated 1.2 k~ Ot-in a given sample. All of the other chemicals used were of reagent grade.
Solutions were prepared in triply distilled water. The pulsing electron generator (2 m. e. v. Van de Graaff), optics, and monitoring equipment have been described in detail earlier (10).

RESULTS
The rates for the lactate dehydrogenase-catalyzed oxidation of NADH by HO2 and 0, were studied with pulse radiolysis in oxygenated sodium formate solutions. The LDH-NADHr solutions were always prepared by diluting a stock solution of freshly dialyzed enzyme, which had been titrated for the total number of active sites with the method of Holbrook (9). Following an electron pulse, the reaction was monitored by observing the disappearance of NADH in the spectral region between 340 and 380 nm.
Because the total light path of the optical setup was 6.1 cm, it was often necessary, due to the high absorbance, to measure the changes in NADH concentration at a wavelength other than the spectral maximum at 340 nm. For a particular set of experiments the pulse length as well as the energy input per pulse was kept constant.
On the average, the variation in the energy input per pulse and hence the initial concentration of superoxide radicals generated was of the order of 4 to 5%.
Because preliminary pulse experiments revealed that the observed rate of the disappearance of NADH followed pseudo-first The absorbance changes were monitored at 370 nm (optical light path 6.1 cm) following a 10.~~ electron pulse (1.9 m.e.v.), which generated 0.60 f 0.03 PM of (HO2 + Oi-) radicals.
All of the measurements were carried out at 23". The solid and dashed line was calculated from Equation II.
order kinetics in the presence of lactate dehydrogenase, a number of solutions with varying amounts of enzyme were studied at pH 8.1 in order to determine the second order rate constant for Reaction 8. The experimental conditions and results are given in Fig.  1. The good agreement between the second order rate constants computed from the kobs values at different enzyme concentrations, supports the assumption that under the given experimental conditions, all of the active sites on the enzyme were occupied by NADH.
The study of the pH effect upon the rates of oxidation of the LDH-NADH complex by HO2 and 02-radicals is shown in Fig.  2. The experiments were limited on the acidic side (below pH 4.5) by the instability of NADH, the denaturation of the enzyme, and the decreasing signal to noise ratio.
Similarly, on the alkaline side, the signal to noise ratio made accurate determinations impractical beyond pH 9.0. The results show that the k&s did not change significantly between pH 9 and pH 7, whereas it increased sharply below pH 7.
In the same series of experiments, the total decrease in absorbance within 20 s after the electron pulse was determined and converted to the amount of NADH oxidized.
Because the amount of superoxide radical generated by the pulse could be calculated from the energy input, the ratio of the number of molecules of NADH oxidized per superoxide radical (i.e. the chain length) for each experimental point can be determined. As can be seen in Fig. 3, the chain length has a maximum at pH 7.2 and it drops off almost symmetrically with increasing either acidity or alkalinity. DISCUSSION The qualitative aspects of the lactate dehydrogenase-catalyzed chain oxidation of NADH by superoxide radicals had been established and discussed in detail in earlier reports (1, 2). The present study was aimed at the establishment of some of the Lactate dehydrogenase binds NADH much more strongly than NAD+.
In this study, based on the over-all magnitude of the dissociation constant for the LDH-NADH complex (11, 12), NADH was always present in excess to saturate all of the active sites of the enzyme.
Hence, the effective concentration of the complex LDH-NADH in Reactions 7 and 8 is equal to the experimentally determined total number of active sites. This assumption was supported by the observation of strict pseudofirst order kinetics over a 4-fold enzyme concentration at constant NADH concentration. The observed pseudo-first order decay of NADH ( Fig. 1) over several half-lives indicates that the effective complex concentration of LDH-NADH is maintained at a constant level over many chain lengths (compare with Fig. 3), the reason being that the rates controlling its replenishment (Reactions 6, 9 and lo), from the excess (0.1 mM) pool of unbound NADH are more rapid than those of its consumption (Reactions 8 and 9). Hence, it may be assumed that one of the essential factors for the chain reaction is the difference in magnitude between the dissociation constants of LDH-NADH and LDH-NAD+. The change in the pseudo-first order kobs with pH is shown in Fig. 2. The plateau region between pH 7 and 9 represents Reaction 8 with little or no contribution from Reaction 7. The second order rate constant calculated was ks = 3.6 f 0.1 X 10" M--l S-'.
As the pH is decreased, the equilibrium (HO2 = 02 + H+) is shifted to the left and hence in favor of Reaction 7. The relatively steep increase in kobs below pH 6, suggests that HO2 reacts more rapidly with the LDH-NADH complex than does 01.
Because the rate constant for Reaction 7 could not be determined in isolation, that is from the corresponding plateau region (See Fig. 2, dashed line between pH 1 and 3), an approximate value for k7 = 1.2 x 10b M-I s-1 was computed by Equation II from the experimental points along the upper slope between pH 4.5 and 5.5. Equation II was derived from the two controlling reactions, 7 and 8, and the dissociation constant of HO2 (Reaction 3, -3)  [ 1 (II) As is apparent in Fig. 2, the value for k? gives a fairly good fit between the computed line and the experimental points. The dashed part of the curve is the extrapolated pH profile in the acid region if both the enzyme and NADH were stable and active. Although the concentration of the unbound NADH in the reaction mixture was about 6 times as high as that of the LDH-NADH complex, the interaction between superoxide radicals and unbound NADH did not contribute significantly to the rate of oxidation.
Land and Swallow (13) estimated that kll << 27 M-1 s-1 NADH + 02--NAD-+ H02- The difference by a factor of lOa between kg and kll dramatically illustrates the catalytic effect of lactate dehydrogenase in this oxidation step. Reaction 9, in which the enzyme-bound radical LDH-NAD. reacts rapidly with molecular oxygen, is the chain propagating step.
The assumption that Reaction 9 is very rapid, is based on the reported rate constant (13, 14) for Reaction 12 of the unbound radical with molecular oxygen: k = 1.9 x 10' M-l 8-l NAD. + o2 l 2 * NAD+ + 02- and on experimental observations made in the present study. An attempt to search for the absorbance of the LDH-NAD. as an intermediate species in the presence of molecular oxygen after the pulse has not been successful.
The signal was of the order of the noise level, which indicated an extremely low steady state concentration and a very rapid Reaction 9. A simiIar experiment carried out in the absence of molecular oxygen showed that the enzyme stabilized the free radical by many orders of magnitude.
The disappearance of the free radical (Reactions 13 and 14) could be observed for several minutes at 405 nm2 yielding the dimer as postulated by Land and Swallow (15) in Reaction 14.
LDH-NAD * SlVd -LDH+NAD* = 5 6 x l o7 M-I s-' NAD. + NAD. k17 'l'his observation confirms that Reaction 9 is fast and precludes the possibility of dissociation in Reaction 13 to be followed by Reaction 12 in the chain mechanism. Hcncc, Reaction 9 leads to the formation of the LDII-X,21)+ complex which dissociates (Reaction 10) and the vacant site is rcplenishcd with another molecule of NADH (Rcaction 6). Thus, the concentration of LDH-NADH remains constant throughout several cycles as indicated by the pseudo-first order kinetics in Fig. 1. The true nature of the termination step has not yet been established due to experimental difficult~ies.
In the absence of appropriate reactants, the superoxide radicals decay by disproportionation.
Various studies (10, [16][17][18] (17) The same studies have also found that when traces of impurities were present in the systems, the superoxide radicals decayed with pseudo-first order kinetics, indicating interaction between the radicals and the impurities.
Hence, one can assume a priori that the addition of an cnzymc to a solution will introduce "impurities" as scavengers for the superoxide radicals. These scavengers could be some reactive groups on the protein, which do not have a dctcctable effect 011 the enzyme activity. A study on the influence of lactate dehydrogenasc on the decay of supcroxidc radicals in the absence of NADH did show some acceleration of the decay rate.
Although It is apparent that with decreasing pH the rate of over-all dccay of superoxide radicals increases rapidly and thus tends to shorten the chain length.
On the othrr hand, a decrease in p1I is accompanied by an increase in the I-10,: 02-ratio and thereby increases the rate of over-all oxidation as indicated in Fig. 2, and therefore the latter factor tends to lengthen the chain.
Conscquently, when the chain length was plotted against pII, an optimal chain length of 18 was found at pH 7.2.
An earlier study (1) had shown that one of the factors controlling the chain length is the dose rate, because it dctermincs the ratio of LDH-NADH : 02-. As expected, the chain length under pulse radiolysis conditions is not as lorlg as that under steady state conditions in Wo y ray studies (1) The rate of superoxidc radical formation in pulse radiolysis is orders of magnitude higher and competition by Reactions 15, 16, and 17 for Reactions 7 and 8 is more efficient.
Termination of the chain reaction by a radical-radical intcraction (Reaction 18) can be ruled out in this system, because Rcac tion 12 is very fast, and the ratio of OZ:OZ-is of the order of 500.