Effects of pH and temperature on the reaction of milk xanthine oxidase with 1-methylxanthine

Article abstract of DOI:10.1039/b005721p

The reaction of milk Xanthine Oxidase (XO) with 1-methylxanthine has been investigated by steady state and stopped flow transient kinetic studies to understand the effect of a methyl group substituent on the purine ring. The pH dependence of the steady state kinetic parameter (V_max/K_m) shows a bell-shaped curve implying at least two ionisable groups are involved in the binding of XO with 1-methylxanthine, the higher pK_a 7.7 corresponding to ionisation of the substrate and the lower one 6.2 to the enzyme (possibly Glu-1261 by analogy to Glu-869 of aldehyde oxidoreductase from Desulfovibrio gigas). The temperature dependence of the steady state and transient kinetic studies suggests the existence of at least one molecular intermediate during breakdown of the enzymesubstrate complex. The thermodynamic parameters of the microscopic rate constants were determined from the temperature dependence studies. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/b005721p

View Article Online / Journal Homepage / Table of Contents for this issue
Effects of pH and temperature on the reaction of milk xanthine
oxidase with 1-methylxanthine
Apurba Kumar Sau,*† Madhu Sudan Mondal and Samaresh Mitra
Department of Chemical Sciences, Tata Institute of Fundamental Research,
Homi Bhabha Road, Colaba, Mumbai 40 0005, India
Received 14th July 2000, Accepted 18th August 2000
First published as an Advance Article on the web 27th September 2000
The reaction of milk Xanthine Oxidase (XO) with 1-methylxanthine has been investigated by steady state and
stopped flow transient kinetic studies to understand the effect of a methyl group substituent on the purine ring.
The pH dependence of the steady state kinetic parameter (Vmax/Km) shows a bell-shaped curve implying at least
two ionisable groups are involved in the binding of XO with 1-methylxanthine, the higher pKa 7.7 corresponding
to ionisation of the substrate and the lower one 6.2 to the enzyme (possibly Glu-1261 by analogy to Glu-869 of
aldehyde oxidoreductase from Desulfovibrio gigas). The temperature dependence of the steady state and transient
kinetic studies suggests the existence of at least one molecular intermediate during breakdown of the enzyme–
substrate complex. The thermodynamic parameters of the microscopic rate constants were determined from the
temperature dependence studies.
unpasteurized cow’s milk by the reported procedure.12,15,16 The
Introduction
activity of the enzyme was measured spectrophotometrically by
monitoring the formation of uric acid from xanthine at 295 nm.
The calculated AFR (activity to flavin ratio) value of the
enzyme was in the range 90–110 which corresponds to 50–60%
functional enzyme.17 The concentration of the enzyme was
determined spectrophotometrically by using its molar absorp-
tion coefficient of 37,800 MϪ1 cmϪ1 at 450 nm.15
The EDTA used for the isolation of XO was removed by
dialysing it thoroughly against the buffer. In all experiments
Millipore milli-Q quality water was used. The experiments were
done in 100 mM Tris buffer at pH > 7.6 and in 100 mM
NaH2PO4 buffer at pH < 7.6.
All spectrophotometric experiments were performed on a
Shimadzu UV-2100 spectrophotometer. A pH meter from
Orion Research was used to measure the pH of the sample. A
temperature controller connected to the UV-visible spectro-
photometer maintained the temperature of the solution inside
the cuvette for the steady state kinetic experiments. The stopped
flow kinetic experiments were performed on a HITECH
SF-61 stop flow machine with four microprocesser-controlled
syringes. The transient reductive half reaction was studied by
loading the enzyme and 1-methylxanthine in two separate
syringes and made anaerobic by prolonged purging with oxygen
free argon gas. The two separate solutions were mixed in a mix-
ing chamber and the reaction was maintained under anaerobic
conditions. The temperature of the syringe reservoir was main-
tained within 0.5 ЊC using a water circulating bath. The meas-
urements were done spectrophotometrically by monitoring the
wavelength of interest. Kinetic data for the rapid kinetics were
analysed by fitting using an exponential function and HITECH
software.
Milk Xanthine Oxidase (Xanthine: Oxygen Oxidoreductase,
E.C. 1.2.3.2.) is a complex metallo-flavo enzyme which catalyses
the oxidation of xanthine in the presence of oxygen to form uric
acid. The enzyme has two independent subunits each contain-
ing four redox centers: a molybdenum() center, two iron–
sulfur (2Fe/2S) clusters and a flavin adenine dinucleotide (FAD)
unit.1–3 The catalytic cycle of Xanthine Oxidase (XO) consists
of two half reactions: reductive and oxidative. The reductive
half reaction (i.e. conversion of xanthine into uric acid) takes
place at the molybdenum center of the enzyme,4,5 while the
oxidative half reaction (i.e. conversion of oxygen into peroxide
and superoxide) takes place at the FAD center.6 In the process
of the turnover, an intramolecular electron transfer (ET) takes
place within the four redox centers of the enzyme. Steady state
and transient kinetic studies of the reaction of XO with its
physiological substrate xanthine and other substrates have been
reported,7–13 and the rates of reaction have been observed to
vary with substrate. Olson et al. have investigated the steady
state and transient kinetics of the reaction of xanthine with XO
and suggested a two-step breakdown of the enzyme–substrate
complex.7 More recently, Mondal and Mitra12 have reported
the temperature dependence of this reaction and have suggested
the formation of at least two molecular intermediates. The reac-
tion of 1,7-dimethylxanthine with XO has been found to be
very slow compared to that of xanthine14 and no intermediate
was reported during the breakdown of the enzyme-substrate
complex. It is found that the reaction of 1-methylxanthine with
XO is similar to that of xanthine but its rate is slower. To
determine the effect of substitution of a methyl group on the
purine ring we have carried out pH and temperature dependent
studies of the reaction of XO with 1-methylxanthine by steady
state and transient kinetics.
Analysis of kinetic data
Steady state kinetics. The reductive half reaction of XO with
its substrate can be described by considering that the substrate
binds to XO very rapidly and forms an enzyme–substrate
complex which subsequently breaks down to form product,18
eqn. (1) where S is substrate. The steady state rate equation (2)
Materials and methods
Xanthine oxidase was isolated and purified from fresh
† Present address: Section of Molecular Genetics and Microbiology,
University of Texas at Austin, Austin, TX 78 712, USA. E-mail:
apurba@mail.utexas.edu; Fax: (512) 471 5546.
k1
k2
(XO)ox ϩ S
(XO)oxؒS
product ϩ (XO)red
(1)
kϪ1
3688
J. Chem. Soc., Dalton Trans., 2000, 3688–3692
This journal is © The Royal Society of Chemistry 2000
DOI: 10.1039/b005721p

View Article Online
1
1
Km
1
=
ϩ
×
(2)
γ0 Vmax Vmax [S]0
can be derived from the above reaction scheme under the condi-
tion of [O2]0 ӷ KmO
,
19,20 where γ0 is the initial rate of the reaction,
2
[S]0 the initial concentration of the substrate, Km and KmO are the
apparent Michaelis constants for the substrate and oxygen
respectively and Vmax is the maximum velocity for the steady
state conversion of the substrate into product. A plot of γ0Ϫ1 vs.
[S]0Ϫ1 will give a straight line which allows us to determine values
2
of Km and Vmax
.
Transient kinetics. The reductive half reaction of XO,
involves transfer of electrons from the substrate to the enzyme
in three consecutive steps. This results in a monophasic
decrease in the enzyme absorbance at 450 nm. In the presence
of an excess of the substrate compared to the enzyme, rate
eqn. (3) can be derived for kobs, where Kd = kϪ1/k1, kobs is
Fig. 1 Lineweaver–Burk plot for XO catalysed steady state formation
of 1-methyluric acid from 1-methylxanthine using UV-visible spectro-
photometry. The formation of 1-methyluric acid was measured at 295
nm. The experiments were performed at various pH (5.8–10.5). The con-
centration of the enzyme was 0.0249 µM. The solid lines drawn through
the experimental data points at pH 5.8(᭺), 6.7(ᮀ), 7.6(᭝), 7.8(᭞),
8.5(᭛) and 9.1(᭜). The values of Km and Vmax were determined from
the slope and intercept of eqn. (2).
1
Kd
1
1
=
×
ϩ
(3)
kobs k2 [S] k2
the pseudo first order rate constant, Kd the dissociation con-
stant of the enzyme–substrate complex and k2 is the rate con-
stant for formation of the product. Olson et al.7 have shown a
reaction scheme for the conversion of the enzyme–substrate
complex into the product for xanthine as a substrate which can
be written as in eqn. (4) where k2Ј and k2Љ are the microscopic
k2Љ
k2Љ
(XO)oxؒXan
(XO)redؒUric acid
(XO)red ϩ Uric acid (4)
rate constants for conversion of the enzyme–substrate complex
into product.
By comparing the rate constant k2 of eqn. (1) with k2Ј and k2Љ
of (4), the following relation (5) for k2 can be derived.18 The rate
1
1
1
Fig. 2 Plot of Vmax/Km vs. pH for the reaction of XO with 1-methyl-
xanthine. The solid line drawn through the experimental data is the
theoretical fit to eqn. (9). The fitted pK1 and pK2 were 6.2 and 7.7
respectively.
(5)
=
ϩ
k2 k2Ј k2Љ
constants can be expressed in terms of the Arrhenius equation
(k = A exp(ϪE/RT)) as in eqn. (6) where AЈ and AЉ and EЈ and
Results
(a) pH dependence
1
1
1
(6)
=
exp(EЈ/RT) ϩ
exp(EЉ/RT)
The steady state XO catalysed oxidation of 1-methylxanthine
to its corresponding uric acid was monitored at 295 nm at
different concentrations of 1-methylxanthine. The oxygen con-
centration of the sample was kept constant at 515 µM. The
k2 AЈ
AЉ
EЉ are the corresponding frequency factors and activation ener-
gies respectively for the two steps, T the absolute temperature
and R the molar gas constant. The expression for k2 shows a
non-linear behavior with temperature. Thus, the temperature
dependence of k2 can be fitted to determine the values of AЈ,
AЉ, EЈ and EЉ.
Michaelis constant (KmO ) for the binding of oxygen to XO is 50
2
µM.15,21 The oxygen concentration was thus about 10 times
12
higher than KmO
.
The steady state kinetic experiments were
2
carried out in the pH range 5.8–9.5. It is to be emphasized that
XO is stable in this pH range. Fig. 1 shows the Lineweaver–
Burk (LB) plot at various pH from which Km and Vmax were
determined. Typical values were found to be 8.3 µM and 7.2
µM minϪ1 respectively at pH 6.8. The plot of the second order
rate constant Vmax/Km vs. pH exhibits a bell shaped variation
with a maximum at pH 6.8 (Fig. 2). A similar pH dependence
was reported for xanthine.13 It suggests that at least two
ionisable groups are involved in the reaction of XO with
1-methylxanthine, which can be explained by considering
Determination of thermodynamic parameters. The thermo-
dynamic parameters associated with binding of substrate to XO
were determined from the van’t Hoff relation, see eqns. (7) and
(8), where ∆H, ∆S and ∆G are the change in enthalpy, entropy
Ϫ∆H
log Kd = ͩ ͪ
R × 2.303
1
∆S
×
ϩ
(7)
(8)
T
R × 2.303
K2
K2
double ionisation of the enzyme, H2E
HE
E where K1
∆G = ∆H Ϫ T∆S
and K2 are the acid equilibrium constants. Assuming that the
singly protonated form of the enzyme is catalytically active, the
expression (9) for Vmax/Km can be derived.13 A fit of the experi-
mental data of Fig. 2 by eqn. (9) gave pK1 = 6.2 and pK2 = 7.7.
and Gibbs free energy for formation of the enzyme–substrate
complex.
J. Chem. Soc., Dalton Trans., 2000, 3688–3692
3689

View Article Online
Fig. 4 Typical stopped flow trace for the anaerobic reduction of XO
(5 µM) with 1-methylxanthine (75 µM) at 450 nm. The solid line drawn
through the experimental trace is the computer fit to the single
exponential function. At = A0 exp(Ϫkobst) ϩ A∞, where At is the absorb-
ance at time t, A0 the amplitude of the absorbance change, kobs the
observed rate constant, t the time in seconds and A∞ the equilibrium
Fig. 3 Logarithmic plots of Km (A) and Vmax (B) vs. 1/T for the
steady state conversion of 1-methylxanthine into the corresponding
uric acid at pH 6.8.
Ϫ1
signal absorbance (offset). The inset in the figure shows a plot of kobs
vs. [1-Methyl Xan]Ϫ1. The solid line drawn through the experimental
data is the fit by eqn. (3).
(Vmax/Km)max
1 ϩ [Hϩ]K1Ϫ1 ϩ K2[H]Ϫ1
(Vmax/Km) =
(9)
This indicates that an ionisable group having pKa of 6.2 must
be deprotonated and that one having a pKa of 7.7 must be
protonated for the reaction to take place. The former pKa may
correspond to a functional group in the active site of the
enzyme and the latter to the substrate.13
(b) Temperature dependence
The temperature dependence of the steady state kinetics of the
reaction of 1-methylxanthine with XO was studied between 14
and 50 ЊC. The data were analysed as described above for pH
dependent measurements, and LB plots were constructed (not
shown here) at each temperature to obtain Vmax and Km. The
logarithmic plots of Vmax and Km vs. temperature show a linear
trend for Vmax and a non-linear trend for Km (Fig. 3). Theore-
tically, both Vmax and Km consist of several microscopic rate
constants.12 Therefore, their logarithmic plots with temperature
might be expected to follow a non-linear trend. In practice,
however, both linear and non-linear trends are possible depend-
ing on the relative magnitudes of the individual microscopic
rate constants in Vmax and Km.12 The temperature dependence
of the logarithmic plots of Vmax and Km for xanthine was found
to be similar to that observed for 1-methylxanthine. The present
observation of a linear trend for log Vmax vs. 1/T suggests that
overall composite rate constants control the turnover rate of
the reaction of XO with 1-methylxanthine. From the linear
trend in Fig. 3B, the activation energy related to Vmax was
determined to be 9.3 kcal molϪ1. The non-linear behaviour of
log Km vs. 1/T (Fig. 3A) is consistent with the theoretical pre-
diction of Km.12 A similar non-linear dependence of log Km with
temperature has been observed for other enzyme–substrate
reactions.22
Fig. 5 (A) Logarithmic plot of k2 vs. 1/T shows the non-linear
behavior for the formation of product at pH 6.8. The solid line drawn
through the experimental data points shows the trend of the plot. (B)
Plot of 1/k2 vs. 1/T for the same experiment. The solid line drawn
through the experimental data is the fit by eqn. (6). The fitted values
are E2Ј 1.6 kcal molϪ1, E2Љ 16.4 kcal molϪ1, A2Ј 3.3 × 102 sϪ1 and A2Љ
7.6 × 1012 sϪ1
.
obtained when 75 µM 1-methylxanthine reacted with 5 µM XO.
A plot of kobsϪ1 vs. [1-Methyl Xan]Ϫ1 gives a straight line which
suggests reversible binding of 1-methylxanthine to XO with 1:1
stoichiometry. Using eqn. (3), typical values of k2 and Kd were
deduced: 6.0 sϪ1 and 9.0 µM at 25 ЊC. Using the values of Kd at
various temperatures the thermodynamic parameters (∆G, ∆H
and ∆S) associated with binding of 1-methylxanthine to XO
The transient kinetics of the reaction of XO with 1-methyl-
xanthine were measured in the temperature range 14–50 ЊC.
Fig. 4 shows a typical stopped flow trace of the reduction of the
enzyme at 450 nm under pseudo first order conditions. The
experimental data could best be fitted by a single exponential
function to obtain kobs. A typical value of kobs = 4.9 sϪ1 was
(XO ϩ 1-methylxanthine
bound complex) were deter-
mined from the linear plot of log Kd vs. 1/T (figure not shown).
3690
J. Chem. Soc., Dalton Trans., 2000, 3688–3692

View Article Online
Using eqns. (7) and (8), we obtain ∆G = 6.9 kcal molϪ1, ∆H =
10 kcal molϪ1 and ∆S = 10.8 cal KϪ1 molϪ1. The Arrhenius plot
of k2 shows a non-linear trend (Fig. 5A). The plot of 1/k2 vs.
1/T also shows a non-linear trend (Fig. 5B). Eqn. (6) was used
to fit the experimental data of Fig. 5B (see the solid line). Eqn.
(6) has four adjustable parameters, and hence their individual
values cannot uniquely be determined by fitting the experi-
mental data. However, an estimate of their values was made by
the least squares fit which gave: E2Ј = 1.6 kcal molϪ1, E2Љ = 16.4
k4. This is consistent with the observation of Olson et al.21 that
k4 is higher than k2 in the case of xanthine as a substrate. Km is
expressed as: Vmax(k2 ϩ kϪ1)/E0k1k2.12 Since Vmax shows a linear
Arrhenius plot with negative slope, the non-linear behavior of
Km must be attributed to the factor (k2 ϩ kϪ1)/k1k2. The
Arrhenius plot of k2 shows non-linear behavior, so the non-
linearity of log Km with 1/T arises from the contribution of k2.
The ratio Vmax/Km determines the specificity of the enzyme
towards the substrate and it was found that this ratio is
independent of temperature for 1-methylxanthine. However, for
xanthine it has been observed that it increases with increasing
temperature.12
We have determined Kd due to the binding of 1-methyl-
xanthine to XO from the transient kinetic measurements. The
thermodynamic parameters associated with the binding at
various temperatures were determined from the temperature
dependence of Kd. The positive value of ∆G indicates that the
dissociation of the enzyme–substrate complex is energetically
unfavorable. The positive values of ∆H and ∆S indicate that
the binding of XO to 1-methylxanthine is an endothermic pro-
cess and the dissociation of the enzyme–substrate complex is
entropy-driven.
kcal molϪ1, A2Ј = 3.3 × 102 sϪ1 and A2Љ = 7.6 × 1012 sϪ1
.
Discussion
The Km and Vmax values for 1-methylxanthine were found to be
8.3 µM and 7.2 µM minϪ1 at 25 ЊC and pH 6.8. The correspond-
ing values for xanthine were reported as 4.0 µM and 15 µM
minϪ1 at 25 ЊC and pH 6.8.18 This indicates that the binding
affinity of 1-methylxanthine to XO is lower than that of xan-
thine. It also suggests that the rate of the reaction of XO with 1-
methylxanthine is slower than with xanthine. These may be due
to the presence of a methyl group at N-1 in 1-methylxanthine
which sterically hinders the binding of 1-methylxanthine to XO.
Vmax/Km exhibits a bell shaped pH dependence as expected from
the double ionisation of the enzyme, considering the singly pro-
tonated form of the enzyme to be catalytically active. Eqn. (9)
gave a good fit to the experimental data of Fig. 2 and the pKas
were found to be 6.2 and 7.7. The pH dependence of Vmax/Km
represents ionisation of the free enzyme and free substrate.
It is to be noted that, with regard to the high pH limb of the
curve, there is no known ionisation of the enzyme in this pH
range.13 Therefore, the pKa 7.7 corresponds to ionisation of the
substrate. In the case of xanthine Vmax/Km also exhibits a bell
shaped pH dependence.13 It is known that xanthine has a pKa of
7.3 which has been attributed to ionisation of its N-1 proton.23
It is suggested that the neutral form of the substrate is required
for binding to the enzyme.13 The presence of the methyl
group at N-1 in 1-methylxanthine replaced the ionisation of
the hydrogen at N-1. Therefore, the pKa 7.7 of 1-methyl-
xanthine arises from ionisation of its N-9 proton. The pKa
of 1-methylxanthine obtained from our studies is not very
much different from that of xanthine. This suggests that the
presence of the methyl group does not seem to have much
effect on the ionisation of the substrate. The implication is
that the protonated (neutral) form of the substrate is required
for binding to the enzyme. This is consistent with the abstrac-
tion of the proton from C-8, as the negative charge of the
ionised substrate destabilises the accumulating negative
charge on C-8 for the deprotonation. The pKa 6.2 must be
attributed to the ionisation of a residue in the active site of
the enzyme. Recently, the crystal structure of aldehyde
oxidoreductase from Desulfovibrio gigas was solved at 1.8 Å,
and on the basis of this structure it was proposed that Glu-869
at the active site abstracts a proton from molybdenum bound
water/hydroxide and thereby facilitates nucleophilic attack
on the substrate.24 Xanthine oxidase has good amino acid
sequence homology with aldehyde oxidoreductase in the
vicinity of their molybdenum center that includes this
glutamate residue (Glu-869 of the D. gigas versus Glu-1261 of
the bovine enzyme). We suggest that Glu-1261 of XO plays a
role analogous to that proposed for Glu-869 of aldehyde
oxidoreductase. Although the observed pKa 6.2 is rather
high for a glutamate it is not surprising. The pKa of Glu-35 in
lysozyme is reported to be 6.5.25
Olson et al.7,21 and Hille and Massey18 suggested the
involvement of one intermediate in the breakdown of the
enzyme–substrate complex (XOoxؒXan → XOred ϩ Urate) in
the reaction of XO with xanthine. Kim et al.13 also studied the
reaction of XO with xanthine and suggested the existence of
multiple intermediates during the breakdown of the enzyme-
substrate complex. The temperature dependent transient
kinetics of the reaction of XO with xanthine has also been
investigated by Mondal and Mitra.12 Their investigations indi-
cate non-linear plots of log k2 (and 1/k2) vs. 1/T. These results
suggest that at least two intermediates are involved during the
breakdown of the XOoxؒXan complex to the product (uric
acid).12 In the present measurements, similar attempts have
been made to obtain further insights into the breakdown of the
XOoxؒ1-Methyl Xan complex to the product (1-methyluric
acid). Eqn. (6) was used to fit the experimental data of Fig. 5B.
Eqn. (4) considers the existence of only one intermediate for the
breakdown of XOoxؒ1-Methyl Xan to the product. Further, a
critical inspection of eqn. (6) and the experimental data in Fig.
5B reveals that both the activation energies (E2Ј and E2Љ) can
not be negative. It is however possible that both the activation
energies are positive or with opposite signs. The results from
our analysis show that eqn. (6) fits very well the experimental
data of Fig. 5B (see the solid line) only when both the activation
energies are considered to be positive (1.6 and 16.4 kcal molϪ1).
It therefore appears that the breakdown of the XOؒ1-Methyl
Xan complex to the product involves the existence of at least
one intermediate.
References
1 R. X. Bray, in The Enzymes, ed. D. P. Boyer, Academic Press, New
York, 1975, vol. XII, part B, p. 300.
2 R. Hille, V. Massey and T. G. Spiro, Molybdenum Enzymes, John
Wiley & Sons, New York, 1985, pp. 443–518.
3 R. C. Bray, Q. Rev. Biophys., 1988, 21, 299.
4 R. Hille, G. N. George, M. K. Eidsness and S. P. Cramer, Inorg.
Chem., 1989, 28, 4018.
5 R. C. Bray, G. Palmer and H. Binert, J. Biol. Chem., 1964, 239,
2667.
6 H. Komai, V. Massey and G. Palmer, J. Biol. Chem., 1969, 244,
1692.
Km and Vmax for the reaction of xanthine with XO are known
to be combinations of several microscopic rate constants.12 The
linear dependence of log Vmax with 1/T for the reaction of
1-methylxanthine with XO suggests that Vmax is controlled by
one rate constant (either k2 or k4).12 Since the Arrhenius plot of
k2 shows non-linear behavior (see Fig. 5A), the linear depend-
ence of log Vmax with 1/T must arise from the contribution of
7 J. S. Olsen, D. P. Ballou, G. Plamer and V. Massey, J. Biol. Chem.,
1974, 249, 4363.
8 J. W. Buntiny and A. Gunasekara, Biochim. Biophys. Acta, 1982,
704, 444.
9 F. F. Morpeth, Biochim. Biophys. Acta, 1983, 744, 328.
10 B. E. Skibo, Biochemistry, 1986, 25, 4189.
11 B. E. Skibo, H. J. Gilchrist and H. C. Lee, Biochemistry, 1987, 26,
3032.
J. Chem. Soc., Dalton Trans., 2000, 3688–3692
3691

View Article Online
12 M. S. Mondal and S. Mitra, Biochemistry, 1994, 33, 10305.
13 J. H. Kim, M. W. Ryan, H. Knant and R. Hille, J Biol. Chem., 1996,
271, 6771.
14 A. K. Sau, M. S. Mondal and S. Mitra, J. Chem. Soc., Dalton
Trans., 1998, 4173.
15 V. Massey, P. E. Brumby, H. Kombai and G. Palmer, J. Biol. Chem.,
1969, 244, 1682.
16 M. S. Mondal and S. Mitra, J. Inorg. Biochem., 1996, 62, 271.
17 D. Edmondson, V. Massey, G. Palmer and L. M. Beachan, J. Biol.
Chem., 1972, 247, 1597.
20 R. Radi, S. Tan, E. Prodanov, R. A. Evans and D. A. Parks,
Biochim. Biophys. Acta, 1992, 1112, 178.
21 J. S. Olsen, D. P. Ballou, G. Plamer and V. Massey, J. Biol. Chem.,
1974, 249, 4350.
22 C. T. Privalle, S. E. Kong and I. Fridovich, Proc. Natl. Sci. U.S.A.,
1993, 90, 2310.
23 A. Albert and D. J. Brown, J. Chem. Soc., 1954, 2060.
24 R. Huber, P. Hof, R. O. Durate, J. J. G. Moura, J. Legall, R. Hille,
M. Archer and M. J. Romabo, Proc. Natl. Acad. Sci. U.S.A., 1996,
93, 8846.
18 R. Hille and V. Massey, J. Biol. Chem., 1991, 266, 17401.
19 E. L. King and C. Altman, J. Phys. Chem., 1956, 60, 1375.
25 S. M. Parsons and M. A. Raffery, Biochemistry, 1972, 11,
1633.
3692
J. Chem. Soc., Dalton Trans., 2000, 3688–3692