Professional Documents
Culture Documents
ABSTRACT: An electron paramagnetic resonance (EPR) study was carried out to examine structural aspects
of vanadium-containing bromoperoxidase from the brown seaweed Ascophyllum nodosum. A t high pH,
the reduced form of bromoperoxidase showed an apparently axially symmetric E P R signal (go = 1.969 and
A,, = 86.8 X lo4 cm-') with 16 hyperfine lines. When the pH was lowered, a new EPR spectrum was formed
(go = 1.970 and A . = 92.6 X cm-') in which the values for gll and g, were hardly affected, whereas
both A l land A , showed a considerable increase. When EPR spectra of the reduced enzyme were recorded
in the p H range from 4.2 to 8.4, it appeared that these changes were linked to a functional group with an
apparent pK, of about 5.4. In D 2 0 this value for the pK, was 5.3. It is suggested that these effects arise
from protonation of histidine or aspartate/glutamate residues near the metal ion. The values for the isotropic
hyperfine coupling constant of the reduced enzyme at both high and low p H are also consistent with a ligand
field containing nitrogen and/or oxygen donor atoms. When reduced bromoperoxidase was dissolved in
D 2 0 or H2170instead of H2160,vanadium(1V) hyperfine line widths were markedly affected, demonstrating
that water is a ligand of the metal ion. Furthermore, it is shown that hydrogen peroxide and bromide were
not able to oxidize reduced bromoperoxidase. Together with previous work these findings suggest that
vanadium(1V) is not involved in catalytic turnover and confirm the model in which the vanadium(V) ion
of the native enzyme only serves to bind both hydrogen peroxide and bromide. After excess vanadate was
added to a homogeneous preparation of purified bromoperoxidase, the extent of vanadium bound to the
protein increased from 0.5 to 1.1, with a concomitant enhancement of enzymic activity. Finally, it is
demonstrated that both vanadate (VOd3-) and molybdate MOO^^-) compete for the same site on apo-
bromoperoxidase. Molybdate, however was unable to restore enzymic activity, and binding of this metal
ion to the V site in the apoenzyme therefore resulted in a loss of halogenating activity.
H a l o p e r o x i d a s e s are known to catalyze the peroxidative containing bromoperoxidases it was demonstrated for the first
oxidation of iodide, bromide, and/or chloride ions, which in time that this transition metal is involved in enzyme-catalyzed
the presence of a nucleophilic halogen acceptor gives rise to redox reactions. The involvement of vanadium in enzymically
a halogenated compound. A variety of haloperoxidases from catalyzed redox reactions has recently been reviewed by Wever
different sources have been investigated, such as bacteria (Van et al. (1987).
Pee & Lingens, 1985), fungi (Morris & Hager, 1966), marine Native vanadium bromoperoxidase contains the vanadium
invertebrates (Jannum et al., 1981), marine algae (Hewson ion in the 5+ oxidized state. An EPR spectrum of vanadi-
& Hager, 1980), and mammals (Bolscher et al., 1984). Many um(1V) is obtained after reduction of the resting enzyme with
of these haloperoxidases contain heme at the active site sodium dithionite (de Boer et al., 1986a,b). This EPR spec-
(Morrison & Schonbaum, 1976). However, the haloper- trum possesses axial symmetry consisting of two sets of eight
oxidases identified so far in brown seaweeds and several red hyperfine lines due to the coupling of the unpaired electron
seaweeds did not possess a heme prosthetic group, as was with the nuclear spin of vanadium ( I = 'I2). Since EPR
verified by Vilter (1983), Itoh et al. (1986), and us (Wever spectral data of reduced bromoperoxidase resemble those of
et al., 1985; de Boer et al., 1986a,b; Krenn et al., 1987). several inorganic oxovanadium(IV) complexes, it was inferred
Reconstitution experiments showed that the transition metal that in the reduced enzyme the vanadium ion is coordinated
vanadium is essential for the halogenating activity for most to an axially terminal oxo ligand. Here, we report on the
of these algal enzymes (Vilter, 1984; de Boer et al., 1986a,b; effects of solvents such as D 2 0 and H2170on EPR spectra of
Krenn et al., 1987). Moreover, when purified enzyme prep- reduced bromoperoxidase at both high and low pL. Resulting
arations were analyzed by electron paramagnetic resonance EPR spectral data are discussed with regard to possible co-
(EPR)' and atomic absorption spectrophotometry, it was ordinating ligands of the vanadium ion in bromoperoxidase.
shown that vanadium was present in the bromoperoxidases
from Ascophyllum nodosum and Laminaria saccharina (de MATERIALS AND METHODS
Boer et al., 1986a,b). With the establishment of vanadium- Bromoperoxidase from the brown seaweed A . nodosum was
purified as described by Wever et al. (1985) with modifications
(de Boer et al., 1986a). Purified enzyme preparations were
This work was made possible by financial support from the Neth-
erlands Technology Foundation (STW). We gratefully acknowledge the dissolved in 0.1 M Trissulfate/O.l M sodium sulfate (pH 8.3).
support by DSM N.V., Geleen, The Netherlands. This work is part of
the research program of the Organization for the Advancement of Pure
Research (ZWO) under the auspices of the Netherlands Foundation for ' Abbreviations: EPR, electron paramagnetic resonance; L, unspec-
Chemical Research (SON). ified isotope of hydrogen; Tris, tris(hydroxymethy1)aminomethane;
* Author to whom correspondence should be addressed. EDTA, ethylenediaminetetraacetic acid; MCD, 2-chlorodimedone.
experimental
conditions" glib g, gor A,ld A, Aoe
H,O. citrate. high DH 1.948 1.979 1.969 160.1 50.2 86.8
I .
I A
3 g-value
2.4
I.IIIl....J....l.l
2.3
9-value
..I
2.2
FIGURE 3: Effect of deuterium and I70-labeled water analogues on
mI = - 5 / 2 parallel resonance line widths of reduced bromoperoxidase.
Bromoperoxidase(in 0.1 M Tris-sulfate 0.1 M sodium sulfate, pH
8.3) was dissolved in H2I60(trace a), H2I 0 (trace b), and D20 (trace4 FIGURE 5: Effect of pD on ml = - 5 / 2 and -7/2 parallel hyperfine lines
c) at 155 pM. Equal intensities were obtained for all three EPR of r e d u d bromoperoxidase. Bromoperoxidase (226-3 10 pM in D20)
spectra. EPR spectra of bromoperoxidase dissolved in H2170(45% was dissolved in 0.1 M sodium citrate/O.l M sodium sulfate. pD
enriched in oxygen-17) were normalized to 100% 170contribution. values: a, pD 6.9; b, pD 6.0; c, pD 5.4; d, pD 4.2. EPR spectra were
Average traces of 15 scans are shown, normalized to a microwave recorded at identical values for temperature (67 K), microwave power
frequency of 9255 MHz. Line widths are 1.87 (H260),2.25 (HZi7O), (2 mW), modulation width (0.63 mT), and receiver gain. Average
and 0.98 mT (D20). Instrument settings: microwave power, 2 mW; traces of 15 scans are shown, normalized to a microwave frequency
modulation width, 0.5 mT; temperature, 60 K. of 9255 MHz.
0.8
0.4
-0 1 - 0 . 4
-0.8
4 5 6 1 4 5 6 1
I . * . I I , I . . I .... I . . .I PH PD
2 4 2.3 2 2
FIGURE 6: Henderson-Hasselbalch plots of the relative intensities
9-value
of the high-pL and low-pL resonances. Intensities were obtained from
different sample preparations in H 2 0 (part a) and D20 (part b) and
were calculated as mentioned under Materials and Methods. For
FIGURE 4: Effect of pH on m, = - 5 / 2 and -7/2 parallel hyperfine lines experimental conditions see Figure 4 and 5. Ib, intensity of the high-pL
of reduced bromoperoxidase. Bromoperoxidase (14C-226 pM in H20) features; Za, intensity of the low-pL features.
was dissolved in 0.1 M sodium citrate/O.l M sodium sulfate. pH
values: a, pH 8.4; b, pH 5.9; c, pH 5.0; d, pH 4.2. EPR spectra were lowering the p H can be ascribed to protonation of a ligand
recorded at identical values for temperature (60 K), microwave power near the paramagnetic center. In D 2 0 only minor effects on
(2 mW), modulation width (1.0 mT), and receiver gain. Average
traces of 15 scans are shown, normalized to a microwave frequency vanadium hyperfine line widths could be detected in going from
of 9255 MHz. high to low pD. Since deuterium hardly contributes to vanadyl
hyperfine line widths (Albanese & Chasteen, 1978), this small
of the EPR spectra. In particular, the parallel ml = and effect is also in accordance with an ionizing group near the
-7/2 hyperfine lines were nearly completely resolved for both paramagnetic species.
forms. This is shown in detail in Figure 4 for reduced bro- When the logarithms of the relative iqtensities of the two
moperoxidase in H 2 0 in the p H region from 4.2 to 8.4. This conformations in either H 2 0 or D20 were plotted as a function
figure clearly shows that the low-pH form dominated the traces of pL, straight lines were obtained (Figure 6 ) . These plots
below p H 5, whereas above this p H the high-pH form was show that the relative intensities of the high-pL and low-pL
most abundant. Hyperfine line widths for both conformations resonances obeyed the Henderson-Hasselbalch equation [pL
were obtained by comparing simulated line shapes with ex- +
= pKa log ([base]/[acid])]. This also suggests that the
perimental curves. The best agreement with experimental appearance of a second set of resonances at low pL was coupled
spectra was obtained by using a Gaussian line shape function to an ionizing group a t the metal site. From the Hender-
with full widths of 1.9 and 2.5 m T a t half-height of the m, son-Hasselbalch titration curves apparent pKa values of 5.4
= (-7/2)11 first derivative absorption curve for the high-pH and f 0.4 and 5.3 f 0.4 could be calculated for sample prepara-
low-pH forms, respectively. Also, when D 2 0 was employed, tions containing H 2 0 and D 2 0 , respectively.
both conformations had equal intensities around pD 5 (Figure In earlier work it was suggested that the vanadium(V) ion
5). Due to the decrease in line width of these parallel ml = of native bromoperoxidase serves as a binding site for the
- 5 / 2 and -7/2 hyperfine lines, resolution was highly improved. substrates hydrogen peroxide and bromide and does not change
Full widths of 0.9 and 1.0 m T a t half-height were obtained valence state during catalytic turnover (de Boer et al., 1986b).
for the high-pD and low-pD m, = (-7/2),1 curves, respectively. The conclusion that the oxidation state of the vanadium ion
In the light of the effects of hydrogen on vanadium(1V) remains unchanged during the peroxidative oxidation of
hyperfine lines, the observed line width broadening upon bromide ions is supported by the observation that hydrogen
EPR STUDIES O N REDUCED VANADIUM BROMOPEROXIDASE VOL. 27, NO. 5 , 1988 1633
c 2 3 4
I M o lYIll lmMl I
i 6 0 t
0'
0
I I
200
I
400
I I -0
E 0 - -
[V(P)I
(yM) 0 10 20 30 40 50
FIGURE 7: Effect of additional vanadium(V) on the specific activity [Mo(PI)l (mM)
of bromoperoxidase. Purified bromoperoxidase (183 pM) was in- FIGURE 8: Effect of molybdate on the reconstitution of apobromo-
cubated with various amounts of vanadate; after incubation for 24 peroxidase with vanadate. Experimental conditions were as mentioned
h at room temperature the specific activity was measured. The amount under Materials and Methods. Specific activity was measured by
of vanadium(V) incorporated in the enzyme was measured by double the bromination of 2-chlorodimedone (pmol of MCD min-' mg-I).
integration of m, = (-'/& hyperfine lines for three different sample The amount of vanadium incorporated in the enzyme is denoted by
preparationsof the reduced enzyme (denoted in the figure with arrows). [V(IV)-BPO]. Insert, EPR spectral intensities of vanadium (IV)
The extent of vanadium binding was (1) 5.8, (2) 11.6, or ( 3 ) 12.4 bromoperoxidase plotted as [Vlfree(l/cv- l), as a function of the
nmo! of V/mg of protein. amount of molybdate added to the apoenzyme.
peroxide is unable to rapidly oxidize reduced vanadium( IV) Fe(II), Cu(II), Nb(V), Zn(II), and Ni(II), were not effective
bromoperoxidase. In this experiment dithionite-reduced (Vilter, 1984; de Boer et al., 1986a,b). Instead, when added
bromoperoxidase was applied to a column of Sephadex G-25 to the apoenzyme, molybdate acted as an inhibitor in the
and centrifuged to remove the excess reducing agent. reconstitution with vanadate (Figure 8). In this experiment
Thereafter, hydrogen peroxide was added to part of the column several sample preparations of apobromoperoxidase were in-
effluent. EPR spectroscopy showed that the signal intensities cubated with equal amounts of vanadate (VOd3-) and mo-
of the dithionite-free reduced bromoperoxidase sample prep- lybdate (Mood2-), which was present in the concentration
arations were identical, both in the absence and in the presence range from 0 to 50 mM. After prolonged incubation (24 h
of hydrogen peroxide (not shown). Also a combination of at room temperature) enzymic activity was measured and the
hydrogen peroxide and bromide did not affect EPR signal samples were reduced with sodium dithionite, after which EPR
amplitudes of vanadium(1V) bromoperoxidase. Previous spectra were recorded. From the spectra, concentrations of
studies have demonstrated that hydrogen peroxide and bromide enzyme-bound vanadium(1V) were determined. In Figure 8
were also unable to reduce native bromoperoxidase (de Boer it can be seen that both activity and the amount of enzyme-
et al., 1986a). The inability of the two substrates to induce
bound vanadium(1V) decreased to the same extent with in-
redox changes confirms the model in which vanadium(V) only
creasing concentrations of molybdate. The reduction in en-
serves to bind hydrogen peroxide and bromide in the per-
oxidation reaction. zymic activity when reconstitution was carried out in the
Bromoperoxidase from A . nodosum as purified by us had presence of molybdate suggests that both Mo(V1) and V(V)
a low content of vanadium(V) (0.4 mol of V/mol of protein; compete for the same binding site on apobromoperoxidase.
de Boer et al., 1986a). Since the enzyme preparation was When two ligands (A and B) compete for the same set of
aggregated, it was not possible to incorporate more vanadium (identical) binding sites (n) on, e.g., a macromolecule, each
in the enzyme [cf. de Boer et al. (1986b)l. When purified with a different association constant (KAand KB),the following
bromoperoxidase was incubated with vanadate under condi- equation holds for the molar fraction of ligand A bound to the
tions of high ionic strength (Le., by adding 0.1 M sodium macromolecule (cA) (Steinhardt & Reynolds, 1969):
sulfate to the Tris buffer), specific activity increased from 66
to 126 pmol of 2-chlorodimedone (MCD) brominated min-'
(mg of protein)-' (Figure 7 ) . Samples were taken in this
experiment (marked with an arrow in Figure 7) to determine
the extent of vanadium binding by EPR. Double integration For molybdenum and vanadium competing for one site per
of mI = (-7/2)11 hyperfine lines of these preparations showed protein molecule (n = 1) this equation becomes
that the amount of vanadium bound to the enzyme increased
from 5.8 to 12.0 nmol of V/mg of protein. Assuming a mo- [vlfree(l/'vV - l ) = l/KV - (KMo/KV)[Molfree
lecular weight of 90000 for the protein molecule (de Boer et
al., 1986a), this value of 12.0 nmol of V/mg corresponds to where [VIfreeequals the amount of vanadium(V) added to
1.1 mol of V/mol of bromoperoxidase. This indicates that apobromoperoxidaseminus the concentration of enzyme-bound
vanadium(1V). 'vv is given by [V(IV)]bound/ [bromoperoxidase],
after adding excess vanadate to purified bromoperoxidase, the
metal ion was present at the active site in a 1:l ratio per -
and [Molfree [MO]addcd,since the binding of molybdenum
hardly affects the amount of free molybdenum in solution.
enzyme molecule.
It has previously been shown that vanadium can be removed When [VIfr,( 1/Ev - 1) was plotted as a function of molybdate
from native bromoperoxidase by treatment with EDTA at pH concentration, a linear curve was obtained (insert, Figure 8).
3.8. Halogenating activity of the apoenzyme reappeared upon This demonstrates that Mo and V indeed compete for the same
adding vanadate while other metal ions, such as Mo(VI), binding site on the a p ~ e n z y m e . ~The line cuts the ordinate
1634 BIOCHEMISTRY DE BOER ET A L .
at the origin, indicating that Kv must be large (>lo6 M-I; correlation between average ligand field strength and isotropic
because of the spread in the intercept an accurate value for hyperfine splitting this suggests an increase in field strength
Kv cannot be obtained by this method). Tight binding of of the ligand environment. This is also in accordance with
vanadium(V) to bromoperoxidase was also suggested by Vilter ionization of a functional group near the vanadium ion. When
(1984). Assuming an upper limit for Kv of lo6 M-I, the the vanadyl cation was used as an EPR spin probe to inves-
association constant for Mo binding is on the order of 14 000 tigate metal-binding sites of metalloproteins, EPR line shifts
M-I. as a function of pH were also employed to study the proton-
ation behavior of first coordination sphere vanadium(1V)
DISCUSSION ligands in serotransferrin and carbonic anhydrase (Chasteen
Since the isotropic g and hyperfine coupling constant A . of et al., 1977; Fitzgerald & Chasteen, 1974).
vanadium(1V) species provide sensitive measures of the Within a protein backbone apparent pKa values for histidyl
metal-binding site, we employed EPR spectroscopy in studying imidazole nitrogens and 0-or y-carboxylic acid side groups
the vanadium-containing bromoperoxidase from A . nodosum. of glutamate and aspartate residues are known to vary between
In the reduced state this enzyme showed an EPR spectrum 5-7 and 4-5, respectively (Brill, 1977). The apparent pKa of
typical of a vanadyl cation (V02+) coordinated to the protein 5.4 indicates that histidine or aspartatelglutamate residues
moiety. At pH 8.3 this axially symmetric EPR spectrum are likely candidates for coordinating ligands. The possible
exhibited a total of 16 hyperfine lines, indicating the presence coordination of vanadium(V) in bromoperoxidase by carboxylic
of one homogeneous metal coordination environment. Re- acid side chains was also derived from 51VN M R spectra of
placing D 2 0 for HzO resulted in a significant narrowing of the native enzyme (Vilter & Rehder, 1987). Alternatively,
vanadium hyperfine line widths. This indicates that ex- it is conceivable that we deal with ionization of a coordinated
changeable hydrogens are present near the vanadium(1V) ion water molecule (Albanese & Chasteen, 1978). The values for
of reduced bromoperoxidase. Since a marked increase in the isotropic hyperfine coupling constant for both the high-pL
vanadium hyperfine line widths was observed when H2I7Owas (87.0 X cm-I) and low-pL (92.1 X cm-') confor-
employed instead of H2I60,water obviously forms part of the mations are also in accordance with a ligand field of moderate
coordination environment of the vanadium(1V) ion. Lowering strength, involving oxygen and/or nitrogen donor atoms
the pL in solutions of reduced bromoperoxidase in either HzO (Boucher et al., 1969; Chasteen, 1981).
or D 2 0 highly affected EPR spectra, resulting in overlapping Bromoperoxidase is able to incorporate the metal ion mo-
resonances with equal intensities at about pL 5.4. This su- lybdenum(V1) instead of vanadium(V). The molybdenum
perposition of resonances can be explained by the occurrence enzyme, however, does not display any haloperoxidase activity.
of two magnetically nonequivalent metal-binding sites, des- Although molybdenum is incorporated in bromoperoxidase,
ignated the high-pL and low-pL resonances. In H 2 0 the it does not have a marked affinity for the enzyme, compared
appearance of a second set of 16 hyperfine lines at low pH to vanadium. This was evident from the high amount of
was accompanied by line width broadening. Moreover, the molybdenum needed to compete with vanadium for the active
relative intensities of both the high-pL and low-pL forms site of bromoperoxidase. Similarly, it was demonstrated that
showed linear Henderson-Hasselbalch curves. From this it phosphate (which is a structural analogue of vanadate) forms
is inferred that the pL-dependent changes in EPR spectra of an inactive complex with apobromoperoxidase (de Boer et al.,
reduced bromoperoxidase are associated with the ionization 1986a).
of a functional group with an apparent pK, of about 5.4. It Earlier it was reported that vanadium was incorporated in
may very well be that protonation of this group is responsible the nitrogenases from Azotobacter uinelandii and Azotobacter
for the observed inhibition of enzymic activity at lower pH chroococcum (Benemann et al., 1972). For N, fixation these
values (Wever et al., 1985). bacteria require a molybdenum-iron cofactor which is a
The isotropic EPR spectral parameters go and A . are largely constituent of the nitrogenase system. It was demonstrated
determined by the vanadium(1V) ligand field strength. For that, under conditions of shortage of molybdenum, vanadium
instance, weak-field VO(H,O)sZ' gives rise to go = 1.964 and was taken up by the nitrogenase enzyme complexes from these
A . = 109.1 X cm-', whereas the strong-field complex Azotobacter species. Although vanadium was present in the
VO(S,CCN)22- yields go = 1.992 and A, = 67.3 X lo4 cm-' nitrogenase system, it did not appear to fulfil a role as a
(Boucher et al., 1969; Chasteen, 1981). From these data it catalyst. Moreover, it was proposed that the effect of vana-
is clear that the spread in go is relatively small, whereas A , dium incorporation could be accounted for by structural
decreases considerably with increasing field strength. On going aspects, such as stabilizing the enzyme and affecting some of
from low to high pH, the A . of reduced bromoperoxidase its catalytic properties (Burns et al., 1971; Benemann et al.,
decreased by about 4.9 X cm-I, and according to the 1972).
Recently, evidence has been presented that both A . uine-
landii and A . chroococcum possess two nitrogen fixation
After reduction with sodium dithionite the sample preparations systems: the conventional molybdenum nitrogenase and an
turned green, obviously due to the formation of paramagnetic
[MoOCI5I2-(Cotton & Wilkinson, 1972). This complex exhibited an alternative nitrogen-reducingenzyme complex. The alternative
intense rhombic signal around g = 1.92, and for this reason it was not nitrogenase complexes from both species were shown to contain
possible to obtain an EPR signal derived from any molybdenum(V)- a vanadium-iron cofactor, analogous to the MoFe cofactor
enzyme complex. This strong EPR signal of the molybdenum complex in the conventional enzymes. These VFe nitrogenases com-
did not disturb vanadium(1V) bromoperoxidase signal intensities which
were obtained from the central m, = hyperfine line, situated around pletely differed from the conventional nitrogenases with regard
g = 2.0. Signal intensities of vanadium(1V) bromoperoxidase were not to substrate specificity and molecular weights (Hales et al.,
affected by the presence of molybdenum, indicating that after addition 1986; Robson et al., 1986; Arber et al., 1987). In contrast
of - 5 mM sodium dithionite to the preparations containing 0-50 mM to vanadium bromoperoxidase the alternative nitrogenase from
molybdate, vanadium(V) bromoperoxidase was reduced prior to molyb- A . uinelandii did not exhibit an EPR signal derived from the
denum(V1). At neutral pH free V02+ (Le., not bound to the enzyme)
will precipitate due to the formation of VO(OH),. Vanadyl hydroxide vanadium ion (Hales et al., 1986). Since structural aspects
is not observable by EPR (Chasteen, 1981). of the vanadium( IV) ion in reduced bromoperoxidase can
EPR STUDIES ON REDUCED VANADIUM BROMOPEROXIDASE VOL. 27, NO. 5 , 1988 1635
readily be studied by EPR spectroscopy, this technique provides De Boer, E., van Kooyk, Y., Tromp, M. G. M., Plat, H., &
a convenient tool to obtain information on the metal-binding Wever, R. (1986a) Biochim. Biophys. Acta 869, 48-53.
site of this vanadium-containing enzyme. De Boer, E., Tromp, M. G. M., Plat, H., Krenn, G. E., &
Registry No. V, 7440-62-2; VO?-,14333-18-7; Mo04-, 14259- Wever, R. (1986b) Biochim. Biophys. Acta 872, 104-1 15.
85-9; bromoperoxidase, 69279-19-2. Fitzgerald, J. J., & Chasteen, N. D. (1974) Biochemistry 13,
4338-4347.
REFERENCES Hales, B. J., Case, E. E., Morningstar, J. E., Dzeda, M. F.,
Aasa, R., & VanngPrd, T. (1975) J . Magn. Reson. 19, & Mauterer, L. A. (1986) Biochemistry 25, 7251-7255.
308-3 15. Hewson, W. D., & Hager, L. P. (1980) J . Phycol. 16,
Ahern, T. J., Allan, G. G., & Metcalf, D. G. (1980) Biochim. 340-345.
Biophys. Acta 616, 329-339. Itoh, N., Izumi, Y., & Yamada, H. (1986) J. Biol. Chem. 261,
Albanese, N. F., & Chasteen, N . D. (1978) J . Phys. Chem. 5194-5200.
82, 910-914. Krenn, B. E., Plat, H., & Wever, R. (1987) Biochim. Biophys.
Albracht, S. P. J. (1984) Curr. Top. Bioenerg. 13, 79-106. Acta 912, 287-291.
Arber, J. M., Dobson, B. R., Eady, R. R., Stevens, P., Hasnain, Lowry, 0. H., Rosebrough, N. J., Farr, A. L., & Randall, R.
S. S., Garner, C. D., & Smith, B. E. (1987) Nature (Lon- J. (1951) J . Biol. Chem. 193, 265-275.
don) 325, 372-374. Morris, D. R., & Hager, L. P. (1966) J . Biol. Chem. 241,
Beinert, H., & Albracht, S. P. J. (1982) Biochim. Biophys. 1763-1 768.
Acta 683, 245-277. Morrison, M., & Schonbaum, G. R. (1976) Annu. Rev. Bio-
Benemann, J. R., McKenna, C. E., Lie, R. F., Taylor, T. G., chem. 45, 861-888.
& Kamen, M. D. (1972) Biochim. Biophys. Acta 264, Orii, Y., & Morita, M. (1977) J . Biochem. (Tokyo) 81,
25-38. 163-1 68.
Bolscher, B. G. J. M., Plat, H., & Wever, R. (1984) Biochim. Penefsky, H. S. (1979) Methods Enzymol. 56, 527-530.
Biophys. Acta 784, 177-186. Robson, R. L., Eady, R. R., Richardson, T. H., Miller, R. W.,
Boucher, L. J., Tynan, E. C., & Yen, T. F. (1969) in Electron Hawkins, M., & Postgate, J. R. (1986) Nature (London)
Spin Resonance of Metal Complexes (Yen, T. F., Ed.) pp 322, 388-390.
11 1-1 30, Plenum Press, New York. Steinhardt, J., & Reynolds, J. A. (1969) in Multiple Equilibria
Brill, A. S. (1977) in Transition Metals in Biochemistry in Proteins (Horecker, B., Kaplan, N . O., & Scheraga, H.
(Kleinzeller, A., Springer, G. F., & Wittmann, H. G., Eds.) A., Eds.) Academic Press, New York and London.
Springer-Verlag, New York. Van Pee, K.-H., & Lingens, F. (1985) J . Bacteriol. 161,
Burns, R. C., Fuchsman, W. H., & Hardy, R. W. F. (1971) 1171-1175.
Biochem. Biophys. Res. Commun. 42, 353-358. Vilter, H . (1983) Bot. Mar. 26, 451-455.
Chasteen, N . D. (1981) in Biological Magnetic Resonance Vilter, H . (1984) Phytochemistry 23, 1387-1 390.
(Berliner, L., & Reuben, J., Eds.) Vol. 3, pp 53-1 19, Plenum Vilter, H., & Rehder, D. (1987) Inorg. Chim. Acta 136,
Press, New York. L7-L10.
Chasteen, N. D., & Hanna, M. W. (1972) J. Phys. Chem. 76, Wever, R., Van Drooge, J. H., Van Ark, G., & Van Gelder,
3951-3958. B. F. (1974) Biochim. Biophys. Acta 347, 215-223.
Chasteen, N. D., DeKoch, R. J., Rogers, B. L., & Hanna, M. Wever, R., Plat, H., & de Boer, E. (1985) Biochim. Biophys.
W. (1973) J . A m . Chem. SOC.95, 1301-1309. Acta 830, 181-186.
Chasteen, N. D., White, L. K., & Campbell, R. F. (1977) Wever, R., de Boer, E., Plat, H., & Krenn, B. E. (1987) FEBS
Biochemistry 16, 363-368. Lett. 216, 1-3.
Cotton, F. A., & Wilkinson, G. (1972) in Advanced Inorganic White, L. K., & Chasteen, N . D. (1979) J . Phys. Chem. 83,
Chemistry, Wiley, New York. 279-284.