Vol. 167, No. 2, 1990 March 16, 1990
ACTIVATION
BIOCHEMICAL
OF THE IRON-CONTAINING
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 813-818
82 PROTEIN OF RIBONUCLEOTIDE
BY HYDROGEN Margareta
Sahlin’,
Britt-Marie
Sjoberg’,
REDUCTASE
PEROXIDE Gabriele Backest, Thomas Loehrt, and
Joann Sanders-Loehrt *Department tDepartment
of Molecular
Biology, University of Stockholm, S-i 0691 Stockholm, Sweden
of Chemical and Biological Technology,
Received
January
11,
Sciences, Oregon Graduate
Institute of Science and
Beaverton, Oregon 97006-l 999
1990
The active form of protein 82, the small subunit of ribonucleotide reductase, contains two dinuclear Fe(lll) centers and a tyrosyl radical. The inactive metB2 form also contains the same diferric complexes but lacks the tyrosyl radical. We now demonstrate that incubation of metB2 with hydrogen peroxide generates the tyrosyl radical. The reaction is optimal at 5.5 mM hydrogen peroxide, with a maximum of 25-30% tyrosyl radical being formed after approximately 1.5 hr of incubation. The activation reaction is counteracted by a hydrogen peroxide-dependent reduction of the tyrosyl radical. It is likely that the generation of the radical proceeds via a ferry1 intermediate, as in the proposed mechanisms for cytochrome P450 and the peroxidases. @1990 Academic Press, Inc.
Ribonucleoside diphosphate reductase (EC 1 ,17.4.1) from Escherichia co/i consists of two different homodimeric proteins, Bl and B2. The active enzyme is a 1:l complex of the two proteins. In its active form, the 82 protein carries two u-0x0 bridged dinuclear Fe(lll) centers and a tyrosyl radical, that is localized to tyrosine residue 122 (l-4). It has long been known that the enzymatic activity is directly correlated to the tyrosyl radical content (5). Although the enzymatic reaction mechanism for conversion of ribonucleotides to deoxyribonucleotides is not known in detail, a substrate radical intermediate has been demonstrated (6). An inactive form of B2, known as metB2, tyrosyl radical (7,, 8). Hitherto, the only pathway to generate
has intact ferric irons but no a tyrosyl radical was via an
02-dependent
The
oxidation
of the diferrous
form (reduced
82).
latter is obtained
either by
introduction of ferrous ions to apoB2, the metal-free form (7), or by chemical (4,8) or enzymatic (9) reduction of the diferric form. Upon exposure of the diferrous form to molecular oxygen the tyrosyl radical appears concomitantly with the oxidation of the irons (8, 10). In this study we show the tyrosyl
radical,
generating
that the addition a B2 species
of H202
to diferric
metB2
that is spectroscopically
leads to formation identical
to active 0006-291X/W
813
of B2. $1.50
Copyright 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.
Vol. 167, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
MATERIALS AND METHODS Preparations of the different forms of protein 82 were performed as described earlier (7, 8, 11). A molar absorptivity of 120,000 M-l cm-l (for 280 nm minus 310 nm) has been used for all B2 preparations. Bovine liver catalase, bovine serum albumin, hydrogen peroxide, peracetic acid and I-butylhydroperoxide were obtained from Sigma. lodosobenzene was prepared as described previously (12). Absorption spectra of Ii-25 uM 82 in 50 mM Tris-HCI, pH 7.6 were recorded on a PerkinElmer Lambda 2 or a Hewlett-Packard 8452A Diode Array spectrophotometer. The amount of tyrosyl radical formed during H202 incubation was determined from difference spectra obtained by subtracting the initial spectrum and using extinction coefficients according to Petersson et al. (10). EPR spectra of 24.5 pM 82 in 50 mM Tris-HCI, pH 7.6 at 20K were recorded on a Bruker ESP 300. Quantitation of radical with EPR was performed as described earlier (10, 13). Catalase assays were performed as described previously (14) using E=43.6 M“ cm-l at 240 nm (15), but with 2-6 uM protein and 5 mM H,O, in 50 mM Tris-HCI, pH 7.6. Catalase control experiments contained: bovine serum albumin, 2-10 uM; Fe*+ or Fe3+, 1 PM; EDTA, 1.5 pM; and glycerol, 0.25 M. Iron analyses were performed as described previously (7), except that an incubation at room temperature for 10 min in 0.4 M HCI was added for extraction of the protein-bound iron. RESULTS Incubation of the metB2 form of the small subunit of ribonucleotide mM H202 generates the tyrosyl radical form: active 82. The formation is evident from the appearance spectrum (1, 10). A spectrum
reductase with 5.5 of the tyrosyl radical
of the typical sharp 410-nm band in the visible absorption of metB2 treated with H202 is shown in Figure lB, together
with the corresponding spectra of active B2 (Fig. 1A) and untreated metB2 (Fig. IC). It is evident that the incubation produces tyrosyl radical without any change in the iron-related
0
01 0
2
2
1 Time
of incubation
3 (hr)
Fiaure 1. Absorption spectra of (A) active B2, (B) metB2 after 99 min of incubation with 5.5 mM H,O, at 25”C, and (C) metB2. Protein concentration was 11 PM. The absorption range was 0.15 absorbance units full scale. An arbitary vertical offset has been used for all spectra in order increase clarity. Fiaure 2, Time- and concentration-dependent reactivation of metB2 by hydrogen peroxide. MetB2 (11 PM) was incubated with hydrogen peroxide at 25”C, and spectra were recorded at the indicated times. Hydrogen peroxide concentrations were: 0 , 2.5 mM; W , 5.5 mM; 3 , IO mM; 0 , 20 mM.
814
Vol. 167, No. 2, 1990
03
HhT1 329
326
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
332
1
2 Time
of
incubation
3 (hr)
Fiaure 3. EPR spectra obtained at 20K with (A) active 82, 1.5 mM tyrosyl radical; (B) metB2 after 168 min of incubation with 5.5 mM hydrogen peroxide at 25”C, and (C) metB2. Protein concentrations in (B) and (C) were 25 PM. Recording conditions were: microwave power, 12.6 uW; microwave frequency, 9.232 GHz; scan time, 671s; modulation amplitude, 1G (A), 1.6 G (B and C); gain 1.25x1 O4 (A) 2x1 O6 (B and C). Fiaure 4. Time dependence of the reactrvation of metB2 by hydrogen peroxide. MetB2 (24.5 uM) was incubated in a cuvette with 5.5 mM hydrogen peroxide at 25°C and absorption spectra were recorded. At the indicated tim.es aliquols were withdrawn and quickly frozen in liquid nitrogen for EPR measurements at 20K. Tyrosyl radical concentration was determined from the A4,e (0) and from the double integrated intensity of the g=2 EPR signal (0).
absorption
bands at 325 nm and 370 nm.
A lower concentration
final concentration of radical (Fig. 2). The radical formation is close to 5 mM, and the 0.29 tyrosyl radical per 82 protein after 1.5 radical can be fitted to pseudo first order the incubation with 2.5 and 5.5 mM H202, higher
concentrations
of H202
of H202
results in a lower
optimal hydrogen peroxide concentration for activation of 11 uM metB2 reaches a plateau of hr of incubation. The generation of tyrosyl kinetics, with a time constant of 5-7~10.~ s-l for However, it is evident from the experiments with
(Fig. 2) that the reactivation
peroxide-dependent inactivation of the protein. Another measure of tyrosyl radical formation
process
in ribonucleotide
is also counteracted reductase
by a
is the
appearance of a g=2 doublet in the EPR spectrum (4). As can be seen in Figure 38, treatment of metB2 with 5.5 mM H202 results in an EPR signal which is similar to that of active 82 (Fig. 3A) and absent from the spectrum of the metB2 starting material (Fig. 3C). The doublet EPR signal generated by incubation of metB2 with H202 shows all the hyperfine characteristics of the tyrosyl radical in active 82. In one experiment the peroxide-dependent formation of the
BIOCHEMICAL
Vol. 167, No. 2, 1990 tyrosyl radical was monitored incubation of 24.5 uM metB2
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
concomitantly with EPR and light absorption (Fig. 4). The resulted in similar time curves. After 3 hr with 5.5 mM H202
the amount of tyrosyl radical per 82 protein was 0.21 as judged by EPR an 0.22 as judged by light absorption. The EPR spectra at 20 K exhibit a g=4.3 EPR signal that is indicative of a small amount of adventitiously bound iron. During the incubation with H202 the adventitious iron increases
about 2-fold.
lost from metB2
during
Iron analysis
the incubation
showed
that not more than 1-2 % of the iron was
with H202. DISCUSSION
Several redox-reactions of the iron center in protein B2 are known (8). It was recently shown that reduced 82 is an intermediate in the enzymatic reactivation of metB2 occurring in E. co/i cells (16). In vitro oxidative reactivation of reduced 82 is achieved by molecular oxygen (7, 8). A possible pathway for this latter reaction is depicted in Figure 5. An initial ferric peroxide intermediate is proposed for the reaction of the iron-tyrosyl site of reduced B2 with molecular oxygen (reaction l), analogous to the oxygenation of deoxyhemerythrin (3, 17). Breakdown to a high valent iron-ox0 intermediate (reaction 2) would lead to the formation of the tyrosyl radical (reaction 3). This latter pathway is analogous to the peroxidase-catalyzed oxidation of phenols (18). The actual source of reducing equivalents for reaction 3 is not known, and it is possible that OH* is produced instead of H20. We have now shown
that the addition
of H202
to the diferric
metB2
species
also leads to formation
of the
tyrosyl radical, a finding which supports the hypothesis outlined in Figure 5. It is likely that the peroxide-dependent production of tyrosyl radical proceeds via reactions 4, 2, and 3 (Fig. 5). The overall scheme with its high-valent metal-oxo intermediate is similar to mechanisms proposed for the heme-containing peroxidases (18) and cytochrome P-450 the dinuclear iron protein, methane monooxygenase (21), and the dinuclear pseudocatalase (22). Wt,y is only 25-30% tyrosyl radical formed during optimal conditions? reaction 5, which will counteract reaction 3 (Fig. 5). All 82 preparations metB2,
and apoB2)
have a low but significant
catalase
activity’
(19, 20), as well as manganese protein, One explanation is tested (active B2,
and O2 gas bubbles
accumulate during the reaction. It was also observed that incubation of active B2 (5.5 yM) with 5.5 mM H202 leads to a 30% decrease in radical content (to 4 FM) over a period of 1 hr at 25°C. This observation suggests that there is a competing radical. The rate of destruction is increased by increasing mM (Fig. 2).
It is likely that the protein
itself is being
process which destroys the H202 concentration
inactivated
by H202
the to lo-20
or other
byproducts.2 1
The specific catalase activity is similar in all forms of 82, but is absent from controls containing bovine serum albumin f Fe2+, or glycerol f Fe3+ and EDTA. Addition of Fez+ to apoB2 did not increase its catalase activity. The low catalase activity (-40 moles H202~min1~mole 82-l) of all B2 preparations, including apoB2, is 10m5 of that observed for bovine liver catalase and 1O-4 of that observed for Mn catalase (22). 2 The catalase activity of 82 also decreases gradually the added hydrogen peroxide has been consumed. 816
with time and levels off before
50% of
Vol.
167, No. 2, 1990
BIOCHEMICAL
Fe(ll) Fe(ll) Tv Reduced
\
02 (‘I
82
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Fe(lll) Fe(III)-OOH I Tyr
I
PeroxIde bntermediate
‘2’
Active 82
Hz02
< ’
(4)
Fe(III) Fe(lll) Tyr Met 82
H2 0
FWQJl intermedlate
Fiaure 5. Proposed reaction pathways for tyrosyl radical formalion and peroxide decomposition by protein 82. Proton stoichiometry omitted for clarity.
Since the generation of the tyrosyl radical appears to involve a ferry1 intermediate, it should be possible to obtain activation of metB2 with other peroxide or oxene donors. For example, the bleomycin-dependent activation of an algal ribonucleotide reductase (23) may be caused by an oxygen-generated peroxide species of the iron-bleomycin complex (24,251). In the present study, metB2 was incubated in the presence of mM concentrations of the oxene donors peracetic acid, I-butylhydroperoxide or iodosobenzene, but all failed to produce a tyrosyl radical signal. The reason may be that the iron center of protein 82 from E. co/i is inaccessible to negatively charged or bulky reagents (8, 26), a hypothesis which is supported by the preliminary x-ray crystal structure of protein 82 (P. Nordlund, B.-M. Sjbberg & H. Eklund, manuscript in preparation). ACKNOWLEDGMENTS We thank Drs. N. J. Blackburn and V. Renganathan for helpful discussions. This work was supported by the Swedish Cancer Society (B.-M.S.), Wallenbergs Jubileumsfond, Resebidrag (MS.) and the U.S. Public Health Service, National Institutes of Health (GM 18865, T.M.L. and JS-L.). REFERENCES 1. 2. 3.
Reichard. P. and Ehrenberg, A. (1983) Science 221, 514-519. Sjoberg, B.-M., and Graslund, A. (1983) Adv. Inorg. Biochem. 5, 87-l 10. Sanders-Loehr, J. (1989) In Iron Carriers and Iron Proteins (T. M. Loehr, Ed.) pp. 374466, VCH, New York. 4. Lynch, J.B., Juarez-Garcia, C., Munck, E. and Que, L., Jr. (1989) J. Biol. Chem. 264, 8091-8096 5. Ehrenberg, A., and Reichard, P. (1972) J. Biol. Chem. 247, 3485-3488. 6. Sjdberg, B.-M., Graslund, A., and Eckstein, F. (1983) J. Biol. Chem. 258, 8060-8067. 7. Atkin, C.L., Thelander, L., Reichard, P., and Lang, G. (1973) J. Biol. Chem. 248, 7464-7472. 8. Sahlin.M., Graslund, A., Petersson, L., Ehrenberg, A., Sjoberg, B.-M. (1989) Biochemistry 28, 2618-2625. 9. Fontecave, M., Eliasson, R. and Reichard, P. (1989) J. Biol. Chem. 264, 9164-9170. IO. Petersson, L., Graslund, A., Ehrenberg, A., Sjoberg, B.-M. and Reichard, P. (1980) J. Biol. Chem. 255, 6706-6712. 817
Vol.
167, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
11. Sjoberg, B.-M., Hahne, S., Karlsson, M., Jornvall, H., Gbransson, M. and Uhlin, B.E. (1986) J. Biol. Chem. 261, 5658-5662. 12. Lucas, H.J., and Kennedy, E.R. (1955) Org. Synth. Coil. 3, 482-485. 13. Sahlin, M., Petersson, L., Graslund, A., Ehrenberg, A., Sjoberg, B.-M. and Thelander, L. (1987) Biochemistry 26, 5541-5548. 14. Beers, R.F., Jr., and Sizer, I.W. (1952) J. Biol. Chem. 195, 133-140. 15. Bergmayer, H.U., Gawehn, K., and Grassl, M. (1970) In Methoden der Enzymatischen Analyse (H.U. Bergmeyer, Ed.), Vol. 1, p. 440, Verlag Chemie, Weinheim. 16. Fontecave, M., Eliasson, R., and Reichard, P. (1987) J. Biol. Chem. 262, 12325-12331. 17. Shiemke, A.K., Loehr, T.M., and Sanders-Loehr, J. (1986) J. Am. Chem. Sot. 108, 2437-2443. 18. Ortiz de Montellano, P.R. (1987) Act. Chem. Res. 20, 289-294. 19. Hrycry, E.G., Gustavsson, J.-A., Ingelman-Sundberg, M. and Ernster, L. (1975) Biochem. Biophys. Res. Commun. 66, 209-216. 20. Murray, R.I., Fisher, M.T., Debrunner, P.G., and Sligar, S.G. (1985) In Metalloproteins (P. Harrison, Ed.) Part 1, pp. 157-206, Verlag Chemie, Weinheim. 21 Fox, B.G., Froland, W.A., Dege, J.E., and Lipscomb, J.D. (1989) J. Biol. Chem. 264, 10023-l 0033 22. Kono, Y., and Fridovich, I. (1983) J. Biol. Chem. 258, 6015-6019. 23. Hofmann, R. and Follmann, H. (1985) 2. Naturforsch. 4Oc, 919-925. 24. Stubbe, J. and Kozarich, J.W. (1987) Chem. Rev. 87, 1107-l 136. 25. Umezawa, H. and Takita, T. (1980) Struct. Bonding (Berlin) 40, 73-100. 26. Kjoller-Larsen, I., Sjoberg, B.-M. and Thelander, L. (1982) Eur. J. Biochem. 125, 75-81.
818
ACTIVATION
BIOCHEMICAL
OF THE IRON-CONTAINING
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 813-818
82 PROTEIN OF RIBONUCLEOTIDE
BY HYDROGEN Margareta
Sahlin’,
Britt-Marie
Sjoberg’,
REDUCTASE
PEROXIDE Gabriele Backest, Thomas Loehrt, and
Joann Sanders-Loehrt *Department tDepartment
of Molecular
Biology, University of Stockholm, S-i 0691 Stockholm, Sweden
of Chemical and Biological Technology,
Received
January
11,
Sciences, Oregon Graduate
Institute of Science and
Beaverton, Oregon 97006-l 999
1990
The active form of protein 82, the small subunit of ribonucleotide reductase, contains two dinuclear Fe(lll) centers and a tyrosyl radical. The inactive metB2 form also contains the same diferric complexes but lacks the tyrosyl radical. We now demonstrate that incubation of metB2 with hydrogen peroxide generates the tyrosyl radical. The reaction is optimal at 5.5 mM hydrogen peroxide, with a maximum of 25-30% tyrosyl radical being formed after approximately 1.5 hr of incubation. The activation reaction is counteracted by a hydrogen peroxide-dependent reduction of the tyrosyl radical. It is likely that the generation of the radical proceeds via a ferry1 intermediate, as in the proposed mechanisms for cytochrome P450 and the peroxidases. @1990 Academic Press, Inc.
Ribonucleoside diphosphate reductase (EC 1 ,17.4.1) from Escherichia co/i consists of two different homodimeric proteins, Bl and B2. The active enzyme is a 1:l complex of the two proteins. In its active form, the 82 protein carries two u-0x0 bridged dinuclear Fe(lll) centers and a tyrosyl radical, that is localized to tyrosine residue 122 (l-4). It has long been known that the enzymatic activity is directly correlated to the tyrosyl radical content (5). Although the enzymatic reaction mechanism for conversion of ribonucleotides to deoxyribonucleotides is not known in detail, a substrate radical intermediate has been demonstrated (6). An inactive form of B2, known as metB2, tyrosyl radical (7,, 8). Hitherto, the only pathway to generate
has intact ferric irons but no a tyrosyl radical was via an
02-dependent
The
oxidation
of the diferrous
form (reduced
82).
latter is obtained
either by
introduction of ferrous ions to apoB2, the metal-free form (7), or by chemical (4,8) or enzymatic (9) reduction of the diferric form. Upon exposure of the diferrous form to molecular oxygen the tyrosyl radical appears concomitantly with the oxidation of the irons (8, 10). In this study we show the tyrosyl
radical,
generating
that the addition a B2 species
of H202
to diferric
metB2
that is spectroscopically
leads to formation identical
to active 0006-291X/W
813
of B2. $1.50
Copyright 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.
Vol. 167, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
MATERIALS AND METHODS Preparations of the different forms of protein 82 were performed as described earlier (7, 8, 11). A molar absorptivity of 120,000 M-l cm-l (for 280 nm minus 310 nm) has been used for all B2 preparations. Bovine liver catalase, bovine serum albumin, hydrogen peroxide, peracetic acid and I-butylhydroperoxide were obtained from Sigma. lodosobenzene was prepared as described previously (12). Absorption spectra of Ii-25 uM 82 in 50 mM Tris-HCI, pH 7.6 were recorded on a PerkinElmer Lambda 2 or a Hewlett-Packard 8452A Diode Array spectrophotometer. The amount of tyrosyl radical formed during H202 incubation was determined from difference spectra obtained by subtracting the initial spectrum and using extinction coefficients according to Petersson et al. (10). EPR spectra of 24.5 pM 82 in 50 mM Tris-HCI, pH 7.6 at 20K were recorded on a Bruker ESP 300. Quantitation of radical with EPR was performed as described earlier (10, 13). Catalase assays were performed as described previously (14) using E=43.6 M“ cm-l at 240 nm (15), but with 2-6 uM protein and 5 mM H,O, in 50 mM Tris-HCI, pH 7.6. Catalase control experiments contained: bovine serum albumin, 2-10 uM; Fe*+ or Fe3+, 1 PM; EDTA, 1.5 pM; and glycerol, 0.25 M. Iron analyses were performed as described previously (7), except that an incubation at room temperature for 10 min in 0.4 M HCI was added for extraction of the protein-bound iron. RESULTS Incubation of the metB2 form of the small subunit of ribonucleotide mM H202 generates the tyrosyl radical form: active 82. The formation is evident from the appearance spectrum (1, 10). A spectrum
reductase with 5.5 of the tyrosyl radical
of the typical sharp 410-nm band in the visible absorption of metB2 treated with H202 is shown in Figure lB, together
with the corresponding spectra of active B2 (Fig. 1A) and untreated metB2 (Fig. IC). It is evident that the incubation produces tyrosyl radical without any change in the iron-related
0
01 0
2
2
1 Time
of incubation
3 (hr)
Fiaure 1. Absorption spectra of (A) active B2, (B) metB2 after 99 min of incubation with 5.5 mM H,O, at 25”C, and (C) metB2. Protein concentration was 11 PM. The absorption range was 0.15 absorbance units full scale. An arbitary vertical offset has been used for all spectra in order increase clarity. Fiaure 2, Time- and concentration-dependent reactivation of metB2 by hydrogen peroxide. MetB2 (11 PM) was incubated with hydrogen peroxide at 25”C, and spectra were recorded at the indicated times. Hydrogen peroxide concentrations were: 0 , 2.5 mM; W , 5.5 mM; 3 , IO mM; 0 , 20 mM.
814
Vol. 167, No. 2, 1990
03
HhT1 329
326
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
332
1
2 Time
of
incubation
3 (hr)
Fiaure 3. EPR spectra obtained at 20K with (A) active 82, 1.5 mM tyrosyl radical; (B) metB2 after 168 min of incubation with 5.5 mM hydrogen peroxide at 25”C, and (C) metB2. Protein concentrations in (B) and (C) were 25 PM. Recording conditions were: microwave power, 12.6 uW; microwave frequency, 9.232 GHz; scan time, 671s; modulation amplitude, 1G (A), 1.6 G (B and C); gain 1.25x1 O4 (A) 2x1 O6 (B and C). Fiaure 4. Time dependence of the reactrvation of metB2 by hydrogen peroxide. MetB2 (24.5 uM) was incubated in a cuvette with 5.5 mM hydrogen peroxide at 25°C and absorption spectra were recorded. At the indicated tim.es aliquols were withdrawn and quickly frozen in liquid nitrogen for EPR measurements at 20K. Tyrosyl radical concentration was determined from the A4,e (0) and from the double integrated intensity of the g=2 EPR signal (0).
absorption
bands at 325 nm and 370 nm.
A lower concentration
final concentration of radical (Fig. 2). The radical formation is close to 5 mM, and the 0.29 tyrosyl radical per 82 protein after 1.5 radical can be fitted to pseudo first order the incubation with 2.5 and 5.5 mM H202, higher
concentrations
of H202
of H202
results in a lower
optimal hydrogen peroxide concentration for activation of 11 uM metB2 reaches a plateau of hr of incubation. The generation of tyrosyl kinetics, with a time constant of 5-7~10.~ s-l for However, it is evident from the experiments with
(Fig. 2) that the reactivation
peroxide-dependent inactivation of the protein. Another measure of tyrosyl radical formation
process
in ribonucleotide
is also counteracted reductase
by a
is the
appearance of a g=2 doublet in the EPR spectrum (4). As can be seen in Figure 38, treatment of metB2 with 5.5 mM H202 results in an EPR signal which is similar to that of active 82 (Fig. 3A) and absent from the spectrum of the metB2 starting material (Fig. 3C). The doublet EPR signal generated by incubation of metB2 with H202 shows all the hyperfine characteristics of the tyrosyl radical in active 82. In one experiment the peroxide-dependent formation of the
BIOCHEMICAL
Vol. 167, No. 2, 1990 tyrosyl radical was monitored incubation of 24.5 uM metB2
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
concomitantly with EPR and light absorption (Fig. 4). The resulted in similar time curves. After 3 hr with 5.5 mM H202
the amount of tyrosyl radical per 82 protein was 0.21 as judged by EPR an 0.22 as judged by light absorption. The EPR spectra at 20 K exhibit a g=4.3 EPR signal that is indicative of a small amount of adventitiously bound iron. During the incubation with H202 the adventitious iron increases
about 2-fold.
lost from metB2
during
Iron analysis
the incubation
showed
that not more than 1-2 % of the iron was
with H202. DISCUSSION
Several redox-reactions of the iron center in protein B2 are known (8). It was recently shown that reduced 82 is an intermediate in the enzymatic reactivation of metB2 occurring in E. co/i cells (16). In vitro oxidative reactivation of reduced 82 is achieved by molecular oxygen (7, 8). A possible pathway for this latter reaction is depicted in Figure 5. An initial ferric peroxide intermediate is proposed for the reaction of the iron-tyrosyl site of reduced B2 with molecular oxygen (reaction l), analogous to the oxygenation of deoxyhemerythrin (3, 17). Breakdown to a high valent iron-ox0 intermediate (reaction 2) would lead to the formation of the tyrosyl radical (reaction 3). This latter pathway is analogous to the peroxidase-catalyzed oxidation of phenols (18). The actual source of reducing equivalents for reaction 3 is not known, and it is possible that OH* is produced instead of H20. We have now shown
that the addition
of H202
to the diferric
metB2
species
also leads to formation
of the
tyrosyl radical, a finding which supports the hypothesis outlined in Figure 5. It is likely that the peroxide-dependent production of tyrosyl radical proceeds via reactions 4, 2, and 3 (Fig. 5). The overall scheme with its high-valent metal-oxo intermediate is similar to mechanisms proposed for the heme-containing peroxidases (18) and cytochrome P-450 the dinuclear iron protein, methane monooxygenase (21), and the dinuclear pseudocatalase (22). Wt,y is only 25-30% tyrosyl radical formed during optimal conditions? reaction 5, which will counteract reaction 3 (Fig. 5). All 82 preparations metB2,
and apoB2)
have a low but significant
catalase
activity’
(19, 20), as well as manganese protein, One explanation is tested (active B2,
and O2 gas bubbles
accumulate during the reaction. It was also observed that incubation of active B2 (5.5 yM) with 5.5 mM H202 leads to a 30% decrease in radical content (to 4 FM) over a period of 1 hr at 25°C. This observation suggests that there is a competing radical. The rate of destruction is increased by increasing mM (Fig. 2).
It is likely that the protein
itself is being
process which destroys the H202 concentration
inactivated
by H202
the to lo-20
or other
byproducts.2 1
The specific catalase activity is similar in all forms of 82, but is absent from controls containing bovine serum albumin f Fe2+, or glycerol f Fe3+ and EDTA. Addition of Fez+ to apoB2 did not increase its catalase activity. The low catalase activity (-40 moles H202~min1~mole 82-l) of all B2 preparations, including apoB2, is 10m5 of that observed for bovine liver catalase and 1O-4 of that observed for Mn catalase (22). 2 The catalase activity of 82 also decreases gradually the added hydrogen peroxide has been consumed. 816
with time and levels off before
50% of
Vol.
167, No. 2, 1990
BIOCHEMICAL
Fe(ll) Fe(ll) Tv Reduced
\
02 (‘I
82
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Fe(lll) Fe(III)-OOH I Tyr
I
PeroxIde bntermediate
‘2’
Active 82
Hz02
< ’
(4)
Fe(III) Fe(lll) Tyr Met 82
H2 0
FWQJl intermedlate
Fiaure 5. Proposed reaction pathways for tyrosyl radical formalion and peroxide decomposition by protein 82. Proton stoichiometry omitted for clarity.
Since the generation of the tyrosyl radical appears to involve a ferry1 intermediate, it should be possible to obtain activation of metB2 with other peroxide or oxene donors. For example, the bleomycin-dependent activation of an algal ribonucleotide reductase (23) may be caused by an oxygen-generated peroxide species of the iron-bleomycin complex (24,251). In the present study, metB2 was incubated in the presence of mM concentrations of the oxene donors peracetic acid, I-butylhydroperoxide or iodosobenzene, but all failed to produce a tyrosyl radical signal. The reason may be that the iron center of protein 82 from E. co/i is inaccessible to negatively charged or bulky reagents (8, 26), a hypothesis which is supported by the preliminary x-ray crystal structure of protein 82 (P. Nordlund, B.-M. Sjbberg & H. Eklund, manuscript in preparation). ACKNOWLEDGMENTS We thank Drs. N. J. Blackburn and V. Renganathan for helpful discussions. This work was supported by the Swedish Cancer Society (B.-M.S.), Wallenbergs Jubileumsfond, Resebidrag (MS.) and the U.S. Public Health Service, National Institutes of Health (GM 18865, T.M.L. and JS-L.). REFERENCES 1. 2. 3.
Reichard. P. and Ehrenberg, A. (1983) Science 221, 514-519. Sjoberg, B.-M., and Graslund, A. (1983) Adv. Inorg. Biochem. 5, 87-l 10. Sanders-Loehr, J. (1989) In Iron Carriers and Iron Proteins (T. M. Loehr, Ed.) pp. 374466, VCH, New York. 4. Lynch, J.B., Juarez-Garcia, C., Munck, E. and Que, L., Jr. (1989) J. Biol. Chem. 264, 8091-8096 5. Ehrenberg, A., and Reichard, P. (1972) J. Biol. Chem. 247, 3485-3488. 6. Sjdberg, B.-M., Graslund, A., and Eckstein, F. (1983) J. Biol. Chem. 258, 8060-8067. 7. Atkin, C.L., Thelander, L., Reichard, P., and Lang, G. (1973) J. Biol. Chem. 248, 7464-7472. 8. Sahlin.M., Graslund, A., Petersson, L., Ehrenberg, A., Sjoberg, B.-M. (1989) Biochemistry 28, 2618-2625. 9. Fontecave, M., Eliasson, R. and Reichard, P. (1989) J. Biol. Chem. 264, 9164-9170. IO. Petersson, L., Graslund, A., Ehrenberg, A., Sjoberg, B.-M. and Reichard, P. (1980) J. Biol. Chem. 255, 6706-6712. 817
Vol.
167, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
11. Sjoberg, B.-M., Hahne, S., Karlsson, M., Jornvall, H., Gbransson, M. and Uhlin, B.E. (1986) J. Biol. Chem. 261, 5658-5662. 12. Lucas, H.J., and Kennedy, E.R. (1955) Org. Synth. Coil. 3, 482-485. 13. Sahlin, M., Petersson, L., Graslund, A., Ehrenberg, A., Sjoberg, B.-M. and Thelander, L. (1987) Biochemistry 26, 5541-5548. 14. Beers, R.F., Jr., and Sizer, I.W. (1952) J. Biol. Chem. 195, 133-140. 15. Bergmayer, H.U., Gawehn, K., and Grassl, M. (1970) In Methoden der Enzymatischen Analyse (H.U. Bergmeyer, Ed.), Vol. 1, p. 440, Verlag Chemie, Weinheim. 16. Fontecave, M., Eliasson, R., and Reichard, P. (1987) J. Biol. Chem. 262, 12325-12331. 17. Shiemke, A.K., Loehr, T.M., and Sanders-Loehr, J. (1986) J. Am. Chem. Sot. 108, 2437-2443. 18. Ortiz de Montellano, P.R. (1987) Act. Chem. Res. 20, 289-294. 19. Hrycry, E.G., Gustavsson, J.-A., Ingelman-Sundberg, M. and Ernster, L. (1975) Biochem. Biophys. Res. Commun. 66, 209-216. 20. Murray, R.I., Fisher, M.T., Debrunner, P.G., and Sligar, S.G. (1985) In Metalloproteins (P. Harrison, Ed.) Part 1, pp. 157-206, Verlag Chemie, Weinheim. 21 Fox, B.G., Froland, W.A., Dege, J.E., and Lipscomb, J.D. (1989) J. Biol. Chem. 264, 10023-l 0033 22. Kono, Y., and Fridovich, I. (1983) J. Biol. Chem. 258, 6015-6019. 23. Hofmann, R. and Follmann, H. (1985) 2. Naturforsch. 4Oc, 919-925. 24. Stubbe, J. and Kozarich, J.W. (1987) Chem. Rev. 87, 1107-l 136. 25. Umezawa, H. and Takita, T. (1980) Struct. Bonding (Berlin) 40, 73-100. 26. Kjoller-Larsen, I., Sjoberg, B.-M. and Thelander, L. (1982) Eur. J. Biochem. 125, 75-81.
818
