Transcript
Biochem. J. (1967) 103, 10 c
lOc
Mossbauer Evidence on the Form of Iron In Xanthine Oxidase By 0. E. JoHNsoN, P. F. KNowiEs and R. C. BRAY Solid State Phyeic8 Divi8ion, Atomic Energy Re8earch E8tabli8hment, Harwell, Berk8., and Che8ter Beatty Re8earch In8titute, In8stitute of Cancer Re8earch: Royal Cancer Ho8pital, London, S. W. 3 (Received 25 January 1967)
The iron of xanthine oxidase is related to that in mitochondrial systems (Beinert, 1965) and in the ferredoxins (Palmer, Sands & Mortenson, 1966; Palmer & Sands, 1966) and has previously been studied by ESR (Palmer, Bray & Beinert, 1964; Bray, Palmer & Beinert, 1964) and by magnetic susceptibility (Ehrenberg & Bray, 1965; Bray Pettersson & Ehrenberg, 1961). Reports (Bayer & Voelter, 1966; Piette, 1967) that iron can be removed from the enzyme have not been confirmed (L. I. Hart & R. C. Bray, unpublished work). The present communication describes the application of M6ssbauer absorption spectroscopy of 57Fe (Phillips, Knight & Blomstrom, 1965; Lang & Marshall, 1966) to studies on the state and bonding of iron in this enzyme. From the results, it is concluded that all eight atoms are affected on full reduction. In the reduced enzyme they could be either simple low-spin ferric atoms, an exchangecoupled ferric-ferrous system or low-spin ferrous interacting with a radical. The preparation of samples and Mossbauer technique are described elsewhere (Johnson, Bray & Knowles, 1967). Samples were stored under liquid N2 when not in use. M6ssbauer spectra were obtained for the enzyme: (i) in the oxidized state, (ii) when reduced enzymically with salicylaldehyde and (iii) when reduced with Na2S204. Preliminary results for oxidized and salicylaldehyde-reduced enzyme are presented elsewhere (Johnson et al. 1967). More recent data confirm the earlier finding that the spectrum is unaffected by salicylaldehyde, i.e. less than 10-20% of the iron is changed by reduction of the enzyme under these conditions. Spectra at 4-20K for oxidized and reduced specimens are shown in Figs. 1(a) and 1(b) and show a small quadrupole splitting (approx. 0.5mm./sec.) and chemical shift (approx. 0.2mm./sec. relative to metallic iron). Spectra at 1*5EK, 77°1 and 1950K were very similar. Spectra were also taken at 1.5'K and 4-2EK in an applied magnetic field of 30kG and showed a small broadening due to the field, but no changes giving evidence for an internal field at the nuclei were obtained, showing that the iron was mainly in a non-magnetic state. The effect of reduction with Na2S204 is shown in Figs. 1(c) and l(d) for spectra at 770i and 4-20K
respectively. These are very different from those of the oxidized and salicylaldehyde-reduced enzyme. At 4 20K the spectrum is complex with several peaks over a broad velocity range from about - 2mm./sec. to + 3mm./sec. It must arise from magnetic hyperfine interaction at the 57Fe nuclei and is observable at this temperature because electronspin relaxation times are long compared with the nuclear precession time (approx. 10-8 sec.). At 770K (Fig. 1c) the relaxation times have become shorter and the hyperfine spectrum has partly collapsed, although the lines are still broad. When a
I I
I
IIJ
I
0
04
(a)~~~~~~~~~~~~~~~~~~~~~~~~~~~~07p
-
0
04 o-1 -+
0
.0
1~~~~~~~~~~~~~p
-
El
044
m
la
0
0 4 I
0 0-4
-6
-4
-2
0
2
4
6
Source velocity (mm.Isec.)
Fig. 1. Mossbauer spectra of xanthine oxidase under various conditions. (a) Oxidized enzyme at 4*20K; (b) enzyme reduced with salicylaldehyde, at 4-2°K; (c) enzyme reduced with Na2S204, at 770 K; (d) enzyme reduced with Na2S204, at 4.20E ; (e) enzyme reduced with Na2S204, at 4-20x in a magnetic field of 0*5kG. Scans over a wider velocity range on samples (a) and (b) failed to reveal further absorption. The vertical lines on the right of each diagram represent mean standard counting errors.
Vol. 103
SHORT COMMUNICATIONS
small external magnetic field (0.5kG) is applied at liquid He temperatures (Fig. le), a six-line pattern characteristic of an effective magnetic field of 180 + 30kG at the nuclei is obtained. This arises because the field effectively decouples the nuclear electronic moments so that they precess about the external field. Spectra in fields from 0-5 to 30kG at temperatures between 1-50K and 4*20K appear the same. The lines are broader than the natural width, probably because of anisotropy of the hyperfine coupling. From the observed field at the iron nuclei, the field at the electrons due to the 57Fe nuclei may be calculated and corresponds to an expected splitting of 18 + 3 G for the electron-spinresonance signal of xanthine oxidase enriched with 57Fe. Shethna, Wilson, Hansen & Beinert (1964) found 22 G splitting for the iron protein of Azotobacter vinelandii. The disappearance of the absorption dip near zero velocity (Fig. le) on applying a small field suggests that the spectrum of Fig. l(d) arises from all (i.e. 80% or more) of the iron atoms behaving in the same way, rather than being the sum of spectra from reduced and oxidized atoms, i.e. under the conditions used all the iron atoms in the sample have been affected by dithionite reduction. This contrasts with reduction with salicylaldehyde, which produced changes in, at most, a small proportion of the iron atoms. The difference was unexpected since both reducing agents gave the characteristic iron electron-spin-resonance signal. However, further electron-spin-resonance work has shown that, though concentrations and reaction times are important, salicylaldehyde always gives weaker signals than dithionite. Conditions for optimum signal development are currently being investigated. At 980K the signal was earlier reported very weak, corresponding to only about one unpaired electron per molecule, i.e. per eight iron atoms (Bray et al. 1964). However, more recent work (J. F. Gibson & R. C. Bray, unpublished work; G. Palmer, personal communication), has shown that the signal, like that from ferredoxin (Palmer & Sands, 1966; Hall, Gibson & Whatley, 1966), is much stronger at lower temperatures, and the possibility even of all the iron atoms of the xanthine oxidase molecule being involved in it under favourable conditions cannot now be excluded. The signal changes markedly between 77°K and 4-201x, and this is presumably connected with the increase in relaxation time that also produces the change in Mossbauer spectrum between Fig. 1(c) and 1(d). From the M6ssbauer data, on a single-ion model, the iron in oxidized xanthine oxidase would appear to be in a low-spin (S = 0) ferrous state. High-spin ferrous is excluded since it typically has a much larger chemical shift (approx. -15mm./sec.), and
110
high-spin or low-spin ferric would show an internal hyperfine magnetic field when an external magnetic field is applied. However, models have been proposed (Gibson, Hall, Thormley & Whatley, 1966; Brintzinger, Palmer & Sands, 1966) where ferric atoms are coupled together by exchange interaction to give a total spin S = 0, and this would be compatible with our results. In the reduced state after treatment with Na2S204 the single-ion picture would exclude siLmple low-spin ferrous, which could not show hyperfine magnetic interaction, and high-spin ferric, which would show a hyperfine magnetic field of 500-600kG. The most likely simple state would appear to be low-spin (S= j) ferric, which would have about the right hyperfine field and relaxation times. This could arise, e.g. if steric changes due to reduction of other centres destroyed exchange coupling between low-spin ferric atoms of the oxidized molecule. Another possibility from the Mossbauer data, again based on an exchangecoupled picture and in this case implying coupling in both oxidized and reduced states, is that reduction affects only one of a coupled pair, so producing a net magnetic moment (cf. Gibson et at. 1966). There would have to be four such pairs of irons in the molecule. A third possibility, not involving coupled irons in either oxidation state, is low-spin ferrous with the hyperfine field arising from a neighbouring free radical, formed during the reduction (probably sulphur), to which the metal is strongly bound. This would be analogous to the nitric oxide compound of haemoglobin (Lang & Marshall, 1966) to the pentacyanonitrosoiron complex described by Beinert, Dervartanian, Hemmerich, Veeger & Van Voorst (1965) and would be related to the model of Blumberg & Peisach (1965). We thank Dr T. E. Cranshaw, Dr J. F. Gibson and Dr D. 0. Hall for helpful discussions and Miss F. Pick and Mr L. Becker for valuable technical assistance. Work at the Chester Beatty Research Institute (Institute of Cancer Research: Royal Cancer Hospital) was supported by grants from the Medical Research Council, the British Empire Cancer Campaign for Research and by Public Health Service Research Grant no. CA-03188-10 from the National Cancer Institute, U.S. Public HIealth Service. Bayer, E. & Voelter, W. (1966). Biochim. biophye. Acta, 118, 632. Beinert, H. (1965). In Non-Heme Iron Proteins: Role in Energy conver8ion, p. 23. Ed. by San Pietro, A. Yellow Springs, Ohio: Antioch Press. Beinert, H., Dervartanian, D. V., Hemmerich, P., Veeger, C. & Van Voorst, J. D. W. (1965). Biochim. biophys. Acta,
96,530.
Blumberg, W. E. & Peisach, J. (1965). In Non-Heme Iron Proteins: Role in Energy Conver8ion, p. 101. Ed. by San Pietro, A. Yellow Springs, Ohio: Antioch Press.
12 c
C. E. JOHNSON, P. F. KNOWLES AND R. C. BRAY
Bray, R. C., Palmer, G. & Beinert, H. (1964). J. biol. Chem. 239, 2667. Bray, R. C., Pettersson, R. & Ehrenberg, A. (1961). Biochem. J. 81,178. Brintzinger, H., Palmer, G. & Sands, R. H. (1966). Proc. nat. Acad. Sci., Wash., 55, 397. Ehrenberg, A. & Bray, R. C. (1965). Arch. Biochem. Biophys. 109, 199. Gibson, J. F., Hall, D. O., Thornley, J. H. M. & Whatley, F. R. (1966). Proc. nat. Acad. Sci., Wash., 56, 987. Hall, D. O., Gibson, J. F. & WVhatley, F. R. (1966). Biochem. biophys. Res. Commun. 23, 81. Johnson, C. E., Bray, R. C. & Knowles, P. (1967). In Magnetic Resonance in Biological Systems. Ed. by Viinngard, T. Oxford: Pergamon Press Ltd. p. 405. Lang, G. & Marshall, W. (1966). Proc. phys. Soc. Lond., 87, 3.
1967
Palmer, G., Bray, R. C. & Beinert, H. (1964). J. biol. Chem. 239, 2657. Palmer, G. & Sands, R. H. (1966). J. biol. Chem. 241, 253. Palmer, G., Sands, R. H. & Mortenson, L. E. (1966). Biochem. biophys. Res. Commun. 23,357. Phillips, W. D., Knight, E. & Blomstrom, D. C. (1965). In Non-Heme IronProteins: Role in Energy Conversion, p. 69. Ed. by San Pietro, A. Yellow Springs, Ohio: Antioch Press. Piette, L. H. (1967). In Magnetic Resonance in Biological Systems. Ed. by Vanngard, T. Oxford: Pergamon Press Ltd. (in the Press). Shethna, Y. I., Wilson, P. W., Hansen, R. E. & Beinert, H. (1964). Proc. N. Y. Acad. Sci. 52, 1263.
