- Email: [email protected]
Oxidant-induced mobilization of zinc from metallothionein
ARCHIVES
OF BIOCHEMISTRY
Vol. 293, No. 1, February
AND BIOPHYSICS
14, pp. 19%199,1992
COMMUNICATION Oxidant-induced
Mobilization of Zinc from Metallothionein
Henry Fliss’ and Michel Mknard Department
of Physiology,
Faculty of Medicine,
University
of Ottawa, 451 Smyth Road, Ottawa, Ontario, Canada KlH
8M5
Received October 22, 1991
Neutrophils which accumulate at sites of inflammation secrete a number of injurious oxidants which are highly reactive with protein sulfhydryls. The present study examined the possibility that this reactivity with thiols may cause protein damage by mobilizing zinc from cellular metalloproteins in which the metal is bound to cysteine. The ability of the three principal neutrophil oxidants, hypochlorous acid (HOCl), superoxide (. O;), and hydrogen peroxide (HzO,), to cleave thiolate bonds and mobilize complexed zinc was compared using two model compounds (2,3-dimercaptopropanol and metallotbionein peptide fragment 56-61), as well as metallothionein. With all compounds, 60 PM HOC1 caused high rates of Znz+ mobilization as measured spectrophotometrically with the metallochromic indicator 4-(2-pyridylazo)resorcinol. Xanthine (500 PM) plus xanthine oxidase (30 mu), which produced a similar concentration of -0;) also effected a rapid rate of Zna+ mobilization which was inhibited by superoxide dismutase but not catalase, indicating that ~0, is also highly reactive with thiolate bonds. In contrast, H,Oz alone was much less reactive at comparable concentrations. These data suggest that HOC1 and -0; can cause damage to cellular metalloproteins through the mobilization of complexed zinc. In view of the essential role played by zinc in numerous cellular processes, Zn2+ mobilization by neutrophil oxidants may cause significant cellular injury at sites of inflammation. 0 1992 Academic Press, Inc.
Neutrophil accumulation at sites of tissue injury is a welldocumented feature of inflammatory reactions and is essential for the phagocytosis and removal of injured cells (1). The neutrophils apparently become activated during this process and, for poorly understood reasons, secrete high concentrations of reactive oxygen metabolites, such as superoxide anion (. O,), hydrogen peroxide (H202), and hypochlorous acid (HOCl). Although the primary function of these oxidants may be to facil1To whom correspondence should be addressed. 0003-9861/92 $3.00 Copyright D 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
itate the phagocytosis and digestion of cellular debris, they can, paradoxically, also cause further injury by damaging healthy surrounding tissues (2). The nature of the damage caused by neutrophil oxidants at sites of inflammation is under intense investigation at the present time in a number of clinical settings. Recent data suggest that the primary injury may be the direct oxidation of cellular proteins (3-5). Additional, indirect, oxidative injury may accrue as a result of the mobilization of transition metals such as Fe’+ or Cu” (6, 7), which can undergo redox reactions to produce the highly reactive hydroxyl free radical (OH *) (8). We have recently demonstrated the possibility of yet another type of direct oxidative injury, the facile mobilization of zinc from cellular metalloproteins (9, 10). The intracellular mobilization of Zn2+ may cause injury as a result of the inactivation of Zn’+-requiring metalloproteins (ll), or as a consequence of aberrant regulatory effects of the mobilized Zn*+ on a number of cellular functions (12). Zn*+ is frequently coordinated to the sulfur atom of one or more cysteine residues in metalloproteins (11). Our previous studies showed that HOC1 is highly effective at mobilizing Zn*+ from metallothionein (MT),2 and other metalloproteins in which the metal is bound through such thiolate bonds (9). The present study was aimed at determining the ability of the other principal neutrophil oxidants, * 0; and H202, to effect similar mobilization of Zn*+ from MT. MATERIALS Materials.
AND A 5.0
mM
METHODS stock solution of 4-(2-pyridylazohesorcinol
(PAR, Sigma) was prepared as described elsewhere (9,13) by dissolving the solid PAR in deionized water with the addition of 1 N NaOH to maintain the pH at 8.8. Horse kidney MT, bovine erythrocyte superoxide dismutase (SOD), bovine liver catalase, buttermilk xanthine oxidase (X.0.), and MT fragment 56-61 (Lys-Cys-Thr-Cys-Cys-Ala) were obtained from Sigma. Xanthine oxidase activity was determined at 37°C by the urate production assay (14) and is reported as International Units (U = 1 pmol urate/min). Zinc-containing MT fragment (Zn-peptide) was prepared by incubating the peptide (1 mM in deionized water) with ZnCl, (0.5 mM) for 30 min at 37°C. APO-MT was prepared by 2 Abbreviations used: MT, metallothionein; PAR, 4-(2-pyridylazo)resorcinol; SOD, superoxide dismutase; X.0., buttermilk xanthine ox&se; BAL, 2,3dimercaptopropanol; TPEN, N,iV,N:N’-tetrakis(2pyridylmethyl)ethylenediamine. 195
196
FLISS AND MfiNARD
dialysis of MT against 50 mM HCl for 24 h at 4°C (15), and its concentration was determined by titration of the cysteine residues (20 mol/ mol MT) with 5,5’-dithiobis(2-nitrobenzoic acid) (16). Zinc-containing MT was prepared by incubating 90 pM apo-MT with 600 p&l ZnClx in Hepes buffer (40 mM, pH 7.0) for 5 min at 37°C. HOC1 was prepared
just prior to an experiment by adding sodium hypochlorite (Fisher) to water or buffer, and its concentration was determined spectrophotometrically (I& = 350 M-‘; (17)). Xanthine and 2,3dimercaptopropanol (BAL) were products of Sigma. The Zn-BAL complex was prepared by incubating BAL (2 mM in deionized water) with ZnCl, (1 mM) for 10 min at 20°C. TPEN (N,N,N’,N’-tetrakis(2-pyridylmethyl)ethylenediamine) was obtained from Sigma and was used as a 20 mM stock solution in dimethyl sulfoxide.
Spectrophotometric Measurement of Oxidant-Znduced Zn’+ Mobilization. Previously published protocols were followed (9, 13). Cuvettes containing 1.0 ml of PAR (100 PM) in Hepes buffer (40 mM, pH 7.0) were brought to 37“C in a temperature-regulated Beckman DU-7 spectrophotometer. Aliquots of Zn-BAL, Zn-peptide, or Zn-MT stock SOlutions were added followed, after 1 min, by aliquots of freshly prepared oxidant stock solutions. The increase in absorbance at 500 nm, which was indicative of Zn-PAR formation (Em = 6.5 X 10’ M-’ cm-’ (9,13)), was recorded continuously with time. Each curve in the figures represents an actual typical recorder tracing obtained from at least three identical runs.
RESULTS The abilities of the three principal neutrophil oxidants, HOCI, . O;, and HzOz, to mobilize thiolate-bound Zn2+ were compared initially with two model compounds, Zn-BAL and Zn-peptide, and subsequently with Zn-MT. Zn2+ mobilization was monitored with PAR, a very sensitive, high affinity, metallochromic indicator (13). Mobilization
of Zn”
from Zn-BAL
BAL has a high affinity for Zn2+ and presumably forms a dithiolate bond with the metal (15). In order to minimize the possibility of unbound Zn2+ interfering with the test, the Zndimercaptopropanol complex was prepared under conditions of a 2:l molar excess of dithiol over Zn2+, thus ensuring that all Zn2+ was tightly bound. The effects of oxidants on Zn-BAL are shown in Fig. 1. HOC1 was much more reactive with the thiolate bonds than H202, with 50 PM HOC1 releasing the bound metal at an initial rate of 1.2 nmol/min as compared to a release rate of 0.46 nmol/min for 1 mM H202. The relative reactivity of the thiolate bonds with -0, was more difficult to assess in view of the uncertainty associated with establishing the steady state concentration of * 0, in the reaction mixture. Superoxide was produced continuously by X.0. using xanthine as substrate (18), whereas HOC1 and Hz02 were added in bolus amounts. The reaction mixtures contained 30 mU of X.0., which was capable of producing urate, and presumably -0; , at a rate of 30 nmol/ min. However, in view of the known rapid spontaneous dismutation of * 0; to H202 the effective concentration of -0, was no doubt considerably lower. In fact, the rate of -0, production as determined by the SOD-inhibitable cytochrome c reduction assay (19) was found to be only 7.3 f 1.2 nmol/min at 37’C (n = 5) under our experimental conditions and in the absence of the thiolate-containing target molecules, and therefore produced approximately 50 nmol (50 PM) -0; during the 6 min of the assay (Fig. 1). With this simplistic assumption it can be argued
TIME (min) FIG. 1. Mobilization of Zn*+ from Zn-dimercaptopropanol (Zn-BAL). Aliquots (10 ~1) of Zn-BAL were added to 1.0 ml of PAR (100 @M) in 40 mM Hepes, pH 7.0, at 37’%, to give a final concentration of 20 pM BAL and 10 pM complexed zinc. One minute later, aliquots (10 al) of freshly prepared oxidant stock solutions were added to give a final concentration of 50 pM HOC1 or 1 mM H,Oz or 500 FM xanthine plus 30 mU xanthine oxidase (X./X.0.). Absorbance at 500 nm was recorded with time. Each curve in the figure depicts one representative recorder tracing of at least three identical runs. Control runs contained Zn-BAL only. TPEN was added to a final concentration of 20 pM.
that 50 nmol of HOCl, * 0,) or H202 can mobilize 4.6, 0.62, or 0.14 nmol, respectively, of the 10 nmol of BAL-bound Zn2+. Since the xanthine/X.O. reaction can produce H202 as well as * 0, (18), the possibility that H202 contributed to the displacement of Zn2+ from Zn-BAL in the xanthine/X.O. reaction mixture cannot be excluded. However, this appears unlikely in view of the SOD and catalase protection experiments below. The addition of TPEN, a strong heavy metal chelator, to the reaction mixtures rapidly decreased Asw ,,,,,to basal levels, confirming that the absorbance at 500 nm reflected Zn-PAR content (Fig. 1). Treatment of PAR alone with HOCl, H202, or xanthine/ X.0. for up to 10 min did not alter the absorbance of the solution at 500 nm. Moreover, the oxidant-treated PAR retained its ability to form a colored complex with an identical Esoo to that obtained with untreated PAR (data not shown). Incubation of Zn-PAR (10 mnol) alone with each oxidant resulted in a small continuous decrease in A500,,,,,ranging from 1 to 2% per minute (not shown). Mobilization
of Zn2+ from Zn-Peptide
Peptide fragment 56-61 of metallothionein contains three cysteine residues and displays a high affinity for Zn2+ (20). As with
dimercaptopropanol,
the metallopeptide
complex
was pre-
pared under conditions of a 2: 1 molar excess of peptide to metal. The relative abilities of HOCl, H202, and -0; to mobilize the bound Zn2+ are illustrated in Fig. 2. As with Zn-BAL, HOC1 was considerably (approximately 750X) more reactive than H202 with the thiolate bonds of the Zn-peptide (initial mobilization rates, 14 nmol/min with 50 pM HOC1 vs 0.36 nmol/min with 1 mM H,O,). Catalase abolished the H202-induced mobilization (Fig. 2A). In the presence of xanthine/X.O. the rate of mobilization
OXIDANT-INDUCED .5 HOC1
MOBILIZATION
OF ZINC
FROM
197
METALLOTHIONEIN
oms of zinc per mole of MT. Sequential addition of 50-nmol aliquots of HOC1 to MT caused the rapid release of Zn2+ at initial rates greatly exceeding those observed with 1 mM HzOz (Fig. 3A). Catalase completely abolished the HzOz-induced mobilization of Zn’+. Mobilization of the metal by xanthine/X.O. (0.26 nmol/min) proceeded at a rate comparable to that obtained with 1 mM HzOz (0.23 nmol/min), and was inhibited by SOD but not catalase (Fig. 3B), showing that -0% is highly reactive with the thiolate bonds in proteins. TPEN restored basal A 5oonmlevels.
A
DISCUSSION .lln’*“““’ 0
2
.4-
4
6
B
6
10
We have shown previously that HOCl, a potent neutrophil oxidant, is very effective at mobilizing Zn2+ from metalloproteins TPEN
.5 -
TF’EN
Zn-PEPTIDE
-‘02
10 TIME
(mm)
of Znx+ from peptide fragment 56-61 of metalFIG. 2. Mobilization lothionein (Zn-peptide). Aliquots (10 ~1) of Zn-peptide were added to 1.0 ml of PAR (100 fiM) in 40 mM Hepes, pH 7.0, at 37°C to give a final concentration of 10 pM peptide and 5 WM complexed zinc. One minute later, aliquots (10 ~1) of freshly prepared oxidant stock solutions were added to give a final concentration of 50 pM HOC1 or 1 mM HrOr (A), or 500 pM xanthine plus 30 mU xanthine ox&se (X.0., B). Two minutes later superoxide dismutase (SOD, final concentration 50 pg/ml) or catalase (CAT, final concentration 50 pg/ml) were added to selected cuvettes. Absorbance at 500 nm was recorded with time. Control runs contained Zn-peptide only. TPEN, final concentration 20 FM. Each curve in the figure depicts one representative recorder tracing of at least three identical runs.
to that observed with 1 mM Hz02 (0.36 nmol/min). SOD, but not catalase, inhibited this mobilization, suggesting that thiolate cleavage was caused almost exclusively by -0; (Fig. 2B). Applying the assumptions used above to estimate superoxide production, it appears that * 0, is approximately 20 times more reactive than Hz02 with the Zn-S bonds in the peptide. TPEN restored As00 nmto basal levels in all reaction mix-
.,I
2
“1
B
0
4
6
8
10
TPEN
was similar
tures.
Mobilization
of .Zn’+ from MT
Metallothionein contains 20 cysteine residues which can complex a variety of heavy metals, most commonly Zn’+, by means of thiolate bonds (21, 22). The Zn-containing MT prepared for these experiments contained approximately seven at-
10
*loI TIME (min)
FIG. 3. Mobilization of Zn2+ from metallothionein (Zn-MT). Aliquots (10 ~1) of Zn-MT were added to 1.0 ml of PAR (100 pM) in 40 mM Hepes, pH 7.0, at 37”C, to give a final concentration of 0.9 pM MT and 6 @M complexed zinc. One minute later, aliquots (10 ~1) of freshly prepared oxidant stock solutions were added to give a final concentration of 50 pM HOC1 or 1 mM H202 (A), or 500 pM xanthine plus 30 mU xanthine oxidase (X.0., B). Additional aliquots of HOC1 were added as indicated (arrows). Two minutes later superoxide dismutase (SOD, final concentration 50 pg/ml) or catalase (CAT, final concentration 50 ag/ml) were added to selected cuvettes. Absorbance at 500 nm was recorded with time. Control runs contained Zn-MT only. TPEN, 20 pM. Each curve in the figure depicts one representative recorder tracing of at least three identical runs.
198
FLISS
AND
in which the metal is bound to the sulfur atom of cysteine residues (9). In view of the fact that neutrophils can also produce other reactive oxidants such as * 0; and H202 at sites of inflammation, the present studies tested and compared the abilities of these oxidants to effect similar mobilization of Zn’+. As the data show, the ability of HOC1 to mobilize thiolatebound Zn2+ far exceeds that of HzOz with model compounds such as Zn-BAL or Zn-containing metallothionein fragment, as well as with a protein, metallothionein. In fact, with the Znpeptide the rate of Zn2+ release by HOC1 is approximately 750 times greater than with Hz02. We and others have shown previously that HOC1 is much more reactive than H202 with proteins in general, and the sulfur-containing amino acids (methionine, cysteine) in particular (3-5,23). The present study found a similarly greater HOC1 reactivity with cysteine residues bound to metals by means of thiolate bonds. Determination of the relative reactivity of -0; with the ZnS bonds was rendered more difficult by the uncertainties involved in establishing the steady state concentration of -0, in the reaction mixtures containing xanthine and X.0. With the aid of the cytochrome c assay for -0, production, and some simplistic assumptions (see Results section), it may be argued that at equivalent concentrations, -0; can cleave approximately 20 times more protein thiolate bonds than HzOz, but may still be an order of magnitude less reactive than HOCl. However, in view of the fact that the steady state concentration of -0; is a function of the square root of its rate of production (24), and must therefore be on the order of only a few micromolar, the relatively high rates of Zn2+ mobilization by this oxidant in the present studies suggest that its reactivity with thiolate bonds may be similar to, or in fact exceed, that of HOCl. The present study therefore shows that of the various neutrophil oxidants produced at sites of inflammation, HOC1 and . O,, but not HzOz, possess the ability to mobilize significant amounts of Zn2+ from cellular metalloproteins such as MT. Metallothionein, a protein that contains 20 cysteine residues which are normally coordinated to heavy metals, may play a central role in intracellular metal regulation (21,22). It has come under intense scrutiny and the reactivity of the MT thiolate bonds with various oxidants had already been examined to some extent. For example, the exposure of MT to y-irradiation or xanthine/ X.0. resulted in the production of hydroxyl free radicals and -0, of which the OH. were much more reactive with the protein thiols (25). Although some metal loss was observed in the study, no analysis of the relative contribution of these oxidants to metal release was presented. Two other studies employing xanthine/ X.0. (26), or neutrophil-generated oxidants (27), with MT also presented evidence of thiolate bond cleavage and metal (Zn’+, Cd2’, and CL?) release. However, in contrast to our study, the thiolate cleavage was apparently caused by H202 but not * 0,. We have no explanation at the present time for this apparent discrepancy. In addition to -0, and Hz02, the xanthine/X.O. reaction can apparently also produce OH., presumably as a result of iron contamination in commercial X.0. preparations (28). However, it is unlikely that OH. contributed to metal mobilization in our xanthine/X.O. studies, in view of the fact that our reactions were done in 40 mM Hepes, a potent OH * scavenger (8, 29). Moreover, since H202 is required for the production of OH * with the Haber-Weiss or Fenton reactions (30), the inability of
MENARD
catalase to protect against the mobilization induced by xanthine/ X.0. suggests that OH * did not contribute to this process. Zinc has been shown to play an important role in a large number of cellular processes such as Ca2+ regulation (31), immune function (32), gene expression (ll), and others (12). In addition to MT, there are numerous other zinc-containing cellular proteins, such as hormone receptors and transcription factors, in which the metal is coordinated solely by cysteine (33). Moreover, there are many other zinc metalloproteins in which the metal is bound partially to the sulfur of cysteine in “zinc fingers” (11, 33). The mobilization of Zn2+ from these metalloproteins by neutrophil oxidants at sites of inflammation may therefore result in significant alterations in cellular function in target tissues and may contribute to tissue injury. The concentrations of HOC1 and -0; utilized in these studies are physiologically relevant, since sites of interstitial inflammation can contain as many as 2.5 X 10s neutrophils/ml of fluid (34) which are capable of producing extracellular HOC1 concentrations of more than 400 pM/min (35), as well as superoxide concentrations of 100-300 fiM/min (36). Our data suggest that significant displacement of Zn2+ from proteins can be achieved with considerably lower concentrations of oxidants (9). Moreover, recent studies in our laboratory have shown that 50-100 PM HOC1 can cause marked intracellular mobilization of Zn2+ in a number of tissues (10). The possible involvement of -0, in this process is of particular interest. Superoxide has recently been shown to be much more reactive with protein thiols and metal-sulfur bonds than had originally been assumed (14, 37). Moreover, it is increasingly suggested that -0, may function as a cellular messenger in a variety of cellular processes (38,39). These two factors raise the intriguing possibility that -0, may assert its regulatory effects through the mobilization of metals such as Zn2+. Evidence that Zn2+ mobilization may play a regulatory role in a number of proteins such as collagenase (40), or tissues such as the brain (41,42), lends support to this hypothesis. Since Zn*+ mobilization has been observed under a variety of toxic conditions (43,44), it is possible that the mobilization of this metal by oxidants such as HOC1 or -0; may play a beneficial regulatory role at sites of injury. In summary, the present study compares for the first time the relative ability of the neutrophil oxidants HOCl, Hz02, and -0; to mobilize Zn2+ from thiolate bonds in metallothionein. It suggests that HOC1 and -0,) but not H20z, are very effective at causing metal release. In view of the important cellular role may have injurious effects. played by Zn2+, such mobilization However, the possibility that Zn2+ mobilization may play a beneficial regulatory role at sites of injury cannot be excluded at this time. ACKNOWLEDGMENTS The excellent secretarial assistance of Geraldine Villeneuve is greatly appreciated. This project was funded by the Defence Research Establishment of Ontario.
REFERENCES 1. Ward, P. A., Warren, J. S., and Johnson, K. J. (1988) Free Radicals Biol. Med. 5, 403-408. 2. Klebanoff, S. J. (1999) in Inflammation: Basic Principles and Clinical Correlates (Gallin, J. I., Goldstein, I. M., and Snyderman, R., Eds.1, pp. 391-444, Raven Press, New York.
OXIDANT-INDUCED
MOBILIZATION
3. Fliss, H. (1988) Mol. Cell. Biochcm. 84, 177-188. 4. Kukreja, R. C., Weaver, A. B., and Hess, M. L. (1990) Am. J. Physbl. 269, H1330-H1336. 5. Schraufstatter, I. U., Browne, K., Harris, A., Hyslop, P. A., Jackson, J. H., Quehenberger, O., and Cochrane, C. G. (1990) J. Clin. Znuest. 86,554-562. 6. Rosen, H., and Klebanoff,
S. J. (1985) Infect. Zmmun. 47,613-618.
I. Schechinger, T., Hartmann, J. 240, 281-283. 8. Miller, D. M., Buettner, Biol. Med. 8, 95-108.
H. J., and Weser, U. (1986) B&hem.
G. R., and Aust, S. D. (1990) Free Radicals
9. Fliss, H., and Mbnard, M. (1991) Arch. Biochem. Biophys. 287, 175-179. 10. Fliss, H., Menard, M., and Desai, M. (1991) Can. J. Physiol. Phurmacol. 69,1686-1691. 11. Vallee, B. L., and Auld, D. S. (1990) Biochemistry 12. Cunnane, S. C. (1988) Zinc: Clinical CRC Press, Boca Raton, FL.
29, 5647-5659.
and Biochemical
13. Hunt, J. B., Neece, S. H., and Ginsburg, 146,150-157.
Significance,
A. (1985) Anal. Biochem.
14. Radi, R., Bush, K. M., Cosgrove, T. P., and Freeman, B. A. (1991) Arch. Biochem. Biophys. 286, 117-125. 15. Klgi, J. H. R., and Vallee, B. L. (1961) J. Biol. Chem. 236, 24352442. 16. Sedlak, J., and Lindsay, 17. Morris,
R. H. (1968) Anal. Biochem. 26, 192-205.
J. C. (1966) J. Phys. Chem. 70, 3798-3805.
18. McCord, J. M., and Fridovich, 5760.
I. (1968) J. Biol. Chem. 243,5753-
19. McCord, J. M., and Fridovich, 6055.
I. (1969) J. Biol. Chem. 244,6049-
20. Yoshida, A., Kaplan, B. E., and Kimura, Sci. USA 76,486-490.
M. (1979) Proc. N&l. Acad.
21. Hamer, D. H. (1986) Anna
Reu. Biochem. 66,913-951.
22. Kagi, J. H. R., and Schlffer,
A. (1988) Biochemistry
23. Vissers, M. C., and Winterbourn, phys. 286, 53-59.
27,8509-8515.
C. C. (1991) Arch. Biochem. Bio-
24. Miki, S., Ashraf, M., Salka, S., and Sperelakis, Cell. Cardiol. 20. 1009-1024.
N. (1988) J. Mol.
OF ZINC
FROM
METALLOTHIONEIN
199
P. J., and Vaslk, M. (1985) Biochim. Biophys. Acta 25. Thornalley, 827,36-44. 26. Thomas, J. P., Bachowski, G. J., and Girotti, A. W. (1986) Biochim. Biophys. Acta 884,448461. 27. Hartmann, H. J., Glirtner, A., and Weser, U. (1985) Clin. Chim. Acta 162,95-103. 28. Britigan, B. E., Pou, S., Rosen, G. M., Lilleg, D. M., and Buettner, G. R. (1990) J. Biol. Chem. 266, 17,533-17,538. B., Gutteridge, J. M., and Aruoma, 0. I. (1987) Anal. 29. Halliwell, Biochem. 166,215-219. 30. HaIhwell, B., and Gutteridge, J. M. C. (1985) Free Radicals in Biology and Medicine, Oxford Univ. Press (Clarendon), Oxford. 31. Vergara, J., Tsien, R. Y., and Delay, M. (1985) Proc. Natl. Acad. Sci. USA 82,6352-6356. 32. Wakabayashi, S., and Shigekawa, M. (1987) J. Biol. Chem. 262, 11,524-11,531. 33. Vallee, B. L., Coleman, J. E., and Auld, D. S. (1991) Proc. NC&. Acad. Sci. USA 88,999-1003. 34. Weiss, S. J. (1989) N. En&. J. Med. 320, 365-376. 35. Kukreja, R. C., Weaver, A. B., and Hess, M. L. (1989) Biochim. Biophys. Acta 990, 198-205. 36. Black, C. D. V., Samuni, A., Cook, J. A., Krishna, C. M., Kaufman, D. C., Malech, H. L., and Russo, A. (1991) Arch. Biochem. Biophys. 286,126-131. 37. Gardner, P. R., and Fridovich, I. (1991) J. Biol. Chem. 266, 14781483. 38. Cross, A. R., and Jones, 0. T. (1991) Biochim. Biophys. Acta 1067, 281-298. 39. Ikebuchi, Y., Masumoto, N., Tasaka, K., Koike, K., Kasahara, K., Miyake, A., and Tanizawa, 0. (1991) J. Biol. Chem. 266, 13,23313,237. H., and Van 40. Springman, E. B., Angleton, E. L., Birkedal-Hansen, Wart, H. E. (1990) Proc. N&l. Acad. Sci. USA 87, 364-368. 41. Assaf, S. Y., and Chung, S. H. (1984) Nature 308, 734-736. 42. Howell, G. A., Welch, M. G., and Frederickson, C. J. (1984) Nature 308,736-738. 43. Oh, S. H., Deagen, J. T., Whanger, P. D., and Weswig, P. H. (1978) Am. J. Physiol. 234, E282-E285. 44. Suntres, Z. E., and Lui, E. M. (1990) Biochem. Pharmacol. 39,833840.
OF BIOCHEMISTRY
Vol. 293, No. 1, February
AND BIOPHYSICS
14, pp. 19%199,1992
COMMUNICATION Oxidant-induced
Mobilization of Zinc from Metallothionein
Henry Fliss’ and Michel Mknard Department
of Physiology,
Faculty of Medicine,
University
of Ottawa, 451 Smyth Road, Ottawa, Ontario, Canada KlH
8M5
Received October 22, 1991
Neutrophils which accumulate at sites of inflammation secrete a number of injurious oxidants which are highly reactive with protein sulfhydryls. The present study examined the possibility that this reactivity with thiols may cause protein damage by mobilizing zinc from cellular metalloproteins in which the metal is bound to cysteine. The ability of the three principal neutrophil oxidants, hypochlorous acid (HOCl), superoxide (. O;), and hydrogen peroxide (HzO,), to cleave thiolate bonds and mobilize complexed zinc was compared using two model compounds (2,3-dimercaptopropanol and metallotbionein peptide fragment 56-61), as well as metallothionein. With all compounds, 60 PM HOC1 caused high rates of Znz+ mobilization as measured spectrophotometrically with the metallochromic indicator 4-(2-pyridylazo)resorcinol. Xanthine (500 PM) plus xanthine oxidase (30 mu), which produced a similar concentration of -0;) also effected a rapid rate of Zna+ mobilization which was inhibited by superoxide dismutase but not catalase, indicating that ~0, is also highly reactive with thiolate bonds. In contrast, H,Oz alone was much less reactive at comparable concentrations. These data suggest that HOC1 and -0; can cause damage to cellular metalloproteins through the mobilization of complexed zinc. In view of the essential role played by zinc in numerous cellular processes, Zn2+ mobilization by neutrophil oxidants may cause significant cellular injury at sites of inflammation. 0 1992 Academic Press, Inc.
Neutrophil accumulation at sites of tissue injury is a welldocumented feature of inflammatory reactions and is essential for the phagocytosis and removal of injured cells (1). The neutrophils apparently become activated during this process and, for poorly understood reasons, secrete high concentrations of reactive oxygen metabolites, such as superoxide anion (. O,), hydrogen peroxide (H202), and hypochlorous acid (HOCl). Although the primary function of these oxidants may be to facil1To whom correspondence should be addressed. 0003-9861/92 $3.00 Copyright D 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
itate the phagocytosis and digestion of cellular debris, they can, paradoxically, also cause further injury by damaging healthy surrounding tissues (2). The nature of the damage caused by neutrophil oxidants at sites of inflammation is under intense investigation at the present time in a number of clinical settings. Recent data suggest that the primary injury may be the direct oxidation of cellular proteins (3-5). Additional, indirect, oxidative injury may accrue as a result of the mobilization of transition metals such as Fe’+ or Cu” (6, 7), which can undergo redox reactions to produce the highly reactive hydroxyl free radical (OH *) (8). We have recently demonstrated the possibility of yet another type of direct oxidative injury, the facile mobilization of zinc from cellular metalloproteins (9, 10). The intracellular mobilization of Zn2+ may cause injury as a result of the inactivation of Zn’+-requiring metalloproteins (ll), or as a consequence of aberrant regulatory effects of the mobilized Zn*+ on a number of cellular functions (12). Zn*+ is frequently coordinated to the sulfur atom of one or more cysteine residues in metalloproteins (11). Our previous studies showed that HOC1 is highly effective at mobilizing Zn*+ from metallothionein (MT),2 and other metalloproteins in which the metal is bound through such thiolate bonds (9). The present study was aimed at determining the ability of the other principal neutrophil oxidants, * 0; and H202, to effect similar mobilization of Zn*+ from MT. MATERIALS Materials.
AND A 5.0
mM
METHODS stock solution of 4-(2-pyridylazohesorcinol
(PAR, Sigma) was prepared as described elsewhere (9,13) by dissolving the solid PAR in deionized water with the addition of 1 N NaOH to maintain the pH at 8.8. Horse kidney MT, bovine erythrocyte superoxide dismutase (SOD), bovine liver catalase, buttermilk xanthine oxidase (X.0.), and MT fragment 56-61 (Lys-Cys-Thr-Cys-Cys-Ala) were obtained from Sigma. Xanthine oxidase activity was determined at 37°C by the urate production assay (14) and is reported as International Units (U = 1 pmol urate/min). Zinc-containing MT fragment (Zn-peptide) was prepared by incubating the peptide (1 mM in deionized water) with ZnCl, (0.5 mM) for 30 min at 37°C. APO-MT was prepared by 2 Abbreviations used: MT, metallothionein; PAR, 4-(2-pyridylazo)resorcinol; SOD, superoxide dismutase; X.0., buttermilk xanthine ox&se; BAL, 2,3dimercaptopropanol; TPEN, N,iV,N:N’-tetrakis(2pyridylmethyl)ethylenediamine. 195
196
FLISS AND MfiNARD
dialysis of MT against 50 mM HCl for 24 h at 4°C (15), and its concentration was determined by titration of the cysteine residues (20 mol/ mol MT) with 5,5’-dithiobis(2-nitrobenzoic acid) (16). Zinc-containing MT was prepared by incubating 90 pM apo-MT with 600 p&l ZnClx in Hepes buffer (40 mM, pH 7.0) for 5 min at 37°C. HOC1 was prepared
just prior to an experiment by adding sodium hypochlorite (Fisher) to water or buffer, and its concentration was determined spectrophotometrically (I& = 350 M-‘; (17)). Xanthine and 2,3dimercaptopropanol (BAL) were products of Sigma. The Zn-BAL complex was prepared by incubating BAL (2 mM in deionized water) with ZnCl, (1 mM) for 10 min at 20°C. TPEN (N,N,N’,N’-tetrakis(2-pyridylmethyl)ethylenediamine) was obtained from Sigma and was used as a 20 mM stock solution in dimethyl sulfoxide.
Spectrophotometric Measurement of Oxidant-Znduced Zn’+ Mobilization. Previously published protocols were followed (9, 13). Cuvettes containing 1.0 ml of PAR (100 PM) in Hepes buffer (40 mM, pH 7.0) were brought to 37“C in a temperature-regulated Beckman DU-7 spectrophotometer. Aliquots of Zn-BAL, Zn-peptide, or Zn-MT stock SOlutions were added followed, after 1 min, by aliquots of freshly prepared oxidant stock solutions. The increase in absorbance at 500 nm, which was indicative of Zn-PAR formation (Em = 6.5 X 10’ M-’ cm-’ (9,13)), was recorded continuously with time. Each curve in the figures represents an actual typical recorder tracing obtained from at least three identical runs.
RESULTS The abilities of the three principal neutrophil oxidants, HOCI, . O;, and HzOz, to mobilize thiolate-bound Zn2+ were compared initially with two model compounds, Zn-BAL and Zn-peptide, and subsequently with Zn-MT. Zn2+ mobilization was monitored with PAR, a very sensitive, high affinity, metallochromic indicator (13). Mobilization
of Zn”
from Zn-BAL
BAL has a high affinity for Zn2+ and presumably forms a dithiolate bond with the metal (15). In order to minimize the possibility of unbound Zn2+ interfering with the test, the Zndimercaptopropanol complex was prepared under conditions of a 2:l molar excess of dithiol over Zn2+, thus ensuring that all Zn2+ was tightly bound. The effects of oxidants on Zn-BAL are shown in Fig. 1. HOC1 was much more reactive with the thiolate bonds than H202, with 50 PM HOC1 releasing the bound metal at an initial rate of 1.2 nmol/min as compared to a release rate of 0.46 nmol/min for 1 mM H202. The relative reactivity of the thiolate bonds with -0, was more difficult to assess in view of the uncertainty associated with establishing the steady state concentration of * 0, in the reaction mixture. Superoxide was produced continuously by X.0. using xanthine as substrate (18), whereas HOC1 and Hz02 were added in bolus amounts. The reaction mixtures contained 30 mU of X.0., which was capable of producing urate, and presumably -0; , at a rate of 30 nmol/ min. However, in view of the known rapid spontaneous dismutation of * 0; to H202 the effective concentration of -0, was no doubt considerably lower. In fact, the rate of -0, production as determined by the SOD-inhibitable cytochrome c reduction assay (19) was found to be only 7.3 f 1.2 nmol/min at 37’C (n = 5) under our experimental conditions and in the absence of the thiolate-containing target molecules, and therefore produced approximately 50 nmol (50 PM) -0; during the 6 min of the assay (Fig. 1). With this simplistic assumption it can be argued
TIME (min) FIG. 1. Mobilization of Zn*+ from Zn-dimercaptopropanol (Zn-BAL). Aliquots (10 ~1) of Zn-BAL were added to 1.0 ml of PAR (100 @M) in 40 mM Hepes, pH 7.0, at 37’%, to give a final concentration of 20 pM BAL and 10 pM complexed zinc. One minute later, aliquots (10 al) of freshly prepared oxidant stock solutions were added to give a final concentration of 50 pM HOC1 or 1 mM H,Oz or 500 FM xanthine plus 30 mU xanthine oxidase (X./X.0.). Absorbance at 500 nm was recorded with time. Each curve in the figure depicts one representative recorder tracing of at least three identical runs. Control runs contained Zn-BAL only. TPEN was added to a final concentration of 20 pM.
that 50 nmol of HOCl, * 0,) or H202 can mobilize 4.6, 0.62, or 0.14 nmol, respectively, of the 10 nmol of BAL-bound Zn2+. Since the xanthine/X.O. reaction can produce H202 as well as * 0, (18), the possibility that H202 contributed to the displacement of Zn2+ from Zn-BAL in the xanthine/X.O. reaction mixture cannot be excluded. However, this appears unlikely in view of the SOD and catalase protection experiments below. The addition of TPEN, a strong heavy metal chelator, to the reaction mixtures rapidly decreased Asw ,,,,,to basal levels, confirming that the absorbance at 500 nm reflected Zn-PAR content (Fig. 1). Treatment of PAR alone with HOCl, H202, or xanthine/ X.0. for up to 10 min did not alter the absorbance of the solution at 500 nm. Moreover, the oxidant-treated PAR retained its ability to form a colored complex with an identical Esoo to that obtained with untreated PAR (data not shown). Incubation of Zn-PAR (10 mnol) alone with each oxidant resulted in a small continuous decrease in A500,,,,,ranging from 1 to 2% per minute (not shown). Mobilization
of Zn2+ from Zn-Peptide
Peptide fragment 56-61 of metallothionein contains three cysteine residues and displays a high affinity for Zn2+ (20). As with
dimercaptopropanol,
the metallopeptide
complex
was pre-
pared under conditions of a 2: 1 molar excess of peptide to metal. The relative abilities of HOCl, H202, and -0; to mobilize the bound Zn2+ are illustrated in Fig. 2. As with Zn-BAL, HOC1 was considerably (approximately 750X) more reactive than H202 with the thiolate bonds of the Zn-peptide (initial mobilization rates, 14 nmol/min with 50 pM HOC1 vs 0.36 nmol/min with 1 mM H,O,). Catalase abolished the H202-induced mobilization (Fig. 2A). In the presence of xanthine/X.O. the rate of mobilization
OXIDANT-INDUCED .5 HOC1
MOBILIZATION
OF ZINC
FROM
197
METALLOTHIONEIN
oms of zinc per mole of MT. Sequential addition of 50-nmol aliquots of HOC1 to MT caused the rapid release of Zn2+ at initial rates greatly exceeding those observed with 1 mM HzOz (Fig. 3A). Catalase completely abolished the HzOz-induced mobilization of Zn’+. Mobilization of the metal by xanthine/X.O. (0.26 nmol/min) proceeded at a rate comparable to that obtained with 1 mM HzOz (0.23 nmol/min), and was inhibited by SOD but not catalase (Fig. 3B), showing that -0% is highly reactive with the thiolate bonds in proteins. TPEN restored basal A 5oonmlevels.
A
DISCUSSION .lln’*“““’ 0
2
.4-
4
6
B
6
10
We have shown previously that HOCl, a potent neutrophil oxidant, is very effective at mobilizing Zn2+ from metalloproteins TPEN
.5 -
TF’EN
Zn-PEPTIDE
-‘02
10 TIME
(mm)
of Znx+ from peptide fragment 56-61 of metalFIG. 2. Mobilization lothionein (Zn-peptide). Aliquots (10 ~1) of Zn-peptide were added to 1.0 ml of PAR (100 fiM) in 40 mM Hepes, pH 7.0, at 37°C to give a final concentration of 10 pM peptide and 5 WM complexed zinc. One minute later, aliquots (10 ~1) of freshly prepared oxidant stock solutions were added to give a final concentration of 50 pM HOC1 or 1 mM HrOr (A), or 500 pM xanthine plus 30 mU xanthine ox&se (X.0., B). Two minutes later superoxide dismutase (SOD, final concentration 50 pg/ml) or catalase (CAT, final concentration 50 pg/ml) were added to selected cuvettes. Absorbance at 500 nm was recorded with time. Control runs contained Zn-peptide only. TPEN, final concentration 20 FM. Each curve in the figure depicts one representative recorder tracing of at least three identical runs.
to that observed with 1 mM Hz02 (0.36 nmol/min). SOD, but not catalase, inhibited this mobilization, suggesting that thiolate cleavage was caused almost exclusively by -0; (Fig. 2B). Applying the assumptions used above to estimate superoxide production, it appears that * 0, is approximately 20 times more reactive than Hz02 with the Zn-S bonds in the peptide. TPEN restored As00 nmto basal levels in all reaction mix-
.,I
2
“1
B
0
4
6
8
10
TPEN
was similar
tures.
Mobilization
of .Zn’+ from MT
Metallothionein contains 20 cysteine residues which can complex a variety of heavy metals, most commonly Zn’+, by means of thiolate bonds (21, 22). The Zn-containing MT prepared for these experiments contained approximately seven at-
10
*loI TIME (min)
FIG. 3. Mobilization of Zn2+ from metallothionein (Zn-MT). Aliquots (10 ~1) of Zn-MT were added to 1.0 ml of PAR (100 pM) in 40 mM Hepes, pH 7.0, at 37”C, to give a final concentration of 0.9 pM MT and 6 @M complexed zinc. One minute later, aliquots (10 ~1) of freshly prepared oxidant stock solutions were added to give a final concentration of 50 pM HOC1 or 1 mM H202 (A), or 500 pM xanthine plus 30 mU xanthine oxidase (X.0., B). Additional aliquots of HOC1 were added as indicated (arrows). Two minutes later superoxide dismutase (SOD, final concentration 50 pg/ml) or catalase (CAT, final concentration 50 ag/ml) were added to selected cuvettes. Absorbance at 500 nm was recorded with time. Control runs contained Zn-MT only. TPEN, 20 pM. Each curve in the figure depicts one representative recorder tracing of at least three identical runs.
198
FLISS
AND
in which the metal is bound to the sulfur atom of cysteine residues (9). In view of the fact that neutrophils can also produce other reactive oxidants such as * 0; and H202 at sites of inflammation, the present studies tested and compared the abilities of these oxidants to effect similar mobilization of Zn’+. As the data show, the ability of HOC1 to mobilize thiolatebound Zn2+ far exceeds that of HzOz with model compounds such as Zn-BAL or Zn-containing metallothionein fragment, as well as with a protein, metallothionein. In fact, with the Znpeptide the rate of Zn2+ release by HOC1 is approximately 750 times greater than with Hz02. We and others have shown previously that HOC1 is much more reactive than H202 with proteins in general, and the sulfur-containing amino acids (methionine, cysteine) in particular (3-5,23). The present study found a similarly greater HOC1 reactivity with cysteine residues bound to metals by means of thiolate bonds. Determination of the relative reactivity of -0; with the ZnS bonds was rendered more difficult by the uncertainties involved in establishing the steady state concentration of -0, in the reaction mixtures containing xanthine and X.0. With the aid of the cytochrome c assay for -0, production, and some simplistic assumptions (see Results section), it may be argued that at equivalent concentrations, -0; can cleave approximately 20 times more protein thiolate bonds than HzOz, but may still be an order of magnitude less reactive than HOCl. However, in view of the fact that the steady state concentration of -0; is a function of the square root of its rate of production (24), and must therefore be on the order of only a few micromolar, the relatively high rates of Zn2+ mobilization by this oxidant in the present studies suggest that its reactivity with thiolate bonds may be similar to, or in fact exceed, that of HOCl. The present study therefore shows that of the various neutrophil oxidants produced at sites of inflammation, HOC1 and . O,, but not HzOz, possess the ability to mobilize significant amounts of Zn2+ from cellular metalloproteins such as MT. Metallothionein, a protein that contains 20 cysteine residues which are normally coordinated to heavy metals, may play a central role in intracellular metal regulation (21,22). It has come under intense scrutiny and the reactivity of the MT thiolate bonds with various oxidants had already been examined to some extent. For example, the exposure of MT to y-irradiation or xanthine/ X.0. resulted in the production of hydroxyl free radicals and -0, of which the OH. were much more reactive with the protein thiols (25). Although some metal loss was observed in the study, no analysis of the relative contribution of these oxidants to metal release was presented. Two other studies employing xanthine/ X.0. (26), or neutrophil-generated oxidants (27), with MT also presented evidence of thiolate bond cleavage and metal (Zn’+, Cd2’, and CL?) release. However, in contrast to our study, the thiolate cleavage was apparently caused by H202 but not * 0,. We have no explanation at the present time for this apparent discrepancy. In addition to -0, and Hz02, the xanthine/X.O. reaction can apparently also produce OH., presumably as a result of iron contamination in commercial X.0. preparations (28). However, it is unlikely that OH. contributed to metal mobilization in our xanthine/X.O. studies, in view of the fact that our reactions were done in 40 mM Hepes, a potent OH * scavenger (8, 29). Moreover, since H202 is required for the production of OH * with the Haber-Weiss or Fenton reactions (30), the inability of
MENARD
catalase to protect against the mobilization induced by xanthine/ X.0. suggests that OH * did not contribute to this process. Zinc has been shown to play an important role in a large number of cellular processes such as Ca2+ regulation (31), immune function (32), gene expression (ll), and others (12). In addition to MT, there are numerous other zinc-containing cellular proteins, such as hormone receptors and transcription factors, in which the metal is coordinated solely by cysteine (33). Moreover, there are many other zinc metalloproteins in which the metal is bound partially to the sulfur of cysteine in “zinc fingers” (11, 33). The mobilization of Zn2+ from these metalloproteins by neutrophil oxidants at sites of inflammation may therefore result in significant alterations in cellular function in target tissues and may contribute to tissue injury. The concentrations of HOC1 and -0; utilized in these studies are physiologically relevant, since sites of interstitial inflammation can contain as many as 2.5 X 10s neutrophils/ml of fluid (34) which are capable of producing extracellular HOC1 concentrations of more than 400 pM/min (35), as well as superoxide concentrations of 100-300 fiM/min (36). Our data suggest that significant displacement of Zn2+ from proteins can be achieved with considerably lower concentrations of oxidants (9). Moreover, recent studies in our laboratory have shown that 50-100 PM HOC1 can cause marked intracellular mobilization of Zn2+ in a number of tissues (10). The possible involvement of -0, in this process is of particular interest. Superoxide has recently been shown to be much more reactive with protein thiols and metal-sulfur bonds than had originally been assumed (14, 37). Moreover, it is increasingly suggested that -0, may function as a cellular messenger in a variety of cellular processes (38,39). These two factors raise the intriguing possibility that -0, may assert its regulatory effects through the mobilization of metals such as Zn2+. Evidence that Zn2+ mobilization may play a regulatory role in a number of proteins such as collagenase (40), or tissues such as the brain (41,42), lends support to this hypothesis. Since Zn*+ mobilization has been observed under a variety of toxic conditions (43,44), it is possible that the mobilization of this metal by oxidants such as HOC1 or -0; may play a beneficial regulatory role at sites of injury. In summary, the present study compares for the first time the relative ability of the neutrophil oxidants HOCl, Hz02, and -0; to mobilize Zn2+ from thiolate bonds in metallothionein. It suggests that HOC1 and -0,) but not H20z, are very effective at causing metal release. In view of the important cellular role may have injurious effects. played by Zn2+, such mobilization However, the possibility that Zn2+ mobilization may play a beneficial regulatory role at sites of injury cannot be excluded at this time. ACKNOWLEDGMENTS The excellent secretarial assistance of Geraldine Villeneuve is greatly appreciated. This project was funded by the Defence Research Establishment of Ontario.
REFERENCES 1. Ward, P. A., Warren, J. S., and Johnson, K. J. (1988) Free Radicals Biol. Med. 5, 403-408. 2. Klebanoff, S. J. (1999) in Inflammation: Basic Principles and Clinical Correlates (Gallin, J. I., Goldstein, I. M., and Snyderman, R., Eds.1, pp. 391-444, Raven Press, New York.
OXIDANT-INDUCED
MOBILIZATION
3. Fliss, H. (1988) Mol. Cell. Biochcm. 84, 177-188. 4. Kukreja, R. C., Weaver, A. B., and Hess, M. L. (1990) Am. J. Physbl. 269, H1330-H1336. 5. Schraufstatter, I. U., Browne, K., Harris, A., Hyslop, P. A., Jackson, J. H., Quehenberger, O., and Cochrane, C. G. (1990) J. Clin. Znuest. 86,554-562. 6. Rosen, H., and Klebanoff,
S. J. (1985) Infect. Zmmun. 47,613-618.
I. Schechinger, T., Hartmann, J. 240, 281-283. 8. Miller, D. M., Buettner, Biol. Med. 8, 95-108.
H. J., and Weser, U. (1986) B&hem.
G. R., and Aust, S. D. (1990) Free Radicals
9. Fliss, H., and Mbnard, M. (1991) Arch. Biochem. Biophys. 287, 175-179. 10. Fliss, H., Menard, M., and Desai, M. (1991) Can. J. Physiol. Phurmacol. 69,1686-1691. 11. Vallee, B. L., and Auld, D. S. (1990) Biochemistry 12. Cunnane, S. C. (1988) Zinc: Clinical CRC Press, Boca Raton, FL.
29, 5647-5659.
and Biochemical
13. Hunt, J. B., Neece, S. H., and Ginsburg, 146,150-157.
Significance,
A. (1985) Anal. Biochem.
14. Radi, R., Bush, K. M., Cosgrove, T. P., and Freeman, B. A. (1991) Arch. Biochem. Biophys. 286, 117-125. 15. Klgi, J. H. R., and Vallee, B. L. (1961) J. Biol. Chem. 236, 24352442. 16. Sedlak, J., and Lindsay, 17. Morris,
R. H. (1968) Anal. Biochem. 26, 192-205.
J. C. (1966) J. Phys. Chem. 70, 3798-3805.
18. McCord, J. M., and Fridovich, 5760.
I. (1968) J. Biol. Chem. 243,5753-
19. McCord, J. M., and Fridovich, 6055.
I. (1969) J. Biol. Chem. 244,6049-
20. Yoshida, A., Kaplan, B. E., and Kimura, Sci. USA 76,486-490.
M. (1979) Proc. N&l. Acad.
21. Hamer, D. H. (1986) Anna
Reu. Biochem. 66,913-951.
22. Kagi, J. H. R., and Schlffer,
A. (1988) Biochemistry
23. Vissers, M. C., and Winterbourn, phys. 286, 53-59.
27,8509-8515.
C. C. (1991) Arch. Biochem. Bio-
24. Miki, S., Ashraf, M., Salka, S., and Sperelakis, Cell. Cardiol. 20. 1009-1024.
N. (1988) J. Mol.
OF ZINC
FROM
METALLOTHIONEIN
199
P. J., and Vaslk, M. (1985) Biochim. Biophys. Acta 25. Thornalley, 827,36-44. 26. Thomas, J. P., Bachowski, G. J., and Girotti, A. W. (1986) Biochim. Biophys. Acta 884,448461. 27. Hartmann, H. J., Glirtner, A., and Weser, U. (1985) Clin. Chim. Acta 162,95-103. 28. Britigan, B. E., Pou, S., Rosen, G. M., Lilleg, D. M., and Buettner, G. R. (1990) J. Biol. Chem. 266, 17,533-17,538. B., Gutteridge, J. M., and Aruoma, 0. I. (1987) Anal. 29. Halliwell, Biochem. 166,215-219. 30. HaIhwell, B., and Gutteridge, J. M. C. (1985) Free Radicals in Biology and Medicine, Oxford Univ. Press (Clarendon), Oxford. 31. Vergara, J., Tsien, R. Y., and Delay, M. (1985) Proc. Natl. Acad. Sci. USA 82,6352-6356. 32. Wakabayashi, S., and Shigekawa, M. (1987) J. Biol. Chem. 262, 11,524-11,531. 33. Vallee, B. L., Coleman, J. E., and Auld, D. S. (1991) Proc. NC&. Acad. Sci. USA 88,999-1003. 34. Weiss, S. J. (1989) N. En&. J. Med. 320, 365-376. 35. Kukreja, R. C., Weaver, A. B., and Hess, M. L. (1989) Biochim. Biophys. Acta 990, 198-205. 36. Black, C. D. V., Samuni, A., Cook, J. A., Krishna, C. M., Kaufman, D. C., Malech, H. L., and Russo, A. (1991) Arch. Biochem. Biophys. 286,126-131. 37. Gardner, P. R., and Fridovich, I. (1991) J. Biol. Chem. 266, 14781483. 38. Cross, A. R., and Jones, 0. T. (1991) Biochim. Biophys. Acta 1067, 281-298. 39. Ikebuchi, Y., Masumoto, N., Tasaka, K., Koike, K., Kasahara, K., Miyake, A., and Tanizawa, 0. (1991) J. Biol. Chem. 266, 13,23313,237. H., and Van 40. Springman, E. B., Angleton, E. L., Birkedal-Hansen, Wart, H. E. (1990) Proc. N&l. Acad. Sci. USA 87, 364-368. 41. Assaf, S. Y., and Chung, S. H. (1984) Nature 308, 734-736. 42. Howell, G. A., Welch, M. G., and Frederickson, C. J. (1984) Nature 308,736-738. 43. Oh, S. H., Deagen, J. T., Whanger, P. D., and Weswig, P. H. (1978) Am. J. Physiol. 234, E282-E285. 44. Suntres, Z. E., and Lui, E. M. (1990) Biochem. Pharmacol. 39,833840.