Inhibition of ceruloplasmin and other copper oxidases by thiomolybdate

Inhibition of Ceruloplasmin and Other Copper Oxidases by Thiomolybdate M. V. Chidambaram,’ G. Barnes, and E. Frieden Department of Chemistry, Florida State University, Tallahassee

ABSTRACT The dietary antagonism between copper and molybdate salts prompted a study of the inhibition of copper enzymes by thiomolybdate (TM). TM strongly inhibited the oxidase activity of five copper oxidases with 250x values in the l-5 pM range. The mechanism of the TM effect on the copper oxidase, ceruloplasmin (Cp) (E.C. 1.16.3.1). was studied in detail. In V,,,,, vs. E plots, TM gave parallel data suggesting irreversibility but a large number of TM molecules per Cp were required. The inhibition of Cp by TM could not be reversed by dialysis. Isolation of TM-inhibited Cp on Sephadex G-10 did not yield any active Cp molecules. Cu(II) did not restore any inhibited oxidase activity. Gel electrophoresis supported the covalent binding of Cp by TM without any extensive change in protein structure. EPR results confirmed that Cu(II) is reduced to Cu(1) after reaction with TM. However, the MO(W) in MO&- did not change in oxidation number. Analysis of the TMCp compound accounted for all six Cu atoms as found in native Cp. The data suggest the covalent binding of sulfide to Cp copper. TM also inhibited the activity of ascorbate oxidase, cytochrome oxidase, superoxide dismutase, and tyrosinase. However, no inhibition of carbonic anhydrase, a zinc enzyme, was observed at 1 mM TM.

ABBREVIATIONS Cp, ceruloplasmin; TM, thiomolybdate; ODA, o-Dianisidine; pPD, p-phenylenediamine; hCp, human ceruloplasmin; bCp, bovine ceruloplasmin; pCp, porcine ceruloplasmin; PAGE, polyacrylamide gel electrophoresis; EDTA, disodium ethylenediaminetetraacetate; DDC, sodium diethyldithiocarbamate; DPD, N,N-dimethyl-p-phenylenediamine; Cyto C. Fe(H), ferrocytochrome C.

INTRODUCTION One of the prime examples of the biological interactions of copper and other metals is the antagonism between molybdate and copper. In a review, Mills [I] pointed Address reprint requests to Dr. Earl Frieden, Department of Chemistry, Florida State University, Tallahassee, FL 32306. ’ Present Address: Department of Biochemistry and Biophysics, College of Agriculture, Texas A&M, College Station, TX 77843. Journal of Inorganic Biochemistry 22, 231-239 (1984) 0 1984 by Elsevier Science Publishing Co., Inc. 52 Vanderbilt Ave., New York, NY 10017

231 0162-0134/&41$3.00

232

M. V. Chidambaram,

G. Barnes, and E. Frieden

out that molybdate strongly inhibits copper utilization when the molybdate content of the diet exceed 1 nag/kg dry weight. The molybdate effect is potentiated by sulfur compounds, especially sulfur-containing amino acids or inorganic sources of sulfur. A likely product of these reactions is thiomolybdate, MoSd2-. An early expression of molybdate antagonism in the rat is the reduction in the copper content and oxidase activity of the plasma fraction containing ceruloplasmin [2]. There is also a decrease in the cellular fractions of the kidney that normally contain copper metallothionein or superoxide dismutase activity. Since many copper enzymes show a high sensitivity to sulfhydryl inhibition, we thought it would be of interest to study the effect of TM on several copper enzymes. Our results show that TM is a strong inhibitor of five copper oxidases. TM inhibited the oxidase activity of ceruloplasmin by forming an irreversible inhibitor-enzyme complex.

MATERIALS

AND METIIODS

Mammalian ceruloplasmins were isolated from bovine, human, and porcine sera by a modification of the method of Zgirski et al. [3]. In this method, instead of the preparative electrophoresis step, a second adsorption on DEAE Sephadex A-50 was used with elution by a NaCl gradient. Purity of Cp preparations was checked by polyacrylamide gel electrophoresis (PAGE). These gels showed only one band when stained with Coomassie Brilliant Blue and for oxidase activity with o-dianisidine (ODA). Another criterion of purity was the absorbance ratio A6r0/A2sc,. This ratio was 0.045 for hCp, 0.050 for bCp, and ,O.OSl for pCp. Stock solutions of Cp were kept in phosphate buffer (0.20 M, pH 7.0) and stored frozen. The concentrated hCp solution which was used for the EPR study was a gift from Dr. J. Witwicki, Serum and Vaccine Lab, Warsaw, Poland. All solutions were chelexed before use. ODA and p-phenylenediamine (pPD) assays employed the procedures of Lehman et al. [4] and Ravin [5], respectively. (NH&MO& (formula weight 260) was prepared according to the method of Hein and Herzog [6]. Blood colored crystals with metallic luster were obtained. The uv and visible spectra of an aqueous solution of (NH&MO& were determined. Two characteristic peaks at h,,, 470 and X,,, 318 with E values of 11250 and 13000 were obtained. This solid was stored in sealed tubes in a nitrogen atmosphere. Stock solutions of 20 mM TM were made up in chelexed water and stored frozen at - 20°C. Secondary dilutions were prepared immediateiy before use. All other chemicals used were of analytical grade. All glassware was soaked in concentrated HNOs and was thoroughly washed in double distilled water.

Cu-Oxidase Activities Ascorbate oxidase (lot No. 112 F-0296, 1400 units/mg), Tyrosinase (lot No. 23F9540, 2600 units/mg), and cytochrome oxidase (lot No. 83F-0476, 25 units/mg) were obtained from Sigma. Ascorbate oxidase activity and its inhibition by TM was measured by a modification of the method of Frieden and Maggiolo [7]. The following conditions were used: 10 nM ascorbate oxidase (0.17 units/ml) was added to 35 PM ascorbate in 200 mM acetate buffer, pH 5.2, at 30°C. Tyrosinase and its inhibition by TM was measured according to the method of Rasper [8]. Two

Inhibition

of Cu Oxidases

by Thiomolybdate

233

methods were employed. In one, the formation of dopachrome at Ad75 was measured in 19 mM phosphate buffer, pH 7.0, 3O”C, where the concentrations were tyrosinase, 42 unitsJ3.5 ml and tyrosine, 257 PM. In the second method [9] catecholase was measured at A265 in 100 mM phosphate buffer, pH 7.0, at 30°C. The concentration of tyrosinase was 2.5 units/ml. Cytochrome oxidase activity and its inhibition by TM was measured according to the method of Wharton and Tzagaloff [lo]. The reaction mixture incubated at 38”C, included 10 mM phosphate buffer, pH 7.0. The concentration of cytochrome oxidase was 109 nM (1.25 units) and ferrocytochrome C (Cyto C. Fe(I1)) was 21 PM. A crude carbonic anhydrase fraction was obtained from lysed rat erythrocytes and measured according to the method of Maren et al. [ 111. For the inhibition of superoxide dismutase, 100 PL of packed human erythrocytes were incubated with 10 Z.JMTM for 1 hr in 3 ml of 20 mM PBS buffer pH 7.4. The cells were washed free of excess TM and lysed. Chloroform ethanol (1.5:2.5 v/v) was added to the lysate and the mixture was centrifuged [12]. The supernatant thus obtained was analyzed for SOD activity using the method of Misra and Fridovich [ 131.

Spectral Studies EPR spectra were recorded at 100 K with a Varian El2 spectrometer, with DPPH used as a field marker. TM solutions were added to concentrated Cp (150 PM) in 100 mM acetate buffer, pH 6.0 and frozen in liquid nitrogen in EPR tubes. Spectral measurements were performed on a Cary 15. MO and Cu concentrations were determined in a Perkin Elmer model 5000 Atomic Absorption spectrophotometer using the model 200 graphite furnace. RESULTS Figure 1 shows a plot of Cp activity versus TM concentration when pPD was used as a Cp substrate. The Zs0r value obtained was 3.0 PM. The mole ratio of TM/Cp was found to be 60 for 50% inhibition and 200 for complete inhibition. For ODA as a substrate the value of Isas was found to be 2.7 PM. Table 1 summarizes the Z5as values for five copper oxidases including the above data. To characterize the nature of TM inhibition as reversible or irreversible, reaction rates were measured for several Cp and one TM concentrations. Linear I’,,,,, versus E plots were obtained as shown in Figure 2. An almost parallel linear rate was obtained for the plot of oxidase activity versus Cp concentration in the presence of 2pM TM. The data with inhibitor intercepted the horizontal axis at a position marked Ei. This represented the number of moles of the enzyme which were irreversibly titrated by TM. A molar ratio of 60 for TM/Cp was obtained.

Reaction of Thiomolybdate with Ceruloplasmin The mechanism of the reaction between TM and Cp was studied in several ways. When an inhibited Cp preparation was dialyzed extensively, no reactivation was observed. pCp (7.61 PM), TM (1.61 mM) in acetate buffer (320 mM, pH 6.0) was dialysed against a large excess of acetate buffer (100 mM pH 6.0) for 24 hr at 4°C. No catalytic activity of Cp towards pPD, ODA, or Fe(I1) was observed. This

234

M. V. Chidambaram, G. Barnes, and E. Frieden

100 F

\

80-

0

5 ..I

2

\

60-

Q, .2

I 500,0=3.0~M

0

\/

\

20-

“\ Oo

I I

I I I 2345678

I

I

“>,

FIGURE 1. Plot of percent relative oxidase activity (V,,,,,) of pCp with pPD as substrate at different concentrations 2.8 mM.

of TM. Conditions:

320 mM acetate buffer,

pH 6.0, I = 37°C. pCp = 50 nM, pPD =

reaction mixture was filtered on a Sephadex G-10 column, previously equilibrated with acetate buffer, pH 6.0. Excess TM was removed and the Cp fraction, collected in the void volume, was inactive. Cu(I1) did not reactivate the inhibited enzyme. 7.56 PM hCp was incubated with 1.52 mM TM and then mixed with a 25or 50-fold molar excess of Cu(I1) ions for 2.5 hr at pH 6.0. After dialysis to remove unbound Cu(II), the resulting Cp fraction did not regain any of its oxidase activity. The electrophoretic mobility of Cp and TM bound Cp were identical on native PAGE. Both exhibited single bands when stained with Coomassie Blue. When stained for ODA activity Cp was positive and TM bound Cp was negative.

Spectral Studies bCp (7.61 PM) and hCp (7.61 PM) were reacted with TM (1.61 mM), and dialysed. Visible absorption spectra (400-650 nm) of the Cp-TM complex were obtained. The TM-Cp complex did not show the characteristic peak at 470 nm of free TM, indicating that no free TM was present. All the TM present was bound to the Cp protein. The absence of a 610 peak in the TM-Cp complex also indicated reduction of Cp-Cu. Further confirmation of Cp-Cu reduction was obtained by EPR. Figure 3 shows a low-temperature EPR spectrum of hCp in acetate buffer pH 6.0 at 9.25 GHz. This spectrum agrees well with those reported previously [14] which reflect the two readily distinguishable spectral components, one with a small parallel hyperfine splitting (70 G) (type 1 copper) and one with a larger parallel hyperfine splitting (160 G) (type 2 copper). The two resolved hyperfine peaks at about 2780 G and 2825 G are assigned to type lb and type la copper, respectively, and the low field line at about 2620 G belongs to type 2 copper. Figure 3 also includes spectra obtained when hCp was treated with various mole ratios of TM. The intensity of the main peak (gl) and the hyperfine lines (gil) became less as the

Inhibition

of Cu Oxidases

TABLE

235

by Thiomolybdate

1. Comparison

of the Inhibition

Thiomolybdate

of Copper Oxidases by

and Cyanide ~___~ Cyanide

Thiomolybdate Substrate

Ref.

Enzyme

Substrate

Lo, (PM)

Ceruloplasmin

PPD ODA

3.0 2.7

DPD

3.0

[241

Cytochrome oxidase

Cyto C. Fe(H)

2.0

Cyto C. Fe(H)

0.01

WI

Superoxide dismutase

Epinephrine

5.0

Superoxide

Ascorbate oxidase

Ascorbate

1.0

Ascorbate

Tyrosine” Tyrosine b

3.0 4.5

Tyrosine

Tyrosinase

n Dopachrome

formation.

b Catecholase

Iso% (PM)

50.0

100. 60.0

WI 1271 [91

activity.

of TM was increased. At a TM:Cp ratio of 67, (Figure 3, spectrum No. 5), 77% of the Cu(I1) signal is lost. Cu(I1) was reduced to Cu(1) and there was no evidence of a MO(V) signal. The decrease in the intensity of all peaks indicates that TM binds to all coppers rather than to a specific copper site such as type la or type lb. concentration

Elemental Analysis Elemental

analyses

were performed

to test for the presence

of various

elements

FIGURE 2. Plot of the reaction rate of pPD (V,.) vs. hCp concentration with and without 2 PM TM. Reaction conditions are the same as in Figure 1. I’,,,.,is the absorbance measured at 530 nm in 1 hr.

‘z

.

0.6 0.5-

I .

hCp /

:/

>E

0.3-

0.2-

nM hCP

236

M. V. Chidambaram, G. Barnes, and E. Frieden

FIGURE 3. ESR spectrum of native hCp (spectrum No. 1) and the TM treated hCp at 77 K and 9.25 Ghz; acetate buffer 100 mM pH = 6.0. The concentration of hCp was 135 PM. Spe.tra 2-5 show the effect of different concentrations of TM on hCp. The concentrations of TM used were No. 2, 3.6 mM: No. 3, 5.0 mM; No. 4, 6.2 mM; and No. 5, 9.0 mM [Spectra 2 and 5 were expanded with 5 x more gain.]

such as total copper and MO. Atomic absorption studies on TM-treated Cp showed all six copper atoms can be detected and accounted for as with native Cp. MO determination established the MO to Cu stoichiometry of 1.6:1. Since each Cp molecule contains six coppers, the ratio of MO to Cp was found to be 10: 1. Copper and Other Metalloenzymes Is TM a general inhibitor for other copper-containing enzymes? Table 1 lists several copper oxidases and their respective Iso% values for TM inhibition. The values obtained are compared to CN ~, a strong inhibitor of most copper enzymes. SOD activity of human red cells and its inhibition by TM is presented in Figure 4. The TM-treated cells showed a marked inhibition of SOD activity. Freshly lysed rat cells were used as a source of carbonic anhydrase and tested by the method of Maren et al. [I I]. At the maximum concentration of TM tested, 1 mM, no inhibition of this Zn enzyme was observed. To test whether TM causes an inhibitory effect on Cp in serum the following experiment was done. 3 ml of 1:1 dilution of porcine plasma containing 1.2 PM pCp was treated with 240 ,uM TM (a 200-fold excess). The control and inhibited samples were dialysed for 18 hr against 100 mM acetate, pH 6.0. Less than 10% activity was detected in the TM treated sample when compared to the control. DISCUSSION Our data clearly establish

TM as a strong inhibitor

of all the Cu-oxidases

tested. As

Inhibition

of Cu Oxidases

237

by Thiomolybdate

0.01

I,, 0

‘j/,

1

100

200

300

I 400

SECS. FIGURE 4. Measurement of SOD activity. Plot of the auto-oxidation of DL-epinephrine with time in 50 mM Na2COI buffer, pH 10.2. Curves are identified as follows: (1) auto-oxidation of DL-epinephrine; (2) inhibition by SOD (obtained from a lysed human red cell fraction); and (3) prevention of the inhibition by SOD after reaction with TM. The number on the curves represent the rate of oxidation measured per 100 sec.

shown m Table 1, the Zsoa values were remarkably

consistent, all in a relatively narrow range of 1.0 to 5.0 PM. In contrast, cyanide inhibition of these enzymes varies greatly with an Z50s much larger for ascorbate oxidase, tyrosinase, and SOD. The unique sensitivity of cytochrome oxidase to cyanide, ZsOs = 10 nM, is also noteworthy. Two zinc-enzymes were tested for their response to TM. NO inhibition of carbonic anhydrase obtained from rat red blood cells could be detected at 1 mM. In the case of purified collagenase 1 mM TM was needed for 80% inhibition 1151. These data suggest the relative insensitivity of Zn-containing enzymes to TM. The mechanism of TM inhibition was studied using Cp-as a typical copper oxidase. The binding of TM to Cp resembles covalent bond formation. The results from Figure 2 can best be interpreted by an irreversible inhibition of hCp by TM. However, the large ratio of TM/hCp of 60 at Ei suggests both specific and nonspecific binding of TM to Cp. Possible explanations for this unusual stoichiometry are discussed later. Evidence for the irreversible binding of TM to Cp and the loss of Cp activity were confirmed by three experiments. In one experiment, the TM-Cp reaction mixture was extensively dialysed. The TM-Cp compound, thus isolated, did not show any catalytic activity towards pPD, ODA, or Fe(I1) ions. In another

238

M. V. Chidambaram, G. Barnes, and E. Frieden

experiment, the TM-Cp reaction mixture was chromatographed on a Sephadex G10 column. Catalytic tests on Cp also were negative. Finally, incubation of TMCp with excess Cu(I1) did not reverse the inhibition of Cp. This establishes the strength of the binding of TM to Cp. It also suggests that the sulfide ligands in TM which were bound to Cp copper are not available for reaction with free Cu(I1). Results from PAGE show that TM on binding to Cp did not cause any fragmentation of the protein. It also indicated the stability of the TM-Cp compound. The lack of staining by ODA for the TM-treated Cp is consistent with the irreversible inhibition of Cp by TM. It is also interesting to note that under serum conditions, Cp activity is irreversibly inhibited. This result is consistent with the observation by Mason et al. [16]. Cp contains blue and nonblue copper atoms. The sensitivity of these Cp-coppers on binding to TM was tested with EPR. These experiments showed a decreased intensity of the Cu(I1) signal due to reduction to Cu(1). It is important to note that Mo(V1) in TM is in the highest oxidation state (4d”) and cannot undergo further oxidation. MO(V), if present in the TM-Cp compound, should show its characteristic hyperfine splitting [17]. Since this was not observed, it can be concluded that coordination of TM with Cp-Cu occurs through the sulfide ligand. The strong absorption of free TM at 470 nm (E = 11,250) is attributed to the MO-S charge transfer [18]. But the TM-Cp isolated after dialysis, had no peak at 470 nm. This strongly suggests that TM is bound to Cp through sulfide. Atomic absorption was used to determine the copper content in TM-Cp. The results quantitatively account for all the six copper atoms in Cp and in TM-Cp. It also confirms that TM does not remove copper from Cp, unlike other chelating agents such as, DDC, CN- , and EDTA, when used at high concentrations [ 191. These data suggest the following reactions for the formation of TM-Cp: S

S

MO&

+ Cu(Il),,Cp +

-> MO :

s4

' c'u(l),,('p

S”

S2- has been excluded as a reactant because of the stability of TM to hydrolysis as indicated by a tl,z of 50 hr according to the following reaction [2O]: MO&~- +H20=MoOS32-

+S2-

+2H’

The oxygen derivatives of MO can also be excluded because MOODY- does not inhibit Cp. Inhibitory activities of other MO sulfur oxy compounds decrease in the following order 12 11: MoSd2- > MoOSj2-

> MoO~S~~- > Mo03S2

The decomposition products of TM are the above sulfur-oxy compounds of MO. As oxygen is progressively substituted for S, the inhibitory activity is reduced. The stability of the aqueous solutions of TM and its interaction with Cu(I1) have been reported [22]. Thus, sulfide appears to be essential for the inhibition phenomenon. MO content in TM-Cp was determined by atomic absorption. The Cu/Mo ratio was determined as 1: 1.6. It is of interest to note a similar range of ratios of Cu/Mo (I .7-2.5) in in vivo experiments, which show antagonism between Cu and MO [23]. Based on a mole per mole basis, 10 MoSd2- molecules have been shown to bind each Cp molecule. If each copper in Cp is bound to one TM, then there must also

Inhibition

of Cu Oxidases

by Thiomolybdate

239

be some nonspecific binding of TM to Cp. Similar nonspecific binding of TM to albumin has been reported. Using 99Mo and anaerobic dialysis, the affinity of TM for albumin has been studied [23]. A mole ratio of MO/Albumin of 13 was found. The high ratios of TM/Cp may be accounted for by the need for a large excess of TM to cause a conformational change in the Cp molecule before irreversible inactivation by TM can occur. Also the nonspecifically bound TM may be the source of electrons in the reduction of Cp-Cu(I1). Further work is needed to establish the stoichiometry with sulfide analysis and to clarify the role of other organic moieties involved in the nonspecific binding of TM to Cp. The isolation of homogeneous TM-Cp would provide useful information. The reactivation of TM-Cp by dye-mediated or electrochemical oxidation of the proposed Cu(I)-Cp intermediate might also be possible. We thank Dr. B. B. Garrett for his help in the use of the EPR spectrometer and helpful discussions. This is paper No. 74 in a series on Copper Metalloproteins. This work was supported in part by NIH grant No. AM 33540-I.

REFERENCES 1.

2. 3.

4. 5. 6. 7.

8. 9. 10. 11.

12. 13. 14.

15. 16.

17.

C. F. Mills, Excerpta Med. Ciba. Fndn. Symp. 79, 49 (1980). C. F. Mills, Chem. Uses of Molybdenum, Proc. Intl. Conf. 4th., H. F. Barry and C. H. Mitchell, Eds., Ann Arbor, MI, p. 134, 1982. A. Zgirski, J. Witwicki, M. Hilewicz, and H. Witas, Acta Biochem. Pol. 25, 361 (1978). H. P. Lehman, K. H. Schosinsky, and M. F. Beeler, C/in. Chem. 20, 1564 (1974). H. A. Ravin, J. Lab. Clin. Med. 58, 161 (1961). F. Hein and S. Herzog, in Handbook of Preparative Inorganic Chemistry, G. Brauer, Ed., Academic Press, New York, 1965, Vol. 2, p. 1416. E. Frieden and I. W. Maggiolo, Biochem. Biophys. Acta 24, 42 (1957). H. S. Rasper, J. Chem. Sot. 125 (1938). M. A. El-Bayoumi and E. Frieden, J. Am. Chem. Sot. 79, 4854 (1957). D. C. Wharton and A. Tzagoloff, Methods Enzymol. 10, 245 (1967). T. H. Maren, V. I. Ash, and E. M. Bailey Jr., Bull. Johns Hopkins Hosp. 95, 244 (1954). M. Tsuchihashi, Biochem. Z. 140, 65 (1923). H. P. Misra and I. Fridovich, J. Biol. Chem. 247, 3470 (1972). P. 0. Gunnarsson, U. Nylen, and G. Pettersson, Eur. J. Biochem. 37, 47 (1973). M. Bond and H. Van Wart, personal communication. J. Mason, M. Lamand, and C. A. Kelleher, Brit. J. Nutr. 43, 515 (1980). D. L. Kessler, J. L. Johnson, H. J. Cohen, and K. V. Rajagopalan, Biochem. Biophys. Acta 334,

867 (1974). 18. 19. 20. 21. 22.

23. 24. 25. 26.

27.

E. I. Stieffel, in Progress in Inorganic Chemistry, S. J. Lippard, Ed., John Wiley & Sons, New York, 1977, Vol. 22, p. 36. J. 0. Erickson, R. D. Gray, and E. Frieden, Proc. Sot. Exp. Med. 134, 117 (1970). M. Harmer and A. G. Sykes, in Molybdenum Chemistry of Biological Significance, W. E. Newton and S. Otuska, Eds., Plenum Press, New York, 1980, p. 401. J. Mason, Irish Vet. J. 36, 164 (1980). N. J. Clarke and S. H. Laurie, Znorg. Chim. Acta 66, L35 (1982). C. F. Mills and I. Bremner, in Molybdenem and Molydenem Containing Enzymes, M. P. Coughlan, Ed., Pergamon Press, Oxford, 1980, p. 512. G. Curzon and J. N. Cumings, in The Biochemistry of Copper, J. Peisach, P. Aisen, and W. E. Blumberg, Eds., Academic Press, New York, 1966, p. 551. M. Dixon and E. C. Webb in Enzymes, Academic Press, New York, 1964, p. 338. G. Rotillo, R. C. Bray, and E. Martin Fielden, Biochem. Biophys. Acta 268, 605 (1972). E. Frieden, Biochem. Biophys. Acta 9, 696 (1952).

Received May 21, 1984; accepted September II, 1984