ANALYTICALBIOCHEMISTRY

189,35-39

(19%))

Quantitation of Cu-Containing by a Cd-Saturation Method Dominik Institute

Received

Klein,

Riidiger

of Toxicology,

March

Bartsch,

and Karl

GSF Research Center, D-8042

Metallothionein H. Summer’ Neuherberg,

of Germany

15,199O

A rapid and sensitive method for determining Cucontaining metallothionein (MT) is described. The main features of this Cd-saturation assay are: high-molecular-weight Cd-binding compounds are denatured with acetonitrile (SO% final concentration), Cu bound to MT is removed with ammonium tetrathiomolybdate, excessive tetrathiomolybdate and its Cu complexes are removed with DEAE-Sephacel, apothionein is saturated with Cd, and excessive Cd is bound to Chelex 100. The thiomolybdate assay is capable of reliably detecting 14 ng MT and thus is particularly suitable for measuring MT in small tissue samples (e.g., biopsies), in extrahepatic tissues, and in cultured cells. Moreover, the combination of the thiomolybdate assay with the recently developed Cd-Chelex assay also makes it possible to determine the portion of MT which binds Cu (Cu load of MT), provided that the amount of non-Cu-thionein exceeds 100 ng, the detection limit of the Cd-Chelex assay. 0 lSS0 Academic Press, Inc.

Metallothionein (MT)’ is a cysteine-rich protein with high affinity for both the nonessential metals Hg and Cd and the nutritionally essential trace elements Zn and Cu (1,2). Investigations on the complex role of MT in Cu metabolism and toxicity are hampered by the lack of a sensitive and yet simple method for quantifying Cu-containing MT. Until now, Cu-thionein was determined mainly by gel filtration chromatography and subsequent metal analysis of the MT fractions. This rather insensitive and time-consuming technique, however, is not suiti To whom correspondence should be addressed at Institut fur Toxikologie, Gesellschaft fiir Strahlenund Umweltforschung, Ingolstiidter Landstrasse 1, D-8042 Neuherberg, FRG. x Abbreviations used: MT, metallothionein; AAS, atomic absorption spectrometry; BSA, bovine serum albumin; CM, carboxymethyl; RIA, radioimmunoassay; 5100, supernatant fraction after centrifugation at 100,000g. 0003-2697/90 Copyright All rights

Federal Republic

$3.00 0 1990 by Academic Press, of reproduction in any form

able for quantifying Cu-thionein in tissues from control animals or for determining MT in small probes such as biopsies. The metal-independent immunological methods (3,4), although sensitive, are problematic because antibodies with constant high titers against MT are not routinely available. These methods, in particular, cannot distinguish Cu-containing MT from MT that contains metals other than Cu. The determination of MT by an Ag-saturation method also has been reported (5). Due to the higher affinity of the protein to Ag than to Cu, this method seemspromising for quantitation of Cu-containing MT. However, the stoichiometry of Ag-thionein is not yet sufficiently established (6). With the known molar ratio of Cd to MT of 7, a Cdsaturation assay without chromatographic steps, as a simple and accurate method for measuring Zn(Cd)-thioneins (7,8), would be superior to present methods for analyzing Cu-containing MT, provided that Cu can be removed from the protein prior to Cd saturation. The present study describes a fast and simple assay for Cu-thionein using ammonium tetrathiomolybdate to sequester Cu from Cu-containing MT (9,lO) and subsubsequent saturation of the apothionein with radiolabeled Cd. MATERIALS

AND

METHODS

Chemicals

lwCd (37 MBq/pg Cd) in 0.1 M HCl was obtained from Amersham Buchler (Braunschweig, FRG). Ammonium tetrathiomolybdate was supplied by Ventron-Alfa (Karlsruhe, FRG). Cd5,Znp-thionein from rabbit liver, CM-Sephadex (40-200 Mm), DEAE-Sephacel (40-150 pm), and bovine serum albumin (BSA), RIA grade, were purchased from Sigma Chemie (Deisenhofen, FRG). Cadmium chloride (standard solution for AAS) was obtained from Aldrich (Steinheim, FRG), and Chelex 100 (100-200 mesh) was supplied by Bio-Rad (Munich, FRG). Cu(CH&N)&104 was synthesized according to 35

Inc. reserved.

36

KLEIN,

BARTSCH,

Hemmerich and Sigwart (ll), recrystallized twice from acetonitrile, dried in a desiccator over silica gel, and stored under argon at 4°C. All other chemicals used were of the highest purity available. Before use, ion exchange resins were washed with 10 vol of 10 mM Tris-HCl, 1M NaCl, pH 7.4, and equilibrated with 10 vol of 10 mM Tris-HCl, 85 mM NaCl, pH 7.4. All solutions were degassed by sonification and saturated with nitrogen at 4°C. Animals Male Wistar rats (inbred strain; Neuherberg, FRG) weighing 180-200 g were fed a standard laboratory diet (Altromin, Lage, FRG) ad Zibitum and had free access to tap water. For MT induction, rats were injected ip on 2 consecutive days with 10 mg Znzf or 3 mg Cu2+ per kilogram of body weight in 0.9% NaCl. Animals were killed 18 h after the last treatment. Human

Liver

Postmortem specimens were obtained from a 41-yearold male after suicide and from five children ages 2.514 months who died of sudden infant death syndrome. Histological examinations revealed no indication of liver disease. Liver biopsies (Menghini technique) from patients with primary biliary cirrhosis and Wilson’s disease were kept at -80°C until measurements. Cells Human foreskin fibroblasts from two children ages 4 and 10 years were kindly provided by the Dermatologische Klinik, LMU (Munich). Cells were grown in Dulbecco’s modified Eagle’s medium (NordVacc, Sweden) containing 10% fetal bovine serum (Gibco, Paisley, Scotland) and cultivated in a humidified environment of 7% COP at 37°C. Confluent monolayer cultures grown on Falcon petri dishes (10 cm in diameter) were incubated 48 h with 10 ml medium containing 0.3-760 pM Cu. For homogenization, the medium was decanted, and after the dishes were washed four times with 5 ml ice-cold PBS, the cells were harvested with a rubber policeman in 0.5 ml of 10 mM Tris-HCl, 85 mM NaCl, pH 7.4, per dish. Cells were lysed by freezing and thawing five times. Preparation

of Cytosols

Rat livers and samples of human livers were homogenized in 4 vol of 10 mM Tris-HCl, pH 7.4, at 0°C with a Potter-Elvehjem homogenizer. Biopsies (3-8 mg wet wt) were homogenized by ultrasonification in 0.4 ml of 10 mM Tris-HCl for 3 s. Tissue and cell homogenates were centrifuged at 100,OOOg (60 min, OOC), and the supernatant fractions (SlOO, cytosol) were used for MT and metal determinations. Cytosolic protein was measured with the biuret method or according to Lowry et al. (12).

AND

SUMMER

Determination

of Metals

Cu and Zn were spectroscopy (AAS) respectively (13). Thiomolybdate

determined and atomic

by atomic absorption emission spectroscopy,

Assay

In a 1.5-ml vial filled with argon, a 0.05- or O.l-ml sample was mixed with the same volume of acetonitrile. After 3 min, 0.5 ml of buffer A (10 mM Tris-HCl, 85 mM NaCl, pH 7.4) and 0.1 ml of CM-Sephadex (66% v/v suspension in buffer A) were added. The mixture was shaken for 3 min under argon and incubated with 0.05 ml of BSA solution (30 mg/ml buffer A, freshly prepared). After 2 min, 0.02 ml of a freshly prepared ammonium tetrathiomolybdate solution (1.3 mg/ml buffer A) was added, and after 2 min, the mixture was shaken with 0.1 ml of DEAE-Sephacel(66% v/v suspension in buffer A) for 3 min under argon. After centrifugation (8OOOg, 5 min), 0.6 ml of the supernatant fraction was incubated with 0.01 ml of ‘OgCd-labeled CdCl, solution (0.66 mM Cd in buffer A, 18.5 MBq/mg Cd) for 5 min, 0.1 ml of Chelex 100 (66% v/v suspension in buffer A) was added, and the mixture was shaken for 15 min. After centrifugation (SOOOg, 5 min), 0.5 ml of the supernatant fraction was incubated with 0.5 ml of acetonitrile for 3 min. Thereafter, the precipitate was removed by centrifugation (SOOOg, 5 min) and 0.9 ml of the supernatant fraction was analyzed for “‘Cd (Minaxi 5500, Canberra-Packard, Frankfurt, FRG). Determination

of the Cu Load of MT

Samples were analyzed with both the thiomolybdate assay and the Cd-Chelex assay (14), which allows the determination of Zn(Cd)-thioneins only. The portion of the MT which binds Cu (Cu load of MT) was calculated from the difference in the results of both assays. In Vitro Preparation and Characterization of Cu-Containing MT Dilutions of the SlOO from liver homogenates or of a stock solution of purified Zn-thionein (14), quantified by the Cd-Chelex assay, were incubated under argon with the same volume of a solution of Cu(CH&N)&104 in acetonitrile for 3 min. The added moles of Cu per mole MT were 6-50 for the purified Zn-thionein, lo-62 for MT in the hepatic SlOO of the Zn-treated rat, and 7-100 for MT in the cytosol of the human liver. The resulting Cu-containing MT (0.1-3 pg) was analyzed by the CdChelex and the thiomolybdate assay with the first addition of acetonitrile omitted. Cu-saturated MT, obtained by incubating purified Zn-thionein with Cu(CH&N)&104 in an added Cu/MT molar ratio of 16, was further characterized after dialysis against 10 mM Tris-HCl, pH 7.4. Gel filtration by HPLC

QUANTITATION

OF

Cu-CONTAINING

37

METALLOTHIONEIN

PI B

100

c

\

rfl

E 80 I

1

E t;

6040-

E

R &h&ob&%& $0 00

10

20

30

200, -15

20

fraction

FIG. 1.

Efficiency of (NH&Ma$ in removing Cu from Cu-containing MT. Cu distribution of gel-cbromatographically separated hepatic cytosol from a Cu-treated rat (A) and of Cu-thionein obtained from purified Zn-thionein (mol Cu added/m01 MT = 16) (B) before (0) and after (A) addition of (NH&MoS, and DEAE-Sephacel.

(Spherogel TSK 2000 SW, 7.5 X 300 mm, 10 pm; mobile phase, 50 mM Tris-HCl, 150 mM NaCl, pH 7.0, saturated with nitrogen; flow rate, 0.5 ml/min; fraction size, 0.5 ml; detection of absorbance at 254 nm; calibrated with Cd5,Zn,-thionein) revealed a single band with the relative elution volume VJV,, = 1.8-2.3, typical for MT. Cu was exclusively found in the MT fractions. Moreover, a 100% exchange of Zn against Cu was confirmed since no Zn could be detected in the MT fractions. EPR spectra of the Cu-thionein exclusively showed EPR-inactive diamagnetic Cu( I). A resonance peak was obtained only after oxidation with hydrogen peroxide (data not shown). RESULTS

Major requirements for the reliable determination of Cu-containing MT with the thiomolybdate assay are (i) the quantitative sequestration of MT-bound copper by ammonium tetrathiomolybdate and (ii) the removal of Cu-thiomolybdate complexes and excessive thiomolybdate without affecting MT. The extent of Cu removed from MT was investigated with gel filtration chromatography: Before and after addition of thiomolybdate and DEAE-Sephacel, hepatic SlOO from a Cu-treated rat and Cu-thionein prepared in vitro were chromatographed and analyzed for Cu. The treatment with thiomolybdate and DEAE-Sephacel resulted in a total removal of MT-bound Cu (Fig. 1). The usefulness of DEAE-Sephacel and 10 mM TrisHCl, 85 mM NaCl, pH 7.4, in removing thiomolybdate and its Cu complexes without affecting MT was verified by the following experiments: After thiomolybdate solutions (100 nmol thiomolybdate with or without Cu(1) or Cu(I1)) with DEAE-Sephacel in buffer A were shaken, neither tetrathiomolybdate nor Cu could be detected in the supernatant fraction by photometry or AAS (results not shown). MT with Cu contents of O-98%, prepared in

is

5

i0

fraction

FIG. 2. Metal distribution of gel-chromatographically separated hepatic cytosol from a Cu-treated rat before and after the thiomolybdate assay. Cu in the native cytosol (O), Cu (a), and Cd (A) of the same sample after the assay.

vitro, and MT of the SlOO of liver homogenates from Znor Cu-treated rats and of human liver were determined with the Cd-Chelex assay in the presence and the absence of DEAE-Sephacel. In all cases, MT analysis was not influenced by DEAE-Sephacel. The specificity of the thiomolybdate assay for MT in biological probes was examined with gel filtration chromatography and metal analysis. The thiomolybdate assay was performed with hepatic SlOO from a Cu-treated rat, and the final supernatant fraction was chromatographically separated. Whereas Cu was virtually absent, Cd was found exclusively in the MT fractions (Fig. 2), additionally demonstrating the effectiveness of acetonitrile in denaturing Cd-binding cytosolic high-molecularweight proteins. Furthermore, up to 0.1 mM, the Cd-binding thiols glutathione and cysteine, which are not denatured by acetonitrile, did not impair the test results (Table l), confirming the high affinity of Chelex toward Cd. The ability of the thiomolybdate assay to quantify MT (0.1-3.0 pg) with Cu contents of O-98% was investigated

TABLE

1

Effect of Glutathione and Cysteine the Thiomolybdate Assay Percentage Concentration

(mM)

0.1 0.5 1.0 5.0 Note. The values in the supernatant present. The values

on

of added

Cd

GSH 0.06 0.09 0.16 0.32 3.03

5 0.03 + 0.03 AZ 0.04 + 0.07 + 0.09

CYS 0.06 0.09 0.25 0.59 3.97

+ 0.03 f 0.03 310.05 k 0.06 -t 0.10

represent the percentages of added Cd remaining fraction of the thiomolybdate assay with no MT given are means f SD from three experiments.

38

KLEIN,

BARTSCH,

AND

SUMMER

2.0 -0 .-: 60

E s 0)

40

z I

20

s

a?

0 0

20

40

Cu (mol

odded/mol

60

60

100

1.0:

:

0.1: : (r = 0.999)

0.01.

'

..I 0.1

0.01

MT)

FIG. 3. Recovery of Cu-containing MT with (open symbols) and the Cd-Chelex assays (closed taining MT was prepared from Zn(Cd)-thioneins Cu(CHsCN)&104. Purified Zn-thionein from rat cytosol from a Zn-treated rat (A, A); human liver + SD (n = 3-6).

1.0

2.0

pg MT per somple the thiomolyhdate symbols). Cu-con(0.1-3 ag) by adding liver (0, 0); hepatic cytosol (Cl, n ); mean

with Cu-thionein prepared by incubating hepatic SlOO from a Zn-treated rat, human liver cytosol, and a stock solution of purified Zn-thionein with Cu(1). The samples were analyzed with both the Cd-Chelex and the thiomolybdate assay. Whereas MT recovery with the Cd-Chelex assay declined with increasing Cu content of MT, recovery with the thiomolybdate assay was 100 +- 10% (Fig. 3). However, for quantitatively recovering purified Cuthionein with a Cu load of more than 30%, addition of 1.5 mg BSA per sample was necessary. Furthermore, complete recovery of purified Cu-thionein added as internal standard to hepatic SlOO of untreated rats requires the presence of CM-Sephadex. At present, the reasons for these sample-specific effects cannot be explained. Because BSA and CM-Sephadex did not influence the quantification of MT in biological samples, and because we attempted to ensure the universal use of the assay, BSA and CM-Sephadex were included in the standard test procedure. With this experimental protocol, the recovery of MT with the thiomolybdate assay was independent of the sample type, MT amount, and Cu content of MT. Linearity and sensitivity of the thiomolybdate assay were determined with various dilutions of the hepatic SlOO from a Cu-treated rat. MT could be reliably determined within the range of 0.014-3.2 pg (Fig. 4). At higher MT contents addition of more thiomolybdate and Cd is required to maintain linearity of the assay. The thiomolybdate assay was used to quantify MT in Cu-treated cultured human fibroblasts and in liver biopsies of patients with Wilson’s disease and primary biliary cirrhosis, diseases associated with increased hepatic Cu levels. Incubation of the fibroblasts with CuClz up to 130 PM did not raise the MT content, although the cytosolic Cu levels were increased about fourfold at 130 PM Cu com-

FIG. 4. containing ag MT/ml)

Linearity and sensitivity of the thiomolybdate assay for CuMT. Dilutions of hepatic cytosol from a Cu-treated rat (60 were analyzed with the thiomolybdate assay (mean f SD,

n= 3).

pared to levels in cells grown in basal medium. As shown by the increase in the Cu load of MT, up to this concentration, Cu was incorporated into endogenous MT. Incubation of the cells with Cu higher 200 PM led to a dosedependent MT induction and a further increase in the Cu load of MT up to 90% at 760 PM Cu (Fig. 5). Biopsies of patients with Wilson’s disease showed high MT contents of 30.2 f 11.7 kg/mg cytosolic protein. Patients with primary biliary cirrhosis revealed hepatic MT levels of 6.5 f 2.9 pg/mg cytosolic protein. In both diseases, the hepatic MT contents correlated with the levels of cytosolic Cu (Table 2). DISCUSSION

The thiomolybdate assay is a fast and sensitive method for determining Cu-containing MT in biological tissues: High-molecular-weight Cd-binding proteins are denatured by acetonitrile, MT-bound Cu is removed by ammonium tetrathiomolybdate, and excessive thiomolybdate and its Cu complexes are eliminated by DEAE-

100 E 60 s

0

200

400 /AA

60

% D

40

B T

2o

2 0

600 cu

FIG. 5. MT and Cu load of MT in cultured human skin fibroblasts. Cells were incubated with CuCl,. MT was analyzed in the cytosolic fractions with the thiomolybdate and the Cd-Chelex assays. The Cu load of MT was calculated from the results of both assays.

QUANTITATION TABLE MT

Control WD PBC

Cu-CONTAINING

2

and Cu Load of MT in Wilson’s Disease (WD) and Primary Biliary Cirrhosis (PBC) Patients cu

Liver specimens

OF

(w&g protein) 135-c 62 (7) 3271 f 1642 (3) 428 f 420 (8)

MT (cc&w protein) 3.8 + 2.3 (7) 30.2 + 11.7 (3) 6.5 f 2.9 (8)

Cu load of MT (%) 14 + 13 (7) 57 2 12 (3) 36 + 21 (8)

Note. The protein corresponds to cytosolic protein. Control samples were obtained postmortem from individuals without indication of liver diseases. Samples from WD and PBC patients were obtained by liver biopsy. The values given are means ? SD, with the number of individuals in parentheses.

Sephacel. Finally, apothionein is saturated with Cd and excessive Cd is bound to Chelex 100. This approach avoids problems inherent in conventional Cd-heme assays, such as those with heating the samples, which may destroy the heat-labile Cu-thionein (Hi), the possible enclosure of MT into heat-denatured hemoglobin, and the removal of Cd from Cd-thionein by excessive hemoglobin (16). Using gel filtration and metal analysis (Fig. l), we were able to demonstrate that ammonium tetrathiomolybdate is effective in quantitatively sequestering Cu from Cu-containing MT, confirming the data of others (9,17). Oxidation of the intermediately formed apothionein was avoided by using nitrogen-saturated solutions and by carrying out the test procedures in an atmosphere of argon until Cd saturation. Moreover, oxidation of the apothionein may be prevented by tetrathiomolybdate due to its highly negative redox potential (18). Although Cd-saturation assays in general measure the Cd-binding capacity rather than the protein itself, we were able to demonstrate by chromatographic analyses and metal determinations with AAS that the thiomolybdate assay is highly specific for MT (Fig. 2). The quality of the assay to recover Cu-thioneins was assessed by using in vitro-prepared Cu-thioneins from different biological sources as standards. These samples exhibit the same characteristics as physiological Cucontaining MT with respect to metal binding, chromatographic behavior, and oxidation state of MT-bound Cu. With its ability to reliably measure 14 ng MT, the thiomolybdate assay, although lo-100 times less sensitive than immunological tests (4,19), is considerably more sensitive than conventional Cd-saturation assays. It is particularly suitable for determining MT in small tissue samples (e.g., liver biopsies of a few milligrams wet

39

METALLOTHIONEIN

weight), extrahepatic tissues, and cell cultures with low MT concentrations. Because it is able to quantify both the Cu-containing MT and, in combination with the CdChelex assay, the Cu load of MT, the thiomolybdate assay is considered superior to the currently available assays for Cu-thionein. We have shown that MT levels in Cu-treated cultured human fibroblasts are increased and that MT is involved in liver diseases associated with Cu accumulation. This provides further evidence that the thiomolybdate assay is a useful and sensitive tool for studies on the contribution of Cu-containing MT to the cellular metabolism and toxicity of Cu. ACKNOWLEDGMENTS The authors thank Drs. G. A. Drasch and J. Eisenburg for supplying the samples of human liver and also thank J. Lichtmannegger for valuable contributions and expert technical assistance. REFERENCES 1. Kiigi, J. H. R., and Kojima, Y. (1987) in Metallothionein J. H. R., and Kojima, Y., Eds.), pp. 25-62, Birkhliuser,

II (Kagi, Basel.

2. Bremner, I. (1987) J. N&r. 117,19-29. 3. Vander Mallie, R. J., and Garvey, J. S. (1979) J. Biol. 8416-8421. 4. Thomas, D. G., Linton, H. J., and Garvey, J. S. (1986) Methods 89,239-247.

Chem.

254,

J. Zmmunol.

5. Scheuhammer, A. M., and Cherian, M. G. (1986) Toricol. Appl. Phnrmacol. 62,417-425. 6. Zelazowski, A. J., Gasyna, Z., and Stillman, M. J. (1989) J. Biol. Chm. 264,17,091-17,099. 7. Waalkes, M. P., Garvey, J. S., and Klaassen, C. D. (1985) Toricol. Appl. Pharmacol. 79,524-527. 8. Dieter, H. H., Miiller, L., Abel, J., and Summer, K. H. (1986) Toxicol. Appl. Pharmncol. 65,380-388. 9. Bremner, I., and Mehra, R. K. (1983) Chem. Ser. 21.117-121. 10. Allen, J. D., and Gawthorne, J. M. (1987) J. Znorg. Biochem. 31, 161-170. 11. Hemmerich, P., and Sigwart, C. (1963) Experientia 19,483-489. 12. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 13. Schramel, P., Klose, B. J., and Hasse, S. (1982) Fresenius’Z. Anal. Chem. 310,209-216. 14. Bartsch, R., Klein, 64,177-180.

D., and Summer,

K. H. (1990)

Arch.

Toxicol.

15. Winge, D. R., and Brouwer, M. (1986) Enuiron. Health Perspect. 65,211-214. 16. Eaton, D. L., and Toal, B. F. (1982) Toxicol. Appl. Pharmucol. 66, 134-142. 17. Kay, J., Cryer, A., Brown, M. W., Norey, C. G., Bremner, I., Overnell, J., Parten, B., and Dunn, B. M. (1987) Biochem. Sot. Trans.

15,453-454. 18. Kelleher, 635. 19. Tohyama, Toxicology

C. A., and Mason, C., Shaikh, 22,181-191.

J. (1986)

Z. A., Ellis,

Znt. J. Biachem.

K. J., and Cohn,

18, 629S. H. (1981)

Quantitation of Cu-containing metallothionein by a Cd-saturation method.

A rapid and sensitive method for determining Cu-containing metallothionein (MT) is described. The main features of this Cd-saturation assay are: high-...
568KB Sizes 0 Downloads 0 Views