BlOMEDTCAL CHROMATOGRAPHY, VOL. 6 , 205-211 (1992)

An Improved Method for Isolation and Identification of Zn-Metallothionein from Cadmium-induced Rat Liver Aihua Pan,* Feng Tie, Binggen Ru, Lingyuan Li and Tong Shen National Laboratory of Protein Engineering, Department of Biology, Peking University, Beijing 100871, China

Zn-metallothioneins (MT-I and MT-2) were isolated and purified from Wistar rat liver induced by subcutaneous injection with cadmium chloride over a short time. Instead of Sephadex G-50 and DEAE Sephadex A-50, new chromatographic media produced by Pharmacia, Sephacryl S-200, S-100and DEAE Sepharose Fast Flow were used in the purification of metallothioneins. The time required for purification with the new method was only 113 that required with the usual method and had the same purification effect and rate of recovery. The number of mercapto groups measured with modified Ellman’s reagent and cysteine as standard is 20 in MT molecules. Zn and Cd concentrations in each fraction were measured by single sweep polarography rather than atomic absorption spectrophotometry. MT-1 and MT-2 contained 6 gram atoms of zinc, but no cadmium. Purified MT-1 and MT-2 were shown hy high performance liquid chromatographic analysis to be highly homogeneous and had an amino acid composition similar to that of Cd-MT.

INTRODUCTION

Metallothioneins are low-molecular-weight, metal- and sulfur-rich proteins which are widely distributed in nature. Metallothionein was discovered by Margoshes and Vallee in 1957. They isolated cadmium- and zinccontaining proteins from equine kidney, an organ known to accumulate cadmium. Kagi and Vallee further purified and characterized this material, which they found to be a low-molecular-weight, metalbinding, cysteine-rich protein (Kagi and Vallee, 1960; 1961a). In 1964, Piscator found that metalothionein was present in increased quantities in the livers of rabbits exposed to cadmium (Piscator, 1964). As of now , metallothioneins have been purified from vertebrates (Shaikh and Luck, 1972), invertebrates (Olafson et al., 1979), microorganisms (Pring and Weser, 1975) and plants. The biological functions of metallothioneins have not been clearly determined, but may include important roles in trace metal metabolism (Bremner, 1987), in protection against and detoxification of heavy metals (Karin, 1985) in the control of cell division and growth (Lebeau et al., 1985), and in the UV response of cells (Schorpp et al. , 1984). In addition, metallothioneins by virtue of their highly unusual structural features, may react sufficiently rapidly with the hydroxyl radicals formed during radiolysis or other forms of oxidative stress, that they may serve as a sacrificial target for oxidative damage (Thornalley and Vasak, 1985). Metalothionein, as an anti-damage material in the human body, has very important significance in medical research and clinical applications. At present, the general procedure for purifying MT from animal tissue by Cd exposure produces cadmium-binding MT, which if used for clinical purposes may cause cadmium toxicity and liver and kidney damage. Therefore there is an urgent need for methods for extraction of

’Author to whom correspondence should be addressed. 0269-3879/92/040205-07 $08.50 01992 by John Wiley & Sons, Ltd.

Zn-binding MT, which is non-toxic, rather than Cd-binding MT. This paper introduces a procedure for purifying Zn-MT without Cd, within a shorter time than previous procedures, by a limited Cd injection. Animal Tissues indued by Metal (Cd or Zn)

c Homogenize

t Centrifuge

t Precipitate with ethanol

c Centrifuge

c Disolve precipitate with 0.01 md/l W - H Q

c Gel filtration

t Ion-exchange

I

I +

+

Fraction 1

t

Fractian 2

t

Concentrate

Conantrate

t

t

Desalt

Desait

t

t

Lyophilizc

t m-

Lyophilize

t 1

MT-2

Figure 1. Schematic diagram of the purification of metallothionein.

Received 6 February 1992 Accepted 20 February 1992

A. PAN E T A L .

206

Figure 2. Gel filtration (Sephacryl S-200)of precipitate with ethanol from liver homogenate of Cd-exposed rats. Column size: 3.0 x 90 cm. Eluted with 0.01 mol/L Tris-HCI, pH 8.6, at a flow rate of 150 mLlh. Monitored by measuring the absorbances at 254, 280 and 220 nm and the concentrations of Cd and Zn by SSP.

During the process of purification and identification of metallothionein Zn and C d concentrations were simultaneously measured by single sweep polarography (SSP) instead of atomic absorption spectrophotometry (AAS).

~~~

Treatment of animals. About five male Wistar rats with body weights ranging from 200 to 25Og were injected subcutaneously with CdCI2 five times. Rats were given 0.3mg of CdCl,/rat on Day 1, 0.6 mg of CdCldrat on Day 2, l.Omg of CdC12/rat on Day 4 and Day 6, and 2.0 mg of CdC12/rat on Day 8. On the second day after the last injection, the rats were killed and their livers were immediately removed.

~

EXPERIMENTAL Materials and equipment. CdC12 was obtained from A. R. Beijing Chemical Factory (Beijing, China), Tris and Ellman's reagent from Sigma Chemical Co. (St Louis, U.S.A.), and Sephacryl S-200, DEAE-Sepharose Fast Flow and Sephacryl S-100 from Pharmacia (Uppsala, Sweden). The UV-240 spectrophotometer was from Shimadzu (Tokyo, Japan), the high speed freezing centrifuge from Hitachi, Tokyo, Japan), the amino acid composition analyser from Beckman (CA, USA), the JP-2 Model oscillopolarographer from Chengdu Equipment Factory (Chengdu, China), the HPLC systems (Model 990) from Waters Associates (Milford, MA. USA), the HPLC column (Bio-Sil SEC 7.5 cm x 7.5 mm) from Bio-Rad (CA, USA).

Isolation and purification of rat liver metallothionein. A schematic diagram of the fractionation procedures is shown in Fig. 1. Livers were cut into small pieces and homogenized. To one volume of the liver homogenate in 0.01 M Tris-HCI buffer, pH 8.6, cold ethanol and chloroform (1.00: 1.05:0.08, by volume) were added with cooling. Three volumes of cold ethanol were subsequently added to the supernatant obtained by centrifugation of this mixture. The solution was allowed to stand at -20 "C overnight. The produced precipitate was collected by centrifugation, dissolved in 20mL of 0 . 0 1 ~ Tris-HC1 buffer (pH8.6) and then charged to a Sephacryl S-200 column (3 x 90 cm). Elution was carried out by use of 0 . 0 1 ~Tris-HC1 buffer (pH8.6) at a flow rate of about 120 mL/h. The fraction containing metallothioneins was applied to a DEAE Sepharose Fast Flow column (3 X 40 cm) equilibrated with starting buffer, 0.01 M Tris-HCI buffer,

ISOLATION AND IDENTIFICATION OF Z:i-METALLOTIIIONEIN

pH8.6. Elution was carried out by a two litre linear salt gradient (limiting buffer, 0.25 M Tris-HCI, pH 8.6). Purified MT-1 and MT-2 were concentrated by lyophilization and applied to a Sephacryl S-100 column (1.6 X 120 cm) equilibrated with 0.01 M ammonium carbonate. Elution was carried out by 0.01 M ammonium carbonate. MT-1 and MT-2 were lyophilized and stored at -20 "C.

Metal determinations. Cadmium, zinc and copper were determined by means of a Philips PU 9200 atomic absorption spectrophotometer, as well as by a Model JP-2 oscillopolarographer. Cadmium. Single sweep polarography was performed using a 5 mL supporting electrolyte of 0.02 M HC1 and KI. The solution was electrolysed at -0.40V for 5 s, resulting in a preconcentration of Cd at the mercury electrode, followed by a cathodic sweep to 0.90 V at a scanning rate of 250 mV/s. The 2nd order derivative of peak current at -0.68V (v5. SCE) was recorded. Zinc. 0.5 mL of 1 M HNO?,20-40 pL fraction solution and 0.5 mL of 3 M HAc were mixed and 50 pL of 4 x 10 -'M a ,

MT-1

10 20

70

80

100

207

a'-dipyridyl and H 2 0 added to 5 mL. The mixed solution was then transferred to a S m L flask and the derivative peak current was then recorded at -1.13 V (vs. SCE).

Molecular weight estimates. Purified MT-1 and MT-2 were dissolved in pH 8.6 10 mM Tris-HCI buffer. The standard proteins and different amounts of MT-1 and MT-2 were determined by an HPLC filtration chromatograph. A 10 pL aliquot of the samples was applied to the filtration column and eluted with 0.05 M NazSOj, 0.05 M NaH,PO, (pH 6.8) at a flow rate of 1mL/min. The standard proteins were detected at 280 nrn and MT-1 and MT-2 at 254 nm. Amino acid composition analysis. Each sample was hydrolysed at 110 "C with 0.5 mL of 5.7 N HCl in vacuum for 24 h. The dried hydrolysates were then dissolved in 1 mL of 0.2 M sodium citrate buffer, pH 2.2. Another sample was oxidized by performic acid at -10 "Cfor 4 h. The amino acid compositions were determined with a Beckman 121 MB amino acid analyser.

MT-2

120

140

160

180

Tube number (10rnl/Tube) Figure 3. Ion exchange chromatography (DEAE Sepharose Fast Flow) of Zn-binding protein fraction separated by gel filtration. Column size: 3.0 x 40 cm. Eluted with the gradient with a mixing chamber containing 1 L of 0.01 mol/L Tris-HCI, pH 8.6, and a reservoir containing 1 L of 0.25 mollL Tris-HCI. pH 8.6, with a flow rate of 120 mL/h. Monitored by measuring the absorbances at 254,280 and 220 nm and the concentration of Cd and Zn by SSP.

A. PAN E T A L

208

MT- I

5

10

15

20

A

2 5 30

Tube number ( Srnl/Tube)

5

10

15

20 25 3 0

Tube number ( Sml/Tube)

Figure 4. Elution profile of desalination of Cd-exposed rat liver MT-l(A) and MT-Z(B) on Sephacryl S-100. Column size: 1 . 6 ~ 120 cm. Equilibrated with 0.01 mol/L ammonium carbonate and eluted with the same solution at a flow rate of 60 mL/h.

Reaction of mercapto groups. Mercapto groups were titrated with 0.2 mM S,S'-dithiobis 2-nitrobenzoic acid (DTNB) in 0.1 M phosphate buffered saline (PBS) (pH 7.3) containing 1mM ethylene diamine-tetraacetic acid (EDTA).

Table 1. Amino acid composition of MT-1 and MT-2 in rat liver induced by cadmium injection Amino acid

MT-1 Probable number Relative molar of residues per quantitiersa molecule

MT-2 Relative molar quantitiesa

Probable number of residues per molecule

Met 1.0(1) 1 1.4(1) 1 4.4(4) 4 4.3(4) 4 ASP 2 2 2.1(2) Pro 1.9(2) Cysb 19.5(20) 20 19.8(20) 20 Ser 10.5(11) 11 9.1(9) 10 Thr 3.9(4) 4 2.0(2) 2 6 4.3(4) 4 GlY 6.1 (6) 8 7 8.4(8) LYS 7.3(7) Va I 2.1(2) 2 1.0(1) 1 Ile 0.1 (0) 0 0.9(1) 1 Glu 1.5(2) 1 3.2(3) 3 a Values represent the average of samples hydrolysed for 24 h. Determined from cysteic acid content in performic acidoxidized protein.

0

MINUTES

5

10

IS

MINUTES

Figure 5. Chromatographic properties of purified MTs as determined by HPLC. HPLC column and buffer: see Experimental Section. MT-1 or MT-2 (20 pg) in a volume of 10 yL were injected into a gel filtration column. Peak 1: dimer of metallothionein; Peak 2: tetramer of metallothionein.

ISOLATION AND IDENTIFICATION OF zn -METALLOTHIONEIN 09-05-1991 14:17:46

SanPlln9 time 2 6 7 mscc*4

- scale

Y

1 . 5 4 AU/FS

S a m p l e name

mt-1

Retention t i m e 8.54 F i n

9.06 nln

~

~

~

End

~

~

200

I

~

"

~

~

'

~

"

"

"

"

'

'

"

220 240 W a u e l e n g t t , 190

[Ell Change

~

'

"

---

"

'

380

"

~

260

'

'

"

~

'

'

"

280

'

~

~

~

'

388

(5nm)

ninb

Digital

Plot

Waters

Figure 6. UV absorption spectra of MT-2 purified from rat liver induced by cadmium. (A) Peak a t 8.34 min retention time on HPLC (tetramer); (6)peak at 9.06 rnin retention time on HPLC (dimer).

.1?-35-19'1 1+:17:43

sa.r1,n.,

II.,

267 *s*c-*

SaaPle name at-1

8.54'

........,.........,..................................

.

, . . ,,

228

230

244

uautlanqth

I . .

i?e

I . .

8

--,

260

280

Z8o nm (snm)

I , . . -

a

B

\

1 4 :17 :4 0 S..Pl,"* 1.m* 267 . + * C ' d

-Y

S l n P L r ..an

'\

at-1

:--/,

\'

\\

i

/

',

21

1

-

.

'\

'11

_---

- scale

.042 , F S

\

-

_____-I

9.eC.

/' __

/"

\\J I

,

'

1 . ' '

I

. I ' ' ' . I

'

I

'

I

I

____

__I.__

210

A. PAN E T A L . 0

attributed to the non-globular shape of the protein. In this experiment, the molecular weight of rat metallothionein is 8500 Dalton. This value comes from HPLC gel filtration (Fig. 5) combined with amino acid composition analysis (Table 1). In Fig. 5 , Peaks 1and 2 are the dimer and tetramer of metallothionein, respectively. Two peaks were analysed with UV absorption spectra (Fig. 6) and derivative spectra (Fig. 7). The results show that there is a remarkable level of absorption at about 220 nm. The optical features of metallothionein reflect both the metal and amino acid composition. Due to the absence of aromatic amino acids, there is no protein absorbence at 280 nm. There are, however, broad absorption bands, which are typical of the metal thiolate (mercaptide) complexes of the constituent metals. These absorptions are manifested as shoulders at about 250 nm (Cd) and 220 nm (Zn). On adding 20 pL of 1mM CdC12 to replace Zn binding in metallothionein, absorption shoulders appeared at 250 nm, but on removal of the metal at low pH, these absorption shoulders disappeared (shown in Fig. 8).

a . n

2

-____

0

-

0

190

220

250

280

Wnvelenglh(nrn)

300

190

220

250

280

300

W.lrclc,,glll(ntll j

Figure 8. UV absorption spectra of rat liver MT (Cd-exposed) containing 6 Zn per molecule. (1) MT-1 (A) or MT-2 (6)(40 pg/ mL) in Tris-HCI, pH 8.6; (2) 1 +20 pL of 1 mM CdCI,, stand for 1 min at room temperature prior to spectroscopy; (3) 2+200 pL of 2 N H2S04, stand for 1 min at room temperature prior to spectroscopy.

Spectroscopic properties. Purified MT- 1 and MT-2 dissolved in PBS (1mg MT/mL) were scanned with ultraviolet from 190 to 300 nm. 40 pL of 2 N HCI was then added, followed by scanning with the same UV as before.

RESULTS

The gel filtration pattern of the precipitate with ethanol from the liver homogenate of Cd-exposed rats is shown in Fig. 2. The ion exchange chromatographic elution profile of the Zn-binding protein fraction separated by gel filtration is shown in Fig. 3. Purified MT-1 or MT-2 was applied to a Sephacryl S-100 column equilibrated with 0.01 M ammonium carbonate (Fig. 4). Absorbance at 250, 280 and 220 nm in all the fractions above were measured with a UV-240 spectrophotometer. 20 and 30mg of MT-1 and MT-2, respectively, were isolated from the livers of five Cd-exposed rats. Zn and Cd concentrations in each fraction were measured by SSP rather than by AAS. Unexpectedly, the content of Cd as well as Cu was negligible. MT-1 and MT-2 contain 6 atom Zn/mol and 20 mercapto groups, but Zn-MTs and Cd-MTs have a similar amino acid composition (shown in Table 1). The chromatographic properites of purified MT-1(A) and MT-2(B) as determined by HPLC are shown in Fig. 5. Native mammalian metallothioneins are a single-chain protein with a molecular weight ranging from 6500 to 7000, depending on the metal composition. When determined by gel filtration, the molecular weight of mammalian metallothioneins is considerably higher, i.e. about 10,000. This discrepancy can be

DISCUSSION

Isolation procedures for metallothioneins have included fractionation by salt and/or organic solvents, heat treatment, gel filtration, ion exchange chromatography and isoelectric focussing. However, gel filtration and ion exchange chromatography have become the most popular methods. In our experiment, the new chromatographic media (Sephacryl S-200, S-100 and DEAE Sepharose Fast Flow, from Pharmarcia-LKB) were used to replace the Sephadex C-SO or (3-75 and DEAE Sephadex A-50 which had been used in the usual procedures for purification of metallothioneins. This makes it possible to separate and purify metallothioneins from the sample quickly (in only one third of the usual time), but has the same separation efficiency and recovery rate as the usual method. Another improvement in our methods was that zinc and cadmium concentrations in each fraction were measured by SSP instead of by AAS, which was used in the usual method (Pan et al., 1991). This improvement has several advantages: (1) cheaper equipment, (2) easier operation, (3) better reproducibility and (4) shorter time. The results obtained by SSP are consistent with those given by AAS, so the metal ion determination in purification and identification of metallothionein can be completely carried out by SSP. In the usual method, metallothioneins were purified from animal tissues induced by large dose injections of Cd, each purified MT molecule bound 4-5 Cd and the production rate was about 1mg MT/g wet tissue. For example, when we purified mouse liver MT, mice were injected with Cd several times (total amount of Cd: 22 mg Cd/kg body weight) in 12 days, and the product rate was 1mg MT/g wet tissue (Ru et al., 1991). In this experiment, we took Zn-MT-1 and Zn-MT-2 (no Cd) from rat livers induced with a limited dose of Cd (total: 17.6 mg Cd/kg body weight) in a shorter time (8 days). There is no difference between the MTs purified by the two methods except metal binding. It is reported that the binding of metallothionein to Cd is 10,000 times as strong as that to Zn. Why then

ISOLATION AND IDENTIFICATION OF Zn-METALLOTHIONEIN

does MT purified from rat liver induced with Cd injection only contain Zn-MT? A possible reason is that MT is synthesized on a large scale when Cd enters the liver, but Cd-binding MT is transported to the kidney because of the body's protective response to toxic metal, leaving only Zn-MT in the liver. Metallothioneins have an important biological and physiological function, and are potentially very important in medical research and in clinical application as a protein drug. Usually, MT purchased from Sigma company contains 4-5Cd, which must be removed by a series of complex procedures before use in medical

211

research. Our experiment indicates that a limited dose of Cd (