Characterization of Cadmium-Binding Complex of Cabbage Leaves' · CHARACTERIZATION OFCABBAGE LEAF Cd-COMPLEX (last measured in H20) was added to individual solutions and after 5 min

Plant Physiol. (1984) 76, 797-8050032-0889/84/76/0797/09/$01.00/0

Characterization of a Cadmium-Binding Complex of CabbageLeaves'

Received for publication March 16, 1984 and in revised form July 13, 1984

GEORGE J. WAGNER*2Biology Department, Brookhaven National Laboratory, Upton, New York 11973

ABSTRACT

The chemical nature of a principal, inducible cadmium-binding com-plex which accumulates in cabbage leaves (Wagner and Trotter 1982Plant Physiol 69:804809) was studied and compared with that of animalmetallothionein and copper-binding proteins isolated from various orga-nisms. The apparent molecular weight of native cabbage complex andcarboxymethylated ligand of the complex under native conditions asdetermined by gel filtration was about 10,000 daltons. Under denaturingconditions their apparent molecular weights were about 2000 daltons.Ligand of native complex contained 37, 28, and 9 residue per cent ofglutamic acid-glutamine, cysteine, and glycine, respectively, and lowaromatic residue, serine and lysine content. The high acidic and lowhydrophobic residue content explain the behavior of complex on electro-phoresis in the presence and absence of sodium dodecyl sulfate. Itsisoelectric point was below 4.0 and it bound 4 to 6 moles cadmium permole ligand in what appear to be cadmium-mercaptide chromophores.The complex was found to be heat stable, relatively protease insensitive,and lacking in disulfide bonds. Attempts to determine the primary se-quence of reduced native complex and carboxymethylated, cleaved ligndusing the Edman degradation procedure were unsuccessful. An electro-phoretic procedure is described for preparative isolation of purifiedcomplex and a method is described for monitoring ligand of complex asits fluorescent dibromobimane adduct.

Vegetable foods are the principal source of dietary Cd in man(20), yet we know little about the forms of Cd which occur inplants or their fates after ingestion in animals and man. Incontrast, much is known about the principal form of Cd whichoccurs in animals and its fate after ingestion (4, 9). In animaltissues, most of the Cd occurs as an inducible, about 6,500 D(10,000 D apparent mol wt by gel filtration under nondenaturingconditions), cysteine-rich protein called Cd,Zn-thionein or moregenerally, metallothionein (9). Cd and Zn are bound to this-protein in mercaptide bonds with cysteine, and some of thesebonds bridge the single polypeptide chain of the ligand to formwhat is thought to be a compact S-shaped molecule (9). Theamino acid sequence of Cd,Zn-thionein has been determinedand shown to be well conserved in several animal and crustaceanspecies (15). The physiological functions of Cd,Zn-thionein arenot established but it is thought to have roles in toxic metal (Cd,Hg, Cu, Ag) sequestration and in Zn and Cu homeostasis (9, 27).Recently, a Cu thionein was isolated from Neurospora and was

' Research carried out at the Brookhaven National Laboratory underthe auspices of the United States Department of Energy.

2Present address: Agronomy Department, University of Kentucky,Lexington, KY 40546-0091.

shown to have sequence homology with Cd,Zn-thionein, tosequester Cd, and to efficiently donate Cu to apotyrosinase (14).Apparently inducible, 10,000 D (1OK), Cd-complexes (appar-

ent mol wt determined by gel filtration) have been isolated fromseveral tissues of several plant species (1-3, 10, 24, 25, 28).Induction of 1OK, Cd-complex of cabbage and tobacco leaveswas shown to be Cd-concentration and time dependent (25).Plant 1OK D Cd-complexes are often the principal Cd forms intissues containing relatively high levels of the metal but they arenot found in all tissues of all plant species which are exposed toCd (23). It is not known if they are principal Cd-forms in plantsexposed to low levels of Cd. Also, it is not known if these littlecharacterized plant Cd complexes are chemically or physiologi-cally related to Cd,Zn-thionein, yet they have been referred to asmetallothionein-like by us and others. Their inducibility andapparent mol wt are metallothionein-like but their highly anioniccharacter is not (1, 24, 25). In another case, the principal Cd-binding constituent of wheat grain is apparently not inducibleby Cd and not highly enriched in cysteine, yet it binds about60% of the Cd in grain (26; G. J. Wagner, unpublished data). Itsfate in animals has recently been tested (26). Cu-binding com-plexes which have relatively high cysteine content but haveamino acid compositions which differ somewhat from that ofanimal metallothionein have been isolated from animal liver(30), yeast (29), Scenedesmus (22), and from Agrostis roots (17).Nonmetallothionein-like Cd complexes have been isolated fromoysters, one ofwhich is high in acidic residues and low in cysteine(20).Here we have examined certain chemical and physical prop-

erties of cabbage lOK,Cd-complex and have compared these tocharacteristics of animal Cd,Zn-thionein and Cu complexes iso-lated from various species.

MATERIALS AND METHODSPlant Growth, Extraction, and Monitoring of Complex. Cab-

bage plants (Brassica capitate L., cv red danish) were grown inHoagland solution and were labeled with 35S as previously de-scribed (25). Leaves were harvested from plants which had beengrown in Cd-free medium for 10 d and subsequently in freshmedium containing approximately 90 ,M Cd as CdSO4 for 20 d.Preparation of extracts, gel filtration chromatography (nonden-aturing conditions), and atomic absorption analysis were aspreviously described (25). Briefly, extracts were made of frozen(-30°C) tissue by homogenization in 4°C, 25 mm K-phosphatebuffer (pH 7.5), 2 mM DTT. Crystals of dry ice were addedduring grinding to displace 02 and to maintain low temperature.The homogenate was filtered through gauze, the filtrate centri-fuged at 15,000g for 5 min, the resulting supernatant extractedwith an equal volume of CHCl3-BuOH (10:1, v/v), and theaqueous phase desalted on G-25 course as previously described(25). Fractions containing 4,000 to 18,000 mol wt materials werepooled, dialyzed (using Spectrapore-3 tubing, 3,500 mol wt cut-

797 www.plantphysiol.orgon April 7, 2020 - Published by Downloaded from

Copyright © 1984 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

7Plant Physiol. Vol. 76, 1984

off-Fisher Sci.), and lyophylized. The resulting material is re-ferred to as G-50 prepared 10K,Cd-complex. Although 10K,Cd-complex was found to be heat stable (see "Results"), we avoidedheat denaturation of contaminants to prevent possible adverseeffects on the secondary structure of the Cd-complex. Exchangebinding of '"9Cd to complex (25, 26) was effected by incubation(at 22C for 15 min) of about 0.1 MuCi of KOH neutralized'°9CdCl2 (4 Ci/gm, New England Nuclear) in -5 ml samplecontaining complex at pH 7 to 8.Samples containing 14C or 3IS were dissolved in Aquasol (New

England Nuclear) containing 10% (v/v) H20 and counted usinga quench corrected liquid scintillation system.Gel Electrophoresis. Three gel systems were utilized to char-

acterize cabbage 10,000 D complex. In all cases, running andstacking gels (1.8 mm thickness) consisted of 20% and 5% (v/v),respectively, acrylamide-N, N'-diallyltartardiaminde (9.38 to 1).The use of this cross link enabled dissolution of the gel byovernight treatment with 2% (w/v) periodic acid at room tem-perature (31). Gel system A was that of Laemmli without SDSand EDTA (13). Gel system B was a modification of that ofBardoff et al. (1) which consisted of running gel: 0.3 M bis Tris/HCI (pH 6.9), 0.056% (w/v) ammonium peroxydisulfate, 0.039%(v/v) TEMED,3 and the stacking gel: 60 mM bis Tris/HCl(pH5.9), 0.25% (w/v) ammonium peroxydisulfate, 0.18%TEMED, and the electrode buffer: 20 mM bis Tris/glutamic acid(pH 6.5). Gel system C contained running gel: 0.15 M Tris/HCl(pH 7.6), 0.028% (w/v) ammonium peroxydisulfate, 0.019% (v/v) TEMED, 1.0% SDS, and the stacking gel: 0.13 M Na cacodyl-ate (pH 5.6), 0.13% (w/v) ammonium peroxydisulfate, 0.1% (v/v) TEMED, 1.0% SDS, and the electrode buffer. 0.15 M Tricine/HCI (pH 6.95), 0.1% SDS. Samples for gels in all cases contained10 mM K2HPO4/KH2KPO4 (pH 7.0), 10% (v/v) glycerol, andelectrophoresis was at 25 mamp/gel for 3 to 5 h.

For electrophoretic-preparative purification of 10K-Cd-com-plex, system B (running gel only) was used. In a typical experi-ment, 65 g of frozen-lyophilized leaves harvested from 30 plantswere used to prepare G-25 coarse desalted material (see above)which was dialyzed and lyophilized. The entire samples wassuspended in buffer plus glyerol, dinitropyridyl aspartate (pKa <3.6) and bromophenol blue (pKa 4.0) were added, and themixture was applied to two tube gels (0.2 x 3 cm), the outlets ofwhich were each enclosed (see Fig. 2) with a 6-cm piece ofSpectarpore-3 dialysis tubing. The lower portion ofthe gels (withentire tubing) were completely submerged in lower electrodebuffer. Electrophoresis was at 35 mamp tube for 3 h. Afterdinitropyridyl aspartate has completely passed through the geland as bromophenol blue began to emerge from the gel (Fig. 2),electrophoresis was discontinued and the tubing with contentsremoved, closed, and the sample was dialyzed against 15 L ofwater with continuous exchange for 6 to 12 h before lyophiliza-tion. The product of this procedure is referred to as electropho-retically purified complex. From 65 g of leaves, 4.5 mg of light-tan colored material was obtained which contained 0.9 ,umol Cd,reflecting a loss of one-half to one-third of Cd of G-50 preparedcomplex. Yield based on amino acid and Cd analysis was 4.3mg.

Isoelectric Focusing and Formation of Fluorescent Adductswith Dibromobimane. The system described by Righetti andDrysdale (18) was used for isoelectric focusing. Samples (0.2 mgin 200 ul 50 mm K-phosphate buffer (pH 7), 6 M urea wereconverted to their fluorescent dibromobimane derivatives byaddition of 1 ul of 50 mM Thiolite DB (Calbiochem-Behring) inacetonitrile. After 10 min at 22C, samples were carefully layeredover gels and under 20 mM NaOH upper electrode buffer. When

3Abbreviation: TEMED, N,NN'N'tetramethyl ethylenediamine;PMSF, phenylmethylsulfonylfluoride.

complex had reached equilibrium (determined with long wave-length UV lamp), gels were removed and photographed (type 55polaroid film) over a 340-nm transilluminator using a Wratten75 filter. Excitation and emission wavelengths ofadducts are 370and 480 nm, respectively. Parallel gels were fixed and stainedand the pH gradient was determined after leaching slices ofparallel gels in freshly boiled, deionized H20 for 12 h. Rat livermetallothionein was a gift of G. Cherian ([4] for methods ofpreparation).Amino Acid Analysis. Automated amino acid analysis (25)

was utilized after hydrolysis of samples in 6 M Ha at 1 10°C for22 h in sealed tubes. Performic acid oxidation (8) was used forcysteic acid and carboxymethylcysteic acid determination andtryptophan was determined after hydrolysis with 4 N methane-sulfonic acid containing 0.2% 3-(2 aminoethyl) indole (19).Carboxymethylation of Lignd. Ligand was carboxymethyl-

ated essentially as described by Fohlman et al. (6). To removemetal, 1.5 mg of electrophoretically purified comlex was sus-pended in 1 ml of 6 M urea, 0.1 N HC (pH 2.4) and desalted onSephadex G-25 coarse equilibrated with Urea-Ha. The voidvolume was dialyzed in Spectrapore-3 tubing (see above) against16 L H20 for 12 h at 4°C and lyophilized. Recovery of ligand(determined by amino acid analysis) was 70%. Sample wasdissolved in N2-gassed buffer (6) and reduced by addition of a 7-fold molar excess of 2-mercaptoethanol. Carboxymethylationwas at pH 8.2 with 3 MCi iodoacetamide-l-'4C (23 mCi/mmol,NEN) for 1 h followed by addition of excess iodoacetic acid. Thereaction was terminated with acetic acid and the products di-alyzed overnight against 0.1 N acetic acid. Radiochemical yieldwas 70%.

Effects of Hydrolases, Heat, and pH. Sensitivity of nativecomplex to various hydrolytic enzymes was tested using 20 to 60mg of G-50 prepared 1OK,Cd-complex. Controls utilized boiledenzymes and Azocoll (Calbiochem-Behring), the latter as a gen-eral protease substrate. Products ofincubations were fractionatedon Sephadex G-50 and monitored as "2Cd. Incubation condi-tions were as follows: Protease IV (Sigma), 2 ml 0.1 M Na borate(pH 7.4), 40 Mg enzyme, reaction at 37°C, 23 h; Thermolysin(Boehringer Mannheim), 2 ml 0.1 M Na borate (pH 7.2), 40 Mugenzyme, reaction at 35°C, 3 h; Cellulysin (Calbiochem Behring),2 ml 10 M K-phosphate buffer (pH 5.8), 3.5% (w/v) enzyme,reaction at 37°C, 60 min; DNaseI, Type I (Sigma), 2 ml 0.1 MNa acetate (pH 5.5), 5 mM MgCl2, 175 Mg enzyme, reaction 25C,60 min; S. aureus protease (Pierce), 200 Ml Tris/Hepes (pH 8),270 Mg enzyme, reaction 37C, 10 min. In all cases but DNaseI,substantial or complete digestion of Azocoll was apparent at theend ofthe incubation period. Boiled controls showed no digestionof Azocoll.Heat stability of native complex was, examined using G-50

prepared, lOK,Cd-complex which was labeled with "09Cd byexchange binding (see above). Unbound '"Cd was removed bydesalting on a G-25 course column calibrated to distinguishlabeled 1OK-Cd-complex from free '09Cd. The presalt fractionswere heated for 2 min at various temperatures and after heatingsampleswere desalted using G-25 course. Percent bound (presalt)"'Cd and "12Cd were determined as per cent of eluted Cdoccurring in presalt fractions.

Aliquots containing G-50 prepared, IOK,Cd-complex werelabeled with '09Cd and exposed to 25 mM Na citrate buffer (pHrange 2.5-6.0) for 5 min and desalted as above. Per cent of 1'12Cdand '09Cd which remained bound was determined as describedfor heat stability experiments.

Effects of Zn, Hg, and Acid on the UV Spectrum of 1OK,Cd-Complex. The absorption spectra between 230 and 290 nm offour solutions, each containing 1 ml of2 mm K-phosphate buffer(pH 7.6) and 5 mg G-50 prepared, lOK,Cd-complex were deter-mined. Subsequently, 10 Ml of 10 mg/ml ZnCl2, HgCl2 or AgNO3

798 WAGNER

www.plantphysiol.orgon April 7, 2020 - Published by Downloaded from Copyright © 1984 American Society of Plant Biologists. All rights reserved.


CHARACTERIZATION OF CABBAGE LEAF Cd-COMPLEX

(last measured in H20) was added to individual solutions andafter 5 min their UV spectra were re-examined. The fourthsolution was titrated with HCI to pH 2.2, the spectrum deter-mined, and 1 ,uCi of I 9CdCl2 was added before the pH wasreturned to 7.6 with KOH and the UV spectrum re-examined.The last sample was subsequently separated on Sephadex G-50fine and the '09Cd profile determined. The last experiment wasrepeated using electrophoretically purified complex.Gel Filtration under Dissociation Conditions. Behavior of na-

tive 10K,Cd-complex, carboxymethylated ligand and rat liverCd-thionein in the presence of 6 M Guanidine HCI (GuHCl)were examined using a 2.5 x 60 cm column containing BioRadA-0.5 m, 100 to 200 mesh resin. Sample and elution mediumconsisted of 6 M GuHCI (Fisher reagent), 0.1 M 2-mercaptoeth-anol and samples and standards were equilibrated with mediumfor 2 to 3 h prior to chromatography. A constant flow rate of0.42 ml/min was delivered by a Masterflex variable speed pump(Cole-Parmer Inst. Co.) and 1.95 ml fractions were collected.The column was standardized with Blue Dextran, 0.5 to 3 mg ofovalbumin, chymotrypsinogen, Cyt c, lima bean trypsin inhibi-tor, insulin (A and B chains), and CnBr fragments of Cyt c. Thefirst three were monitored as A280 and Cyt c as A409. All otherstandards, and also Cyt c, were labeled with fluorescamine priorto chromatography. In each case, 100 Al of 1% w/v fluorescamine(Roche Fluram, Fisher Scientific) in acetone was rapidly addedto and mixed (5 s) with 300 Al H20 containing 0.5 to 1.5 mgprotein. Acetone was removed with an N2 stream and the samplediluted with 3 ml 6 M GuHCl, 0.1 M 2-mercaptoethanol andchromatographed. Mixing experiments in which unreacted andfluorescamine reacted Cyt c and CnBr-Cyt c fragments werecombined and run simultaneously showed that fluorescaminelabel had little effect on gel filtration (see "Results"). CyanogenBromide fragments I, II, and III of Cyt c (Type III-Sigma) wereprepared essentially as described by Steers et al. (21). Excitationand emission used to monitor fluorescamine adducts were 380and 480 nm, respectively.

RESULTSFigure 1 describes the G-50 fine elution profile obtained after

isolation of 1OK,Cd-complex from cabbage leaves. Typically,fractions 68 through 95 were pooled, dialyzed, and lyophilizedto yield what is referred to as G-50 prepared 10K,Cd-complex(see "Materials and Methods"). Calibration of the column indi-cated that components in fraction 80 had an apparent nativemol wt of 10,000. The A250/A280 ratio of material in the peak Cdfraction was 1.5. However, protein as measured using the BioRad protein assay (Bio Rad Labs., Richmond, CA) was relativelylow in this region suggesting the presence of nonprotein, UV-absorbing material. In the experiment described in Figure 1,tissue obtained from plants labeled with 35S (25) was includedduring homogenization and 35S in the 10,000 D region wascoincident with 112Cd. In a similar experiment (results notshown), and as reported perviously (25), where '09Cd was intro-duced to an extract after CHCl3-BuOH extraction, '09Cd and"'-Cd were found to have identical elution in the region of10,000 D.The inset of Figure 1 shows results obtained after native gel

electrophoresis of fractions recovered after G-50 chromatogra-phy. Fractions labeled A through H were independently pooled,dialyzed, lyophilized, and a portion of each was separated usinggel system A without SDS and EDTA. As shown, slow migratingproteins were prominent in the near void volume fractions Athrough C and were reduced in the near 10K fractions D throughF, while stained material at the electrophoretic front (visible onthe unstained gel as a tan-colored band in fractions B throughG) was maximal in fractions E and F and absent from fractionA. A parallel gel was vertically sliced into 2-mm segments, the

segments were leached overnight in 20 mm K-phosphate buffer(pH 7.4), and the residual gel was dissolved with 2% (w/v)periodic acid. Ninety per cent of the "5S recovered from the runof fraction E occurred in the leach solution ofthe segment at theelectrophoretic front. Another 5% was in the correspondingdissolved gel segment (data not shown). In a third parallel gel,'09Cd was added (to effect exchange binding) to an aliquot offraction E prior to lyophilization. After electrophoresis, the gelwas scanned with a radiochromatogram scanner. Most of the'09Cd had been lost during electrophoresis (found in the upperelectrode buffer-glycine is a divalent cation chelator) but a traceoccurred at the electrophoretic front (data not shown).The above results using native gel electrophoresis suggested

the lOKCd-complex had a highly anionic character and Cd wasmostly lost on electrophoresis in gel system A. In an attempt toresolve the mol wt ofcomplex we examined the migration of"Slabeled, G-50 prepared 1OK,Cd-complex using gel systems A andC containing SDS and EDTA. After electrophoresis the gels weresliced, slices were dissolved, and an aliquot was used to determine35S. In both systems, essentially all of the "S was associated witha still tan-colored band which migrated with the electrophoreticfront. In contrast, low mol wt standard proteins (insulin A andB chains and bovine trypsin inhibitor mol wt, 2500, 3500, and6200, respectively) migrated more slowly (data not shown). Aportion of the dissolved gel slice corresponding to the electro-phoretic front of the above described SDS (system A) gel wasseparated on G-50 fine and 35S eluted as a 10,000 D component(not shown). These results indicate that the charge characteristicsof 1OK complex were not altered by the presence of SDS.Gel system B, essentially that used by Bartloff et al. (1) to

retard migration oftomato root 10,000 D Cd-complex, was usedfor preparative electrophoretic purification ofcomplex. Figure 2diagramatically represents the procedure used in which advan-tage was taken of the highly anionic character of lOK,Cd-com-plex to cause complex to migrate off of the gel and be collectedwhile contaminants remained in the gel. When this system wasused for preparative electrophoretic purification, migration oftan color associated with complex was obscured by residualpigments in extracts. Therefore, material eluting after dinitro-pyridyl aspartate and before bromophenol blue was collected.Staining the gel with Coomassie blue showed that slow migratingcomponents (shown diagramatically in Fig. 2; also Fig. 1 inset-wells B and C) were retained in the gel, as were residual pigments.Gel filtration ofelectrophoretically purified complex showed thatits apparent nondenatured mol wt was 10,000.To verify the anionic character ofcomplex, isoelectric focusing

was performed after labeling ligand with the thiol reactive reagentdibromobimane. Rat liver metallothionein was also analyzed.The dibromobimane adduct of G-50 prepared, electrophoreti-cally purified, cabbage complex and '4C-carboxymethylated li-gand migrated as a single band with a pl of below 4.0 (near theend of the pH gradient) while adduct(s) of metallothioneinoccurred as at least three bands having pls of 4 to 4.75 (Fig. 3).Metallothionein bands may represent isometallothioneins as re-cently separated by reverse phase HPLC (11) or molecules whichvary in the number of acidic mercaptide chromaphores present.With both cabbage lOK,Cd-complex and metallothionein, noeffort was made to remove Cd prior to adduct formation. Sincemuch Cd is removed from cabbage complex during electropho-retic purification, more ligand cysteine sulfurs were presumablyavailable for adduct formation in the former than in the latter.Detailed studies are underway to examine the effects ofremovingmetal on dibromibimane adduct formation and on the isoelectricpoints of cabbage complex and animal metallothionein.The amino acid composition of electrophoretically purified

1OK,Cd-complex and carboxymethylated ligand (determined inthis work) and of rat liver Cd,Zn-thionein (30), Agrostis Cu

799



Plant Physiol. Vol. 76, 1984

FIG. 1. Elution profile (Sephadex G-50,nondenaturing conditions) of cabbage leaf,soluble components to prepare G-50 pre-pared lOKCd-complex (see "Materials andMethods" for details). Leaf material from33S-labeled-Cd-treated plants was includedin the preparation to allow comparison of35 and "2Cd. Protein ( wa-)ws deter-mined with the BioRad reagent. Inset showselectrophoresis of fractions A through Husing gel system A (without SDS or EDTA).Proteins were stained with Coomassie blue.

complex (17), and Neurospora Cu complex (14) (reported byothers) are shown in Table I. Outstanding features are the lowlysine and serine content, the relatively low nonpolar residuecontent, and the extremely high glutamic acid-glutamine contentof cabbage complex relative to Cd,Zn-thionein. Acid hydrolysisconverts glutamine to glutamic acid and asparagne to asparticacid so these are not distinguished here. Agrostis rootCu complexalso had high glutamic acid-glutamine content while Neurosporacomplex had very high serine and glycine and no glutamic acidor glutamine. Neurospora and cabbage complex had 28 residueper cent cysteine while that for Cd,Zn-thionein and Agrostis Cucomplex were 24 to 34 and 18 residue per cent, respectively.Treatment ofG-50 prepared, electrophoretically purified cabbagecomplex with 6 N HCI at room temperature followed by removalof HCI and analysis indicated that strong acid conditions notsufficient to cause hydrolysis of peptide bonds did not degradecomplex and release free amino acids (not shown).

Carboxymethylation of G-50 prepiad, electrophoreticallypurified, reduced and nonreduced ligand yielded stoichiometricconversion of cysteine to carboxymethyl cysteine (Table I), in-dicating that native complex lacks disulfide bonds. Gel electro-phoresis of '4C-carboxymethylated ligand using gel system A orC with SDS showed that essentially all the '4C migrated with thestill, tan-colored band at the electrophoretic front. This resultsuggests that tan-colored component/contaminant is not coor-dinated with cysteine sulfur. Carboxymethylated ligand is ex-pected to be highly anionic. An amino acid analysis was madeof HCO digested, '4C-carboxymethylated ligand in which thecalibrated column eluate was collected and analyzed for radio-activity. About 90% of the '4C applied as an HCI hydrolysate of'4C-carboxymethylated ligand eluted with the RF of carboxy-methyl cysteine (not shown).

Results of attempts to reduce the mol wt (degrade) native Go50 prepared, 10KCd-complex by treatment with Protease IV,

800 WAGNER

i7.1iIS

*~ 4

"-Ik-M.4l.:l..-4

... .!' V.




Cd (pjg)

...................................................,..................................................................................__........

STAINED......GEL........

FIG. 2. Diagrammatic representation of the method used for prepar-ing electrophoretically purified complex. Center, tube gel with Spectra-phore 3 tubing attached to cathodal end (during electrophoresis lowerend of the gel and the entire tubing were immersed in electrode buffer).Crosshatched areas in the gel (center figure) represent green and brownpigments which are retained in the gel at termination of electrophoresisand bromophenol blue is shown emerging from the gel into the tubing.Right, Coomassie blue staining pattern after termination of electropho-resis. Left, an actual "2Cd profile obtained in a typical run at the pointof terminating electrophoresis and the migration of bromophenol blueand dinitropyridyl aspartate are shown.

thermolysin, cellulysin (contains protease, carbohydrase, andother activities), and DNAse I indicated that the complex isrelatively insensitive to these hydrolases. In contrast, S. aureusprotease which cleaves at the carboxyl side of aspartic acid andglutamic acid residues (7) cleaved ligand into at least threefragments, separable on G-50 fine (2.5 x 24 m, elution wih 0.5%(w/v) NH4 carbonate/NH4OH(pH 8.5), 1% butanol-notshown). No free amino acids were released by S. aureus protease.

Analysis of the heat stability of complex indicated that about90% of in vivo associated "2Cd and exchange-bound "'Cd re-mained bound after boiling complex for 2 min (Fig. 4). Possibledenaturation without loss of Cd would not have been detectedin these experiments. Analysis of pH sensitivity indicated thatabove pH 6.8, Cd remained bound to complex while below pH6.8 the Cd content decreased linearly with reduced pH withabout 90% dissociation at pH 2.8 (Fig. 5). Separation ofpH 2.8treated complex on G-50 fine indicated that solubilized Cd actedas free metal. In both heat and pH sensitivity experiments, invivo associated "I2Cd and '09Cd (introduced by exchange binding)behaved similarly, further suggesting that exchange binding canbe used to monitor 10K complex.The effects of Zn, Hg, Ag, and acid on the dissociation of Cd

from G-50 prepared complex was tested to examine the possiblepresence of thiolate chromophores with Cd. The absorptionspectrum between 230 and 290 nm ofnative complex was typicalof CdZn-thionein (9), Agrostis Cu complex (17), and tomatoroot Cd complex (1). As with Cd,Zn-thionein and yeast Cucomplexes, Hg and Ag but not Zn effected A250 (not shown).Acidification caused loss of A250 which was partially restored byreadjusting the pH to 7.4. If recovery ofA250 upon neutralizationof acidified native complex is due to rebinding of Cd to reformthiolate chromophores, it should be possible to cause incorpo-ration of '09Cd into complex during neutralization. We testedthis and found that addition of 1 uCi of '09Cd before readjustmentofpH and subsequent chromatography of the solution on G-50fine did yield 10K, "09Cd-complex. This result suggests that eitherreformation ofCd-mercaptide chromaphores occurs under these

pH 'C(cpm x 102)3 4 5 6 2 6 1014

FIG. 3. Isoelectric focusing patterns of dibromobimane-labeled cab-bage leaf, IOKCd-complex and rat liver metallothionein. Gel A wasloaded with 100 jsg dibromobimane-labeled, electrophoretically purifiedcabbage complex and gel B with 3 x 103 cpm of '4C-carboxymethylatedligand. Gel C was loaded with 100 ;&g dibromobimane-labeled rat livermetallothionein. The pH and '4C profiles were obtained from gel B. Afterremoval of ampholines and Coomassie blue staining of a parallel gel togel B, a single weakly stained band was observed with a pI of <4.0 (notshown).

conditions or "09Cd was bound to other sites on the ligand. Whilefurther study is needed to clarify the nature ofCd binding, resultswith Hg, Ag, Zn, and acidification-neutralization are consistentwith the presence of thiolate chromaphores.Gel filtration of G-50 prepared and '4C-carboxymethylated

ligand under denaturing conditions indicated that the apparentdenatured mol wt of ligand was about 2000 (Fig. 6). In contrast,performic acid oxidized rat liver metallothionein had a mol wtof 6300 as reported by Kagi et al. (9). Two components resultedafter gel filtration of a S. aureus digest of '4C-carboxymethylatedligand. These had apparent denatured mol wts of 1000 and 400.Fluorescamine-labeled calibration standards including CNBrfragments of Cyt c proved useful for monitoring standards (par-ticularly insulin, which has low A280) after chromatography inthe presence of GuHCl and allowed the use of reagent gradeGuHCl which has relatively high A280. Comparison of Cyt c andCNBr fragments of Cyt c with their respective fluorescamineadducts indicated that there was little effect ofadduct formationon elution of these standards under denaturing conditions (Fig.6).Attempts to determine the primary sequence of intact, 70%

HCOOH treated (cleaves metallothionein specifically [12]), re-duced-carboxymethylated ligand, and the three S. aureus frag-ments of reduced-carboxymethylated ligand utilized an auto-mated Edman degradation procedure (Beckman 890C sequencer,Beckman Inst.). None ofthese samples yielded phenylthiohydan-

801

-! ----- -I,-- --- zA5rj




Table I. Amino Acid Composition ofLigands ofCabbage LeafJOK,Cd-Complex, Rat Liver Cd,Zn-thionein(29), Neurospora Cu Thionein (14), and Agrostis Root Cu-Complex (16)

Data are expressed as residue % of each amino acid and cysteine was determined as cysteic acid afterperformic acid oxidation.

Cabbage Leaf IOK,Cd-complex

Residue G-50 Electro- 14C_ Cd,Zn- Neurospora/Cu- Agrostis/Rootphoretically carboxy- Thionein Thionei CuComplexpurified methylated

Lys 2.3 1.0 0.8 12.0 4.0 3.3His 5.3 0.8 0.5 0 0 0.8Arg 1.5 0 0 0 0 1.6Asp-Asn 4.4 4.2 3.2 7.0 11.7 10.0Thr 1.9 1.5 1.5 5.7 0 2.9Ser 2.3 1.9 3.1 13.4 28.0 8.2Glu-Gln 27.4 39.0 37.3 3.6 0 19.0Pro i3.6 1.7 2.7 3.4 0 5.8Gly 9.6 11.1 11.6 10.0 24.0 11.9Ala 2.3 2.1 1.8 4.9 4.0 6.0Cys 20.0 28.9 0 34.0 28.0 18.4Val 2.6 0.9 2.2 3.6 0 4.2Met 1.5 0.7 1.1 1.6 0 0.7Ile 1.6 3.4 3.9 0.3 0 1.5Leu 1.9 1.9 1.5 0 0 3.6Tyr 0.8 0.9 0.5 0 0 0.6Phe 0.7 0 0.3 0 0 1.5Trp 0 0 0 0Carboxymethyl

cysteine 27.1

I I I I

I II

I

60 70 80 90 100Temperature (OC)

110 120

FIG. 4. Effect of temperature on the mol wt of cabbage leaf lOK,Cd-complex, in vivo-associated "2Cd and '09Cd l introduced

by exchange binding.

toin amino acids and in all cases sample was completely re-covered from the sequencer cup as determined by amino acidanalysis and '4C determination.

DISCUSSION

The apparent native mol wt, inducibility, high cysteine con-tent, and other features of higher plant 10,000 D, Cd-complexeshas led to the tentative conclusion that they are structurally andprobably functionally similar to animal metallothionein. Herewe report on characterization ofa 1OK,Cd-complex isolated fromleaves ofcabbage plants exposed to relatively high Cd in nutrientsolution and define several features of this complex which arenot metallothionein-like. These features describe a moleculewhich is perhaps unique in terms of its charge and shape. The

100

.o 80

60 -

C/40 -

a..

20

/X 1 1 1 12.8 3.6 4.4 5.2 6.0 6.8 7.6

pHFIG. 5. Effect ofpH on Cd-binding to cabbage leaf IOK,CC-complex,

in vivo-associated "2Cd O-) and '09Cd - * introduced by ex-change binding.

unique characteristics may influence the physiological func-tion(s) (yet unknown) of the ligand and the fate of the complexand bound Cd after ingestion in animals.Cabbage 10K,Cd-complex has a low isoelectric point of <4.0

(Fig. 3) which is consistent with its migration in native gelelectrophoretic systems (Fig. 1) and its high glutamic acid-gluta-mine content (Table I). The high acidic residue content reportedhere also explins the tight binding ofcabbage lOK,Cd-complexto DEAE exchange resins (25). In contrast, the isoelectric pointsof two primary isometallothioneins from rabbit liver were re-ported to be 3.9 and 4.5 (16) with combined aspartic acid-glutamic acid content (primarily aspartic acid) of about 11

c

m 100c00

r_ 80e

0a.

60

802 WAGNER




X D X*E0

F

1.0 _ 4

0.4 5I , , , , , I\

180 220 300

ELUATE (ml)

FIG. 6. Gel filtration of cabbage leaf lOK,Cd-complex, its derivatives,and rat liver metallothionein in the presence of 6 M GuHCl, 0.1 M 2-mercaptoethanol. Protein standards detected as A280 (0) and as fluores-cence adducts (x) were: A, lima bean trypsin inhibitor, B, Cyt c CnBrfragment I; C, insulin, B chain; D, insulin, A chain; E, Cyt c CnBrfragment II; F, Cyt c CnBr fragment III. Samples were: 1, performic-acid-oxidized rat liver metallothionein; 2, '4C-carboxymethylated ligandof cabbage complex; 3, G-50 prepared, lOKCd-complex detected as

"2Cd; 4 and 5, '4C-carboxymethylated ligand after S aureus digestion.

residue per cent compared to 40 residue per cent (mostly glu-tamic acid-glutamine) for cabbage lOKCd-complex. LikeCd,Zn-thionein, Neurospora Cu thionein is reported to have low(H12 residue per cent) acidic residue content (14) while that ofAgrostis Cu complex is about 29 residue per cent (17). Also,10,000 mol wt Cd-complex isolated from tomato roots haselectrophoretic properties (in gel system B) which are similar tothose reported here for the cabbage leaf complex (1). Therefore,higher plants, 1OK metal binding proteins that have been studiedin terms of their charge characteristics appear to have a highly

acidic nature. We note that an about lOKCd-complex isolatedfrom tobacco leaves (25) has electrophoretic behavior like thatof cabbage lOKCd-complex (G. J. Wagner, unpublished data).Metal coordination in metallothionein is thought to involveseine or basic residues together with cysteine, with the negativecharge of the mercaptide metal complexes neutralized by juxta-posed basic residues (12). Cabbage lOKCd-complex has lowseine and basic residue content but very high acidic residuecontent raising the question ofhow mercaptide metal complexesmight be stabilized in this molecule.The possibility that cabbage lOKCd-complex may consist of

aggregates of glutathione (glutathione has a mol wt of 307 andcontains Glu-Cys-Gly in equimolar ratio) or a similar moleculewas considered. Cd-glutathione is suggested to be the form ofCdexcreted in the bile (4). We have observed that while no freeamino acids are released from lOKCd-complex by acid treat-ment at room temperature, similar treatment of 35S labeledcomplex reduces the mol wt ofthe ligand (G. J. Wagner, unpub-lished data). This observation, along with the incomplete under-

standing of the behavior of the ligand under dissociating condi-tions, and the lack ofevidence showing the absence of y-glutamyllinkage (characteristic of glutathione) in the lOKCd-complex,leaves open the possibility that this complex may consist ofaggregates of Cd-glutathione or some related sulfhydryl com-pound. Analysis of G-50 prepared lOKCd-complex for sugarand phosphate indicated an absence of substantial carbohydrateor nucleic acid (data not shown).Cabbage lOKCd-complex does not behave like metallothi-

onein under the denaturing conditions examined here. Theelution of both cabbage complex (25) and Cd,Zn-thionein (9)during gel filtration under nondenaturing conditions corre-sponded to that of globular proteins having apparent mol wtabout 10,000. In the presence of GuHC1 a value of 6500 wasobtained for Cd,Zn-thionein (9; Fig. 5) while that for cabbagecomplex was about 2000 (Fig. 5). Several possibilities couldexplain these results. First, complex may consist ofa single foldedpolypeptide chain-as is the case for Cd,Zn-thionein-but itmay not denature in the presence of GuHCl and may thereforehave a lower apparent mol wt under what are normally denatur-ing conditions. Second, it may exist as a single linear chaininhibition of folding due to charge repulsion (high glutamic acidcontent) would not be surprising. However, such a structure isnot entirely consistent with Cd-mercaptide chromaphore or clus-ter stability. Sedimentation rate experiments are underway toexamine these possibilities. Third, the ligand may not be a singlepolypeptide chain but an aggregate of smaller species. Again wenote that acid treatment results in loss of metal and a reductionin the apparent nondenatured mol wt of 35-labeled ligand. Afourth possibility is that proteolysis occurs during isolation, butfragments only dissociate in the presence of neutral, GuHCl(causes some loss of metal) or after removal of metal by acidtreatment. To test the last possibility, a preparation was made inwhich all isolation solutions contained PMSF (5 mm) to preventpossible proteolytic cleavage of ligand. Addition of Azocoll tothis extract showed that proteolysis (of that substrate) was essen-tially arrested. Behavior of PMSF-isolated, lOKCd-complex inthe presence and absence ofGuHCl was like that described abovefor preparations made without PMSF. We note that the apparentnondenatured mol wt of 1OKCd-complex as determined by gelfiltration in 0.5 M NaCl, 25 mm K-phosphate buffer (pH 7.4) or0.5% NH4HCO3/NH40H (pH 8.4) was 10,000. The lack ofsensitivity of native complex to Protease IV, thermolysin, andcellulysin protease is interesting, but the methods used probablywould not have detected minor proteolysis and perhaps frag-ments are not readily released.Our lack of success in attempts to determine the primary

sequence of the ligand of cabbage lOKCd-complex (applyingmethods successfully used to sequence metallothionein) mayreflect the unique composition-structure of this complex. Recentstudies suggest that aminoethylated ligand may be more usefulthan carboxymethylated ligand for sequence studies.

Table II summarizes the properties of cabbage lOKCd-com-plex and compares these with properties of rat Cd,Zn-thionein.Like and unlike characteristics are grouped. It is emphasized thatcertain like characteristics, i.e. metal inducibility, are poorlyunderstood for the plant complex.Cd/Zn-thionein is induced by Cd, Zn, Hg, Ag, Au, Cu, and

several glucocorticoid hormones but also by various physiologicalstimuli (4, 27). To date, only Cd is known to induce 10,000 Dplant complexes which bind Cd. This is a crucial point becausefunctionally metallothionein is thought to serve as a Zn or Custorage form (can donate Zn to apoproteins of Zn-requiringenzymes ([14, 27]) and perhaps only sequester toxic metals afterchallenge. Until we know if Zn and/or Cu can induce ligand oflOKCd-complex, it is premature to suggest metallothionein-likecharacter in terms of inducibility. We have grown cabbage plants

803




Table II. Comparison ofProperties ofAnimal Metallothionein and Cabbage LeafJOK,Cd-complex

Cabbage-Leaf lOK,Cd-

Property MetAnimal Reference Complex and RelatedPropertyMetallothionein Refere Higher Plant lOK,Cd- Reference

ComplexLike properties:Induced by Cd, Zn, Hg, Au, Ag, Cu, 4, 9, 27 Cd, others? 24'

stress, hormonesApparent native mol wt About 10,000 9, 15 About 10,000 24'Cys content 24-34 residue % 9, 15 29 residue % bCd/peptide ratio 7 to 1 9 4to6to I bChromaphore stability Hg, Ag > Cu > Cd > 9 Hg, Ag > Cd; Zn b

Zn binding?Heat stable Yes 4,9 Yes bCd binding Acid labile 4 Acid labile 22Intracellular location Cytosol 4 Cytosol 22Present in biological No 4 Not in vascular 24

fluids sapDisulfide bonds No 9 No b

Unlike properties:Charge characterAmino acid content Glu-Asp, 11 residue % 9 Glu-Asp, 40 resi- b, 24

due %CIsoelectric point >3.9 and 4.5 <4.0 b

Apparent mol wt in 6500 9 2000 bGuHClA2s0/A280 10 9 1.4 b

UJnordered structure Yes 9, 29 ?Directed to kidney Yes 4 ?

a Tomato root Cd-complex (1) has similar property.b This report.c Assumes no glutamine or asparagine. Glutamine and glutamic acid or asparagine and aspartic acid are not

distinguishable after acid hydrolysis.

in Cd-free, high Zn medium and also in the presence of lowlevels of65Zn and have not observed a lOK,Zn-complex after gelfiltration. However, amino acid analysis ofthe 1OK region oftheG-50 profile (high Zn experiment) suggests the possible presenceof ligand (G. J. Wagner, unpublished data). Perhaps Zn caninduce the ligand but Zn-complex is unstable. Dabin et al. (5)concluded that Zn and Cd are bound to different ligands in riceroots.The apparent mol wt and cysteine content of metallothionein

and lOK,Cd-complex are comparable as are the Cd/peptideratios (assuming 61 residues and a single polypeptide chain anda ratio of 6 to 1 in cabbage complex). This ratio was determinedhere after examining six preparations for "2Cd and ligand (thelatter determined by amino acid analysis) content in the peakCd-containing fraction (fraction E, Fig. 1) after G-50 chromatog-raphy. Chromaphore stability as determined by the effect ofcompeting metals on A250 (mercaptide chromaphore) appear tobe similar and both Cd forms are relatively heat stable. In both,Cd binding is acid labile and an intracellular location (cytosol)for both soluble Cd forms is suggested. Little or none of eitherCd form is present in extracellular fluids and neither appears tohave disulfide bonds.

Metallothionein is shown to have unordered structure andnothing is known about plant lOK,Cd-complexes regarding sec-ondary structure. Cd,Zn-thionein may have a different fate inanimals than in inorganic Cd and we do not know the relativefate of inducible, plant lOK,Cd-complex. Comparative feedingstudies using CdC12 and a noninducible Cd-complex in wheatgrain suggest that the fate in mice of these forms is similar (26).

Features of cabbage lOK,Cd-complex which differ from thoseof metallothionein include its anionic nature (extremely high

acid residue content, low pI, and unusual electrophoretic behav-ior) and its behavior on gel filtration in the presence ofGuHCa.Also it has much lower A25o/A280 than metallothionein, but theA280 is not due to the occurrence of aromatic residues (Table I).This property and the tan color ofpurified complex are probablydue to the presence of artifactually bound (during isolation)phenolics as first suggested by Bartloffet al. for tomato root Cd-complex (1). It is also possible that tan color and A280 areassociated with an impurity. It is unlikely that phenolics or someother 280 nm absorbing component is an integal part of thecomplex-essentially all (95.6%) of the weight of electrophoret-ically purified complex is accounted for by its amino acid andnietal content (see "Materials and Methods"). Also, recent resultsshow that by inhibiting polyphenol oxidase during extraction oftissue, A280 and tan color is virtually eliminated while Cd isprimarily recovered as 1OK anionic complex (G. J. Wagner,unpublished data).

In summary, cabbage leaf IOK,Cd-complex has certain prop-erties in common vwith metallothionein but others which aredifferent. Metallothioneins having sequence homology occur indiverse eucaryotic species and are therefore expected in higherplants. However, until the structures of cabbage, lOK,Cd-com-plex, and related molecules are better understood and theirinvolvement in micronutrient (Zn and Cu) metabolism is estab-lished, we should avoid the conclusion that they are structuallyand functionally like metallothionein or that their fate afteringestion is like that of Cd-metallothionein.

Since submitting this manuscript, Rauser (Plant Physiol 198474: 1025-1029) described the isolation of a highly anionic Cd-binding protein from Agrostis roots.

Acknowledgments-We especially wish to thank M. Elzinga, N. Alonzo, and H.

804 WAGNER




Kycia for assistance and helpful discussions. We thank P. Mulready and J. Cutt fortechnical assistance in portions of the work.

LITERATURE CITED

1. BARTLOFF M, E BRENNAN, CA PRICE 1980 Partial characterization of a

cadmium-binding protein from the roots of cadmium-treated tomato. PlantPhysiol 66: 438-441

2. CASTERLINE JL, NM BARNETT 1982 Cadmium-binding components in soybeanplants. Plant Physiol 69: 1004-1007

3. CATALDO DA, TR GARLAND, RE WILDUNG 1981 Cadmium distribution andchemical fate in soybean plants. Plant Physiol 68: 835-839

4. CHERIAN MG, RA GOYER 1978 Metallothioneins and their role in metabolismand toxicity of metals. Life Sci 23: 1-10

5. DABIN P, E MARAFANTE, JM MousNy, C MYTTENAERE 1978 Absorption,distribution and binding ofcadmium and zinc in irrigated rice plants. PlantSoil 50: 329-341

6. FOHLMAN J, D EAKER, E KARLSSON, S THESLETT 1976 Taipoxin, an extremelypotent, presynaptic neurotoxin from the venom of the Australian snaketaipan. Eur J Biochem 68: 457-469

7. HALL TC, T MA, BU BUCHBINDER, JW PYNE, SM SUN, FA BLuSS 1978Messenger RNA for Gl protein of french bean seeds: cell-free translationand product characterization. Proc Natl Acad Sci USA 75: 3196-3200

8. HIRs CHW 1967 Determination of cysteine as cysteic acid. Methods Enzymol11: 59-62

9. KAGI JHR, Y KOJIMA, MM KISSLING, K LERCH 1980 Metallothionein: anexceptional metal thiolate protien. In Sulfur in Biology, Ciba FoundationSymposium 72, Excerpta Medica, pp 223-237

10. KANETA M, H KIKICHI, S ENDO, N SUGIYAMA 1983 Isolation of a cadmium-binding protein from cadmium-treated rice plants. Agric Biol Chem 47:417-418

11. KLAUSER S, JHR KAGI, KJ WILSON 1983 Characterization of isoproteinspatterns in tissue extracts and isolated samples ofmetallothioneins by reverse-phase high pressure liquid chromatography. Biochem J 209: 71-80

12. KoJIMA Y, C BERGER, BL VALLEE, JHR KAGI 1976 Amino-acid sequence ofequine renal metallothionein-lB. Proc Natl Acad Sci USA 73: 3413-3417

13. LAEMMLI VK 1970 Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 227: 680-685

14. LERCH K 1980 Copper metallothionein, a copper-binding protein from Neu-

rospora crassa. Nature 284: 368-37015. LERCH K, D AMMER, RW OLAF50N 1982 Crab metallothionein-primary

structure of metallothioneins I and 2. J Biol Chem 257: 2420-242616. NORDBERG GF, M NORDBERG, M PISCATOR, 0 VESTORBORG 1972 Preparation

of two forms of rabbit metallothionein by isoelectric focusing. Biochem J126: 491-498

17. RAUSER WE, NR CURVETrO 1980 Metallothionein occurs in roots of Agrostistolerant to excess copper. Nature 287: 563-564

18. RIGHETrI P, JW DRYSDALE 1971 Isoelectric focusing in polyacrylamide gels.Biochim Biophys Acta 236: 17-28

19; SIMPSON RJ, MR NEUBERGER, TY Lio 1976 Complete amino acid analysis ofproteins from a single hydrolysate. J Biol Chem 251: 1936-1940

20. SPIVEY Fox MR 1983 Cadmium bioavailability. Fed Proc 42: 1726-172821. STEERS E, GR CRAVEN, CB ANFINSEN 1965 Evidence for nonidentical chains

in the j-galactosidase of E. coli K12. J Biol chem 240: 2478-248422. STOKES PM, T MALER, JR RIORDAN 1977 A low molecular weight copper

binding protein in a copper tolerant strain of Scenedesmus acutiformis. InDD Hemphill, ed, Trace Substances in Environmental Health-XI, pp 146-155, University of Missouri Press, Columbia, MO

23. WAGNER GJ 1979 The subcellular site and nature of cadmium in plants. InDD Hemphill, ed, Trace Substances in Environmental Health-XIII, pp115-123, University of Missouri Press, Columbia, MO

24. WAGNER GJ 1982 Characterization of a cadmium binding complex fromcabbage leaves. Plant Physiol 69: S-44

25. WAGNER GJ, MM TROrrER 1982 Inducible cadmium binding complexes ofcabbage and tobacco. Plant Physiol 69: 804-809

26. WAGNER GJ, E NULTY, M LEFEVRE 1984 Cadmium in wheat grain: its natureand fate after ingestion. J Toxicol Environ Health. In press

27. WEBB M, K CAIN 1982 Functions of metallothionein. Biochem Pharmacol 31:137-142

28. WEIGEL Hl, HJ JAGER 1980 Subcelular distribution and chemical form ofcadmium in bean plants. Plant Physiol 65: 480-482

29. WESER V, H RuPP 1979 Copper-thionein and other metal-sulfur protiens. InJHR Kagi, M Nordberg, eds, Metallothionein. Birkhauser Verlag, Basel, pp

221-23030. WINGE DR, BL GELLER, J GARVEY 1981 Isolation ofcopper thionein from rat

liver. Arch Biochem Biophys 208: 160-16631. YOUNG RB, M ORcUrr, PB BLAUWIEKEL 1980 Quantitative measurement of

protein mass and radioactivity in N,N'diallyltartradiamide cross linkedpolyacrylamide slab gels. Anal Biochem 108: 202-206

805



Documents

Characterization of Cadmium-Binding Complex of Cabbage Leaves' · CHARACTERIZATION OFCABBAGE LEAF Cd-COMPLEX (last measured in H20) was added to individual solutions and after 5 min