Comparison of the solution conformations of human [Zn7]-metallothionein-2 and [Cd7]-metallothionein-2 using nuclear magnetic resonance spectroscopy

.J. Xol.

Niol. ( l9!,n) 225. 433-443

Comparison of the Solution Conformations of Human [Zn,]-metallothionein-2 and [Cd,]-metallothionein-2 using Nuclear Magnetic Resonance Spectroscopy Barbara A. Messerle’j-, Andreas SchPffer*$, Milan VaGk* Jeremias H. R. Ktigi and Kurt Wiithrich’ ‘Institut fib Molekularbiologic and Biophysik EidgevSssische Technische HochschulP-Hiinggrrbrrg (‘H-80.93 Ziirich. SwitxAxnd “Riochemisches Institut der Unicersitiit Ziirich. Wintwthurerstrassr (‘H-80.57 Sirich. 8r&wlnnd (Received 17 October 1991: accepted 14 January

1.90

1992)

The solution struct,ure of native human [Zn,]-metallothionein2 has been compared with the previously determined structure of human [Cd,]-metallothionein-2. The cwmparison \+:as based on cwmplete sequence-specific ‘H nuclear magnetic resonance assignments for human IZn,]-metallothionein-2 obtained using the sequential assignment method. The sevontiary struc%ure was found to be very similar in the [Zn,]- and [Cd,]- forms of the prot,An. Oni) seven amide protons in [Zn,]- metallothionein-2 were found to have exchange rates Iowc~r than k 0.2 min ’ at pH 74 and lO”C, which corresponds slowly to the results of amide prot,on exc*hange studies with t,he [Cd,]- form of the protein. Finally, the ‘H- ‘H (1ixtanc.e constraints determined from nuclear Overhauser enhancement spectroscopy for human IZn,]-metallothionein-2 were checked for compatibility with the [C!d,]-metallothionein-2 structure. Overall, although no direct method is available for identifying the metal po1ypeptidr co-ordinative bonds in the Zn2+ -containing protein. these measurements provided several independent lines of evidence showing that thcl [Zn7]- and IC‘tl,]- forms (It human metallothionein-2 have thr same molecwlar architectjure, h’c~,ywo~~s: human metallothionein-2: resonance spe&oscop+v;

[Zn,]-metallothionein structure; nuclear protein structure: amide proton exchange

1. Introduction Mammalian metallothioneins (MTS;) are a homologous family of metalloproteins consisting of a polypeptide chain of 61 or 62 amino acid residues, t I’rrsent ad(lress: Ijepartment of Organic C’hrmistry. The university of’ Sydney. KG$’ 2006, Australia. z Prrsrnt address: (*iha-(;eigr J,t,d.. Agricultural 1)ivision. Bawl. S\zitzerlantl. * 4 Abbreviations usrd: MT. metallothionein; n.m.r.. nuclear magnetic. resonance: TPPI, time-proportional phase incarementation: XOE. nuclear Overhauser enhanchrment; (‘OSY. P-dimensional correlated spe&oscwp; dQF-. ~-quantum-filtered: REJATiEJH’OSY. l-dimensional relayed coherence transfer spectrosc.opy: E.COSY, d-dimensional exclusive correlat,ion spwtroswpy: TOCSY, Z-dimensional total c.orrelation spectroscopy: K’OESY, %-dimensional nuclear (kerhauser enhancement spectroscopy; p.p.m.. parts per million: EXAFS. extended X-ray absorption fine structure: TSP. sodium %,l-t,etradeutero-:~-trimethvl-sil~l lwopionate.

mayneti(,

incaludiny PO strictly conserved cysteinyl residues that serve as thiolate ligands for seven bivalent post-transit.ion metal ions, usually zinc or a mixturf> of zinc and cadmium (Gigi & Svh$tt‘er. 198X). R~ewntlv. the threw~imensional structures of rat [(‘d,]-Mk? (Schultze rt nl., 198X). rabbit j(‘d,]-MT& (Arwniev et tel.. 1988) and human i(‘d,]-MT2 (Jlcwerlc rjt trl.. 1990n) have been determined with hornonuc*lcar and heteronuclear t~~o~dimensional n.m.r. spwtroswpy. using protein samples reconstitutrtl with the ‘12Ctd or ‘13(‘d isotopes. In parallel to their close seciwnw homology (KGgi & Kojima, 1987). the spatial solution structures of the t)hree metwllothioneins are very similar (Messrrle rt (II., 199On). This includes ident’ical c*admiun-i o-t hiolate cwordinative bonds and metalLsulfllr (aluster topologies in all three proteins, and nearly identical polypept,ide srcwndary strucatures and global polylq)t idr folds. The recently redetermined crystal strnvture of native [C’d,.Zn,]-51T2 isolated from the Ii\-cr of rats r~spowd to subtoxic tlosfxs of’ cwlminrri

434

B. =1. Messerle

salts (Robbins et al., 1991) also corresponds in ail essential features to the n.m.r. structures in solution; this is in contrast to a previously reported crystal structure of the same protein (Furey et al.. 1986)) which required revision. With regard to studies on structure-function correlations on the basis of n.m.r. structures of metallothioneins, which have for technical reasons to date been obtained for all-cadmium derivatives of MT rather than for the native forms (which contain primarily or solely zinc(I1) in most mammalian tissues and in cultured cells; K&gi & Kojima, 1987), it is of paramount’ interest to extend the structural studies to zinc(II)-containing forms of t’he proteins. There is increasing evidence that the principal physiological functions of the metallothioneins are in the regulation of the intracellular free zinc(I1) concentration and in the modulation of the man) zinc(II)-dependent processes in cellular growth, differentiation and repair (Kggi $ Schiffrr. 1988), possibly through interactions wit)h zinc finger transcription factors (Zeng et al., 1991h). Knowledge of the structure of zinc-containing metallothioneins in solution should improve our understanding of their chemical reactivity and hence their role in these regulatory mechanisms. As mentioned above, the structure of native rat [(Id,,Zn,]-MT2 was determined in crystals (Robbins et nl.. 1991), and n.m.r. studies of this form of the protein provided evidence that it contains the same metal-sulfur cluster architecture as [(‘d,]-.&IT% (Va%k et al., 1987). IZn,]-MT2 has not been crystallized and, owing to the lack of suitable n.m.r.observable isotopes of zinc, a direct and indepen dent determination of its solution structure hv the methods developed for the cadmium(II)-cont&ning MT forms (Neuhaus it nl., 1984; Frey it al.. 1985: Wiirgiitter et nl.. 1988: Wiithrich. 1991) is not possiblt~. Therefore. we have based struot’ural studies of zinc( II)-containing metallotllionc,irls on comparisons of detailed struct,ural feat.urrs measurable by ‘H r1.m.r. sprct~roscopy with the corrtbsponding properties of the cadmium form. The present paper reports the results of such a comparative study of naturally occurring hurnan [ Zn, ]-MT:! and reconstituted human [(‘d,]-MT2 in solution.

2. Materials and Methods The isolation of human liver MT2 and the samplr preparation for n.m.r. measurements have been described (Messerle et nl., 1990~~). Proton n.m.r. spe&ra were recorded on Bruker AM600 and AM500 spectrometers. Throughout, the pure phase absorption mode was used with TPPI (Redfield & Kunz. 1975; Marion & Wiithrich, 1983). Standard procedures were used for the following experiments. which were needed for the determination of the sequence-specific resonance assignments and the collection of NOE distance constraints and other conformational constraint,8 in [Zn,]-MT2: 2&F-COSY (Rance et al., 1983: Neuhaus et al., 1985), S-quantum spectroscopy with a mixing time of 30 ms (Wagner & Zuiderweg, 1983: Otting &. Wiithrich.

et al

1986). TOCSY

weiler &

with a mixing

Ernst.

time of 80 ms (Braunsch-

1983: Bax

8

Davis.

1985).

RELAYED-COSY with a mixing time of 25 ms (M’agner. 1983), NOESY with mixing times of 40 ms and 150 ms (Anil-Kumar et al.. 1980)and E.COSY (Griesinger ct al.. 1985). The typical data size was 512 x 2048 points in the time domain, with zero fillin-g to 2048 x 4096 points. For the acquisition of NOESY spectra, zero-quantum c.oherence was suppressed using a modification of the technique described by Rancr et al. (1985) (Ott’ing e!t al.. l99O). For the NOES+ data a cosine window was applied as thv weighting funct,ion prior to Fourier transformation. Baseline distortions were eliminated using a polynomial tit of order 3. Sequence-specific assignments for human I&I, I-MT:! were obtained using the standard sequential assignment method (Billeter et al.. 1982: Wagner & FTiithric-h. 1982: W’iithrirh. 1986). The NOE: int’ensities in thr SOES\r spectra were determined using the program EAS\. fin thv integration of the cross-peaks (Ercles it 01.. 19Hl). Thr relationships betwern peak int.rnsity and upper liistanc*tL limit were determined using similar arguments to those described in d&ail by Arseniev it ml. (198X; WC’ also Wiithrich, I986). Amide proton exchange experiments were performed at p2H 74 in 20 tnM-[*H,,]-Tris HCI, at 10°C’. The protein concentration was 8 m.v. and the solutions c,clntalnvd 20 miv-K(‘l. The determination of the amide proton excbhangt, rates from the n.m.r. data for human [Zn7]-MT2 was achieved in the same way as deserihrd by Mrssrrlv rot al. (199Oh) for human ICd,J-MT2.

3. Results The procedures used for the ‘H n.m.r. assignments closely followed those used for rabbit [Cd,]-MT2a (Neuhaus et nl.. 1985; Wagner et (~1.. 1986) and human [Cd,]-MT2 (Messerle et al., 199Oa). The one difference in thta assignment procedure was the lack of a n.m.r.-observable zinc isotope. so that no direct, spectral assignrnents c*ouldbe made of tht metal-to-polypeptide co-ordinative bonds. Except for th(b two srquentia.1 c.onnectivities mentioned below, t’he assignment of the resonancesduta t,o both labile and non-labile protons of human 1Zn, ]-MT:! was achieved without reference to t)hr assignments previously obtained for human [(:d, ]-MT2. The spin system identifications were obtained using %QFCOSY, KEIAYE:I)-(1OSY and TO(XY, and the sequential assignments using NOE:SY with a mixing t,ime of 150 milliseconds (Wiithrich, 1986). As had been found necessary for thr othrr tnetallothioneins. NOIXSY spectra were acquired at’ different) temperatures, i.e. I()“(! and 7°C’. in order to resolve arnhiguities in assignments of overlapping amide proton resonances. Thtl amide proton resonances of’ thtx residues Cys24 and Glu23 were overlapping at. both temperatures, and the amide proton resonance ot Asp1 1 was not, observed. so that the sequential assignment, to GlylO was not possible. The ensuing ambiguit,i&

with

the

were eventually

resolved

by comparison

spectra of human [Cd,]-MT2. The sequence-specific ‘H n.m.r. assignments are listed in Table 1. For the non-labile protons the assignments are complete, with the exceptions Ser12 and Ser32, where only a range of shifts is given for 0~6%

n.m.r. Structures

of Human

(Zn J- and [Cd ,]-metallothionein-2

435

Table 1 ‘H chemical s hzjEts, 6 (p.p.m.),

for human [Zn,]-MT2 Chemical

Amino acid residue Met1 Asp2 Pro3 *4sn4 cys5 Serti CYS7 Ala8 Ala9 GlylO Asp11 SerlP Cysl3 Thr14 CynlA Ala16 Gly17 Serl8 Cysl9 LysPO CysPl Lys22 Glu23 cys24 Lys25 Cys26 Thr27 Ser28 cys29 Lys30 Lys31 Ser32 cys33 (‘ys34 SW35 (‘~“36 cys3i Pro38 Va139 Gly40 (‘ys4 1 Ala42 Lyu43 (‘vs44 A”la4.i Gln46 Glq4i Q-448 IlelY ( ‘*ys.50 Lyn51 My52 Ala53 Ser54 Asp.55 Lys56 (‘w5i S&5X (‘KS59 C&60 Ala61

NH

H”

Hfl

8.41 8.36

442 491 437 4.73 &4 4.75 417 442 4.32 392, 4.23 4.67 4.58 4.35 4.69 4.35 4.02 376. 426 4.56 426 471 403 407 435 431 4.35 4.15 402 422 435 435 450 444 447 504 427 435 521 4.71 3.78 365. 4.10 4.07 413 426 444 412 4.54 3.5i. 434 4.22 467 $71 435 3.86. 406 446 $62 4.39 4.71 5.25 470 456 462 m

1.97,2.04 259, 3.16 190, 2.30 2.83, 2.95 m, 3.44 387, 416 2.97, 300 1.41 1.40

8.67 7.45 901 &92 8.92 8.35 8.45 8.19 7.94 8.46 909 7.75 872 804 828 907 830 868 902 902 957 853 897 834 7.42 7.44 8.41 884 8.14 8.33 8.99 %46 7.07 8.58 894

7.12 933 821 8.07 7.16 8.11 7.48 %90 7.23 8.89 8.67 8.63 8.28 %24 8.57

7.76 8.46 835 814 7.09

at pH 7.0

shift? Others

2.69, 2.79 3.864923 3.07, 307 456 2.98, 346 1.34 3.82, 387 m, 292 1.75, 1.81 3.06. 3.55 1.82; 1.94 1.75, 1.87 2.80, 3.18 1.95, 2.02 3.08, 3.25 425 3.96, 3.96 2.81, 3.04 1.76, 1.84 1.76, 1.87 3.92, 397: 3.25, 308 3.48, 362 3.83, 3.95 2.69, 3.11 3LJ, 3.14 2.05, 2.29 1.92 3.13, 3.22 1.50 2.04, 2.10 2.61, 3.79 1.54 %Ol. 2.41 2.84, 2.95 2.21 240, 3.16 1.78. 1.88

CYH,:

2.54, 2.61; C’H,:

2.11; S-Ac-CH,:

CYH,: Y’H,:

1.91, 2.01; CdH,: 7.04, 7.96

386,

3.86

VH,:

1.27

VH,:

2.09. 2.09

CYH,:

1.29

WH,: 1.80, 2.07; CdH,: CYH3: @91, @96

381,

3.81

CYH2: 2.35, >35;

7.65

CYH,:

N’H,:

1.01; CYH2: 0.95, 1.69; CdH,:

2.04

092

1.43 387, 393 2.66, 2.ii 1.75,

1.81

355, 3.62 389, 3.99 312, 3.26 2.62, 3.17 1.40

The amide proton chemical shifts were measured at lO”C, and all others at 25°C. t Chemical shifts are in p.p.m. relative to internal TSP. For methylene groups 2 chemical shifts are givm wherever 2 resolved signals were observed, or where the presence of 2 degenerate signals had unambiguously been established. The chemical shifts are underlined where they deviate from the chemical shifts of the corresponding protons in [Cd,]-MT2 by more than @lo p.p.m. $ A range of chemical shifts is given for those methylene groups that showed strong-coupling fine structure patterns.

K56 K20 2.0 K43

E a a

K56

K20

It shifts

was found that. on the whole. the chemical of cwrresptmdirig labile arid noir~labilr prototls of’ hllt~lan I(‘d,]-MT:! and IZn,]-MT2 tliffrrrd hy ltw than 0.1 p.p.m. This is indic~atctl in ‘I’ablt~ I ahcw thtb chemitd shift has heen underlined if this tiifkrtv~tv is greater t)han 0.1 p.p.m. All hut two of thew largest. tkria~tions of chemical shift, hetwt~cn the two protrins \vt‘rt‘ olx3rrved for cystrinyl proton rwoiianc’w: t hct only vxwptions bring IAys20 and Ser35. whit+ arc both locatIed between two (‘ys residuw. Thrrr is thus a clrar indication that these chrrni(d

n.m.r.

Structures

of Human

[Zn 7]- rind [(‘rZ,]-m~fallothionein-’

137

(4 -

--

-

-

do\,,,

-ma0~0

-2%

dY,*

a

dzy,.,,+3, (b)

--

LIZ

-

4 -h-,i,+

---==-

byr.l’ ,, AZ!

--

d

-cIb-. h4

-hth4

MDPNCSCAAGDSCTCAGSCKCKECKCTSCKKSCCSCCPVGCAKCAQGCICKGASDKCSCCA , 10 20 30

_ff

-3,

40

---he

-3

50

d,:, -

60

Figure 2. Survry of the sequential and medium-range bavkhone SOI& obtained at lO’(” for (a) human [%n7 J-MT?. and (1)) human [(‘d, I-MTZ. In a semiquantit~ative analysis of the SOES\’ spwtra. thp intensitirs of thr sequential NOI% wrre c~lassifirtl as “strong“ ot “Heak“. which is rrpresented by thick or thin bars. rwprc*tively. The lwations of’ :<,,, heliws (S,“) and turns (t) or half-tlrrns (h) are also intlicaated for hoth protrins.

guounly assigned for hot.h proteins were residues (‘ys50. Tle49, Ala45, Cys44 and Cys37. Tao additional slowly exchanging amide protons were observed that could not he conclusively assigned in one of t*he two fi)rms, but in both cases the possible assignments included the unambiguous assignment made in the case of the other form of the protein. Rgure 3 affords a visual csomparison of the rat’es of exchange of those amide protons that have been unambiguousl,v assigned in t)he two prot,eins. It is clear that in all Ijut one case. Jle19. these are virtually identicaal. The faster exchange rate of the amide prot,on of Tle49 (Table 2) and the abovementioned extra half-turn between Ser35 and C’ys37 in [Zn,]-,MT% as compared to [Cd,]-MT% are consistent with a local strurtural difference affwtinp both (‘ys37 and (‘ys50. These residues are known to bind to the same metal position in [(Al,]-MTQ. i.e. (‘d \‘I1 (MesserIc rf (11.. 199Oa). Overall. the slowl) excshanging amide protons observed for huma,n IZn,]-SIT2 indicaate t)hat there are hydrogen bonds I)etwcvn the same residues as in human [Cd,]-MTP. ant1 hence that the local secondary strueturr elements are similar in both forms of the protein. In

KSCCSCCPVGCAKCAQGCICKGASDKCSCCA 35 40 45 50 Amino acid sequence

Figure 3. Plot of the amide proton c-on&ants c’ersus the amino acid sequence of human [Zn,]-MT2 (a). arid [Cd,]-MT2 indicat,e k, values larger t,han I x 10-l

55

60

exchange rates in the cc-domain (a). The arrows min ’

additioll. just as in [Cd,]-MT%. only one slowi) exchanging amide proton was observed in the /?-domain for human I&,]-SIT%. which indicates t.hat in the zinc(II) protein the P-domain also has greater flexibility than the cc-domain. Information on the orientation of’ individual amino acid side-chains relative to the protein backbone and on the local eonformat,ion of more peripheral parts of individual side-chains was obtained using hot h spill-spin coupling constants and SOE distanc,v constraints. which were used as input for the program HARM ((jiintert et c/l., 198!)). X com~~arison of’ corresponding (‘“H~m(‘DH coupling wnstarrts in t hv two forms of the prott4n (Table 3) sho\f,s that the greatest difference was only 1 Hz, indicating identical local orientation rtllative to the

Table 3 (‘ortt~unri.son~ of correspondiny in human [Zn ,]-MT2

cou,pling con.stcznts 3J,, and /(‘C, ,I-MT2t

3 4--

3

3 -t-

n.rrc .r. Structures

of Human

[Zn

been inferred from EXAFS measurements of the rabbit and sheep liver proteins (Hasnain et al., 1987), and is attributed to the evolutionarily conserved arrangement of the metal-binding cysteinyl residues in the sequence. As suggested by a variety of spectroscopic studies, the same folded structure can provide for near-isomorphous binding of either seven moles of mercury(II), lead(I1). bismuth(TTT), cobalt(TT) or iron (VaGk & K&gi. 1983; K%gi & Kojima. 1987). The fact that metals of widely differing size such as zinc (“covalent” atomic radius 1.31 A: 1 a = &l nm) and cadmium (“covalent” atomic radius 1.48 A) can be accommodated in the interior of the globular domains without causing gross conformational changes in the enfolding protein is in it,self an astonishing property of the metallothioneins. The reduct’ion in cluster volume upon replacement of cadmium by zinc is about 18yo as inferred both from EXAFS measurements of the average (:d-S and Zn-6 bond distances in [Cd?]-MT and [Zn,]-MT (Hasnain et al., 1987), and from the differcrystallographically determined ences of the volumes of the S, octahedra of the adamantane-like cages in the henzenethiolate (SPh-) complexes [Cd,(SPh),o]2, i.e. 31.78 a3, and [Zn(SPh),,]‘-, i.e. 36.13 A3 (Costa et al., 1983). We attribute the ability of the folded polypeptide chain to adjust to this volume reduction without compromising the mode of metal co-ordination to the adaptability of the peptide loop structures interspaced between the metal-bound cyskinyl residues. These loops make up the major portion of the protein that encloses the metal-thiolate core, and they can be characterized either as turn or half-turn secondary structure element)s (Fig. 2). Their presence allows the polppeptide chain to follow contractions or expansions of the core with only minimal changes in bond angles and side-chain orientation of the residues in the loops. As is expected for such rather unconstrained open structures. the only noticeable conformational differences that are assignable to differences in the two metal forms manifest themselves primarily at metal-bound cysteinyl residues and some of their immediate neighbors in the interior of the domains, for example, in the form of n.m.r. cbhemical shift changes (Table 1; Fig. 1). The ability of metallothioneins to accommodate met,al species of different size and chemical reactivit’y without compromising its overall molecular architecture and dynamic features, which is implicated by the present study, offers a ready explanat ion as to why many MT preparations are heterogeneous in metal composition. wit,h zinc(I1). cBadmium( IT). copper(T) and other metal ions occurring in varia,ble and often non-stoichiometric proportions (KBgi & Kojima, 1987). The relatively irtdiscriminat’e mode of met’al selection, which appears to be guided mainlv by the relative affinity for thiolate cao-ordination sites and the availability of the met)al ions in the tissue, has been used in support of t,he hypothesis that metallothioneins serve primarily a non-specific protective function in

& and [Cd ,]-metallothionein-2

441

limiting the intracellular concentration of reactive heavy-metal ions. They would thus shield cellular structures from the harmful influence of toxic metals such as cadmium, mercury, platinum, bismuth. silver and gold (Nordberg & Kojima, 1979; Webb: 1987a). Tn addition to such non-specific protective functions. the more recent findings, that metallothioneins are ubiquitous tissue components that occur irrespective of the presence or absence of toxic metal ions, that MT biosynthesis is tightly regulated by the activation state of the cell and by certain hormones, cytokines, growth factors, oncogenes and tumor promoters, and that Zn2+ is t)he natural metal ronstitutent in most tissues and cultured cells, imply that metallothioneins may have a morespecific major role in fundamental zinc-related cellular processes (Bremner, 1987; Biihler & K%gi, 1974; Hamer, 1986; KB;gi & Kojima, 1987: Karin, 1985). One hypothesis is that [Zn,]-MT might be an essential part of a metallo-regulatory system governing zinc-dependent processes in cell growth, differentiation and repair processes. This is supported by the programmed regulation of metallothionein mRNA levels in the course of embryogenesis (Nemer et al., 1984) and at different stages of fetal (Andrews et al., 1984) and perinatal (Webb, 19876) development,, and is also strongly sustained by the recent observation that the apoform of metallothionein readily removes zinc from zinc finger t’ranscription factors such as Spl (Zeng et al.. 1991a) and Xenopus TFIIIA (Zeng et al., 1991b)$ thereby abrogating the functionality of these compounds in transcriptional activation. It t,hus appears that [Zn,]-MT in conjunction with thionein supplied by cont’rolled biosynthesis fulfills an important, function in regulating the flow of zinc(I1) within the cell and thereby modulating the action of zinc-dependent processes in response to signals for cell activation in proliferation and differentiation, Financial support Xationalfonds (projects gratZrfullp acknowledged.

from the Schweizerischer 3125174.88 and X.160.88) is

References Andrews. 6. Ii.. Adamson. E. D. & Gedamu. I,. (1984). The ontogeny of expression of murine metallothiowith the a-fet,oprotein gene. nein: comparison Deoelop. Biol. 103. 294-303. Anil-Kumar. Ernst. R. R. & Wiithrich. K. (1980). A two-dimensional nuclear Overhauser enhancement (2D NOE) experiment for the elucidation of complete proton-proton cross-relaxation networks in biological marromolerules. Riochem. Biophys. RPX. (‘ommun. 95, l~m6. Arseniev. A.. Schultze. I’.. Wiirgiitter. E.. lsraun, W., Wagner. G., VaHBk, M., KGgi. .J. H. R. & Wiithrich, K. (1988). Three-dimensional struct,ure of rabbit liver [Cd,]-metallothioneim2a in aqueous solution determined by nuclear magnetic resonance. J. Mol. Rio/. 201. 637-657. Rax. A. R- Davis. I). C. (198.5). MLEV Ii-based two-

dimensional homonuclear magnetization transfer spectroscopy. J. Magn. Reson. 65, 335 -360. f3ifleter. M., Braun, \V. & Wiithrich. K. (1982). Sequential resonance assignments in protein ‘H nuclear magnetic resonance spectra. Computation of sterically allowed proton-proton distances and stat,istical analysis of proton-proton distances in single cryst,al protein conformations. .I. Mol. Biol. 155, 321L346. Braunschweifer. L. & Ernst,. Et. R. (1983). Coherencae t,ransfer by isotropic mixing: application to proton correlation spectroscopy. .J. Mayn. Reson, 53. 521 -528. Rremner. I. (1987). Nutritional and physiological sign& rancr of metallothionein. In Mrtallothionein IJ (Kiigi. ,J. H. R. & Kojima, Y.. eds). Experientia Suppl. 52, pp. 81-107, BirkhB;user, Base]. Biihfer, R. H. 0. & KB;gi, J. H. R. (1974). Human hepatic metaflothioneins. FEBS Letters, 39, 229.-234. (:osta. T.. Dorfman. *J. R., Hagen. K. S. X: Hofm, R. H. (1983). Synthesis and stereochemistry of mono-, bi-. and tetranucfear manganese thiofates. Inmy. (‘hem. 22, 40914099. Eccles, C.. Giintert. P., Bifleter. M. CG Wiithrich. K. (1991). Efficient analysis of protein 2D NMR spectra using the software package EASY. .I. Riow~ol. ~VMR,

1. I I l-130. Frey,

M. H., Wagner, G., Va%k, M., Serensen, 0. W., Neuhaus, D., WiirgGtter, E., KB;gi, J. H. R.. Ernst, R. R. 8: Wiithrich, K. (1985). Pofypeptideemetal cluster connectivities in metaffothionein-2 by novel ‘H-l ’ 3Cd heteronucfear t’wo-dimensional NMR experiments. .I. Amer. Chem. Sot. 107. 6847-6851. Furey, N. F., Robbins, A. H., Cfancy, L. I,.. Winge. D. R., Wang, B. C. & Stout, C. D. (1986). Crystal structure of Cd,% metalfothionein. Science, 231. 704 -710. Griesinger, C., Sorensen, 0. W. & Ernst. R. R. (1985). Two-dimensional correlation of connected NMR, transitions. J. AWW. Chem. b’oc. 107. 6394-6396. Giintert, P.. Braun. W.. Biffeter, M. & Wiithrich. K, (1989). Automated stereospecific ‘H NMR assignments and their impact on the precision of protein structure determinations in solution. .I. Amer. Them.

sot. 111, 399774004. Hamer, 11. H. (1986). Metallothionein. Annu. Ru?:. Biocham. 55, 913m9.5f. Hasnain, S. S.. Diakun, G. P., Abrahams. I.. Ross. 1.. Garner, (‘. I)., Bremner. I. & VaBBk. M. (1987). studies EXAFS of metaffothionein. In Metallothionein, II (KLgi, J. H. R. & Kojima. Y.. eds), Experientia Suppl. 52. pp. 227-236. Birkhluser. Basel. KZgi. tJ. H. R. $ Kojima, Y. (1987). Chemistry and biochemistry of metalfothionein. In Metallothionein JI (KIgi, J. H. R. & Kojima, Y., eds). Experientia Suppf. 52, pp. 25-61, Birkhguser, Base]. KB;gi, J. H. R. & SchPffer, A. (1988). Biochemistry of metaffothionein. Biochemistry, 27, 8509-8515. Karin, M. (1985). Metaflothioneins: proteins in search of function. Cell 41, 9-10. Marion, D. & Wiithrich, K. (1983). Application of phase sensitive two-dimensional correlated spectroscopy (COSY) for measurements of ‘H--‘H spin-spin coupling constants in proteins. Biochem. Riophys. Res. Commun. 113, 967-974. Messerfe, B. A., Schiiffer, A., VaRBk, M., K%gi, J. H. R. & Wiithrich, K. (1990a). Three-dimensional structure of human [ ‘13 C d 7] -metaffothionein-2 in solution determined by nuclear magnetic resonance spectroscopy. .J. Mol. Biol. 214, 76.5779.

Messerlr. 13. A., BOH, M., SchB;ffer. :\., Vaklk. 11. I\+I. J. H. li. & Wiithrich. K. (199Oh). Amid(b f)roton exchange in human rnet,alfothionrirr-2 rnrasllr~~tl I)) nuclear magnetic8 resonance spect,rosc~ofq .I .Ilr,/. Biol. 214, 781 -786. Nemer, M.. Travagfini. E. (‘., Rondinell, E. & I)‘Afonzo, .J. ( 1984). Developmental regufat,ion. induct,ion. and emf~rgonic~ tissue specifirity of sea urcahin rnc~taflo~ thioncxin gene expression. I)P~IP~o~). Hio/. 102. 47 f 482. Neuhaus. II., Wagner. C:., \‘ai&k. M.. Kltgi. .I. H. R. k Wiithrich. K. (1984). “%d-‘H spin-spin c~ouplings in homonuclear ‘H correlated spect,rosc~opg of metallot,hionein. Identification of the cysteine ’ H sf)in syst,ems. Eur. .I. Biochem. 143, 659-66i. Neuhaus, I).. Wagner. G.. VaBBk. M.. Kiigi. .I. H. K. & Wiithrich. K. (1985). Syst,ematic application of highresolution. phase-sensltlvr two-dimensional ’ I-i-xMf< tee-hniques for the ident,ification of thca amino acid prot’on spin systems in flroteins: rabbit mt%allothionein-2. Eu,r. ,I. Blochem. 151. 2.57 -273. Nordberg. M. & Kojima. Y. (f97!)). Metaflothionrin and other low molecular weight, metal-binding proteins. In MetalZothioneirL (Ktigi. .I. H. It. 8 xordberg. M.. eds), Experirnt,ia Suppl. 34. pp. 41 I Ii. f in NOES\ and soft NOESY. .I. Mqn. Reson. 89, 423 -GO. Otvos. .I. I). 8r Armitage. I. M. (1980). St,ruc*turcl of the met,af c&lusters in rabbit liver metallothionrain. /‘TO/,. Sat. Acad. Sci.. I:.S.d. 77, 7094 -7098. Rance. M., Smensen. 0. W., f%odenhausen. (4.. IVagnrr. G.. Ernst. R. R. & Wiithrich, K. (1983). Imf)rov~~d spectral resolution in (QSY ‘H NM10 sJ)rc+ra of proteins ??ia douf~le quantum filtering. ljiorhen, Hiophys. Jbs. (:owt;mun. 117. 479 485. Ranre. M.. Bodenhausen, (i.. Wagner, (i.. Wiithrichh. K. & Ernst,, It. R. ( 1985). A systematic apJ)roach to t hc suppression of .J (Lross peaks in PI> exchange and 21) NOE spectroscsoyy, ,I. Magr~. Reson. 62. 497 510. Redfield. A. (:. & Kunz. S. I). (1975). Quadrature Icourirr NMR detection: Simple multiplex for dual dr~trc%ion and discussion. J. ;Ilag/t. Rrson. 19. 25Obd51. Robbins. A. H.. McRee. 1). E., Williamson. M.. (‘trllet,t, S. A.. Xuong. N. H.. Furey. \V. F.. Wang. It (‘. k Stout. (‘. 1). (1991). Refined csryst,al structure of Cd,% metalfothionein at 2.0 4 resofutiorl. .I. iClo/.

Biol. 221. 1269-1293. Schuftze. P.. Wiirgijtter, E.. Braun. W.. LVagnrr. (i.. VaALk. M.. Kiigi. .J. H. R. & Wiithrirh. K. (1988) C’onformation of I(:d,l-metal1othionein-P from rat liver in aqueous solution determined by nuclear mapnetica resonance spectroscopy. .J. Mol. Uiol. 203. 251 26X. VaB&k, M. $ KGgi, J. H. H,. (1983). Spectroscopic& properties of metaffothionein. In Metal Ions in Biological Systems (Sigef, H.. ed.), vol. 15, pp. 213-273. Dekker. New York. VaLBk. M.. Wiirgiitter, E.. Wagner, G., KB;gi. .I. H. R. & Wiithrich. K. (l&7). Metal co-ordination in rat, liver metalfothionein-2 prepared with or without reconstitution of the metal clusters, and comparison with rabbit liver metallothionein-2. .I. Mol. Biol. 196. 7 I 1 m~7I!). Wagner. (:. (I 983). Two-dimensional relayed coherence

n.m .r. Structures

of Human

(Zn +

transfer spectroscopy of a protein. J. Map. Reson. 55. 151-m156. Wagner. G. & Wiithrich, K. (1982). Sequential resonance assignments in protein ‘H nuclear magnetic resonance spectra. Basic pancreatir trypsin inhibitor. I. Mol. Hiol. 155, 347-366. G. Cy- Zuiderweg, Wagner, E. R. P. (1983). Two-dimensional double quantum ‘H NMR spectroscopy of proteins. Biochem. Biophys. Rrs. Commun.

113. 854-860. Wagner. G.. Xeuhaus. D.. Wiirgiitter, E.. VaR$k, M., Klgi, J. H. R. & Wiithrich, K. (1986). Srquenc*r-specsific ‘H-NMR assignments in rabbitIi ver metaliothionein-2. Eur. J. Biochem. 157,

275”89. < 1 Webb.

M. (1987~~). Toxicological significance of metallothionein. In Metallothionein ZZ (KLgi, ,J. H. R. & Kojima. Y.. eds), Experirntia Suppl. 52. pp. 104-134. Birkhtiuser. Basel. Webb, M. (198%). Metallothionein in regeneration, reproduction and development. In Metallothionein II (KB;gi. .J. H. R. & Kojima, Y., eds), Experientia Suppl. 52. pp. 483498, BirkhLuser, Basrl.

Edited

by P.

and (Cd J-metallothionein-%

443

Wiirgtitter, E.. Wagner, G., Va%k, M.. Kigi. ,J. H. R. & Wiithrich, K. (1988). Heteronuclear filters for twodimensional ‘H NMR. Identification of the metalbound amino acids in metallothionein and observation of small heteronuclear long-range couplings. .I. Amer. Chem. Sot. 110, 2388-2393. Wiit)hrich, K. (1986). NMR of Proteinw and h’uclric Acids. Wiley, sew York. Wiithrich, K. (1991). Determination of the three-dimensional structure of metallothioneins by nuclear magnetic resonance spectroscopy in solution. In Methods in Enzymology, vol. 205, Metallobiochemistry, Part B: Metallothionein and Related Molecules (Riordan, J. F. & Vallee, B. I~.. eds), pp. 502-520. Academic Press. New York. Zeng, ,J.. Heuchel, R., Schaffner, W. & KPgi, J. H. R. (1991a). Thionein (apometallothionein) can modulate DN4 binding and transcription activation by zinc finger containing factor Spl. FEBS Letters. 279, 31&312. Zeng. ,J.. Vallee, B. L. & K%gi, ,J. H. R. (1991b). Zinc transfer from TFIIIA fingers to thionein clusters. l’ror. Sat. Acnd. AX., i-.S.A. 88. 9984 9988.

E.

Wright