Conformational Changes Induced by Binding of Divalent Cations to ...

Vol. 261,No.

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Q 1986 by

The American Society of Biological Chemists, Inc

Issue of July 5 , pp. 8883-8887,1986 Printed in U.S.A.

Conformational Changes Induced by Binding of Divalent Cations to Calregulin" (Received for publication, January 15, 1986)

Navin C. KhannaS, Masaaki TokudaS, and David Morton WaismanB From the Department of Medical Biochemistry, The University of Calgary, Calgary, Alberta, T2N 4N1, Canada

Scatchard analysis of equilibrium dialysis studies have revealed that in the presence of 3.0 mM MgClz and 150 mM KCl, calregulin has a single binding site for Ca2+ with an apparent dissociation constant (apparent &) of 0.05 pM and 14 binding sites for Zn2+ with apparent &(zn2') of 310 pM. Ca2+ binding to calregulin induces a 5% increase in the intensity of intrinsic fluorescence and a 2-3-nm blue shift in emission maximum. Zn2+ binding to calregulin causes a dose-dependent increase of about 250%in its intrinsic fluorescence intensity and a red shift in the emission maximum of about 11 nm. Half-maximal wavelength shift occurs at 0.4 mol of Zn2+/molof calregulin, and 100%of the wavelength shift is complete at 2 mol of Zn*+/molof calregulin. In the presence of Zn2+ and calregulin the fluorescence intensity of the hydrophobic fluorescent probe 8-anilino- 1-napthalenesulfonate (ANS) was enhanced 300-400% with a shift in emission maximum from 500 to 480 nm. Half-maximal Zn2+-inducedshift inANS emission maximum occurred at 1.2 mol of Zn2+/molof calregulin, and 100%of this shift occurredat 6 mol of Zn2+/molof calregulin. Of 12 cations tested,only Zn2+and Ca2+produced changes in calregulin intrinsic fluorescence, and none of these metal ions could inhibit the Zn2+-inducedred shift in intrinsic fluorescence emission maximum. Furthermore, none of these cationscould inhibit or mimic the Zn2+-inducedblue shift inANS emission maximum. These results suggest that calregulin contains distinct and specific ligand-binding sites for Ca2+ and Zn2+.While Ca2+ binding results in the movement of tryptophan away from the solvent, Zn2+causes a movement of tryptophan into thesolvent and the exposure of a domain with considerablehydrophobic character.

cium-binding protein (2, 6). Similar studies have identified calregulin as a major Ca2+-bindingprotein of bovine heart. Immunoblotting and Ouchterlony techniques have demonstrated the existence of calregulin in the 100,000 X g supernatants from a variety of bovine tissues (6). Calregulin was quantitated in various bovine tissue extracts by radioimmunoassay (4) and shown to be present in all tissues except erythrocytes. It was detected in particularly high amounts in pancreas (540 pg/g of tissue), liver (375 pg/g of tissue), and testis (256 pg/gof tissue). Radioimmunoassay also established that while 80% of calregulin was soluble, 20% of this protein was associated with particulate fractions and required nonionic detergent for solubilization. Immunofluorescent studies have indicated that calregulin is localized primarily in a system of cytoplasmic organelles that are virtually indistinguishable from immunocytochemically stained endoplasmic reticulum (6). The physiological function of calregulin remains unknown. In the present report, the metal ion-binding properties and metal ion-dependent conformational changes of calregulin are documented. It is established that calregulin binds Caz+ and Zn2+at distinct and specific sites, and the binding of these metal ions induces significant changes in its conformation. EXPERIMENTALPROCEDURES

Materials All chemicals were reagent grade, unless specified otherwise. Deionized water was used throughout. Chelex 100 was obtained from BioRad. WaClZ (20 mCi/mg of calcium) and &ZnC12(10 Ci/g of zinc) were purchased from Amersham Corp. HEPES,'MOPS,salts of acid various other metal ions, and 1-anilinonapthalene-8-sulfonic (l,8-ANS) were obtained from Sigma. Methods

Previous studies in our laboratory (1, 2) have shown that chromatography of bovine liver 100,000 X g supernatant by DEAE-cellulose and analysis of resultant fractions by the Chelex 100 competitive calcium binding assay (3) resolved four peaks of calcium binding activity. Recently we have purified and characterized a M, 63,000 protein responsible for one of the major calcium binding activity peaks of bovine liver. This protein was previously referred to ascalregulin (1, 4, 5) or CAB-63 (6) and based on physical, chemical, and calcium binding properties, was established as a novel cal*This workwas supported by a grant from the Alberta Heart Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18U.S.C. Section 1734 solelyto indicate this fact. $Fellows of the Alberta Heritage Foundation for Medical Research. § To whom all correspondence should be addressed.

Purification of Calregulin-Calregulin from bovine liver was purified according to theprocedure of Waisman et al. (6). Calcium Binding of Calregulin-Calcium binding of purified calregulin was determined by equilibrium dialysis. Calregulin was first dialyzed overnight against 1000 volumes of a solution containing 150 mM KCl, 10 mM MOPS (pH 7.1),3 mM MgC12, 1.0mM dithiothreitol, and 0.1 mM EGTA to remove bound Ca2+ from the protein. The dialyzed protein was then used for equilibrium dialysis as follows. A 0.5-ml portion of protein was dialyzed with shaking for 48 h a t 4 "C against 100 ml of the same solution containing varying amounts of CaCI, and 45Caz+(5 aCi) to achieve the desired free Ca2+concentration. The solutions outside and inside the dialysis tubing were removed, the absorbance a t 278 nm determined, and the protein concentrations calculated from extinction coefficient values. Aliquots of these solutions were subjected to liquid scintillation spectrometry, and the total Caz+ concentration was calculated from the contaminating Ca" (0.5 PM) determined by atomic absorption, plus the

' The abbreviations used are: HEPES, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonicacid; ANS, l-dimethylaminonaphthalene5-sulfonate; MOPS, 3-(N-morpholino)propanesulfonicacid; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraaceticacid.

8883

Conformational Studies on Calregulin

8884

amount of Caz+added. The association constants for metal and H+ binding to EGTA were based on values measured by Fabiato (7). Zinc Binding of Calregulin-Zinc binding of purified calregulin was also determined by equilibrium dialysis. Calregulin wasdialyzed against 1000 volumes of 10 m M MOPS (pH 7.1) containing 3 mM MgC&and 150 mM KCl. Half-ml aliquotsof dialyzed calregulin were incubated in the dialysis bags for 48 h at 4 "C against the same buffer containing various concentrations ofZnC1, and *Zn2+ (5 pCi). The solutions inside and outside were removed and analyzed for protein and '%nZ+ as described for calcium binding. Metal IonSpecificity-The effect of various divalent cations onthe binding of tracer %a2+ by calregulin was analyzed by equilibrium dialysis. The dialysis conditions were the same as described for calcium binding, and the buffer contained 10 mM MOPS (pH '7.1), 150 mM KC1, and 2 PM "CaC12, without or with salts of various divalent cations as listedinTable I. Contamination of Ca2+, as determined by atomic absorption, in all the salt solutions was found to be 0.5 gM. Fluorescence Measurements-All the fluorescence measurements were performed in a 650-10M Perkin-Elmer spectrofluorometer in the normal mode in water-jacketed cuvette holders at 25"C. The fluorescence measurements were carried out with protein solutions in 10 m M MOPS (pH 7.1) containing 3 mM MgCI,, 150 mM KC]. Zn2+ titrations were performed by adding small aliquots from a stock solution ofZnC1, to protein solutions and to a blank solution containing no protein. The samples were allowed to equilibrate for 10 min after each addition. Titration measurements were made at an excitation wavelength of 286 nm and an emission wavelength of 345 nm. The binding of the hydrophobic fluorescent probe I,8-ANS was monitored by its fluorescence enhancement. ANS was added sequentially in small aliquots from a stock solution (10 mM in water) to cuvettes containing calregulin or buffer only, and thefluorescence of the solution was measured at emission wavelength of 480 nm by exciting at 360 nm. The effects of Ca2+and Zn2+on the binding of ANS to calregulin were evaluated by performing titrations andtaking spectra in the absence and presence of Ca" and Zn2+. All titration data were corrected for the fluorescence of the blank and for dilution.

Zinc-binding Properties of Calregulin-Saturation curve for zinc binding to calregulin is shown in Fig. 2. The protein binds about 10 mol of Zn2+/molof protein in the presence of 3 mM MgCl, and 150 nMKC1 with half-maximal binding at about 120 RM. At Zn2' concentrations above 1.0 mM, calregulin precipitated from solution. The Scatchard analysis of Znz+binding (Fig. 2, inset) indicates an apparent Kd of 310 p M and a maximal binding capacity of 13.8 mol of Znz'/mol of calregulin. Furthermore, the presence of 1mM Ca'* had no effect on Zn2+binding (data not shown). Metal Ion Specificity-The effects of various divalent cations on the binding of tracer 45CaZ' by calregulin was determined by equilibrium dialysis. As shown in Table I, S9+ was the only divalent cation, other than Ca2+which competed for the binding of tracer "CaZ+ to calregulin. This suggests an absolute specificity of the calcium-binding sites of calregulin for Ca2+.Ba2+,M P , Cd", Mn2+,and Co2+at 1 mM concentration had no effect on 45Ca2+binding by bovine calregulin. On the other handZn2+,Fez+,and NiZ+ at 1mM concentration enhanced the 45Ca2+ binding to calregulin by 20-3096, whereas at lower concentrations of these ions, this effect was not 12)

oc

RESULTS

Calcium Binding Properties of Calregulin-Fig. 1 presents FIG. 2. Saturation curveof Zna+binding to calregulin. The a saturation curve for calcium binding to calregulin. Scatchard equilibrium dialysis was carried out asdescribed under "Experimental (8) analysis of the data (Fig. 1, inset and legend) reveals that Procedures." Dialysis buffer contained 10 mM MOPS (pH 7.1), 150 in the presence of 3 mM MgCI, and 150 mM KCl, calregulin mM KCl, 3 mM MgC12, and various concentrations of ZnClz in the binds 1 mol of Ca2+/mol of protein with an apparent K d of presence of %ZnC12. Inset, the Scatchard analysis of the same data; 0.05 RM. Calcium binding studies werd also performed in the V, mol of Zn" bound/mol of calregulin; C, free [Zn"']. absence of MgClz (data not shown) but no significant difference was observed between the values of binding of Ca2' to TABLEI calregulin in the presence or absence of MgC1,. In an earlier Effect of variolls catwns on *'Ca2+ binding by bovine calregulin study we erroneously reported that calregulin bound 3 mol of M MOPS Determined by equilibrium dialysis. Buffer mixture, (pH 7.1), 0.15 M KC1,2 X lo-' LIM CaC12. Ca2+/mo1of protein (6). Periodic group and added cation m

.2

-

None (control) IIA CaZ+ SrZ+

A

V

1 -

8

0E

~

.8

v

P .6

z

3

:m ,+ . 24 N 0

f

t?L!

-

.I

100 10-~ 10-3

10-4

-

0 .OL

10 0

1

FREE

Percentage of control

M

A

a:

-

Concentration

LO

CALCIUM

0.4

100

0.8

1.2

1000

(uM)

FIG. 1. Calcium binding of calregulin. The experiment was carried out by equilibrium dialysis as described under "Experimental Procedures." Conditions: 150 mM KCl, 10 mM MOPS (pH 7.1), 3 mM MgC12, 1 mM dithiothreitol, and 0.1 mM EGTA. Inset; Scatchard plot of data; V , mol of Ca2+bound/mol of calregulin; C, free [Ca"].

10-~

BaZ+ M P IB cuz+

10-~

10-3 10-~

13 15 33 100

88 100 52

LIB Zn2+ Cd" IIIB La3+

vm

Mn2' VI11 Fez+ coz+ Ni2+

lo9

125 100

88 10-3

100 127 100

10-3

130

8885

Conformational Studies onCalregulin observed. A 50-fold excess of Sr'+ over Ca" inhibited *Ca'+ binding by 70%. Cu'+ at 1 mM concentration caused a 50% inhibition of 45Ca'+binding to calregulin; however, the protein solution became turbid on addition of Cu'+, and addition of 5 mM EDTA did not reverse turbidity. This suggests that the inhibitory action of Cu'+ on 45Ca'+ binding to calregulin may be due to the irreversible denaturing effect of Cu2+ on the protein. Effect of Calcium on Intrinsic Fluorescence of CalregulinFig. 3 shows the intrinsic fluorescent emission spectra of calregulin. When excited at 286 nm (where predominantly tryptophan absorbs) the tyrosine contribution to theemission spectra (emission maximum, 303 nm) was negligible, and the emission maximum at 334 is typical of tryptophan. The calcium-saturated protein displays a maximum of fluorescence at 331 nm. In the presence of 3 mM M P and 150 mMKC1, Ca'+ removal by EGTA results in a decrease in intensity of about 5%, and the fluorescence maximum shifts to 334 nm. The 3-nm red-shifted spectrum of Ca2+-freecalregulin indicates that ca'+ removal results in a transfer of one or more tryptophan residues into a highly polar environment exposed to water (9). Effect of Zinc on the Intrinsic Fluorescence of CalregulinThe fluorescence emission spectra of calregulin at several Zn'+ concentrations is presented in Fig. 4. Addition of Zn2+ causes a dose-dependent increase in intrinsic fluorescence intensity of about 2.5-fold and red shift of about 11 nm in Amax. The fluorescence intensity continues to increase up to 1 mMZn", whereas the shift in the A,, is saturated by about 64 pM Zn". Half-maximal shift in X,, of intrinsic fluorescence of calregulin occurs at 8 M Zn'+ (Fig. 4,inset). Addition of 2 mM EDTA to theprotein solution containing 1 mM Znz+ almost completely reversed the zinc-induced increase in fluorescent intensity and the wavelength shift of the emission spectrum of calregulin (Fig. 4b). The Zn'+-induced changes in intrinsic fluorescence of calregulin were abolished by heat treatment (80 "C for 2 min)or by addition of 6 M urea suggesting that the native conformation of calregulin is required for its interaction with Zn'+. Zinc-induced Enhancement of ANS Fluorescence-ANS is an anionic amphiphile used as a probe for hydrophobic regions of proteins (10). Fig. 5 shows the zinc-induced enhancement of ANS fluorescence by calregulin. Calregulin blue-shifted the ANS emission spectra from 520 to 500 nm. Addition of zinc sequentially blue-shifted the ANS emission spectrum from 500 to 480 nmand produced a 3-4-fold enhancementin

80 c

5 70

z. e 60

c n

-k 50

5i 40 z

30 w

Y 20

W

U

LI)

g 10 0

3

2 0

300

350 WAVELENGTH(nm)

FIG. 4. Increased intrinsic fluorescence of calregulin upon binding Zna+.Emission spectra of 1 pM calregulin in 10 mM MOPS (pH 7.1), 3 mMMgC12, 150 mMKC1 at excitation wavelength of 286 nm. The protein solution contained no Zn2+(a);125 p~ ZnZ+( c ) ; 250 p~ Zn2+ (d); 500 PM Zn2+ (e); 1 mM Zn2+ ( f ) ; 1 mM Zn" + 2 mM EDTA (b). Inset shows the titration curve of Zn2+-inducedshift of tryptophan emission maximum at an excitation wavelength of 286 nm.

W &

u a

Eurn : (I-

0

3

U

-

e20

460 500 YIO 580 WAVELENGTH ( nm I

620

FIG. 5. Effect of calregulin and Zn2+ on the fluorescence emission spectraof ANS. Samples consisted of 25 p~ ANS in 10 mM MOPS (pH 7.1), 150 mM KCl, 3 mMMgC12 (a);in addition to 1 p~ calregulin (b);125 pM Znz+( c ) ; and 125 FM Zn2++ 250 p M EDTA (d). Inset shows the titration curve of Zn2+-inducedshift in ANS emission maximum by calregulin. The excitation wavelength was 360 nm.

fluorescence intensity. The zinc-induced effect on ANS emission spectrum was reversible by EDTA. Fig. 5 (inset) shows the titration curve of the Zn'+-induced shift in ANS fluorescence maximum by calregulin. Approximately 32 PM zinc was required to produce 50% of the maximal effect of the zincinduced shift in the ANS emission maximum, and 100% of this shiftoccurred at 233 p M Zn2+.These spectral changes are consistent with removal of ANS from an aqueous environment to hydrophobic binding sites on the protein. Metal IonDependence of Conformational Change-The data presented in Figs. 4 and 5 are consistent with a model in which Zn'+ binding to calregulin results in the exposure of a hydrophobic domain. In order to examine the specificity of this conformation event, the effect of several metal ions on the conformation of calregulin was documented. Table I1 300 350 400 WAVELENGTH In m) shows that, of all the metal ions tested, only Ca'+ and Zn2+ FIG. 3. Intrinsic fluorescence changes of calregulin associ- caused an enhancement of the intrinsic tryptophan fluoresThe fluorescence emission spectra of cence intensity of calregulin. While the association of Zn2+ atedwithCaa+binding. calregulin (2 FM) wasrecorded as described under "Experimental Procedures."The excitation wavelength was 286 nm and the buffer with calregulin also enhanced the fluorescence intensity of contained 10 mM MOPS (pH 7.1), 3 mM MgCl,, 150 mM KC1 in the ANS, CA2+had little effect on this phenomenon. Other dipresence of 0.1 mM EGTA (-) or 0.2 mM Ca2+(- - -). *, arbitrary valent cationssuch as M%+,Mn", Ba'+, Sr2+,Ni'+, Co'+, and units. Cd2+had no effect on the intrinsic tryptophan fluorescence

Calregulin on

8886

Conformational Studies

TABLE I1 Effect of various divalent cations in tryptophan and ANS fluorescence of calremlin

cation

Trp fluo-

ANS fluo-

Trp fluo-

ANS fluo-

rescence" rescence' rescence" rescence" 1 mM

mn,

334 nm

mn, 520nm

m.u

345 nm

hEM

480 nrn

None 100 100 100 100 (control) ZnZtb 170 229 122' 119' Caz+ 105 99 100 101 100 100 103 M e100 100 100 98 101 Mn2+ 100 101 99 99 BaZ+ Sr2+102 98 99 100 Ni2+ 100 100 101 96 co2+100 99 99 100 98 101 100 101 Cd2+d 20 14 9 11 CU2+< 12 6 Fez+' 16 7 Fluorescence calculated as per cent of control in presence of 10 mM MOPS (pH 7.1) + 150 mM KCl. Zn2+,125 p ~ . Total Zn2+,250 pM.

Cd", 250 pM. e

Values unaltered with addition of 2 mM EDTA.

intensity of calregulin or the ANS fluorescence intensity. Cu2+ and Fez+ dramatically reduced the intrinsic tryptophan or ANS fluorescence intensity, but this effect was not reversed by addition of excess EDTA. This suggests the possibility of irreversible denaturation of calregulin upon binding to these metal ions. All the divalent cations mentioned above (except Ca" and Zn2+)were unable to altereither the X,of calregulin intrinsic tryptophan fluorescence or, , ,X of ANS fluorescence (data not shown). Furthermore, various metal ions were tested as inhibitors of the Zn2+-induced conformational changes. Accordingly, calregulin was incubated with 125 p~ ZnClz,and theresultant increase in the intrinsic tryptophan fluorescence intensity and ANS fluorescence intensity was normalized to 100%. Subsequent additions ofCa", M e , Mn2+, Ba2+, Sr2+, Ni2+, Co2+, and Cd" didnot produce any change in the Zn2+induced enhancement of tryptophanor ANS fluorescence intensity. Again, the inhibitory effects of Cu2+and Fez+ on the &?+-enhanced calregulin and ANS fluorescence intensity seemed due to the irreversible denaturation of calregulin by these metalions. None of the metal ions (except Zn") altered the X,,,of ANS fluorescence (data not shown). The results of Table I1 suggest that Ca2+and Zn2+produced significant changes in calregulin conformation, as measured by intrinsic tryptophan fluorescence. Zn2+binding alone produced dramatic changes in ANS fluorescence. Both the changes in tryptophanand ANS fluorescence are specific since other metal ions do not either inhibit or mimic the effects of Caz+and Zn2+on calregulin conformation. Fig. 6 compares the zinc binding (Fig. 2) with the zincinduced change in intrinsic emission maximum (Fig. 4, inset) and thezinc-induced change in ANS emission maximum (Fig. 5 , inset). The conformational changes (ANS fluorescence and tryptophan fluorescence) saturated at Zn2+ concentrations below the concentration of Znz+required for occupation of all the Zn2+-bindingsites. This suggested that ZnZ+binding to only a fraction of the total sites was adequate to induce the conformational changes. DISCUSSION

Evidence exists indicating that zinc ions, a t relatively low concentrations, affect diverse functions of various cells (11).

FIG. 6. Titration curve of conformational changes of calregulin in the presence of Zn2+.Data is co-plotted from Fig. 4 (inset) (U Zn2+-induced ), shift in intrinsic X, of calregulin; Zn2+-induced ), shift in ANS Amax by calregulin; Fig. 5 (inset)(M Fig. 2 (A-A),

binding curve of Zn2+to calregulin.

Recently a role for Zn2+ in sea urchin sperm motility and acrosomal reactions has been suggested (12). Furthermore, Alitalo et al. (13) have demonstrated that micromolar concentrations of Zn2+stimulate the phosphorylation of two polypeptides of M , 54,000 and 57,000 from mouseepithelial membrane vesicles. At least two Ca2+-bindingproteins, calmodulin and SlOO protein, have been shown to bind Zn2+ (14). Calmodulin binds Zn2+ with two sets of sites with Kd(zn2') ranging from 80 to 300 WM. The SlOO protein also exhibits two sets of Zn2+-bindingsites with Kd(Zn2+)between 0.1 and 1 pM. The binding studies described in this report establish two distinct divalent cation-binding sites on calregulin, a single site for ca2+with an apparent Kdof 0.05 p M and about 14 sites for Zn2+with an apparent Kd of 310 pM. Since calregulin precipitates at zn2+concentrations greater than 1.0 mM it is unlikely that more than 10 Zn2+sites can be bound under our assay conditions. Although the cytosolic free concentration of Zn2+is unknown the total cytosolic concentration of Zn2+has been estimated at about 200 p~ (15), and the cytosolic free Ca2+concentration is known to be about 50 nM (16). It is, therefore, reasonable to propose that under physiological conditions calregulin binds both Ca2+and Zn2+.The Ca2+ and Zn" sites oncalregulin appear relatively specific. Ca2+binding is unaffected by a 500-fold excess of Zn2+,M F , Mn2+, or Fez+ (Table I), whereas Zn2+binding as monitored by changes in intrinsic fluorescence is unaffected by a 10-foldexcess of Ca*+, M e , or Mn2+ (Table11). Intrinsic protein fluorescence and thehydrophobic fluorescent probe ANS have been used to analyze conformational changes in calregulin during interaction with Ca2+and Zn2+. Ca2+binding to calregulin results in a modest increase in its intrinsic fluorescence and a blue shift in emission maximum of about 3 nm. In contrast, Zn2+binding to calregulin causes a dramatic increase in its intrinsic fluorescence and a 11 nM red shift in emission maximum. Furthermore, Zn2+binding alone results in a large increase in ANS fluorescence. Both the changes in tryptophan and ANS fluorescence are specific, since other metal ions such as K', M$+, Mn", and Fez+ do not either inhibit or mimic the effects of Ca2+and Zn2+on calregulin conformation. It is known that fluorescence intensity is affected by movement of charged groups along with hydrophobic changes in the microenvironment, whereas the shift in Amax is uniquely sensitive to hydrophobic changes in the microenvironment (9). Therefore, we used shift in, , , ,X rather than fluorescence intensity as an index of changes in hydrophobicity of calregulin. Fig. 6 shows that Zn2+-inducedchanges in intrinsic fluorescence emission maximum and ANS fluorescence do not parallel Zn2+binding. The binding of a single Zn2+to calregulin, at about 20 WM Zn2+,produces 80% of the fluorescence change

Conformational Studies on Calregulin and 55% of the ANS fluorescence change. Considering that the Zn2+-bindingsites of calregulin are homogenous noninteracting sites (Fig. 2 ) this result suggests that a very large conformational change occurs upon the binding of a single Zn2+. The formation of a Zn2+/calregulin molar ratio of 2 produces 100% of the shift in intrinsic fluorescence of calregulin and 70%of the ANS emission maximum shift. At 6 mol of Zn*+/mol of calregulin the ANS fluorescence change is saturated. This suggests that the binding of 2 mol ofZn"/ mol of calregulin results inthe majority of the Zn2+-dependent conformational change in calregulin. Our study clearly demonstrates that Ca2+and Zn2+induce differentconformational changes in calregulin. The Zn2+dependent conformational change appears to involve an increased hydrophobicity of the protein, as illustrated by the Zn2+-dependentincrease in ANS fluorescence and red shift in fluorescence emission maximum. Two calcium-binding proteins, calmodulin and SlOO protein, have been shown to bind Zn2+ (14). In contrast to calregulin, human calmodulin and SlOO protein do not show Zn2+-dependentincreases in hydrophobicity as monitored by the hydrophobic probe 2-p-toluidinylnaphthyline-6-sulfonate,but do show Ca2+-dependent fluoreschanges in 2-p-toluidinylnaphthyline-6-sulfonate cence. The Ca2+-dependentincrease in hydrophobicity represents the mechanism by which calmodulin activatesits target proteins (17), and it is possible that a similar Zn2+dependent mechanism might exist for calregulin. Although highly speculative, it is also possible that calregulin represents a site for the coordinated regulation of cellular processes by

8887

Ca2+and Zn2+ REFERENCES 1. Waisman, D. M.(1983)Fed. Proc. 42, Suppl. 7,2614 2. Waisman, D. M., Smallwood, J., Lafreniere, E., and Rasmussen, H. (1984)Biochem. Biophys. Res. Commun. 119,440-446 3. Waisman, D. M., and Rasmussen, H. (1982)Cell Calcium 4,89105 4. Khanna, N. C.,and Waisman, D. M. (1986)Biochemistry 25, 1078-1082 5. Khanna, N. C., Tokuda, M., and Waisman, D. M.(1986)Methods Enzymol., in press 6. Waisman, D. M., Salimath, B. P., and Anderson, M. J. (1985)J. Biol. Chem. 260, 1652-1660 7. Fabiato, A. (1981)J. Gen. Physioi. 78,457-497 8. Scatchard, G. (1949)Ann. N. Y. Acad. Sci. 51,660-672 9. Lakowicz, J. R. (1983)in Principles of Fluorescence Spectroscopy, Plenum Publishing Corp., New York 10. Fairclough, R. H.,and Canter, C. R. (1978)Methods Enzymol. 48.347-379 - - , - - . - .11. Chrapil, M., Weldy, P. L., Stankova, L., Clark, D. S., and Zukoski, C.F. (1975)Life Sci. 16. 561-572 12. Clapper, D. L., Davis, J. A:, Lamothe, P. J., Patton, C.,and Epel, D.(1985)J. Cell Biol. 100,1817-1824 13. Alitalo, K., Keski-Oja,J., and Bornstein, P. (1983)J. Cell. Physiol. 115,305-312 14. Baudier.J., Haglid, K.. Haiech, J.. and Girard., D. (1983) . . Biochem. Biophys. R e s T C o k u n . 114,i138-1146 15. Thiers, R. E., and Vallee, B. L. (1957)J. Biol. Chem. 226, 911920 16. Snowdowne, K.W., and Borle, A. B. (1984)Am. J. Physiol. 247, C396-C408 17. LaPorte, D. C., Wierman, B. M., and Storm, D. R. (1980)Biochemistry 19,3814-3819