The high-affinity calcium binding protein of sarcoplasmic reticulum. Tissue distribution, and homology with calregulin

Biochtrm~a et B, ophyslca Acra q82 (19B9) 1-8

1

Elsexaer BBAMEM 74~6

The high-affinity calcium binding protein of sarcoplasmic reticulum. Tissue distribution, and homology w~th calxegulin L a r r y F h e g e l ~, K a m b e d y B u r n s ~, M m h a [ O p a s ~- a n d M a r e k Machalak t i Heraaqe Car~h¢~oa~zlarDtrease Group Dcp~tm¢~reo} Pedml~ts a~td t]~hem~ry Um~er~tv of 4~er~a, Edmonro~ Alberta and 2 Department o/ Aaa~om> Umvers~ryo/Toromo Toronto Onfurz~ i Canuda) (Recer,ed 3 Jartuar'¢ [q@q)

Key ~ords Cakmm bmdmgprotettL Sarcoplastmc rcttcelum, Endop|a~n~c retlt.L|.lh3 Calreguhn homology

The 55-kDa hlgh-~finity calcium binding protein (HAC~P) was first ~dentff|ed and isolated item skeletal muscle sarcoplasmic retlealum (SR) Using polydonal anlibodies raised against the HACBP ,~lated from skelelal muscle we have identified this protein in cardiac and smooth muscle as well as in non-muscle cells. Allh0egh the 5~-kDa protein has a size, properties a~3dIocalizaUon simdar to that of calsequeslrin, the two proteins are tmmmm~ogieallyd~sfim-'t.The NHz-terminal sequence el uterine HACBP is also completely different from that of ,.al.,,equest~-in but it is identical to that of rmbbit liver calmgulln, u reeendy identified oak|urn binding protein lnd,rect ,mm.tmed[luereseenees~in|n~ of frozen sections and culture eeUs from • ~ariety of tissues slmws that the 5S-kDa protein [oc~[zes ~dolahmn~y to junctional SR and T - t u l l e meas in skdetal m ~ l e , to SR in smoolh and cardiac muscle cells, and to ER in a variety of non-muscle cells These data show tbat the i~rotein is present in a withe variety of nssues and suggest that ~t is a protein common re, ~ h s~rcoplasmie and endol#asm~c ~eticulm~ memlwanes.

lnO'oductlon The sarcop|asrmc rettculum (SR) membrane plays a key role m the regudaiton of the contractmn-relaxatton cyd¢ of skeletal muSCle [1-4] The SR (Ca 2. + MgZ*)dependent ATPase removes calcmm from the cytosol causing a decrease m free calcmm concentration and mlUatmg relaxauon Calsequestrm [5], a major calcmm binding protein m SR membranes, acts to store caIcmm wttlun the SR membrane lowering the free calcium ¢oncentraUon wtth_m the Sit, and thereby lowenng the cakaum gradmnt against which the (Ca 2+ + Mg2+)dependent ATPase must pump lnv~t~gat~ons on the protein composition of the skeletal muscle SR have revealed the presence of a bagh ~fimty calcium binding protmn (HACBP) of ap-

Abm'evt~Uons SR, smoopl~u¢ ~eueulum, HACbP, lugh-affimty calcium binding ?rattan EIL endoplasmm reuculum EDTA, ~thyl enedmfmnetetraacetm a~ad, EGTA, edtylene$1y¢ol bts(fl-arnln~thyl cther)~traa~ettc a~d, PBS, phosphate-bufkrcd ~hnc, SDS-PAGE, 8Oc.~,.~ d~'4cc31 :Jra'* ~--~:ly~,-yltmlde ~,~1 ~'leeetopMreqn~ Plugs ] 4-ptpetaTaaednethan~ulfomc acid. Correspondence M Mchelak, Heritage C~&ovascular 13~¢.ase ~roup, D ~ a t t t ~ a , of P~h,~t~c.s, ~]',L He,~*ag.~M©dlr.a] Rt..~ch

Bldg., Edmmlten,Alberta Canada, T6G 2R7

parent molecular w e i g h t of 55(~00, which was hrst identified and punfmd by OstwaId and MacLennau [6] in the presence of 100 mM KCI, it binds calcmm wnh a much ]usher afftmty than calscquestnn (K d 2 5-4/~M vs 1 raM) but wRh a lower capacity (1 tool of calcium per real of protein vs 40) [5,6] The HACBP Js a tmoor component of the SR membrane locahzed to the lumen of the SR membranes [7] The HACBP ts present m hghL intermediate and [~eavy SR fractions (free and junctional SR) and tt may be increased m fractious camched m transverse tubules [7] In tbas study, we have identified the HACBP m cardiac, smooth muscle and m non.muscle cells Using mlmunofluorescenee microscopy techniques the HACBP has been locahzcd to g R m skeletal masCte and to ER membranes m non-muscle ussues Our resuhs also show that calreguhn a calcium binding protein of unkaxown funcUon recently ,denthqed m bovine hver [8] ts probably identical ~o the HACBP A pmhmmar) report of tbas work has appeared elsewhere [9]. Experimental prooedures MaWrJa~

DEAE-Seph~

was ~ u ~ d l ~ t from Phm~n~&

Affi-Gel Blue ~ff~.Ge! 10, gel ¢l~trcphorcsls reagenls and molecular weight markvr~ were from Bto-Rad

~0~2736/S9/$03 50 © 108q ElsevierSciencePub~hets B V (Blomed~cal D;vlslor0

Nttrocellulose and polyvmyhdene dlfluonde membranes were obtmned from Mdhpore Horseradish peroxadase hnked rabb~t anu-goat anlabody and goat ante-rabbit antibody were from Bzo-Can Sclentafic Ltd and Bachringer Mannhe~m, respeettvely Pipes was from Sigma D~slaase wa.~ purchased from Collaborahve Research Trypsm-EDTA, fetal calf serum, pemedhn, fung~zone, streptomycin, and alpha mmunal essenaal medtum were purchased from Olbco Bovine hearts were obtained from a local slaughterhouse Oten from adult female rabbas were ob~.mned tmmedmtdy after term, and were a generous gift of Dr G D Lopasehuk, Umverstty of Alberta Purified bovine hver calregulm and rabb~t anti-bovine calregulm at~ubody were a generous gift of Dr D M Watsman, Umverstty of Calgary

Preparation of membrane fractton~ g a b h t skeletal muscle SR vesicles and cat&ae SR • esleles were tsolated as described by Metssner [10] and Pegg and Michalak [11], respecu~ely U t e n n e nueroseines were isolated by howogemzmg the r a b h t uterus vath a Polytron PT-10 m a buffer eontalmng 120 m M NaC1 10 m M mmdazole-HCl (pH 7 4), 0 5 m M phenyimethylsulfonyl fluoride (PMSF) The homogenate was centrifuged at 1600 × g for 10 rmn and the superaatant was saved The peUet was rehomogemzed m the ~ame buffer and centrifuged as above and both supernatants were combined These were centrifuged at 1 0 0 0 0 × g for 15 nun apd pellets were dl~c~rded The supernatant was pelleted at 1000{30 x g for 1 h and the maerosomes were suspended m 250 m M sucrose, 10 m M " m s (pH ~' 0), frozen m hqmd N 2 and stored at - 7 0 * C

Partral pueaficauon of HA CBP HACBP we8 isolated from bovine hearts, rabbR skeletal muscle, and rabbtt uterus by selextlv¢ ammorattm sullate premp~ tahon, follo~ved by fracttonatton by DF.AE-Sepharose Thts procedure was !tdenttoal to that prewously described for the punficatxon of eal~equestnn [12] and was developed m lhe laboratory of Dr K P Campbell, Umverstty of Iowa Tlus method revolves dtrect extractmn of the HACBP from whole muscle komogcnates ar.d ehmanates the need to prepare a SP fraction Membrane-bound protems are soluh~lLzed by this homogemzatlon procedure [12] At the final stage of p u n h c a u o n an anamomum sulfate prectpttate contmmng both HACBP and ealsequesmn was dissolved m 100 m M potassmm phosphate buffer (pH 7 1), contatmng 1 m M EGTA, d~alyzed agmnst the same buffer for approxtmately 20 It, and apphed to z 2 5 × 8 em column of DEAE-Sepharose eqmhbrated with 100 m M potassium pnc~phat¢ (pH 7 I), eon~.~.,,±ng ! m M EGTA and 50 m M NaCI The eclurmx was washed v, tth mmal buffer and protein fraettons (approx 5 ml/fracnon) were eluted vath a linear gradient of 0 05 to 1 0 M NdCI [12] The fract:ons e.,Lnched m H A L B P (200-250

m M NaCI) were tdentthed by ammunostalmng with anIL-HACBP anhbody, ahd pooled

8DS-PAG£, transfer of proteins to nitrocellulose, and reaction w~th anttbodwes SDS-PAGE (10~$ acrylarmde) was t a m e d out according to L a e m m h [13] The gels were stained with Coomassze blue a t w~th the cauomc oarbocyanme dye 'Stems-Air [14] Densltometnc scans of SDS-PAGE were earned out on a Beckman DU-65 spectrophotometer Protem samples were transferred to mtrocelhlose or to polyvmyhdene dtf[uonde membranes [17] The standards used were Bxo-Rad 1o,,, range molecular weag.ht proteins, phosphorylase b (97400), bovme serum al b u m m (66 200), ovaibumm (42 "/00), earbome anhydrase (31000), soybean ttz~sm ~ahlbltor (21500), and lyso-

zyme (14400), or Btc,-Rad prestame,l ,Jlarkers;phosphorylase b (135000), bovine serum albumin (75000), o v a l b u m m OO000), carbome anhydrase (39000), soybean trypsin mhtbttor (27000), lysozymc (17000) Immunoblottmg was earned out m the presence of 1% nulk powder Protein bands reactive with anttbod~es were v~suahzed by developing a peroxtdase reaetton

]~reparauon of anubodtes For h,nmun.w.atlort H A C B P was purified from skeletal muscle SR vesicles by the procedure described by Ostwald and MacLenuan [6] Calsequestrm was p u n h e d from skeletal and cardtac muscle hcmogenate.s by the a m m o m u m sulfate precapltatton method [12], Anttbodles against H A C B P and skeletal muscle calsequestrm were raksed m mature goats by the procedures described eaxher [7] Anttbody against cardiac calsequestrm was raised m rabbits as descnbed earher [ 1]

Immunofluoreacence mwroacopy The tissues dzssected from newborn rabbtts were etther fixed and processed for cryonuerctomy as described [16], or used to estabhsh prmmry cultures For txssue culture the Ussues were freely chopped, incubated at 37 ° C for 30 n u n m Dtspase, follow0d by 20-30 n u n m~ubatton with trypsm-EDTA Ceil suspcnstons were, decanted from over larger pteees of undigested ttssues, centrifuged, incubated with pure fetal calf serum for 10-15 ram, and plated onto coUagen-coated cover shps m alpha mnumal es~nttal medtum supplemented with 10% total fetal calf serum and c.ontammg 100 IU pemeflhn, 0 25 /~g fangtzone and 100 ~g streptomycin per ml For ~mmunoflucr.ocenee nueroseopy eeUs on co cer shps were fixed for 10 n u n m 3 7% formaldehyde m PBS extracted for 2 n u n with 0 1% Triton X-100 Jrt a buffer contatmng I m M EGTA, 4% (w/v) poly(ethylene glycol) 8000 (PEG 8000), 100 m M Plpes (pH 6 9) Stmnmg was cart~.ed out for 30 n u n at room temperature on Lells or 6 - 8 p,m frozen seettons with the nntt-

body agmnst HACBP (dduted 1 50 in PB¢3) followed by 10 nun wash m PBS and mcubauon for 30 mm at room temperature vath rabbtt antt-goat FlTC-conjugated lg(3s (Nordm) chluted 1 30 m PBS For absorptton experiments 3~-pA ahquots of ~ntabody against HACBP (dduted 1 SO m PB$) were incubated [or l h at 37"C with 12 ~tg of purified utenne HACBP~ centrtfuged and the supematant used for tmmunofluoreseence s m t m g r As an adchttonal control, the ~mmur~ofluoresc~¢nce procedure was performed w~th the onussmn of the anUbody against HACBP After the final wash tn PBS, the cover shps were mounted m Vmol 205S (St Lawrence Chermcal, Toronto, Ontario) conta~mng 0 25% 1,4-chaZOblCyClo{2,2]octane (Polysctences) and 0002% p-ph~nylunedmnune (Fisher Sctemttfic) to prevent photoblea¢lung A ZelSS Phototmcroscope 111 equipped wtth an eptfluorescence condenser and selective FITC filters was used for ob~ervaUon and photography Photographs were re.corded on TMa×-400 negaUves (Kodak) pushed to 1600 ASA with TMax developer

Mfscel/aneaus procedures NH2-termmal sequence anal)sis of HACBP was carned out using parttally purified utenne HACBP (from the DEAE-Sepharose c o l u m , Flg~ 2) separated on SDg-PAGE slab gels and transferred to poly(vmyhdene chfluondo) membranes [17] Automated protein sequence analysts [18] of the poly(vmvhdene dtfluonde) membrane dectroblotted protein was earned out on an Apphed Btosystems MOd~I 470A gas-hqmd phase protom sequencer connected ondme to an Apphed Biosystoms Model 120 A HPLC. usmg the ¢urrem protocols of Apphed Btosystems for both instruments Chromatogram.~ were recorded on a Spectra Phystcs Model 4270 recording integrator. All ehermcals used for protein sequence analysis were from Apphed Btosystems, Foster

A

B

Ctty. CA Two mdt.p~,1dent sequence a'~alysts were performeLi wtth rcpetmve y,elds of about 93% Protein was deternuncd by the method of Lowry et al []9], using howne serum albuman as a standard Results

ldem,ficatwn and isolation of HAC~Ii Ftg 1 shows the protein composmon and peroxadase stalmng of mtrocellulose ~hcets reacted ruth HACBP anttbody All three ttssues skeletal muscle SR. cardtac SR and utenne rmcrosomes, coatmned a 55-kDa protern immunoreactlve with the HACBP anttbody rinsed against the skeletal muscle form of this protein (Fig 1B) The anubody against skeletal muscle calseqLte~tnn, v.hen tested agmnst th¢ same memhra:~ fracUon did riot react wtth HACBP and reacted only with skeletal muscle SR (Flg~ 1CI An addmonal sertes of h~gh molecular ~clght proteins (> 130 kDa) crossreacted m skeletal muscle SR vath the anu-calsequestrm antabody. however, thesr relataon to calsequestnn is nc,t know~ yet For the paxtml purtflCatton of HACBP we adapted a procedure descnbed for the punflcatmn of calsequestnn [12]. whtch takes advantage of the bagh solubihty of HACBP in ammomum ~ulfato soluttons. We have shown earher that HACBP was not prec]pttated by a m m o m m sulfate concentrattons as high as 65~ saturahon [20] However, tt could be prectpttated together with calsequestnn by ammomum sulfate concentrauons above 90% saturatmn at pH 5 0 A parttally purified preparatton of HACBP could then be obtained after DEAE-Sepharose chromatography of the ammomnm sulfate prectpRated protetn fracttons This method chromates completely the need to prepare SR frachons prior to the lsolatlon of the HACBP

~.

kDa

C

75

50

1

2

3

4

2

:

1

2

3

Ftg. 1 SDS-PAOE alxl iffU~luliOb]Otaxlalysl~of utenne ancrosomm (]L cardtac muscle SR (2), and sk¢lalal mtt~l¢ SR (3) Membram: baCtlorts

were prcpar~ as deacnt~ under Experimental Pro¢~ures (A) CoomassL¢blue stmmngof polyonyhde~ dtfl.onde membrane [17], (B) unmunolxscU~atyof SK protons ",~th antibody prepaz~d against HACBP. and (C) anmunoreacu~tyof SR pmtetm ,oath anUbodyprepared againstskel~tl musclecalseqL~tnn

kDa

A

B

C

hg 2 Identlfltalmnof the HACBPin DF_.AE-Scpharosefracuo~sof arttm,m u m ~ulfate extract of rabbtt uterine muscle Ammomma s~lph,~t¢extraeUonand DEAE-SephaJ~sechromatograph~were earned out as described uader Expexa3tental Prooedure~ (A) Protein composmonof HL/gCBPc,oBtammgfraetton(Coomassteblue staining) (B) td¢.ntlficaUonof HACBPconta.uuagfracboaby tmmunosiammg {mtlocdlulom mem'aranes) (C) Stature8 ot the HACBP contaJmng fracaon v,~th"Stares-All'

Fig 2 shows the partially purified HACBP from the rabbtt uterus by ammomum sulfate precipaanon and DEAE-Sepharose chromatography The peak HACBP comammg fractions elmed at 200-250 mM NaCI The HACBP appeared to be present m relatively large amoums m ut=rtts The uterine protein was strongly mtmunoreacu~e with the anl~-skeletal muscle anubody mdtcatmg sigmhcant homology between the two proterns Wtth the ennehed uterine HACBP fractions, we observed that the protein stained blue ~ath the cattomc earbocy,mme dye "Stains.All' (Fig 2C) A smaller molecular wetght protein also stained blue (Fig 2C) and hkely represents a degradat)on ~roduct of the pUlafied ltACBP Using me same procedure the HACBP has also been partially punfted ?veto the bovine heart and rabbtt skeletal muscle {data not sho~n) Stmdar to the merme HACBP, the eardtac and skeletal muscle HACBP stamed blue gath 'Slams-All' (data aot shown) The bovine car&at and skeletal muscle HACBP were of the same apparent molecular welght as the uterine protein and duted from DEAE-$epbarose prtor to eardtac and skeletal muscle calsequestnn (200-250 mM NaC1 vs 309-350 mM NaCI)

lmracellular Iocahzatmn of H4 CBP Longttudlnal frozen sections of newborn rabbtt gracfl~s muscle conststenfl~ dir'~layed a striated, double-banded pattern of st~mmg with the antibody

against HACBP i,Ftg 3A) HACBP-pomtave bands comprising a doublet were found closely apposed to both stdes of a Z-hue and occupy a total length of 1 1 -I- 0 2 fxm vath salcomere length of 2 9 pm + 0 1 pm Narrow unstamed bands separating the two HACBP-posattve bands m the doublets corresponded t o Z-lines as vasualized with phase contrast rmcroscopy (Ftg 31)) Absorptton of the anttserum wtth partmlly purified HACBP abobshed the speclhc stmnmg (Fig 3B) Ormsston of the HACBP anuserum from the statmng procedure abolished staimng completely (Fig. 3C) Staining of frozen sections of newborn rabbtt reteSt,he (Fig 3G-I) showed heavy mtracellular labelling of intestine ep~thehum and underlying tt smooth muscle cells (Fig. 3G) Heavy stmmng of hver paranehyma was also observed v,,tL anla-HACBP anlabody (FIg 3M) AbsorpUoo cf the ~pecihe antiserum with the antigen as well as t~e onasslon of the specfftc antiserum from the s ~ m n g p-occdure abohshed the staining of both intestine (Fig 3H, I) and hver (Ftg 3N, O) frozen s~ttons To fatahtate mtraeellular loeahzatlon of HACBP, we have stained spread cells from primary cultures of a variety of ttssues from newborn rabbtts in spread cells, HACBP loealmed exclustvely to a permuclear system of membranes corresponding to ER m smooth (Fig 4A) and cardiac muscle (Ftg 4B), fibroblasts (Fig 4B), intestine epithehum (Fig 4C), and e.ar&ae endothehum (not shown)

ldenttficatton of NH2-termmal sequence of HA CBP To obtain mformauon about the primary protean structure of HACBP, we sequenced the NH2.terrmnal portion of uterme HACBP The following eight N H 2termanal remdues were obtmnm:l from t~o independent analysts N H 2-Olu-Pro-Val-Val-Tyr-Phe-Lys-Glu(EPVVYFKE) Yhe sequence exactly matches the NH2-termanal sequence of rabbtt hver calreguhn as reported by Khamm et al [21]

Homology of calregul,n to HACBP In order to further detemnne the relattonslup between the HACBP and ealregulm, we tested the two proteins for cross-reactaxaty w~th thetr respective antibodies Rabbtt antt-bovme calreguhn amtbody cross-reacted wath uterine and skeletal muscle HACBP (Ftg 5A) Our polyclonal amtbody against rabbtt skeletal muscle HACBP, however, cross-reacted only very weekly with bovine hvar ealregutm (F18 5B) Discussion

We have tdemlhed and locahzed HACBP m skeletal, cardiez and smooth muscle as web as m a vaiaety of non-muscle cells The HACBP is present m a crude rmcro ,omal fraction of the uterus and m cardiac SR vestries The rabbit uterus, a tissue htghly enriched in

Specific

antlseram

Absorbed

antiserub

~econdazy

antxbody

Fig 3 Locahzation of the HACBP m r;x~en se~tzom of ~d~taZ mtLgC]e(8ZaGLlza)(A-F), mtmtmc ((3-1.) a,qd hvcx"(M-S) of a~'~wbom rabble I~ maraanofluoret, cance erector,copy 1~ lov~tu(Lmed r,c~tzo~ of must.le (A) HACBP is p r ~ m t m s|n~,ted pattern ¢,omutm8 of double bo.'lds ¢|o,%-~ appor, cd to both sides of each of Z-ln~, (D) Abs0~Llon of the anlnbody with HACBP abolusltcs the sp~:lic stmamg (B) Omzssim of the antnbody a~1 HACBP t'~ulled m r.o stmran8 (C) (E) and (F) arc phase coatrmt tmag~ of (B) and (C"~,~specb~ely HACBP |s Im~eat mtracellulm- m the lnle~tme eptthehum and tmd~|ym s st ~mooth nm.w.lcs (G) as well a~ m hvcr belmto~y~e~ (M) (33 and (P) are pb&r,e contrast mmges of (G) t a d (M), gespectlvely Ab~;~rplton of the antibody wtth HAC.BP ~x~lssh~ the speafic statmng m both h'~r (H~ K) and mte~tme (M, P). Seconda~ ant,body ~ no st~nmE when reed w~thout the specific a m ~ b ~ j (mlCSlme, I, L, hver, O, S) 5r.ale drdmer, -- 13/tin.

[.

,,

_

!

A

B

"~ .o)

. .

C

F,g 4 [mm.nofluoresceme loeabzatton of ~ ¢ HACBP m ,mltm'¢d smooth (A) and eardm~ (B) muscle c¢llg azld mt~tm¢ eguLbebalc~]ls (C) Arrow m (B) show~ Ihe dJslnbutton of the HACBP m a fibroblasL In all fl'waeceils HA CBP is present m a i~nsuckalr ~ystem of membrm~s composing ER ScaledlVlSlon ItO Frn =

smooth muscle [22], contains high amounts of the HACBP The molecular weight of the HACBP from cardiac, skeletal and smooth muscle estimated by SDS-PAGE m Laemmh gel system [13] is approx 55 000 HACBP and calsequestrm are sma]ar m site and localization, and both brad Ca ~+ but m different amouuts and with different affimttes [5,6] The present results show that these two protems are deafly thsUnct although present m the same membrane systems Polyclonal antthodies against either skeletal or e,trdtac muscle cMsequestrm do not recogmze HACBP and antibodies agmost the HACBP do not r¢cograzc either the skeletal or cardiac form of calsequestrm In addmon, the HACBP from skeletal muscle rmgrates slightly fa,,ter on Laemmh SDS-PAGE than ealse.questrm (55 kDa vs 63 kDa). and the skeletal and cardiac forms of the HACBP can be separated from calsequestrm by DEAE-Sepharose chromatography Another protem

A

B

kDa

pre~¢nt m the SR membranes of skeletal and cardiac muscle and of size smula~ to HACBP on SDS-PAGE as the 53-kDa mLrmsm #ycoprotem [7,23] This protein ts also dffferem from HACBP The intrinsic glycoprotem is of shghtly smaller apparent molecular weaght m SDS-PAOE, and the HACBP antibody d o ~ not crossreact with putafmd 53-kDa glycoprotem Anubodms agmnst tlus glycoprotein also do not react w,th HACBP (Fhegel. L , MacLetman, D H and Miehalak, M . unpubhshed observations). The HACBP ts also not glycosylat0d and does not brad any conconavaim A [7]. Thus, the HACBP is dafferent from either calsequestrm or the mtnnsm glycoprotem of the SR membranes Partially purified HACBF can be obtmned from tassue homogenates after ammomum sulfate preetpRatton and DEAE-Sepharose This method, originally developed for the punfication of car&ac and skeletal musele calsequestnn [12]. completely ehrnmates the need to prepare SR membranes poor to the ~so]aUon of the HACBP Partially purified preparations of utenne HACBP couture a protein band of 55.kDa, which slams blue voth the carbocyamne dye 'Stares-All' confia'rmng its Ca z+ binding properties, however, the staining vath 'Slams-All" ta relauvely weak m comparison to that of calscquesmn [14~ Tlus may bc an mthcatton of its lower Ca 2- binding capacity compared to calsequestnn

[5,61

1

2

3

4

1

2

3

4

Fig .5 lmmunoblottmg of HACBP and c,dtegulm voth Oolyclonal

anllbodtes against calRguhn (A~ and HACBP (B} SDS-PAGEwas earned o.! m 1O~oacohrmd¢ slab gel a0cordmgto the procedureof Laemmh [t3] Lane 1 bo~qx~ehvercalfeguhn, lane 2. rabbR utcrm¢ HACBP, lane 3, rabbfl skeletal muscle HACBP, lane 4, preslame6 molecular weight markers 5 tag o[ proteinw~ applied ca ¢ach lane

Our data show that the HACBP ]ocalmes predormnanfly to the area occupied by both junctmmal SR and T-tubule membranes m skeletal muscle and to SR m smooth and cardiac muscle cells, and to ER m a variety of non-musele cells m ~tro These data clearly show that the HACBP is present m a wide variety of ttsstms and suggest that tt ts a protein common for both SK and ER membranes It has been shown easher that the HACBP Is present in the isolated free and junctional SR vesicles, and tt may be increased m fractions ennched m T-tubules [7] Although HACBP-posmve slammg m found closely apposed to both sides of a Z-ImP., the resolutmn of tmmunocytochen'astry at an optical

m t c r o s c o p y level ts n o t sufhctent to d ] s t m g m s h w h e t h e r a n h b o d y s t m m n g ts l o c a l g e d in the j u n c t i o n a l S R or m the T-tubule, or m b o t h in the course of these studms, we have m a d e a n unexpected &~eovery Based o n N H : - t e r r m n a l a m i n o aczd sequence of r a b h t uterus H A C B P a n d the ~equenee reee,~tly r e p o r t e d for r a b b i t hver c a l r e g u h n [21], w e conclude that c a l r e g u h n ts either ~dent~cal to the H A C B P o r It ~s a t r m n m a u m h o m o l o g o u s over a s~gmfic a n t p o r t , o n of the protein C a l r e g u l m is a cal,.,u~: b m d m g protein of u n k n o w n function recently ~dennfied m bovine h v e t a n d m variety of o t h e r tissues [8,21] St[mint to the H A C B P [6] it b m d s 1 reel of c a l c m m p e r reel of protein v a t h a n a p p a r e n t K a = 0 05 a M [24] T h e a m i n o acid ¢omposatton of b o t h calreguhn [21] a n d the H A C B P [25] are r e m a r k a b l y s m u l a r Calregulm, hke the H A C B P , ]s associated with hver E R m e m b r a n e s [8] In o r d e r to c o n f i r m the ~dentlty of the ~ o proteins0 we teste~ polyelonal antabod~cs raised a g m n s l the t w o p r o teens for eross-reaeUxaty agmnst one a n o t h e r The r a b b i t a n U - b o w n e calregulm a m [ b o d y cross-reacted with p u n fled H A C B P f r o m all three muscle Ussues In c o n t r a s t the g o a t a n u - r a b b x t H A C B P a n t [ b o d y reacted strongly wroth purified H A C B P , b u t did not react well wxth purified bovine h v e r e a l r e g u h n (FLg 5) T l ~ )nay be du~ to differences m the p r o t e i n f r o m o n e species to a n o t h e r In feel, w e n o t e d m o u r s t u d y a decrease m the s e n s m w t y o f o u r a n n - H A C B P a n u b o d y , w h e n used against bo'~ate c a r d i a c protein, tn c o m p a r i s o n to rabbit tissues K h a n n a et at. [21] also n o t e d species differences between v e n o u s f o r m s o f c a l r e g a h n It should not b e ruled out, however, t h a t only a It[rated porlaon of the t w o p r o t m n s p r i m a r y sequences axe ideal[eat U a m g the polyclonal a n n h o 6 y a g h a s t the H A C B P , we h a v e recently isolated a partial length e D N A clone encochng f o r the C O O H - t e r m m a l of the H A C B P f r o m a r a b b i t skeletal muscle h g t l l expression h b r a r y (FhegeL L , B u m s , K , M a c L e n n a n , D H , a n d M]chalak, M , u n p u b h s h e d observations) T h e e D N A of the clone, hybridizes to m R N A of a p p r o x 2 5 k b m length, which ~s the expected stze o f m K N A c o d i n g for rims protein, 0resumang 3 ' a n d 5 ' u n t r a n s l a t e d regions t o t a l h n g a b o u t 1000 b p T h e C O O H - t e r m m a l 28 amino-send of the H A C B P d e d u c e d f r o m the e D N A clone ~s also nt,h m acidic residues, wl~ch ~s s u m l a r to the C O O H - t e r m m a l o f skeletal muscle c a i s e q u e s l n n [261 The f u n c t m n o f these ae~chc residues is n o t k n o w n , b u t they m a y b e revolved m either ca)cram b i n d i n g o r m a y h a v e some structural function s u c h as a n c h o n n g the p r o t e t a to the meanbran¢ The four C O O H - t e r r m n a l a m i n o acids of the H A C B P , - L y s - A s p - G l u - L e u - C O O H d e d u c e d f r o m o u r e D N A d o n e , c o r r e s p o n d to the sequence described for other E R - a s s o c m t e d calcium b i n d i n g proteins, t e , e n d o p l a s m m (100 k D a ) [271, pr01em dtsulplud¢ lSOmerase (61 k D a ) [28], as w e l l as a farealy o f stress-red u c e d proteins [29] T h e ternunal f o u r a m i n o - a c i d s

- L y s - A s p - G l u - L e u - C O O H may be a p a r t of a speclhC s~gnal that causes reteutmn of these proteins m the E R [30] It ]s possible that H A C B P belongs to the s a m e farmly of ER-assoctated, ealemm b i n d i n g p r o t o n s , b u t besides being a b u n d a n t to non-muscle cells, ,t is also present in muscle tissues Acknowledgments We w o u l d like to thank D r D H M a c L e n n a n ( U m ',erstt2~ of T o r o n t o ) for a generous gift o f a n n b o d y a g a i n s t the H A C B P W e t h a n k D r A O Jorgeusen (Umverstty of T o r o n t o ) for p r o w d m g us w~th frozen sect,ons of rabbLt gracd]s muscle We would h k e to t h a n k D r D W a i s m a n (University of Calgary), for a generous gift of calreguhn, a n d a n t [ b o d y against c a l r e g u h n W e a c k n o w l e d g e the superb techmcal asseslance of Ken Wtas~eh,~, and the excellent secretarial asmstance o f Cole[re Eth~ r This ~ o r k was s u p p o s e d b y the A l b e r t a H e a r t a n d & r o k e F o u n d a t i o n , the Medical Research Courted of C a n a d a , a n d the Alberta H e n tage F o u n d a u o n for Medical Research L F ~s a Postdoctoral Fellow o f the Alberta Heritage F o u n d a t i o n for Medical Research M O ~s a Scholar of the Medical Research C o u n c d M M is a Scholar of the Mechcal Research C o u n c d a n d the Alberta Heritage F o u n d a t i o n for Mechcal R e s e a r c h

References I ~ t ~ M Yam~moIo f ,m0 Tonom~ra Y 11978)Phy~ao! P~v 55 1-7q 2 lkemolo, N (t982) Annu Roy Physm) 44, 297-317 3. Manono~t A (1984)Ph~/sto] Ray 64 1240-1320 4 Mlchalak M (1985} m The Enzyme..gof B:ologmal Membranes (M,~to~ost A N ed) Vol 3 pp 115-155 Plenum Press. Ntw York and London 5 Mac:Lent-an El H Campbe~d K P and ReJihmeler R A F (1983) m Calmum and Cell Functmn {Chang. WY ed } Vol 4 pp 151-173, n.cadem~cPr~s.~ blew York 60stwa].d TJ ,rod Mackcnnan, D H 1197't) J Broi Chem 249 97,1-979 7 Mtchal,r&, M, Campbell, K P and MacLcaman, D H (1980} J atol Chem 255, 1317-1326 8 We[smart D M Sahmath B P and Anderson M J 0985) J B~ol Chem 260 10.52-1~1~0 9 Flnegel L Opas M and Mtchalak M (1~88)amphys J 53 ~38a 10 Me ~ner q O984)J B~ol Chem ~59, 2365-23"/4 11 PeM W and M l e h ~ M 0987)Am J Phyaol 252 H22-H31 12 Slupsky J R Ohnlshi M Cal'penter M R and Renhmemr R A F (1087) Bioehermstry 26, 6539-6'344 13 Laemmh U K (1970) Nature (Lo~don) .~.7, 680~685 14 Campbell, K P MacLcnnan D H a~l Jorg¢nsen A 0 (1983) J ale] Chum 25B, 11267-11273 15 Towbm H $taehelm T and Gordon, ] (1979)Proc Ned Ac.ad Sol USA 76, 4350-4354 16 Jocgensea A O Kalmas V and MarLennan, D H (1979) J Cell B~ol 80, 3?2-384 17 Matsudmrs, P (lag7) J aml Ch~arL262, 10035-10038 [~ HeWIg]g, R M , l-.Lmkapdler MW, Hood LE a~d Dreyer, WJ (1981l J Bml C'hem. 256, 7990-7997

19 Lowr,j O.tl Rosebrough lCJ Farr A L and Randall, R J (l~2~t~ J ~h,.n Chcm 193 265-275 20 Mmhalak, M and MacLeanan, D H (1980) J BIol t h e m 255 1327 1334 21 khanna N C, Takuda M and Watsman t D M (1987)Blm:hem J 242 245-251 22 Dared E E. Grover A K and Kwan C Y 0982) Fed P~0¢ 41, 2~91s-2904 23 Campbell K P MacLennan D H. J o f g ~ l ~ A O and Mmtzer. M C {1983)J B*o] Clam 258 1197-1204 24 Khanna. 1NC Tokuda M and Watsman, D M (19861 J Bml Cbe, 1 261 ~,883-gSg7

25 MacLennan, D H . Ytp, C C . lies. G H and Seeman, P {1972) Cold Spnng Harbor ~lymp Quant Btol 37, 469-477 26 Fhcgcl, L Ohntskt M Carpenter, M R , Khanna, V K. Relthmeter, R A F and MacLennan, D H (19871 Pro¢. Natl Aead. Set L1SA :[]4, 1167-1171 27 Kc~h. G L E (191[7) J Ceil ~ l 87~491 tg2 28 Edman. J C Elhs, L, Blacker, R W , Roth, P..A and Rutler, w $ (1985) Naluf¢ 317. 267-270 29 Munro S and Pelham. H R B (1986} Cell 46. 291-300 30 Munro, ~ and Pelham, H g B (1987} Ceil 48, 899-907