Vol. 171, No. 1, 1990
August 31, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 26-32
THE CHICKEN OVIDUCT AND EMBRYONIC RED BLOOD CELL TRANSFERRIN RECEPTORSARE DISTINCT MOLECULES INDRAVATHAMMA
POOLA, *ANNE B. MASON AND JOHN J. LUCAS+
Departmentof Biochemistry and Molecular Biology, StateUniversity of New York Health Science Center, Syracuse,New York 13210 *Departmentof Biochemistry, The University of Vermont College of Medicine, Burlington, VT 05405 Received
We recently describedan estrogen-inducibletransferrin receptorfrom the chicken oviduct. We now report on the comparisonof the oviduct transfer-tinreceptorwith the transferrinreceptor obtainedfrom chick embryo red blood cells. Westernblot analysisrevealsthat rabbit polyclonal antibodiesraisedagainstone receptor do not crossreact with the heterologousreceptor. Furthermore, peptidemap analysesof either affinity purified, native -labelledtransferrin receptors(dimers)or dissociated,andrepurified monomersobtainedfrom oviducts and embryonic red blood cells yield distinct patterns. Therefore, the estrogen-modulatedoviduct transfenin receptorappearsto be structurally distinct from the iron-modulatedred cell transferrin receptor. 0 1990
Iron is an essentialelementfor normalcellular function andproliferating cells have an increasediron requirement. Complex systemsfor acquiring and storing iron have evolved and vertebrate cells acquireiron via receptor-mediatedendocytosisof serumdifenic transferrin. The structure,function and regulationof mammaliantransferrinreceptorshave beenextensively reviewed (l-3). Generally, the transferrin receptoris a disulfide linked homodimericglycoprotein with an n/rr 180KDa. However, there is speciesvariability in receptor fine structureasjudged by sequencecomparisonsand immunologicalcrossreactivity (4-6). The transferrinreceptor is strongly regulatedby iron availability, with iron deficiency eliciting increasedreceptorexpression (7). This regulation is mediatedby iron regulatory elements(IREs) in the 3’-non-translatedregion of the receptor mRNA which affect mRNA half-life (8,9). Receptorexpressionis alsomodulated by severalgrowth factors (10,ll). We have recently purified andcharacterizedan estrogen-inducibleglycoprotein (m 91 KDa) from henoviduct membranesthat exhibits propertiesof a transferrin receptor. Thus, it is a subunitof a disulfide linked homodimer(M-r 180 KDa), which can be purified by affinity chromatographyon an immobilizedovotransferrin column (12). To further characterizethe oviduct transferrinreceptorwe have comparedit with the chick embryo red blood cell tmnsferrin receptoremploying immunologicalcrossreactivity experimentsandpeptidemappinganalyses. Chick embryo red blood cells have a high numberof transferrin receptorsasthey have a high iron + To whom correspondenceshouldbe addressed. 0006-291x/90 $1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 171, No. 1, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
requirementfor hemoglobinsynthesis.The avian red blood ceil receptorhasbeenpartially characterized(13,14). Our resultsindicate that the estrogen-induciblechicken oviduct transfer& receptor is distinct from the embryonic red blood cell receptor. This result is unique; in other organismsthe transferrin receptorsfrom different tissuesappearto be identical (15).
Na[I251]( 100mCi/ml) was from AmershamCorp. Aft?-Gel- 10 and SDS-PAGE protein molecular weight standardswere from Bio-Rad. Ovotransfexrin and heparinwere from Sigma ChemicalCo.. Ovotransferrin wascoupledto Affi-Gel at 15 mg/ml gel and equilibratedwith 2.5M excessof ferrous ammoniumsulfate asdescribedpreviously (16). [125I]protein G (2-15 uCi/ug) and StaDhvlococcusaureusV8 proteasewere from ICN Immunobiologicals. Desferrioxaminewas from CIBA-Geigy. Fertilized White Leghorn eggswere obtainedeither from the Poultry Science Dept., Cornell University, Ithaca, N.Y. or Oliver Merrill and Sons,Londonderry, New Hampshire. Isolation of Chick Embrvo Red Blood Cells: Chick embryo red blood cells were prepared essentiallyasdescribedby Mason and Brown (17). Fertilized eggswere incubatedat 37’C, 80% humidity for 14 days. The shellswere crackedopen at the top and the Y shapedvein wasruptured with a curved forceps. A smallportion of PBS (0.1 M, pH 7.4) containing 200 [email protected]
of heparin wasaddedto aid the blood flow. The embryo wastipped onto a nylon screenwhich wasrubber bandedto a beakerto collect the blood. Typically, one dozen embryosyield about 3 rnL of blood. The blood wascentrifuged at 2000 rpm in a BeckmanJ-21 rotor at 4’C, the supematantwas decanted,and the red blood cells were suspendedin 5 volumesof Earle’ssalt solutioncontaining 2.5 mg/mL BSA. The cells were incubatedat 37°C under5% CO2 -95% air for 10 min with shaking. The cells were centrifuged and the incubationswere repeatedthree times. &I ‘fication of 0 ‘duct Transfenin Receptor: Hen oviduct membraneswere preparedas pr$ously de&red (18). The oviduct transferrin receptor waspurified asdescribedpreviously with minor modifications (12). Oviduct membranes(approximately 15 mg protein) were solubilized in 2 ml of TBS (0.05 M Tris-HCl, pH 7.4), 1 mM PMSF, and 1% NP-40 and loaded on the ovotransferrin Affi-Gel column (2 mL gel). The column waswashedwith 200 mL of TBS, 1 mM PMSF, and 0.1% NP-40 followed by 50 mL sodiumacetate(0.15 M, pH 5.0), 1 mM PMSF, and 0.1% NP-40. Iron wasremoved by washingthe column with 50 mL of citrate buffer (0.05 M, pH 5.0), 1 mM PMSF, 0.1% NP-40 and 50 @ml desfenioxamine. The transferrin receptor was eluted with 10 mL of either glycine-NaOH (0.1 M, pH lO.O), or NaHC03 (0.1 M, pH 8.0) containing 1 mM PMSF, 0.1% NP-40, and 50 ug/mL desferrioxamine. No difference was found in the eluted product usingeither eluant. The eluatewas concentratedto 200 j,tL in a Centricon microconcentrator. Purification of Red Blood Cell Transfetrin Recentor: Chick embryo red blood cells (8-10 mL) were lysed in PBS (pH 7.4), 1 mM PMSF and 0.5% NP-40 (70-90 mL) (19). The lysate was held at 4’C for 30 min, andthen centrifuged at 800 XG for 10 min at 4’C. The supematantwas centrifuged at 100,000XG for 60 min in a BeckmanType-30 rotor at 4’C. The resultant supematantwas loadedon an ovotransferrin Aft?Gel column (2 mL) and the column waswashed with 100 mL of lysis buffer. The column wassequentiallywashedwith 25 mL of sodiumacetate (0.15 M, pH 5.0), 1 mM PMSF, 0.1% NP-40, and 50 mL sodiumcitrate (0.1 M, pH 5.0), 1 mM PMSF, 0.1% NP-40 and 50 u&nL desfetrioxamine. The transfen-inreceptor was eIuted with 15 mL of NH4HC03 (0.1 M, pH 8.0), 1 mM PMSF, 0.1% NP-40 and 50 pg/mL desferrioxamine. The eluatewasconcentratedto 500 l.tL in a Centricon microconcentrator. Purification of the red blood cell receptoryielded a mixture of dimeric and monomeric specieswhereaspurification of the oviduct receptor primarily yielded dimers. ElectronhoresisandWestern Blotting: Polyacrylamide gel electrophoresis(SDS-PAGE) was conductedin a slabgel apparatusasdescribedby Laemmli (20). The proteinswere transblotted onto nitrocellulose(Whatman) asdescribedby Towbin a al, (21). Blocking and antibody treatmentswere performedasdescribed(22). The antigen-antibodycomplexesweredetected using [I25I]protein G and autoradiographyemploying Kodak AR X-Omat film and Dupont 27
Vol. 171, No. 1, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Cronex intensifying screensat -7O’C (23). Molecular weight standardsincludedrabbit muscle myosin (200 KDa), E. coli P-galactosidase (116 KDa), musclephosphorylaseb (97 KDa), bovine serumalbumin (66 KDa), ovalbumin (44 KDa), carbonic anhydrase(31 KDa), soybeantrypsin inhibitor (24.4 KDa) and lysozyme (14.4 KDa). ImmunologicalCrossReactivitv Exneriments: The rabbit polyclonal antiserumagainstthe oviduct transferin receptor hasbeendescribedin an earlierreport (12). Antiserum againstthe chick embryo red blood cells wasproducedin rabbits by subcutaneous iniection of 15 ug of receptor in Freund’scomplete adjuvant. Serumwasobtainedafter 5 boosterinjections of 15pg of antigenin incompleteadjuvant at - 10 day intervals. The serafrom multiple bleedingswaspooled and precipitated with 45% ammoniumsulfate. The pellet was dialyzed againstPBS. The antibody preparationwastreated by multiple passesover an ovotransferrin Affi-Gel column until an ELISA assayrevealed the completeabsenceof ovotransferrin antibodies. The antiserawere usedto probe Western blots of the oviduct and red blood cell receptors. The purified transferrinreceptorsfrom embryo red blood cells and oviduct 1251using Iodobeads(Pierce) accordingto the manufacturer’s instructions. The radiolabelledreceptorswere separatedfrom free 1251-by chromatography through a column of SephadexG-50 (12 cm X 1 cm), equilibrated with Tris-HCl buffer (0.06 M, pH 6.8), 1 mM PMSF, 0.1% NP-40. The radiolabelledreceptorswere further purified by SDSPAGE on 9% gelsin the presenceof 5% p-mercaptoethanol. Portionsof the gelscorrespondingto dimeric or monomerictransferrinreceptorswere excisedwith a razor bladeand electrophoretically eluted into a dialysis bag. The radiolabelledtransferrinreceptorswere digestedwith varying amountsof S. aureusV-8 proteaseat 37°C in 0.125 M Tris-HCl (pH 6.8), and 1% SDS. The partially hydrolyzed receptorswere subjectedto SDS-PAGE on gradientgels. The gels were stainedbriefly with Coomassiebrilliant blue to visualize the protein standards,destained,dried and exposedto Kodak X-Omat AR film usingDuPont Cronex intensifying screensat -7OOC.
Previous resultsfrom this laboratory demonstratedthat an estrogen-inducible91 KDa chicken oviduct glycoprotein exhibited propertiesexpectedof a transferrin receptor (12). The overall propertiesof the 91 KDa protein are similar to thoseof transferrinreceptorsidentified in chicken erythroblaststransformedwith avian erythroblastosisvirus and on chick embryo red blood cells (13,14). In order to determinethe relatednessof the red blood cell and oviduct transferrin receptors,the following immunologicand peptidemappingexperimentswere undertaken. Purified receptorsfrom henoviduct membranesand embryonic red blood cells were probed by Western blotting usingspecific polyclonal antisera. As shownin figure lA, an antiserum directed againstthe oviduct transferrin receptordoesnot crossreact with the red blood cell transfertin receptor. Similarly, the anti-embryonicred blood cell transfer-r-in receptor antiserum doesnot show any crossreactivity with the oviduct transferrin receptor (Fig. 1B). The conditions employedfor probing the blots were chosento enhancethe possibility of detecting crossreactivity. For example, the antiserumdirected againstthe oviduct receptorwas usedat a dilution of 1:750, althoughthe antiserumwill detect the receptor at ng levelsat a 1:2OOCl dilution on Westernblots of oviduct membranes.In additionalexperiments(not shown)the antiserumdirected againstthe oviduct receptordid not cross-reactin an ELISA assaywith the red blood cell receptor nor did the antiserumdirected againstthe red blood cell receptorcross-reactwith oviduct membranes.The completelack of cross-reactivity by the antibodiesagainstthe heterologousreceptorsdemonstrates that the two receptorshave distinct epitopes. Theseresultssuggestedto us that the two receptors might be distinct molecules. 28
171, Na. 1, 1990
-205k c116k c97k
Lack of Immunolo&4 reactivitv between oviduct and embryonic red blood cell msferrtn rew The intact, unlabelled receptors (10 wg) were subjected to 12% SDS-PAGE under nonreducing conditions, uansblotted and probed with antisera and 1251-protein G as described in Materials and Methods. Panel A: The receptors were probed with a 1:750 dilution of anti-oviduct transfer& receptor (anti-91 KDa) antibodies. Lane 1, embryonic red blood cell receptor; Lane 2, oviduct receptor. Panel B; The receptors were probed with a 1:200 dilution of anti-embryonic red cell transfenin receptor antibodies. Lane 1, oviduct receptor, Lane 2, embryonic red cell receptor (note the prevalence of monomers).
of the oviduct
and red blood cell transfertin
further studiedusingpeptide maps. Purified, radiolabelledtransfertin receptorsfrom hen oviducts
and embryonic red blood cells were subjectedto partial proteolytic digestionswith varying amountsof S. M V8 proteaseand the resultantpeptideswere analyzed by SDS-PAGE. Peptidemapsgeneratedfrom dimeric receptorsyielded two distinct patterns(Fig. 2). In eachcase the releasedpeptideswere predominantly under40 KDa. Although the peptidepatternsobtained from the dimeric receptorsareclearly different from eachother the differenceswere confirmed by peptidemap analysisof the respectivemonomers. Becauseneitherof the dimeric receptorscould be completely reducedto monomersby treatmentwith disulfide bondreducing agents,the peptide analyseswere performedon radiolabelledmonomersisolatedby SDS-PAGE andelectroelution. The resultsin figure 3 demonstratethat the monomersalsoyield distinct peptide maps. It is interestingto note that althoughthe patternsaredistinct there are at leastsevenpeptideswhich have 29
116k 97k 66k
3lk 21.5k 14.4k
Pentide mans of transferrin receotors under non-reducing conditions. 1251~labelled dimeric receptors were digested with S. aureus V8 protease for 2 min. as described in Materials and Methods. The digestions were terminated by adding an equal volume of 10% SDS and 2% glycerol. Samples were electrophoresed on 6-18% gradient SDSPAGE. In each panel, lanes l-7 were digested with 0, 0.2,0.4,0.6,0.8, 1 and 1.2 ug of protease respectively. Panel A, oviduct receptor; Panel B, embryonic red blood cell receptor.
similar mobilities. Theseresultssuggestthat althoughthe two transferrinreceptorsaredifferent, they may sharesomelirnited primary peptidestructure,perhapsat the ligand binding site. Although it is clear that transfertin receptorsfrom different speciesdiffer from eachother (4-6) the presenceof distinct transferrin receptorsfrom two different tissuesof the samespecies hasnot previously beenreported. Thus, the humanplacentalandred blood cell receptorswere shownto be identical by peptidemapanalysis(15). In addition, humantransfer-r-in receptorsfrom various tissueculture cell lines seemto be identical despitesomemicroheterogeneity,which is probably due to differencesin post-translationalmodifications (24.25). The existenceof two distinct transfer-r-in receptorsseemsto suggestthat the receptorsmay play different roles in the tissuesin which they are located. In this regardit shouldbe noted that evidencehasaccumulatedfor a uniquerole of uansferrinand the transfenin receptorin the central nervous system(recently reviewed by Espinosade 10sMonteros a &, (26)). The suggestionthat the oviduct and red blood cell receptorsmay have uniquefunctions is further strengthenedby our earlier observationthat the oviduct receptoris regulatedby estrogen(12), whereasthe red blood 30
171, No. 1, 1990
A 1 1 ib
Peotide maps of transfertin recentor monomers. t*sI-labelled monomeric receptors were digested with varying amounts of S. aureus protease in 0.125 M, Tris-HCI (pH 6.8), 1% SDS and 2% P-mercaptoethanol The digestions were terminated by adding an equal volume of 10% SDS, 2% glycerol and 2% g-mercaptoethanol. The products were electrophoresed on 9-18% gradient SDS-PAGE. In each panel, Lanes 1-7, were digested with 0,0.2,0.4,0.6,0.8, 1 and 1.2 ug of protease respectively. Panel A, embryonic red blood cell receptor monomer; Panel B, oviduct receptor monomer.
cell receptor is regulatedby iron (7). In either case,the uptakeof iron-laden transferrin from the serumis presumedto be the major function, although this point remainsto be firmly established for oviduct cells. The well-known increasedrequirementfor iron by proliferating cellsprobably accountsfor the apparentlygreaterincorporationof radioactive precursorsinto the 91 KDa protein observedby DeRosaand Lucasat earlier stagesof estrogentreatmentof chicks (18). At later stagesof estrogenizationthe initial burst of cell proliferation haspassedand the receptor presumablyfunctions only to maintain iron homeostasis. Although evidence from a limited numberof mammalianorganismssuggeststhat the transfertin receptorfrom different tissuesis identical there are severalother receptorsthat exhibit structurally distinct molecular forms. Thus, the atrial natriuretic factor receptor showsdistinct forms in rat smoothmuscle,kidney andLeydig cells (27). Similarly, two unique forms of the tumor necrosisfactor receptor have beenobservedin humancell lines (28). Presumablythese distinct receptor moleculesplay individualized rolestailored to the particular cell’sneeds. Additional studiesare underway to further define the uniquestructural and functional propertiesof the oviduct and red blood cell transferrinreceptorsand to elucidateany cell-specific roles for thesevital molecules. ACKNOWLEDGMENTS This work was supportedby grantsHDO9046(JJL) andDK/ML31729 from the National Institutes of Health to Dr. R.C. Woodworth who supportedABM. We thank Dr. 31
171, No. 1, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Caroline Enns, SyracuseUniversity for numeroushelpful discussionsand Debra Knapp for her excellent typing of the manuscript. REFERENCES
t-16. ii: 13. 14. 15. 16. 17. 18. 19. 2 22. 2 25. 26. 27. 28.
Heubers, H. and Finch, C.A. (1987) Physiol. Rev. 67, 520-581. Trowbridge, I.S., Newman, R.A., Domingo, D.L. and Sauvage,C. (1984) B&hem. Pharmacol. 33, 925-932. Kuhn, L.C. (1989) Schweiz. Med. Wschr. 119, 1319-1326. Stearne, P.A., Pietersz, G.A. and Goding, J.W. (1985) J. Immunol. 134, 3474-3479. Roberts, K.P. and Griswold, M.D. (1990) Mol. Endocrin. 4531-542. Trowbridge, I.S., Lesley,.J. and Schulte, R. (1982) J. Cell Physiol. 112, 403-410. Matia, E., Rao, K., Shapiro, D.S., Sussman,H.H. and Klausner, R.D. (1984) J. Biol. Chem. 259, 2689-2692. Klausner, R.D. and Harford, J.B. (1989) Science 246, 870-872. Mullner, E.W. and Kuhn, L.C. (1988) Cell 53, 815825. Davis, R.J., Faucher, M., Racamiello, L.K., Carruthers, A. and Czech, M. (1987). J. Biol. Chem. 262, 13126-13134. Tanner, L.I. and Lienhard, G.E. (1987) J. Biol. Chem. 262, 8975-8980. Poola, I. and Lucas, J.J. (1988). J. Biol. Chem. 263, 19137-19146. Schmidt, J.A., Marshall, J. and Hayman, M.J. (1985) Biochem. J. 232, 735-741. Schmidt, J.A., Marshall, J., Hayman, M.J. and Beug. H. (1986) Exp. Cell Res. 164, 71-78. Enns, C.A. and Sussman,H.H. (1981) J. Biol. Chem. 256, 12620-12623. Graham, G. and Bates, G.W. (1976). J. Lab. Clin. Med. 88, 477-486. Mason, A.B. and Brown, S.A. (1987) B&hem. J. 247, 417-425. DeRosa, P.A. and Lucas, J.J. (1982). J. Biol. Chem. 257, 1017-1024. VanDriel, I.R., Stearne, P.A., Grego, B., Simpson, R.J. and Goding, J.W. (1984). J. Immunol. 133, 3220-3224. Laemmli, U.K. (1970). Nature (Lond) 227, 680-685. Towbin, H., Staehlin, T. and Gordon, J. (1979). Proc. Nat. Acad. Sci. USA 76, 43504354. Hawkes, R., Niday, E. and Gordon, J. (1982). Anal. Biochem. 119, 142-147. Young, R.A. and Davis, R.W. (1983). Science 222, 778-782. Panaccio, M., Zalcberg, J.R., Thompson, C.H., Leyden, M.J., Sullivan, J.R., Lichtenstein, M. and McKenzie, I.F. (1987). Immunol. Cell Biol. 65,461-472. Nikinmaa, B., Enns, C.A., Tonik, S.E., Sussman,H.H. and Schroder, J. (1984). Stand. J. Immunol. 20, 441-447. Espinosade 10sMonteros, A., Pena,L.A. and de Vellis, 1. (1989) J. Neurosci. Res.24, 125-136. Pandey, K.N., Pavlou, S.N. and Inagami, T. (1988). J. Biol. Chem. 263, 1340613413. Brockhaus, M., Schoenfeld, H.J., Schlaeger,E.J., Hunziker, W., Lesslauer,W. and Loetscher, H. (1990). Proc. Natl. Acad. Sci. USA 87, 3127-3131.