Identification of a region within the cytoplasmic domain of the ...

THEJOURNAL OF B I O ~ I C CHEMISTRY AL 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269,No. 29,Issue of July 22,pp. 19014-19020, 1994 Printed in U.S.A.

Identification of a Region within the Cytoplasmic Domain of the Interleukin-6 (IL-6) Signal Transducer a 1 3 0 Important for Ligand-induced Endocytosisof the IL-6 Receptor* (Received for publication, February 18, 1994,and in revised form, May 3, 1994)

Elke DittrichS, Stefan Rose-John$,Claudia GerhartzS, Jiirgen MullbergS, Tanja StoyanS, Kiyoshi YasukawaO, Peter C. HeinrichS, and Lutz Graevem From the Unstitute of Biochemistry, Rheinisch-Westfalische Technische Hochschule Aachen, 52057 Aachen, Germany and the §Biotechnology Research Laboratory, Tosoh Corporation, 2743-1 Hayakawa, Ayase, Kanagawa 252, Japan

Interleukin-6(IL-6)exerts its action via a cell surface receptor complex consisting of two subunits, the IL-6 receptor and the signal transducer gp130. We have studied the role of both transmembrane proteins for IL-6 internalization and ligand-induced down-regulation of cell surface receptors. Co-expression of wild-type and mutant forms of both the IL-6 receptor and gp130 in transiently transfected COS-7 cells revealed that gp130 is essential for efficient endocytosis and receptor downregulation. Whereas the cytoplasmic domain ofthe IL-6 receptor is not significantly involved in the internalization process, deletion of the corresponding domain of gp130 resulted in an almost complete loss of the ability to endocytose IL-6. Mutants with different truncations within the intracellular domain of gp130 revealed that a 10-amino acid sequence TQPLLDSEER is crucial for efficient internalization. Since this sequence contains a putative di-leucine internalization motif,we suggest that a di-leucine motif directs the receptor mediated endocytosis of the IL-6 receptor complex.

binding of APRF to responsive elements in acute phase genes (11).APRF as well as the tyrosine kinase JAK 1associate with the signal transducer gp130 (12). After binding t o its receptor IL-6 is rapidly internalized and degraded by rat hepatocytes and human hepatoma cells HepG2 (13, 14). In HepG2 cells this internalization leads to a loss of IL-6 binding sites at the cell surface indicatingthat the IL-6R is down-regulated by itsligand (14). This down-regulation might play an importantrole as a protection against overstimulation. For differentreceptor transmembraneproteins, e.g. the transferrin, thelow density lipoprotein and theasialoglycoprotein receptor, it has been shown that specific sequences within the cytoplasmic domain are importantfor efficient internalization (for a recent review see Ref. 15). Substitution of single amino acids within such sequence motifs and structural predictions impliedthat an important characteristicof such motifs is the formation of a “tight p-turn” (16, 17). This structure might be recognized by adaptor proteins which mediate the accumulation of these receptors in clathrin-coated pits (15). Recently, quite different sequence motifs important for interInterleukin-6 (IL-6)’ is a multifunctional cytokine synthe- nalization andlysosomal targeting were found in the intracelsized by many different cells after appropriate stimulation. It lular domains of the y- and 6-chain of the T-cell receptor complex, the cation-dependent mannose 6-phosphate receptor, the acts on a wide spectrum of target cells and exerts multiple cation-independent mannose 6-phosphate/insulin-like growth functions during the immuneresponse, hematopoiesis, neural factor I1 receptor, and the interferony-receptor (18-21). These differentiation, and the acute phasereaction (1-5). IL-6 acts via a cell surface receptor complex composed of two motifs are characterized by a leucine-leucine or a leucine-isosubunits: an 80-kDa IL-6 binding protein(IL-6 receptor) anda leucine sequence. To better understand the mechanism of ligand-induced in130-kDa signal transducing protein, gp130 (6-8). Binding of IL-6 to theIL-6 receptor (IL-6R)induces thehomodimerization ternalization and down-regulation of the IL-6R, we searched of two gp130 molecules thereby transducing thesignal into the for possible internalization sequences in the cytoplasmic docell (9). In the277-amino acid cytoplasmic domain of gp130, a mains of both, the IL-6R and the signal transducergp130. Here we demonstrate, for the first time, that the signal 61-amino acid juxta-membrane region was found to be suffitransducer gp130 is essential for efficient internalization of cient for mediating a proliferation signalintransfected within thecytoplasmic tail of BAFB03 cells (10). This region contains two short segments IL-6. A 10-amino acid long region which are conserved between different members of the hema- gp130 was identified as a short sequence which contains a topoietic cytokine receptor family. Recently, it was found that possible di-leucine internalization signal motif. By contrast, the transcription factor APRF is rapidly phosphorylated in a the cytoplasmic domain of the IL-6R plays only a minor role for number of cells upon IL-6 stimulation. This activation induces ligand internalization. EXPERIMENTALPROCEDURES * This work was supported by grants from the Deutsche Forschungscalf intestinal phosphatase, T4Materials-Restrictionenzymes, gemeinschaft and the Fonds der Chemischen Industrie. The costs of DNA ligase, and protease inhibitors were purchased from Boehringer publication of this article weredefrayed in part by the payment of page Mannheim.Tran[35S]-label (44 TBq/mmol) was obtained from ICN charges. Thisarticle must thereforebe hereby marked “aduertisernent” (Meckenheim, Germany). DMEM was from Life Technologies, Inc. Rein accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondenceshould be addressed:Institute of Biochem- combinant human IL-6 was prepared as described by Arcone et al. (22). istry, RWTH Aachen, Pauwelsstr. 30,52057Aachen, Federal Republic of The specific activity was 1.5 x lo6 B-cell stimulatory factor-2unitdmg of protein. IL-6 which was used for iodination was kindly provided by Germany. Tel.: 241-8088837;Fax: 241-8888428. T. Kishimoto and T. Hirano (Osaka, Japan), specific activity was 5 X ’The abbreviations used are: IL-6, interleukin-6; IL-6R, interleu- Dr. lo6 B-cell stimulatory factor2 unitdmg of protein. The IL-6R-cDNA kin-6 receptor; APRF, acute phase response factor; PBS, phosphatebuffered saline; DMEM, Dulbecco’s modified Eagle’s medium; PAGE; was isolated as described elsewhere (23). The polyclonal monospecific antiserum against the IL-6R was prepared by injecting the extracellupolyacrylamide gel electrophoresis.

19014

Signal Dansducer IL-6-R

WT

1:-

gp130 and IL-6 Internalization

82

A 82

GP130

19015

200 -

200

100.

100 -

-

WT

A105 A126

.

69

-

46.

46

-

69

A136 A251 n

c

A 275

:--1

FIG.1. Scheme of the structure of wild-type IL-6R and gp130 and the different deletion mutants. The extracellular domain is indicated by dotted lines. The black box represents the transmembrane domain. The cytoplasmic domain is shown as a gray box. Numbering of the cytoplasmic amino acids started after the transmembrane domain. The thick and thin bars within the cytoplasmic tail are locations of putative internalization signal sequences. Thick, black bar, tyrosinecontaining motif; thin, black burs, di-leucine motifs. lar domain of the IL-6R expressed in E. coli into rabbits. The generation and characterization of monoclonal antibodies against human gp130 are described in (24). Rhodamine conjugated anti mouse and antirabbit antibody was obtained from Dakopats (Hamburg, Germany). Restriction enzymes NotI, BsaI, and AccI were obtained from New England Biolabs (Schwalbach, Germany). Exonuclease I11 was purchased from Applied Genetechnology Systems (Heidelberg, Germany). N-Succinimidyl-3-[4-hydroxy-5-'2"Iliodophenylpropionate(BoltonHunter reagent; 74 TBq/mmol) was purchased from Amersham International (Amersham, UK). A frequently used buffer was PBS (137 mM NaCl, 2.7 mM KCl, 10.1 mM Na,HPO,, 1.76 mM KH,PO,, pH 7.2). Cell Culture-COS-7 cells (ATCC CRL 1651) were grown in DMEM a t 5%C02 in a watersaturated atmosphere. All cell culture media were supplemented with 10% fetal calf serum (Seromed, Berlin, Germany), streptomycin (100 mditer), and penicillin (60 mg/liter). Expression VectorsStandard cloning procedures were performed as outlined by Sambrook et al. (25). The expression vector pCDM8-IL-6R was constructed by inserting the 1.4-kb Sal1 fragment, containing the entire region of the IL6R excised from the vector pExIRl (23) into the XhoI site of the expression vector pCDM8 (26). The expression vector pSVLgpl30 was kindly provided by Dr. Tetsuya Taga (Osaka, University). TFuncation Mutants-The IL6R mutant (IL6RA82) was generated by an exonuclease I11 digest. First, a linker 5'-GGCCGCTGAATTCAGGCCGGGCCCATAGCTAGTTAGAGC-3',containing one EcoRI site, one ApaI site,and threestop codons, was introduced into the NotI site of the polylinker of the pCDM8 vector. Digestion of the plasmid with ApaI resulted in a vector that was protected from exonuclease I11by an 3"overhang. The plasmid was then sequentially digested with EcoRI, exonuclease 111, and S1 nuclease a s described previously (27). Sequence analysis revealed that thenew 3'-end of the insert abuts theprotected sequence from the linker and that theregister is such that this introduces a termination codon two amino acids after the cDNA sequence. The gp130 deletion mutants lacking different parts of the cytoplasmic domain were obtained by digesting the gp130 cDNA with different single cutting enzymes. The mutant gp130A275 was generated by a BsaI digest, the mutant gp130A251 by a BstEII digest, the mutant gp130A136 byan AccI digest, the mutantgp130A126 by a n Em1 digest and the mutant gp130A105 by a ScaI digest. After ligation with a XbaI-linker 5"TAGTCTAGACTA-3' or 5'-ATAGTCTAGACTA-3' containing two stop codons, the blunt end fragments were cloned into the expression vector pSvL using the filled XbaI and BamH I sites of the polylinker. Mutants gp130A105 and A251 contain two additional nonauthentic amino acids a t the C terminus, each. Dansfection-Transfection of cells was carried out using the Gene PulserTMfrom Bio-RadLaboratories (Munchen, Germany). 2 x lo6 cells in 0.8 ml of DMEM were co-transfected with 1.3 pg of pCDM8 vector containing the IL6R wild-type or mutant cDNA plus 28.7 pg of pSVL vector containing gp130 wild-type or mutant cDNAs (molar ratio 1:16)

FIG.2. Expression of wild-type proteins and deletion mutants in COS-7 cells. COS-7 cells (2 x lo6) were transfected with (A) the IL6R cDNA or the cDNA coding for a deletion mutant 882or with ( B ) the gp130 cDNA or cDNAs coding for different deletion mutants (see Fig. 1). Cells were metabolically labeled with 150 pCi of Tran[35Sl-label in methioninekysteine-free medium for 3 h andchased for 1h in normal medium. Proteins were immunoprecipitated from cell lysates with a polyclonal antiserum against the IL-6R and gp130, respectively. The precipitated proteins were separated by SDS-PAGE and visualized by fluorography. CO, COS-7 cells transfected with either pCDM8 (in A) or pSVL (in B ) control vectors only. using a voltage of 230 V and a capacity of 960 pF. Three days after transfection cells were used for further studies. Pulse-Chase Experiments-Cells were metabolically labeled with [35Slmethionine/cysteinefor 3 h and chased for 30 min in normal medium. Cell lysis was performed in 10 mM "is-HCI, pH 7.4,66 mM EDTA, 1% Nonidet P-40, and 0.4% sodium deoxycholate in the presence of proteinase inhibitors (2 pg/ml aprotinin, 0.5 pg/ml leupeptin; 0.7 pdml pepstatin, 35 pg/ml phenylmethylsulfonyl fluoride). Supernatants and lysates were pretreated with Pansorbinn' (Calbiochem-Novabiochem Corp., La Jolla, Ca). SDS was added to a final concentration of 0.3%. After incubation with the appropriate antisera for 2 h a t 4 "C the immune complexes were precipitated with protein A-Sepharose preincubated with cell lysate of nonlabeled cells. Proteins were separated on 10% SDS-polyacrylamide gels (28) and visualized by fluorography (29). Immunofluorescence Staining-Approximately lo5 COS-7 cells grown on coverslips for 72 h were fixed with 2% paraformaldehyde as described elsewhere (30). Cells were treated with a V200 dilution of an ILGR-specific rabbit antiserum or a 0130-specific mouse antiserum ( A M 64) for 20 min. Antibodies bound to the cell surface were detected using a 1/200 dilution of rhodamine-conjugated anti-rabbit or antimouse IgGfor20 min. Coverslips were mounted on slides with MowiolT" 4-88 (Calbiochem-Novabiochem Corp.) and analyzed using fluorescence microscopy. A 500-fold magnification was used to photograph the cells. Competition Assay-COS-7 cells were transiently co-transfected with one of the two pCDM8 derived plasmids carrying the wild-type or mutant IL-6R cDNA and with one of the pSVL derived plasmids carrying the wild-type or mutant gp130 cDNAs.Transfected cells were plated in duplicate wells. Three days after transfection, cells were washed twice with ice-cold binding medium (DMEM without bicarbonate containing 0.2% (w/v) bovine serum albumin and 20 mM Hepes buffer, pH 7.0). Then cells were incubated with 100 PM 1z51-IL6for 4 h a t 4 "C in the presence of increasing amounts of nonlabeled I L 6 (0.1-1000 n ~ )After . 4 h medium was removed and cells were washed three times with PBS containing 1mM MgCl,, 0.1 mM CaCI,, and 0.2% bovineserum albumin. Cells were subsequently solubilized in 1ml of 1M NaOH and radioactivity was measured in a y-counter. The binding affinity of I L 6 to the IL6R complex was calculated by nonlinear regression using the computer program InPlot 4.0TM(Graphpad, SanDiego, CA). Internalization Assay--Transfected cells were incubated with 1 n~ lZ5I-IL6for 2 hat 4 "C in thepresence of a 200-fold excessof nonlabeled ligand. Internalization was initiated by warming up the cells to 37 "C. After different times of incubation, cells were set on ice again for 2 h. Cells were washed three times with PBS containing 1mM MgCl,, 0.1 mM CaCI,, and 0.2% bovine serum albumin. Surface-bound 'z51-IL6 was determined after subjecting the cells to 0.5M NaCWCl, pH 1, for 3 min followedby an additional wash with PBS. Internalized 1251-IL6was determined after lysis of the cells in 1 ml of 1 M NaOH.

19016

Signal fiansducer gp130 and IL-6 Internalization

FIG.3. Cell surface expression of wild-type and m u t a n t IL-GR or a 1 3 0 proteins in transfected COS-7 cells. COS-7 cells were transiently transfected with either wild-type or mutant IL-6R or gp130. Cells were seeded on glass coverslips and incubated for 3 days. After paraformaldehyde fixation,IL-6R protein was detected withan IL-6R-specific polyclonal antibody andgp130 protein with the monoclonal antibody AM64. A 500-fold magnification was used to photograph thecells. A, pCDM8 and B , pSVL control vectors; C, wild-type IL-6R D,IG6R deletion mutant A82; E , wild-type gp130; F J ,gp130 deletion mutants A275 ( F ) , A251 ( G ) ,A136 (H), A126 (I),and A105 ( J ) .

Expression vectors for wild-type IL-6R and gp130 or the different mutantswere transiently transfected into COS-7 cells by Construction and Expression of Deletion Mutants for the IL-6Rand the signal nansducergp130-studies with the hu- electroporation and expression was studied three days later. man hepatoma cells HepG2 have shown that binding of IL-6 to Cells were metabolically labeled with [35Slmethionine/cysteine its receptorinduces internalization of ligand and receptor thus for by a l-h chase. lysates were prepared and leading to a down-regulation of IL-6 surface binding sites (14). subjected to an immunoprecipitation with polyclonal antibodTo study therole ofthe signal transducergp130 in this process ies directed against the IL-6R and gP130, respectively. Immuand toidentify internalization signalsequences in thecytoplas- noprecipitates were analyzed by SDS-PAGE and visualized by mic domains of either the IL-6R or gp130, we transiently ex- fluorography (Fig. 2). From COS-7 cells transfected with the pressed i n c o s - 7 cells wild type and mutant proteins with IL-6R cDNA a protein of 82 kDa was immunoprecipitated. Dea different truncations of their cytoplasmic tails. letion of 82 amino acids from the C terminus resulted in For the IL-6R we constructed a deletion mutant in which all protein of approximately 74 kDa. In COS-7 cells which were of the 82 cytoplasmic amino acids were removed. This mutant transfected with the control pCDM8 vector no IL-6R protein was designated IL-6RA82 (Fig. 1).For the signal transducer could be detected (Fig. 2 4 ) . In COS-7 cells transfected with gp130, five different mutants were constructed by deleting 105, either the complete a 1 3 0 cDNA or the different truncated 126, 136, 251 and 275 amino acids of the cytoplasmic domain cDNAs, deletion mutant proteins with decreasing molecular which comprises 277 amino acids (Fig. 1).These mutants were weights were precipitated. The increased mobility of the muconstructed on the basis of convenient restriction sites and the tant proteins corresponded to the decreasing size of the cytolocation of four putative internalization motifs: three of the plasmic tails (Fig. 2B). In cells transfected with the control di-leucine type (Fig. 1, thin bars) and one of the tyrosine type vector pSVL two faint bands of 145/150 kDa were detected. bar). (thick bands These detected alsowere in all gp130 transfectants (Fig. RESULTS

Signal Dansducer gp130 and IL-6 Internalization

e WT/A275 A82/WT

+A82IA275

1

10

100

1000

10000

fold excess of IL-6

-t-

-"--

+

1

10

100

WT/A25 1 WT/A136 WT/A105 WT/A126

1000

10000

fold excess of 11-6 FIG.4. Competition of binding of '"I-IL-6 to the IL-6R complex in the presence of increasingconcentrations of nonlabeledIL-6. Three days after transfection cells were washed twice with ice-cold binding medium. Then cells were incubated with 100 PM "'1-IL-6 for 4 h a t 4 "C in the presence of increasing amounts of nonlabeled IL-6 (0.1-1000 nM). After 4 h medium was removed and cells were washed three times with ice-cold PBS in order to remove all nonbound ligand. Finally, cells were lysed in 1M NaOH, and cell-associated radioactivity was measured in the y-counter. Data are expressed as a percentage of control incubations containing only labeled ligand. Values are themean of two independent experiments.

2B).They possibly represent endogenous gp130. Only 5-10% of COS-7 cells expressed the transiently transfected gene (see below). Therefore, in thosecells which are positive, the expression of the transfectedgene is at least ten times higher than expression of endogenous gp130. The cell surface expression of wild-type and mutantproteins was studied by indirect immunofluorescence. 5 1 0 % of transfected COS-7 cells showed a bright staining of the cell surface (Fig. 3, C J ) . Cells transfected with the control vectors gave only a very weak signal (Fig. 3, A and B ) . When cells were co-transfected with cDNAs for both, the IL-6R and the signal transducer gp130, double immunofluorescence stainingrevealed that all cells positive for one protein also expressed the other protein (data not shown). Binding Affinities of Wild-type and Mutant Receptor Complexes-The IL-6R complex is a two-component system. Since gp130 by itself does not bind IL-6 (7), its role in the internalization process can only be studied in the presence of the IL-6R. For this purpose COS-7 cells were always transfected with the cDNAs for both subunits. A molar ratio of 1:16 (pCDM8-1L-GR:pSVL-gpl30)was used in all transfections. For wild-type proteins this ratio resulted in the expression of a receptor complex with a KD of 6.33 x 10"O M (see below). To demonstrate that IL-6R complexes consisting of either wild-type or mutant IL-6R and gp130, respectively, bind IL-6 with similar affinities, competition studies wereperformed. Cells were incubated with100 PM lz5I-IL-6for 4 h at 4 "C in the presence of increasing amounts (up to 10,000-fold excess) of nonlabeled ligand. As shown in Fig. 4 A , receptor complexes containing both wild-type subunits ( W T / W T )displayed the same affinity as receptor complexes that contain either trun-

19017

cated IL-6R (A82/WT) or truncated gp130 (WTlA275)or both (A82/A275). The same was true for complexes that consist of wild-type IL-6R and different mutants of a 1 3 0 i n which increasing parts of the cytoplasmic domain have been deleted (Fig. 4B). Control cells ( C O )showed no binding of IL-6 (Fig. 4 A , filled triangles). Dissociation constants (K,) and receptor numbers percell (R,)were in the range ofK, = 3.3-6.5 x 10"' M and R, = 12-51 x lo3 as determinedby nonlinear regression. These results indicate that thecytoplasmic domain of both, the IL-6R and the signal transducer gp130, are not important for the formation of high affinity binding sites. Znternalization Studies with Wild-type and Mutant Receptor Complexes Lacking Cytoplasmic Domains-Internalization and ligand-induced down-regulation was first studied in cells expressing the wild-type IL-6R complex (WT/WT). Three days after transfection,cells were incubated with1nM lZ5I-IL-6for 2 h at 4 "C. Temperature wasshifted t o 37 "C for different times as indicated after which the cells were set on ice again for another 2 h. This procedure allowed for the determination of the number of surface binding sites after thedifferent incubation times at 37 "C. Surface-bound IL-6 was eluted by a high salt/low pH wash for 3 min, and internalized IL-6 was determined after solubilization of the cells in 1M NaOH. As shown in Fig. 5 A , WT/WT cells internalize IL-6 efficiently during the6-h time course (filled squares).After 2 h 80% of the IL-6 initially bound was detected intracellularly, and surface receptors were down-regulated by 40% (open squares). At later timepoints the amount of internalized IL-6 remained constant and the number of surface binding sites was 50% of the initial value. Cells transfected only with thecontrol vectors pCDM8/pSVL showed neither binding nor internalization of IL-6 (Fig. 5A, triangles). This result suggests that IL-6R an complex capable of endocytosis and down-regulation isreconstitutedin transfected COS-7 cells. In the cytoplasmic domains of several transmembrane proteins sequence motifs have been found that are important for efficient internalization (31). In order to search for such an internalization signal within the cytoplasmic domain of the IL-6R, we analyzed the deletion mutant A82 in which the comthe plete cytoplasmic tail was removed (Fig. 1). Thismutant when expressed together with wild-type gp130 (A82/WT) displayed similar internalizationkinetics as WT/" cells (Fig. 5B). This result suggests that the cytoplasmic tail of the IL-6R is not essential for internalization. Next, we studied the role of the intracellular domain of the signal transducer gp130. We first deleted 275 amino acids of this domain which left behindonly the two juxta-membraneous amino acids (Fig. 1). Internalization of IL-6 by a receptor complex containing this mutant, and the wild-type IL-6R (WT/ A275) was severely impaired (Fig. 5C). Within the first 30 min only 9% of the surface-bound IL-6 was internalized in these cells compared t o 42% in WT/WT cells. At no time point the intracellular amount of IL-6 in WT/A275 cells was larger than 20% of surface-bound IL-6. Furthermore, essentially no downregulation of IL-6 surface binding sites was observed during the 6-h incubationperiod. When both truncation mutants(A82/ A2751 were co-expressed, internalization and down-regulation were the same as withWT/A275 cells (Fig. 50). These results strongly indicate that the cytoplasmic domain of gp130 is essential for an efficient internalization process. Localization of a Region within the Cytoplasmic Domain of gp130 Important for EficientEndocytosis of ZL-6-To localize the region within the cytoplasmic tail of gp130 that is responsible for the efficient endocytosis of IL-6, deletion mutants of gp130 lacking 105, 126, 136, and 251 amino acids from the C terminus (Fig. 1) were expressed together with the wild-type


Signal Dansducer

19018

A 140

140

- WT/WT

co

'20-

. A82/WT

120-

0

,0

0

60

120

180

time

240

300

,

0

360

time

(min) 140

- WT/A2 7 5 120-

240

l

300

360

(min)

. A82/A275

120-

100"

0

Y m

U #-I

100" 80-

60-

60-

-

0

Y m

80-

40

i

60 180 120

40-

2o 0

0

60

120

180

time

240

300

360

(min)

0

60 180 120

time

240

300

360

(min)

FIG.5. Internalization of surface-bound '=I-IL-6 b y wild-type and mutant IL-6R complexes. Three days after transfection cells were washed twice with ice-coldbinding medium.After incubation with 1 I"of '261-IL-6 for 2 h at 4 "Ctemperature was shifted to 37 "C forthe indicated times. Cells were set on ice again for 2 h and then washed three times with PBS to eliminate nonbound ligand. Surface-bound '=I-IL-6 (open symbols) was determined after subjecting cells to 0.5 M NaCVHC1, pH 1, for 3 min. Internalized lZ6I-IL-6(closed symbols) was measured in the y-counter afterlysis of the cells in 1 M NaOH. Data are expressedas a percentageof initial binding of IL-6 at 4 "C. Values arethe mean of two to four independent experiments.

IL-6R in COS-7 cells. Mutants WTIA105 and WT/A126 both internalized IL-6 as efficiently as WT/WT cells (Fig. 6, A and B ) . During the first 30 min 47 and 43%, respectively, of initially bound IL-6 was accumulated within these cells. The numberof surface binding sites was reduced by 50-60% within 2 h (Fig. 6, A and B ) . However, a further deletion of only 10 amino acids (WT/A136) resulted in an almostcomplete loss of the ability to efficiently endocytose IL-6 and to down-regulate IL-6Rs (Fig. 6C). Within the first 30 min only 11%of initially bound IL-6 was internalizedby this mutant. MutantWT/A251 which lacks all but 26 amino acids of the intracellular domain of gp130 surprisingly showed a better internalization of IL-6 than WT/ A275 and WT/A136 cells (19% of initially bound IL-6 within 30 min). Nevertheless, compared t o WT/WT cells, endocytosiswas still impaired, and down-regulation was weak (Fig. 6D).

very inefficiently (data not shown). Co-expression of gp130, however, resulted in an increased bindingaffinity of about KD = 6.3 x 10"O M. In these cells, 42% of initially bound IL-6 was internalized within 30 min (Fig. 5). Deletion of the cytoplasmic domain of gp130 did not affect binding affinity (Fig. 4) but reduced internalization efficiency by 79%. These observations signal transducstrongly suggest that not the binding but the ing component of the IL-6R complex mediates theefficient uptake of IL-6. The binding affinities measured for the reconstitutedreceptor complex in COS-7 cells (K, = 3.3-6.3 x 10"O M) are by one order of magnitude lower than the highaffinity binding measured in human hepatomacells HepG2 (K, = 6 x lo-" M) which naturally express 1,000-2,000 receptordcell (14). The reason for this discrepancy is not clear. In a recent report a similar finding was described by Gearing and co-workers (32). They DISCUSSION M whenthey co-expressed in found an aftinity of 3.3 x This study demonstratesthat the signal transducer a 1 3 0 is COS-7 the IL-6R together with a cDNA coding for a gp130 molecule missing the C-terminal209 amino acids. The failure essential for efficient internalization of IL-6 and for ligandinduced down-regulation of the IL-6R. In support of this con- t o completely reconstitute high affinity IL-6R complexes in deletion of the cytoplasmic domain of COS-7 cells might imply that a third component is needed for clusion is the finding that gp130, but notof the IL-6R, severely reducedthe endocytosis of full affinity-conversion. In thecase of the IL-2R system such a third component was recently identified and found to be essenIL-6 and the concomitant loss of surface binding sites. The IL-6R when expressed alone in COS-7 cells bound IL-6 only tial for high affinity complex formation and internalization M and internalized the ligand (33). with a low affinity of K, = 1x

19019

Signal Dansducer gp130 andIL-6 Internalization

B 140-

1 40

120-

120-

. WT/A126

. WT/Al05

g a -

10080-

zE

-

40 60

0

Y

20-

0

.

0

,

.

,

.

24060180 120

time

,

.

,

.

[

.

I

.-cm, .-E .-

10080-

6040-

u0

20-

$

0'

360 300

0

.

I

,

,

24060180 120

(min)

.

I

time

.

.

I

I

360 300

.

I

(min)

D 1 40

. WT/A136

140

-

120-

a

u

.-g .C n -m .-c, .-c .-

100-

100-

a

TI

80-

6040I I

rc

20-

0

. WT/A25 1

120-

0

24060180 120

time

300

360

(min)

80-

-

6040-

I

20-

.

.

0' I 1 0 24060180 120

.

time

I

. ,

.

I

360 300

.

I

(min)

FIG.6. Internalization of surface bound 1261-IL-6 by IL-6R complexes composedof wild-type IL-6R and mutantgp130. Three days after transfection cells were washed twice with ice-cold binding medium. After incubation with 1 nM of lZ5I-IL-6for 2 h at 4 "C temperature was shifted to 37 "C for the indicated times. Cells were set on ice again for 2 h and then washed three times with PBS to eliminate nonbound ligand. Surface-bound '261-IL-6 (open symbols) was determined after subjecting cells to 0.5 M NaCVHCl, pH 1, for 3 min. Internalized lZ5I-IL-6(closed symbols) was measured in the y-counter after lysis of the cells in 1 M NaOH. Data are expressed as a percentage of initial binding of IL-6 at 4 "C. Values are the mean of two to four independent experiments.

HepG2 cells that overexpress the IL-6R after stable transfection, display binding sites withtwo affinities, a high affinity of KD= 2 x lo-" M and a low affinity of KD= 5 x 10"' M. The low affinity is comparable to the affinity measured in transfected COS-7 cells. When the internalization and down-regulation of high and low affinity receptor complexes in HepG2 cells was studied, no significant difference was found (14). This indicates that theprocess of internalization can be studied inCOS-7 cells despite the lower affinity of the IL-6R complex for IL-6. The cytoplasmic domain of the IL-6R does not seem to be crucially involved in IL-6 internalization. The mutant lacking the complete cytoplasmic domain (A821when co-expressed with gp130 internalized IL-6 as efficiently as the full-length protein (Fig. 5 ) . Whenexpressed alone, however, internalization by mutant A82 or by wild-type IL-6R was very inefficient (data not shown). Recently, it was shown that thecytoplasmic tail of the IL-6R is also dispensablefor transducing a mitogenic signal (8). Furthermore, soluble IL-6Rs which lack both the transmembrane and the cytoplasmic domain were demonstrated to act agonistically on cells expressing gp130 when IL-6 was present (7, 8, 34). These data suggest that the cytoplasmic domain of the IL-6R has neither a function for internalization nor for signal transduction. However, we recently demonstrated that the humanIL-6R is constitutively phosphorylated and that this phosphorylation can be stimulated by 4p-phorbol 12-myristate 13-acetate (35).Although the phosphorylation sitewas not identified, it is probably localized within the cytoplasmic portion of the receptor. The significance of the phosphorylation of the IL-6R is currently unknown. The study of several mutants of gp130, in which increasing parts of the intracellular domain were deleted, resulted in the

identification of a 10-amino acid region (amino acids 142-151; Fig. 1)that is crucial for the efficient internalization of IL-6. This is the first time that a function has been assigned to this part of the signal transducer gp130. Murakami andco-workers (9) recently found that the 61-amino acid juxta-membraneous region of gp130 is sufficient for generating a growth signal in BAFB03 cells. From our studies it is very likely that such a protein would not mediate efficient internalization. This indicates that internalization not is a prerequisite for signal transduction. We conclude that two different regions of the gp130 molecule are involved in the signal transduction and the endocytotic process, respectively. The region spanning aminoacids 142-151 has the sequence TQPLLDSEER. This sequence does not contain the well defined tyrosine containing internalization signal which was described for many constitutively internalizing receptors such as the low density lipoprotein receptor, transferrin receptor, or asialoglycoprotein receptor (31). These sequencemotifs are believed t o form a so-called tight p-turn which is recognized by adaptor proteins of the coated pit complex (16, 17). There is a putative tyrosine containing internalization sequence in the cytoplasmic tail of gp130 from amino acid 118-121, YSTV (Fig. 1).However, this sequence by itself is not sufficient for mediating efficient internalization (Fig. 6C, WTIA136). Recently, another sequence motif has been described for a number of transmembrane proteins, among others the y- and &chain of the T-cell receptor, the cation-dependent mannose 6-phosphate receptor, the cation-independent mannose 6-phosphatehsulin-like growthfactor I1 receptor, andtheinterferon-y receptor (18-21). This motif is characterized by two adjacent leucines or a leucine-isoleucine. This sequence was

19020

Signal Dansducer


described as a lysosomal targeting signal (18,20).However, in for oligonucleotide synthesis and for the design of Fig. 1, and Marcel some receptor systems it also functions as an internalization Robbertz for most skillful help withthe artwork.We are grateful toDr. signal (19).For the transferrinreceptor it was reportedthat a Tetsuya Taga for providing the expression vector pSVL-gpl30. di-leucine motif can replace the tyrosine-containing internalREFERENCES ization sequence (15). The 10-amino acid region identified in 1. Kishimoto, T. (1989) Blood 74, 1-10 the cytoplasmic tail of gp13Owhich is crucial forinternalization 2. Hirano, T., and Kishimoto, T. (1990)Handb. Exp. Pharmacol. 95-1, 633-665 3. Heinrich, P. C., Castell, J . V., and Andus, T. (1990) Biochem. J . 265, 621-636 of IL-6 and down-regulation of the IL-6R does contain a di4. Sehgal, P. B., Grieninger, G., and Tosata, G. (1990) Ann. N. E Acad. Sci. 557, leucine motif. This strongly suggeststhat this di-leucine motif 1-583 5. Van Snick, J. (1990) Annu. Reu. Immunol. 8, 253-278 is recognized in COS-7 cells as a n internalization signal. Fu6. Yamasaki, K , Taga, T., Hirata, Y., Yawata, H., Kawanishi, Y., Seed, B., Tanture studies using site-directed mutagenesis will be necessary iguchi, T., Hirano, T., and Kishimoto, T. (1988) Science 241,825-828 t o corroborate this assumption. The identification of the di7. Taga, T., Hibi, M., Hirata, Y., Yamasaki, K., Yasukawa, K., Matsuda, T., Hirano, T., and Kishimoto, T. (1989) Cell 58,573-581 leucine motif as an internalization signal sequence in gp130 8. Hibi, M., Murakami, M., Saito, M., Hirano, T., Taga, T., and Kishimoto, T. would imply that receptorsbelonging to the hematopoietic re(1990) Cell 63, 1149-1157 9. Murakami, M., Hibi, M., Nakagawa, N., Nakagawa, T., Yasukawa, K , Yamanceptor family are also endocytosed via “coated pits.” ishi, K., Taga, T., and Kishimoto, T. (1993) Science 260, 1808-1810 Interestingly, mutant A251, although lacking the 10-amino 10. Murakami, M., Narazaki, M., Hibi, M., Yawata, H., Yasukawa, K., Hamaguchi, acid di-leucine motif, was more efficient in mediating the inM., Taga, T., and Kishimoto, T. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 11349-11353 ternalization of IL-6 than mutants A136 and A275, respec11. Wegenka, U. M., Buschmann, J., Lutticken, C., Heinrich, P. C., and Horn, F. tively. At the moment we can only speculate that within the (1993) Mol. Cell. Biol. 13, 276288 26-amino acid cytoplasmic tail of this mutant a second weak 12. Liitticken, C., Wegenka, U. M., Yuan, J., Buschmann, J., Schindler, C.,Ziemiecki, A,, Harpur,A. G., Wilks, A. F., Yasukawa, K., Taga, T., Kishimoto, internalization motif is exposed. This signal might not be recT., Barbieri, G., Pellegrini, S., Sendtner, M., Heinrich, P. C., and Horn, F. ognized in receptor forms with a longer cytoplasmic domain, (1994) Science 263,89-92 13. Nesbitt, J. E., and Fuller, G. M. (1992) J . Biol. Chem. 267, 5739-5742 e.g. due to steric hindrance. 14. Zohlnhoefer, D., Graeve, L., RoseJohn, S., Schooltink, H., Dittrich, E., and A model can be envisioned for two-component receptor sysHeinrich, P.C. (1992) FEBS Lett. 306,219-222 tems in which one subunit is specific for binding the ligand, 15. Trowbridge, I. S., Collawn, J. F., and Hopkins, C. R. (1993) Annu. Reu. Cell Biol. 9: 129-161 whereas the second subunit is necessary for efficient internal- 16. Collawn, J. F., Stangel, M., Kuhn, L. A,, Esekogwu, V., Jing, S., Trowbridge, I. ization. For our receptor system this would mean that the S.. and Tainer. J . A. (1990) Cell 63. 1061-1072 IL-6R when expressed alone internalizes slowly and recycles to 17. Ban& A., and Gierasch,L. M. (199i) Cell 67, 1195-1201 18. Chen, W. S., Lazar, C. S., Lund, K.A., Welsh, J. B., Chang, C. P., Walton, G. M., the cell surface. In contrast, after association of the IL-6R with Der, C. J., Wiley, H. S., Gill, G. N., and Rosenfeld, M.G. (1989) Cell 59, 33-43 gp130 internalization is more efficient, and IL-6R complexes 19. Letourneur, F., and Klausner, R.D. (1992) Cell 69, 1143-1157 are targeted into thelysosomal pathway. This would lead t o a 20. Johnson, K. F., and Kornfeld, S. (1992) J. Cell Biol. 119, 249-257 net loss of surface binding sites. Such a model is consistent with 21. Farrar, M. A,, and Schreiber, R. D. (1993)Annu.Reu. Zmmunol. 11, 571-611 22. Arcone, R.,Pucci, P., Zappacosta, F., Fontaine, V., Malorni, A., Marino, G., and our findings in COS-7 cells. Ciliberto, G. (1991) Eur: J . Biochem. 108, 541-547 Currently, it is not clear if signal transduction is necessary 23. Schooltink, H., Stoyan, T., Lenz, D., Schmitz, H., Hirano, T., Kishimoto, T., Heinrich, P. C., and Rose-John, S. (1991) Biochem. J. 277,659-664 for the down-regulation of the IL-6R. Interestingly, gp130 is not T., Taga, T., Miki, D., Futatsugi, K , Yawata, H., Kishimoto, T., and only involved in the signal transduction of IL-6 but also of 24. Saito, Yasukawa, K. (1993) J . Immunol. Methods 163, 217-223 leukemia inhibitory factor, oncostatin M, ciliary neurotrophic 25. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, factor, and interleukin-11 (32, 36, 37). In all of these receptor NY complexes additional specific subunits bind the respective li26. Seed, B. (1987) Nature 329,84W42 gand. This raises the question whetherall of these complexes 27. Henikoff, S. (1987) Methods Enzymol. 155, 156165 28. Laemmli, U.K. (1970)Nature 227,680-685 use gp130 as an “internalization mediator.” A recent report 29. Chamberlain, J. P. (1979) Anal. Biochem. 98, 132-135 demonstrating that incubation with leukemia inhibitory factor 30. Graeve, L., Drickamer, K., and Rodriguez-Boulan, E. (1989) J . Cell Biol. 265, 121G1224 also down-regulates IL-6R surface expression supports this no- 31. Hopkins, C. R.(1992) lFends Biochem. Sci. 17, 27-32 tion (38). 32. Gearing, D. P., Comeau, M. R., Friend, D. J.,Gimpel, S. D., Thut, C. J., McGourty, J., Brasher, K. K , King, J. A,, Gillis, S., Mosley, B., Ziegler, S. F., The physiologic significance of the internalizationof growth and Cosman, D. (1992) Science 255, 1434-1437 and differentiation factors is not clear. The internalization of 33. Takeshita, T., Asao, H., Ohtani, K , Ishii, N., Kumaki, S., Tanaka, N., Munakata, H., Nakamura, M., and Sugamura, K (1992) Science 257,379482 IL-6 could be a mechanism to down-regulate surface-receptor 34. Mackiewicz, A,, Schooltink, H., Heinrich, P. C., and RoseJohn, S. (1992) J . to expression after a cell has received the signal and thus Immunol. 149,2021-2027 prevent overstimulation. Since IL-6 is a cytokine only tempo- 35. Mullberg, J., Oberthiir, W., Lottspeich, F., Mehl, E., Dittrich, E., Graeve, L., Heinrich, P. C., & Rose-John, S. (1994) J . Immunol. 52, 4958-4968 rarily released during acute inflammation and trauma, effiN. Y., Nye, S. H., Boulton, T. G., Davis, S., Taga, T., Li, Y., Birren, S. J., cient endocytosis might also be important for the clearance of 36. Ip,Yasukawa, K., Kishimoto, T., Anderson, D. J., Stahl, N., andYancopoulos, G. D. (1992) Cell 69, 1121-32 this cytokine from the circulation (39). Continuous high levels , T., Taga, T., Tsang, M. L., Yasukawa, K., Kishimoto, T., and Yang, Y. C. of circulating IL-6 led to severe diseases in transgenic mice 37. Y h(1993) J. Immunol. 151,2555-2561 (40). Future studies will have to clarify the physiological and 38. Yamaguchi, M., Michishita, M., Hirayoshi, K., Yasukawa, K., Okuma, M., and Nagata, K. (1992)J . Biol. Chem. 267, 22035-22042 immunological importance of the endocytic process for IL-6 and 39. Castell, J. V., Geiger, T., Gross, V.,Andus,T., Walter, E., Hirano, T., Kishimoto, other cytokines. T., and Heinrich, P.C. (1988) Eur. J. Biochem. 177,357-361 Acknowledgments-We thank Wiltrud Frisch and Hildegard Schmitz-Van de Leur for excellent technical assistance, Dr. Peter Freyer

40. Suematsu, S., Matsuda, T., Aozasa, K., Akira, S., Nakano, N., Ohno, S., Miyazaki, J.,Yamamura,K., Hirano, T., andKishimoto, T. (1989)Proc.Natl. Acad. Sci. U. S. A. 8 6 , 7547-7551