Barbara Schwartz, Jerome A. Langern, and Sidney Pestka. From the Department of Molecular Genetics and Microbiology, University of Medicine and Dentistry of ...
THEJOLIRNAI OF BIOLOGICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.
Vol. 269,No. 27,Issue of July 8, pp. 18102-18110, 1994 Printed in U.S.A.
Expression of a Functional Human TypeI Interferon Receptor in Hamster Cells: Application of Functional Yeast Artificial Chromosome (YAC) Screening* (Received forpublication, March 7, 1994, and in revised form, April 28, 1994)
Jaemog Soh$, ThomasM. Mariano, Jin-KyuLimPn, Lara Izotova, Olga Mirochnitchenko, Barbara Schwartz, JeromeA. Langern, and SidneyPestka From the Department of Molecular Genetics and Microbiology, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, New Jersey 08854-5635
The previously cloned human interferon d p (Hu-IFN- Baglioni, 1981; Pestka et al., 1987; Flores et al., 1991). Cloning d/3;"ype I interferon) receptorcDNA appears to be onlyof a human IFN-OJP receptor (IFN-aR1) cDNA was reportedon one component of a receptor complex since expression the basis of rendering mouse cells sensitive t o Hu-IFN-aB2 of the cDNA in mouse cells confers sensitivity only to (UzB et al., 1990). However, when mouse cells were transfected Hu-IFN-aB2,but a monoclonalantibodyagainstthis with the Hu-IFN-aR1cDNA, they did not exhibit binding and cloned receptor subunit inhibits biological activities of antiviral protection with Type I IFN subtypes other than HuHu-IFN-cuA, Hu-IFN-aB2, Hu-IFN-w, and Hu-IFN-P. HereIFN-aB2, and Chinese hamster ovary (CHO-K1) cells trans(YAC) con- fected with the cloned Hu-IFN-aR1 cDNA displayed no inducwe report thata yeast artificial chromosome taining a segment of human chromosome21 introduced tion of 2'-5'A synthetase in response t o Hu-IFN-a4 and Huinto Chinese hamster ovary (CHO) cells confers upon IFN-aB2 (Revel et al., 1991). Similarly, human cells transfected homologous cloned Mu-IFN-aR1 receptor cDNA these cells a greatly enhanced response to Hu-IFN-cuA withthe showed antiviral protection only with Mu-IFN-all. However, and Hu-IFN-aB2 as well as an increased response to HuIFN-w, Hu-IFN-d/D(BgZ),and Hu-IFN-p. These re- the expression of thisMu-IFN-aR1 cDNA inmurine cells of class I " I C an- (L1210 R101; cells resistant to typeI IFNs) lackingmRNA for sponses were measured by induction tigens and by protection against encephalomyocarditis this Mu-IFN-aR1receptor component showed antiviral protection in response all to I Mu-IFNs tested(Uz6 et al., 1992). virus and vesicular stomatitis virus. Furthermore, these In affinity cross-linking experiments, '251-IFNreceptor comof Hu-IFN-cuA cells exhibit specific high affinity binding and Hu-IFN-aB2, Hu-IFN-P, and Hu-IFN-w. The results plexes with M , = 80,000 (Hannigan et al., 1986),210,000 indicate that all the genes necessary to reconstitute a (Colamonici et al., 1992),260,000 (Vanden Broecke and Pfeffer, biologically active TypeI human IFN receptor complex 1988), or 300,000 (Raziuddin and Gupta,1985) were observed are located within the human DNA insert of this YAC in addition to the major complex which migrates as a broad clone. a doublet a t band witha M , = 140,000-150,000 or sometimes as 110,000 and 130,000 (Faltynek et al., 1983; Colamonici et al., 1992). These observations suggestthat other subunits, compoType I human interferons (the multiple Hu-IFN-a' species, nents, or accessory proteins areinvolved in ligand binding and one Hu-IFN-P, and one Hu-IFN-w) are a family of cytokines subsequent signal transduction of all Type I IFNs (reviewed by that induce a variety of physiological responses. These effects Colamonici and Pfeffer (1991) and Mariano et al. (1992)), as of described for the IFN-7receptor (Jung et al., 1987; Cook et al., include antiviral and antiproliferative activities, stimulation cytotoxic activity in lymphocytes, natural killercells and mac- 1992,1994; Soh et at., 1993, 1994a). Bazan (1990a, 1990b) rophages, modulation of cellular differentiation, and stimula- proposed that the IFN receptors as well as other cytokine retion of class I MHC antigensandothersurfacemarkers ceptors of the same superfamily are composed of two folding (Lengyel, 1982; Pestka et al., 1987). Like most cytokines and domains thatcomprise the ligand binding site which, at least in growth factors, the actions of Type I IFNs are mediated by some cases, resides in the crevice between the folds. The priinteractionwith specific cell-surface receptors(Friedman, mary cytokine-receptor interaction was suggested to involve 1967; Aguet, 1980). Competitionbinding studies demonstrated one face of the ligandwhile another face of the bound cytokine that Type I IFNs share the same receptor complex, whereas can interact withaccessory binding components. Type I1 IFN (IFN-7) binds to a distinct receptor (Branca and Somatic cell genetic studies with human x rodent hybrid cells containing various combinations of human chromosomes * This study was supported in part by United States Public Health have provided evidence that the presence of human chromoService Grant R01 CA46465 from the National Cancer Institute (to some 21 confers sensitivity of the rodentcells to various human S. P.). The costs of publication ofthis article were defrayed in part by the payment of page charges. This article must therefore be hereby marked Type I IFNs (Tanet a l . , 1973; Slate et al., 1978; Epstein et al., 1982; Raziuddin et al., 1984). It was also demonstrated that "advertisement" in accordance with 18 U.S.C.Section 1734 solely to indicate this fact. antibodies to human chromosome 21-encoded cell-surface com$' Supported in part by the Becton Dickinson Company as a desig- ponents were able t o block the actionor binding of Hu-IFN-a to nated Becton Dickinson Scholar. 8 Supported by funds from the Departmentof Molecular Genetics and cells (Revel et al., 1976; Shulman et al., 1984). Furthermore, Langer et al. (1990) demonstrated that 3x1s irradiation-reMicrobiology. llSupported in part by the Foundation of UMDNJ. duced hamster x human somatic hybrid cells containing about The abbreviations usedare: Hu-IFN, human interferon;EMCV, en- 3 mb of chromosome 21q around 21q22.1 (Jung, 1991; Soh et cephalomyocarditis virus; Mu-IFN, murine interferon;YAC, yeast arti- al., 1994a) were able to bind [32PlHu-IFN-aAand generate a ficial chromosome; VSV, vesicular stomatitis virus; MHC, major histocompatibility complex; kb, kilobaseb);mb, megabaseb); PFGE, pulsed- complex of about 150 kDa whencross-linked tn the cell surface. In addition, the cloned Hu-IFN-aR1 receptor gene was mapped field gel electrophoresis; PBS, phosphate-buffered saline.
me
18102
Functional IFN-aJ p Receptor Identified by YAC Screening
18103
nutrient mixture(Life Technologies, Inc.) supplemented with 10% dialyzed fetal calf serum (Jung et d., 1990). Construction ofYACs Containing the Antibiotic G418 Resistance Marker-Transformation of yeast cells was carried out as described (Rose et al., 1990) with the following modifications. Fifty milliliters of each YAC clone were grown to a n absorbance of 1.0 uniuml at 600 nm in AHC liquid media, harvested, and suspended in5 ml of 1 M sorbitol. Fifty microliters of 10 mg/ml Zymolyase 20T (ICN) were added to the cell suspension, which was then incubated at 37 "C for 15-30 min to obtain greater than 95% conversion to spheroplasts. The spheroplasts were washed three times with 1M sorbitol, collected by centrifugation (500 x g, 5 rnin), and resuspended inSTC medium (1M sorbitol, 10 m~ Tris-C1, pH 7.5, 10 mM CaCl,) a t a density of 5 x lo8 spheroplasts/ml. Five micrograms of linearized plasmid p J S l (Soh et al., 1993, 1994b) and 10 pgof salmon testesDNA as a carrier were put into 100 pl of the spheroplast suspension in a polypropylene tube, and the mixture was incubated for 5 min at room temperature. After addition of 4 ml of polyethylene glycol (PEG) solution(20% w/v PEG 8000, 10 mM CaCl,, 10 mM Tris-C1, pH 7 . 9 , the mixture was incubated for 10 min at room temperature, then centrifuged (500 x g, 5 min) to pellet the spheroplasts. The spheroplasts were resuspended in pl 300 of SOS medium (1 M sorbitol, 5 mM CaC1, in YPD medium consisting of 10 g of yeast extract, 20 gof peptone, and20 g of glucose per liter) and incubated for 20 minat 30 "C with shaking. Finally, the spheroplasts were mixed with 6 ml of top agar and the mixture was overlaid on a synthetic dextrose minimal medium plate supplemented with adenine sulfate (10 pg/ml) and histidine (20 pg/ml) to selectLys' transformants. Electrophoresis and Hybridization Procedures-Agarose plugs were prepared as described (Smith et QZ., 1987). Yeast cells were grown t o stationary phasea t 30 "C in 50 ml ofAHC media, washedtwice with 25 ml of 50 mM EDTA (pH 8.0) by centrifugation at 3500 x g for 5 min, followed by resuspension to a density of 1 x 10" cellslml in 50 mM EDTA. An equal volume of 1% (w/v) Insert agarose (FMC BioProducts) in 50 mM EDTA (pH 8.0) and20 pl of 10 mg/ml Zymolyase 20T per ml of agarose solution were added tothe cell suspension, and the mixture was put into the plugmold. Spheroplasts were madeby pushing plugs out of the mold into 1 ml of 0.5 M EDTA (pH 8.0) containing 7.5% P-mercaptoethanol per plug and incubating overnight at 37 "C with gentle shaking. The solution was removed and the plugs were rinsed twice with 50 m~ EDTA (pH 9.25). The spheroplasts were lysed by transferring to Solution ESP (0.5 M EDTA (pH 9.251, 1% Sarkosyl (IBI), and 1mg/ml proteinase K (Sigma)) and incubating for 24 hat 50 "C. The agarose plugs were analyzed by pulsed-field gel electrophoresis (PFGE) (CHEF, OWL Scientific Plastics, Inc.). The gels measured 13 x 13 cm, consisted of 120 ml of 1%agarose in TBE (0.045 M Tris borate, pH 8.0, 2.5 mM EDTA), and were run at 170 Vfor 24 hat 15 "C with a pulse time of 70 s. The DNAs were blotted onto Nytran (Schleicher& Schuell) by standard Southern blot procedures. DNA probeswerelabeledwith [a-3ZPldCTP (DuPont NEN, 3000 Ci/mmol) with random hexadeoxynucleotides as primers (Feinberg andVogelstein, 1983). Fusion and Dansfectzon-The spheroplasts fromYACs F136C5.neo.3 and F136C5.neo.9 were fused to CHO-K1 and 16-9 cells by procedures already described (Pavan et QZ., 1990; Soh et al., 1994b). Thirty-five milliliters of each clone were grown inAHC media to stationary phase. EXPERIMENTAL PROCEDURES After being washed twice with 20 ml of 1 M sorbitol, cells were resusCells a n d Media-Three YAC clones, F136C5and F143C3 (from Drs. pended in 5 ml of a solution which contained 1 M sorbitol, 100 mM Chumakov and Cohen, Centre #Etude du Polymorphisme Humain, sodium citrate (pH 5.81, 10 mM EDTA, and 30 mM P-mercaptoethanol. Paris, France; screened by probe 524-5P) and B49F1 (Washington Uni- Eighty microliters of a Zymolyase 20T stock solution (10 mg/ml) were versity Library; screenedby probe IFNAR), were obtained throughthe added, and the mixture was incubated at 37 "C for 20 min until 95%of chromosome 21 Joint YAC ScreeningEffort(directed by Dr. David the cells were spheroplasts. The spheroplasts were pelleted and washed Patterson, Eleanor Roosevelt Institute, Denver, CO). All YACs were twice in 5 ml of a solution of 1 M sorbitol in 10 mM Tris-C1 (pH 7.5), derivedfromSaccharomyces cereuisiae AB1380 (MATa, ura3, trpl, resuspended in 5 ml of the same solution, and counted. While the ade2-1, canl-100, lys2-1, and h i s 3 a n d grown in AHC media. YACs spheroplasts were being washed, the CHO-K1 or 16-9 cells were harF136C5.neo.3 and F136C5.neo.9 are derivatives ofYAC F136C5 that vested by trypsinization, resuspended in serum-free F-12 medium, and contained an integrated neomycin phosphotransferase gene (neo) as counted. Aliquots containing 4x lo7 spheroplasts were placed in 15-ml described (Soh et al., 1993, 199413). Chinese hamster ovary K1 (CHOtubes and centrifuged a t 500 x g for 5 min. The supernatant was reK1) cells were obtainedfrom the American Type Culture Collection. The moved from the spheroplast pellet, and 5 ml of the cell suspension 16-9 cell line is a human x hamster hybrid containing the long arm of containing 2 x lo6 mammalian cells was added carefully. The cells were human chromosome 6 and a transfected HLA-B7 gene (Soh et al., 1993). then centrifuged at 500 x g for 5 min. The supernatant wasremoved, CHO-Kl/aRc4 and 16-9/cuRc5 cell lines were obtained by transfecting and 50 pl of serum-free F12 medium was added t o resuspend the pelpVADN123 (from Gilles Uze), a plasmid containing the Hu-IFN-aR1 lets. Five hundred microliters of 45% PEG 1500 solution (Boehringer cDNA, into CHO-K1 and 16-9 cells, respectively. CHO-KUcuRyS-4 and MannheimBiochemicals)containing 5% dimethyl sulfoxide, 10 p~ 16-9/aRy9-2 cell lines were obtained by fusing the spheroplasts from P-mercaptoethanol, and 5 mM CaC1, were added to themixed cell susYAC F136C5.neo.9 with CHO-K1 and 1&9 cells, respectively. These pension. Thecells were mixed by tapping the tubebriefly, incubated for transfected and YAC-fused cell lines were maintained inF-12 nutrient 2 min at room temperature, and diluted with 5 ml of serum-free F12 mixture (Sigma) containing 10% fetal calf serum (Sigma) and 450 pg/ml medium. Thiscell suspension was lefta t room temperature for 20 min, antibiotic G418 (Life Technologies, Inc.). Human x hamster hybridcells, then centrifuged at 600 x g for 6 min. The resulting pellet was resus153B7-8, containing human chromosome 21q were grown in F-12D pended in 50 ml of complete F12 medium (F12 plus 10% heat-inacti-
to the 3x1s region (21q22.1) (Lutfalla et al., 1990). The paradoxical observation that CHO 3x1s cells could bind Hu-IFNCUA, whereas the expression of the cloned receptor cDNA in mouse cells did not confer binding to Hu-IFN-olA and expression in CHO cells did not induce 2,5'-A synthetase activity in response to H u - I F N - d , suggested that the 3x1sregion of human chromosome 21 contains other subunits of the receptor complex (Langer et al., 1990; Revel et al., 1991). This assumption was supported by the identification of two separate components following immunoprecipitation of [12511H~-IFN-aA: receptor complexes from 3x1s cell extracts with anti-IFN-a receptorantibody (Colamonici and Domanski, 1993; Colamonici et al., 1990, 1992). These two subunits seem to differ from the cloned Hu-IFN-aR1. The monoclonal antibodies against one subunit (110 kDa) and the recombinant Hu-IFN-aR1 receptor block the biological activity of Type I interferons while monoclonal antibodies againstthe second subunit do not (Colamonici and Domanski, 1993; Benoit et al. 19931, suggesting that the Type I IFN receptor consists of at least three different subunits. In addition, antibodies to the Hu-IFN-aR1 block the activity of various Type I interferons on human cells (UzB et al., 1991; Benoit et al., 1993). All these observations indicate that the Hu-IFN-aR1molecule is only one component of the Type I receptor. Yeast artificial chromosome (YAC) cloning techniques allow the cloning of DNAfragments up to2000 kb (Burkeet al., 1987; Chumakov et al., 1992a). The large insert size can not only facilitate physical mapping of chromosomes (Chumakov et al., 1992b), but also permits expression of very large genes not possible by conventional cloning procedures and regions of chromosomes containingmultiplegenes (Soh et al., 1993, 1994b; Cook et al., 1994). We already demonstrated that YAC clones can be used for expression and phenotypic mapping of genes afterfusion of appropriate yeast spheroplasts with mammalian cells (Soh et al., 1993; Cook et al., 1994) in the absence of any specific DNA mapping or sequence information. In this way, we isolated a YAC clone which contains the gene for an accessory factor required for the function of the human interferon-? receptor (Soh et al., 1993). Here we report that the expression of a YAC clone containing the gene for the cloned Hu-IFN-aR1 receptor subunitrendershamster cells much more sensitive to Hu-IFN-olA and Hu-IFN-aB2 andsomewhat more sensitive to Hu-IFN-o and Hu-IFN-p. Thus, this YAC clone must contain multiple genes required to reconstitute a fully functional Hu-IFN-dp (Type I) receptor.
18104
Functional IFN-a 1p Receptor Identified by YAC Screening
vated fetal bovine serum)and plated at lo6 mammalian cells per 150-mm plate. Thirty-six hours later, plates were washed with phosphate-buffered saline (PBS) to remove dead cells and yeast, then refed with complete F12 mediumcontaining 450 pg/ml antibiotic G418. Cells were fedas necessary, and resistantcolonies typically appeared after 10 to 14 days. The G418-resistant colonies were pooled and expanded for further analysis. After initial assay of pools of colonies, five subclones from each pool were obtained by the limiting dilution technique. CHO-K1 and 16-9 cells were also transfected with Hu-IFN-aR1 cDNA clonepVADN123, as described (Hibino et al., 1992). Five individual G418-resistant colonies from each transfection were tested for various functions and binding activity. Cytofluorographic Analysis of Cells for Expression of Class I MHC Surface Antigens-Cells were seeded in 24-well plates at a density of about 25,000 cells/well(1ml/well) and were treated with the indicated concentrations and types of IFN for about 72 h by which time the cells were nearly confluent. Cells were trypsinized, transferred to 1.5-ml tubes, and washed with complete F12 medium. HLA-B7 antigens on 16-9 cells were detected by incubating the cells with 15 pl of culture supernatant from the hybridoma line producingmousemonoclonal anti-HLA antibody W6/32 (Jung et al., 1988; Hibino etal., 1992)for 30 min at 4 "C. Cells were washed with complete mediumand resuspended in 15 pl of fluorescein isothiocyanate-conjugatedgoat anti-mouse IgG (Cappel) diluted t o 80 pg/ml and incubated for 30 min at 4 "C, after which they were washed with complete mediumand resuspended in 200 pl of cold complete medium for the immediate analysis of live cells. If cells had to be fixed for future analysis, they were washed twice with PBS, resuspended in 15 pl of 3% (w/v) paraformaldehyde in PBS, and incubated from 1to 16 h at 4 "C. The fixed cells were washedwith PBS and finally resuspended in 200 pl of PBS. Samples were analyzed on a Coulter Epics Profile Cytofluorograph. For each analysis, 10,000 events were accumulated and analyzed on CytoLogic software. Interferons and Antiviral Assay-Hu-IFN-aA, Hu-IFN-orAID, and Hu-IFN-p were prepared as previously reported (Staehelin et al., 1981; Rehberg et al., 1982; Moschera et al., 1986), and Hu-IFN-w was a gift from Dr. G. Bodo (Ernst-Boehringer Institute fur Arzneimittelforschung, Austria). Hu-IFN-aA-Pl and Hu-IFN-aB2-P were prepared as described (Li et al., 1989; Wang et al., 1994). Hu-IFN-P activity was measured by a cytopathic effect inhibition assay on human WISH cells with vesicular stomatitis virus (VSV); the activity of all other Type I human interferons was measured on bovineMDBK cells with VSV (Familletti et al., 1981). Parental and transfected CHO-K1 and 16-9 cells were also assayed for resistance to encephalomyocarditis virus (EMCV)orVSV infection by a cytopathic effect inhibition assay (Familletti et al., 1981). Genetically engineered phosphorylatable Hu-IFN-aA (Hu-IFN-aAP1) and Hu-IFN-aB2 (Hu-IFN-aB2-P) were phosphorylated with the catalytic subunit of bovine heart cAMP-dependent protein kinase and [y-32P]ATPas described (Li et al., 1989;Wang et al., 1994). The specific activity of the labeled IFN was 3-5 x lo6 cpdpmol at the time used. IFN Binding-For binding studies, cells were treated with trypsin and collected from 75-cm2 tissue culture flasks. Binding of IFNs to cells was performed in a volume of 0.1 ml containing 0.5-1 x lo6 cells as noted in theappropriate legends to the figures and tables. The L3'P1IFN bound to cells was separated from the unbound [32PlIFNby sedimentation through a cushion of 10% sucrose in PBS (Langer et al., 1986; Langer and Pestka, 1986). Nonspecific binding of [32PlHu-IFN-aA-P1 and [32P]Hu-IFN-aB2-Pwas determined by the addition of a 200-fold excess of unlabeled recombinant Hu-IFN-aA. The nonspecific binding was subtracted from the total radioactivity in each case to yield the countsimin bound specifically. Forcompetition studies, cellswerecollectedfrom tissue culture flasks and resuspended at 2 x lo6 celldm1 for Daudi cells and 8 x lo6 cells/ml for CHO-Kl/aRyS-4 and 16-9/aRy9-2 cells.To 100 plof cells was added 10 pl of nonradioactive Hu-IFN-as to produce final concentrations in the range of 2 x 10"' M to 4 x lo-@M and 1 pl of [32PlHu-IFN&-PI (final concentration, 1x 10"" M; 0.8 x lo6 cpm). Afterincubation at room temperature for 1h with intermittent shaking, bound L3'PlHuIFN-aA-Pl was measured as described above. Covalent Cross-linkingof P2PIHu-IFN-d-Pland P2PlHu-IFN-aB2P To Cell-SurfaceReceptors-Cells were harvested, pelleted, and resuspended in Dulbecco's modified Eagle's mediumcontaining 10%fetal calf serum at a density of 1x lo7 cells/ml. Cells wereincubated with 5 x lo6 cpdml of [32P]Hu-IFN-aA-P1 or [32P]Hu-IFN-aB2-Pat room temperature for 1 h in a volume of 0.2 ml with or without a 200-fold excess of unlabeled IFN-aA. The cells werewashed and resuspended in 0.1 ml of cold PBS (pH adjusted to 8.0 with 1 M potassium borate), and the chemical cross-linker disuccinimidyl suberate (Pierce)in dimethyl sulf-
oxide was added to a final concentration of 500 (Rashidbaigi et al., 1985; Langer et al., 1986).After incubation on ice for 30 min, 50 mM Tris-HC1(pH 8.0) was added toquench the reaction for 5 min. The cells were washed with ice-cold PBS and pelleted. [32PlIFN:receptorcomplexes were extracted with 0.1 mlof 1%(v/v) Triton X-100 in PBS containing protease inhibitors (Ronnett et al., 1984). The detergent extracts were then analyzed on 7.5%polyacrylamidegels inthe presence of SDS (Laemmli, 1970). Gels weredried under vacuum and autoradiographed. RESULTS
YAC F136C5 Contains the Gene for the Cloned Hu-ZFN-alp Receptor (Hu-IFN-&l)--TWo YAC clones (F136C5 and F143C3) were identified by screening with a sequence tagged site (STS) made from the 524-5P (021358) probe located in the vicinity of 21q22.1. YAC B49F1 (IFNAR YAC) was identified by apolymerase chain reaction primer pair generated from the exon VII-intron VI junction of the Hu-IFN-aR1gene (Mariano etal., 1992). Physical mapping data showed that the IFNAR, GART, and 021958 loci werelocated in the same 400-kb MluI fragment, and, furthermore, that the 021958 and ZFNAR loci mapped within a 170-kb MlullNotI fragment (Tassone et al., 1990; Cheng et al., 1993). Therefore, it is likely that theYACs screened with the 524-5P (021358) probe contain the entiregene for the cloned Hu-IFN-aR1 receptor if the YACs are not chimeric. To test this point, yeast chromosomal DNAs from the F136C5, F143C3, and B49F1 YACs were digested with EcoRI restriction endonuclease, and the blot was probed with cDNA for the Hu-IFN-aR1 receptor. Exon I and part of the next long intron, and a fragment of about 5 kb corresponding to the gene segment encoding the intracellular domain of the protein,were missing from YAC B49F1.2As shown in Fig. lA, the F136C5 YAC contained four EcoRI fragments hybridizing t o the cDNA, which indicates that the YAC contains the entire 30-kb Hu-IFN-aR1 receptor gene, based on the sequence and mapping of the gene (Lutfalla et al., 1992; Mariano et al., 1992). Therefore, two additional bands, 15 kb and 4.8 kb, in YAC F136C5 must correspond t o exon I and part of the intronI sequence, and to the intracellularregion of the cloned Hu-IFN-aR1 gene, respectively. Integration of the neor Gene into YACs-AYAC integration plasmid pJSl for the 3. cereuisiae strain AB1380 commonly used for YAC library construction was described (Soh et al., 1993, 1994b). After transformation ofYAC F136C5 with pJSl linearized with ClaI restriction endonuclease, 12 Lys' transformants were selectedfor further analysis. In order confirm to that the plasmid is targeted into theYAC, agarose plugs from these 12 clones were run on PFGE gel and theblot was probed with a neor gene fragment. Four of twelve transformants had the neo' gene targeted into the YAC while one clone had theneo' gene integrated in yeast chromosome I1 (data not shown). As shown in Fig. 1B, YACs F136C5.neo.6 and F136C5.neo.10 have a reduced size compared to the original YAC (430 kb), indicating that a fairly large genomic fragment located between two Alu-targeted sites was deleted (Soh et al., 1994b). Therefore, YACs F136C5.neo.3 and F136C5.neo.9, which had no apparent deletions, were selected for fusion with CHO-K1 or 16-9 cells. After selection of antibiotic G418-resistant colonies following fusion between yeast spheroplastsfrom YACs F136C5.neo.3 or F136C5.neo.9 on the one hand and CHO-Kl or 16-9 cells on the other, the resistant colonies were pooled and expanded for further analysis. The pools of colonies from YAC-fused CHO-K1 and 16-9 cells were referred to as CHO-KUaRy3 or CHO-Kl/ aRy9 and 16-9/aRy3 or 16-9/aRy9, respectively, where 3 and 9 represent different neor gene-targeted YACs. J. Soh, unpublished data.
Functional IFN-a I p Receptor Identified by YAC Screening
A
B
Kb
18105
TABLEI Interferon concentrations for50% protection (ED,5,) of cells from EMCV infection The values in the table represent the IFN titer (units/ml) for 50% protection of cells (ED,,). Hu-IFN-olA/D(Bgl) is a chimeric Hu-IFN-a that is active on CHO and most mammalian species tested (Rehberg et a positive control. The two sets of cell lines (i.e. al., 1982) and is used as derived from CHO-K1 and from 16-9 cells) were not assayed at the same time so the precise end points between the two sets are not directly comparable. However, each set of cells was assayed a t one time. Cell line
Hu-IFN aRy9-2 aRc5
- 815
aAlD 88 aA aB2 0 177
-
P
CHO-Kl aRy9-4
88 442 >10,000 177 11
CHO-K11 aRc4
313 10,000 11
CHO-K11
44 2 111 44 8
16-9
16-91
625 442 2,500 28 1,768 >10,000 7,072 >2,000 >2,000 55 55
16-9/
28 55 707 10,000 >10,000 0 177 1,414 1,414 P 10 10 olA/D
1
CHO-K1 aRy9-4
CHO-Kl1
39
aA
78 10,000 >10,000 >2,000 >2,000 111 55
16-91
55 78 78 500 4
to greater than200-fold increases in sensitivity to H u - I F N - d and Hu-IFN-aB2. There was even an increase insensitivity to Hu-IFN-oLAID that interacts well with the hamster Type I receptor. Furthermore, in both of these YAC-containing cells there was a significant increase in sensitivity t o Hu-IFN-o; 16-9/aRy9-2 cells showed a significant increase insensitivity t o Hu-IFN-P (>ll-fold). Decreased sensitivity of 16-9 cells compared to the parental CHO-K1 cells to Hu-IFN-o wasobserved (>2,000 versus 177 units/ml). When the ED,, of the IFNson cells challenged withVSV was measured, cells with the YAC F136C5.neo.9 showed a substanAntiviral Protection of Cells by Interferons-The CHO-K1/ tial increase in sensitivityalltothese interferons, ranging from 100-fold (Table 11).There was little aRy9 cells (CHO-K1 cells fused with YAC F136C5.neo.9) about %fold to greater than showed increased sensitivity to Hu-IFN-d and Hu-IFN-aB2 or no increased sensitivity of the comparable cells containing as measured by antiviral protection against EMCV when com- the cloned Hu-IFN-aR1 cDNA. pared with parentalCHO-K1 cells or CHO-K1 cells transfected All these antiviral data with16-9 cells are graphically sumwith the Hu-IFN-aR1 receptor cDNA construct pVADN123 marized in Fig. 2. In all cases,sensitivity of 16-9cells containing (CHO-Kl/aRc4). Also, 16-9/aRy9 cells showed a similarinthe aYAC was greater than parental 16-9 cells or 16-9 cells concreased sensitivity toHu-IFN-d and Hu-IFN-aB2. After fus- taining the Hu-IFN-aR1cDNA. The sensitivity of cells to the ing the spheroplasts from YAC F136C5.neo.3 to CHO-K1 and to interferons wasexpressed as the reciprocal of the ED,, (Fig. 2). Induction of Class I MHC Surface Antigens-Subclones of 16-9 cells (CHO-Kl/aRy3 and 16-9/aRy3), pools of G418-resistant cells (data not shown) showed the same phenotype as CHO-K1 and 16-9 cells fused to YAC F136C5.neo.9 or transCHO-Kl/aRyS and 16-9/aRy9. The subclones (CHO-Kl/aRy9-4 fected with the Hu-IFN-aR1 cDNA were tested for class IMHC and 16-9/aRy9-2 derived from CHO-Kl/aRyS and 16-9/aRy9, induction. Dueto therelatively highendogenous background of respectively)wereassayed for antiviral protection against hamster class I MHC antigens on the CHO-K1 cells, treatment EMCV (Table I) and VSV (Table 11). CHO-Kl/aRc4 and 16-9/ of transformed CHO-K1 cells showed little or no hamster class aRc5 were individual cell lines from CHO-K1 and 16-9 cells, I MHC antigen induction as detected withmonoclonal antibody respectively, transfected with Hu-IFN-aR1 receptor cDNA K204 which reacts with mouse MHC class I antigens andwhich (pVADN123). Five subclones eachfrom CHO-K1 and 16-9 cells also reacts with hamster MHC class I antigens on CHO-K1 transfected with the Hu-IFN-aR1cDNA were tested, andall of cells (data not shown). However, subclones isolatedfrom transformed 16-9 cells, which express the human HLA B7 antigen, them showed the same sensitivityas the parental cells. The CHO-K1 and 16-9 cells transfected with the Hu-IFN- showed very good HLA B7induction (Fig. 3 and Table 111).Fig. aR1 cDNA showed a slight increase inprotection from EMCV 3 shows treatment of cells with 100 unitdm1 of Hu-IFN-d, in response to Hu-IFN-aB2, but no increased sensitivity in -aB2, and -w, which gave the most significant increases in the response to Hu-IFN-d, -o,-& or -AD (Fig. 2; Table I). How- HLA-B7 antigen expression. Table I11 summarizes theHLA-B7 ever, cells containing the YAC F136C5.neo.9 exhibited 89-fold induction as a function of IFN concentration for all of the Type
18106
Functional IFN-a I p Receptor Identified by YAC Screening
Antiviral Activity (EMCV)
Antiviral Activity (VSV) 0.015
".UT
( aYAC B 0.03
aYAC
0.010 -
aYAC
0.02 -
-
0.005 0.01 -
% r
0.00
Hu4FN-d
EG
% F
I-
0.000
Hu-IFN-cY.B~
Hu-IFN4
Ea
Hu-IFNIxES
0.25
C
aYAC
0.20
0.25
-
0.20
-
aYAC
0.15 0.15 0.10
0.10 0.05
0.00
Hu-IFN-9
IE
Hu-1FN-P
nYAC
0.001 -
I
0.005
0.004
-
0.003
-
0.000
Hu-IFN-oI
Hu-IFN-oI
FIG.2. Antiviral activityof interferons. A , the data in the figure represent thereciprocal of the IFN titer (unitdml)for 50% protectionof cells (ED,,) against EMCV. The 16-9 cell line is ahuman x hamster hybrid containing the long arm of human chromosome 6 anda transfected HLA-B7 gene (Soh et al., 1993). &I denotes 16-9 cells stably transfected with the plasmid containing the IFN-aR1 cDNA aYAC denotes 16-9 cells containing YAC F136C5.neo.9.The data for both Hu-IFN-orAand Hu-IFN-aB2 are shown here with the 16-9 cells.Similar data were obtained with parental CHO-K1 hamster cells. B, the reciprocal of the IFN titer (units/ml) for 50% protection of cells (ED,,) against VSV. The experiments were performed as described in A except that VSV was used instead ofEMCV. The data for both Hu-IFN-orA and Hu-IFN-aB2 are shown. C, the reciprocal of the IFN titer (units/ml) for 50% protection of cells (ED,,) against EMCV. The experiments were performed as described in A except that Hu-IFN-P was used instead of the a interferons. The value for the (ED,,)-l for the cells containing the aYAC is >0.20 shown in the figure as maximal protection was obtained a t 5 unitshl of Hu-IFN-P, the lowest Hu-IFN-rJconcentration tested: therefore, the end point (ED,,) was
