BACKGROUND OF INVENTION

Field of Invention

The present invention relates to the molecular

10 cloning of a gene encoding a polypeptide which modulates cellular response to type I interferon, such polypeptide being a putative component of the interferon type I rece tor system, and expression of the gene in a suitable hos The present invention relates to the polypeptide itself,

15 active analogs thereof and processes for obtaining poly¬ peptides with such properties.

Information Disclosure Statement

Human and murine cells can be induced to make 20 three classes of interferon (IFN) designated alpha, beta, and gamma on their basis of their antigenic properties an on the type of cell producing them. These IFN's in turn induce a number of changes in human cells which result in establishment of an antiviral, anti-tumor, and/or anti- ²⁵ cellular state, and cause a number of alterations in the cell membrane including induction and/or increased expres sion of major histocompatibility complex (MHC) antigens. Lindhal, P. et al (1973), "Enhancement by interferon of the expression of surface antigens on murine leukemia L

30 1210 cells," Proc. Natl. Acad. Sci. (USA). 70. pp. 2785- 2788; and Fellous, M. et al (1982), "Interferon-dependent induction of mRNA for the major histocompatibility anti¬ gens in human fibroblast and lymphoblastoid cells," Proc. Natl. Acad. Sci. fUSA) . 79, pp. 3082-3086. Most, or all,

35 of these changes are triggered by the signal generated by interaction of IFN's and their cell surface receptors.

Direct ligand binding studies and binding site competition studies as well as indirect immunological and somatic cell genetic studies demonstrate that the three classes of IFN's interact with interferon sensitive cells through specific binding to one of two types of high affinity receptors on the cell surface (reviewed in Rubinstein, M. and Orchansky, P. (1985) , "The Interferon Receptors," CRC Critical Reviews in Biochemistry. 2. p. 249). All human IFN-alpha's and human IFN-/3 (Type I IFN's) bind to a chromosome 21-encoded Type I receptor, while human IFN-gamma (Type II IFN) binds to a chromosome 6-encoded Type II receptor which requires a chromosome 21 encoded gene product in order to confer sensitivity. The IFN receptor required for MHC induction and the IFN recep tor which triggers antiviral state (AVS) induction share common antigenic determinants and are encoded on human chromosome 21.

The human type I IFN receptor has been estimate to have a molecular weight of 95-140 kDa, based on experi ments in which cell membrane having bound ¹²⁵ I-IFN are cross-linked and run on SDS-PAGE (Razziudin, A. et al. (1984) , Proc. Natl. Acad. Sci. (USA) . 81, pp. 5504-5508 and Thompson, M.R. et al. (1985) , J. Biol. Chem.. 260. pp. 563-571) . There is some evidence that the receptor is a glycoprotein and the lower values may be closer to the size of the polypeptide itself.

No admission is made that any reference cited herein qualifies as pertinent prior art. All publication cited herein are hereby incorporated by reference.

SUMMARY OF THE INVENTION Despite the considerable interest in the inter¬ feron system, and the cloning of the alpha and beta inter feron genes several years ago, the Type I receptor has neither been isolated from nature and purified, nor pro¬ duced by recombinant DNA techniques.

It has not been possible to isolate IFN-α,/3 receptors by direct protein purification because of thei scarcity (a few thousand molecules per cell) and of the absence of cell lines with an amplified number of recep- tors. To overcome this difficulty, the present inventor approached the problem by DNA-mediated gene transfer. B transforming mouse cells with human genomic DNA and selecting cell clones which have acquired increased abil ty to respond to human alpha and beta IFN, the gene for protein component of the Type I IFN receptor system was successfully isolated. The present inventors chose to u the increase in H2 antigens of the cell surface as a mea to isolate human IFN responder cells.

The present invention relates to a DNA sequenc comprising the DNA sequence encoding a protein which modulates cellular response to Type I interferon and is thus a putative component of the human interferon Type I receptor system. In a preferred embodiment, it relates the cDNA of Fig. 1 comprising the coding region 899-3253.

The present invention also relates to a DNA vector including the DNA of the invention, and preferabl to an expression vector consisting essentially of the plasmid pSVE3-L7 containing the cDNA of Fig. 1 fused to the Simian Virus (SV) 40 early gene promoter.

It further relates to host cells, preferably eukaryotic, e.g. mammalian cells, transformed with a DNA of the invention in a manner allowing the host cell to express said human IFN type I receptor-related poly¬ peptide, and to the polypeptide thus expressed.

BRIEF DESCRIPTION OF THE DRAWINGS

Fiσure 1 is the nucleotide sequence of the putative Type I interferon receptor component cDNA clone L7. The coding sequence is identified and the predicted translation into amino acids is given.

Figure 2 is a schematic illustration of a preferred protocol for cloning proteins which modulate cellular response to Type I interferons, such proteins being putative components of the human Type I IFN recepto system.

Figure 3 is a restriction map of the L7 cDNA illustrating the structural relationship between three molecules used to raise antibodies, and the L7 protein.

Figure 4 is a bar chart showing the amplifica- tion of the L7 cDNA expression in CHO cells by metho- trexate (MTX) expressed as the response of cells to human IFN, (i.e., fold increase of (2'-5') A synthetase) . The old increase for CHO-pSVL7 clones selected with each MTX concentration was calculated by comparison to control CHO DHFR clones subjected to same treatment. The (2'-5') A synthetase increase in non-amplified clones (0 MTX) was two-fold over CHO-DHFR, and in the ordinate, this value i taken as 1 for comparison to the amplified clones.

Figure 5 is a graph plotting (2'-5*) A synthe¬ tase activity in CHO-pSVL7 transformants (at 300 nM MTX) against the activity in CHO-DHFR controls for cells induced by either human IFN-3 (squares) or hamster IFN (crosses) . Figure 5B shows an expanded plot of the data in Figure 5A.

Figure 6 is a FACS analysis of Daudi cells stained with Rhodamine-conjugated Protein A in the pres¬ ence of rabbit anti-L7 (PSN) antibodies. The abscissa is fluorescence arbitrary units and the ordinate is the number of cells. The left panel is light scattering showing there are no differences in cell size with or without anti-L7. The right panel shows Rhodamine fluores cence: the two curves peaking at 60 are cells without rabbit antiseru or with normal (non-immune) rabbit serum the curve peaking at 110 is from cells with anti-L7 anti- serum. The displacement shows that all the Daudi cells have L7 on their surface.

Figure 7 is a bar graph showing the binding of human _IFN-α to Daudi cells and the effect of different additions on this binding. Cold IFN-0 competes, and so do antibodies to cells with chromosome 21 (<chr.21), but anti-L7 antibodies (<L7) do not inhibit. NRS and NMS ar normal rabbit and mouse serum respectively.

DETAILED DESCRIPTION OF THE INVENTION

Receptors for cytokines or other protein hor- mones are protein molecules on the surface of cells whic interact with specificity with a ligand (i.e., the cyto- kine or hormone) present in the extracellular environmen and transfer a signal into the intracellular compartment in order to effectuate the biological action of the ligand. Known receptors for cytokines (e.g., interleuki 2 (IL-2), interferon-7 (IFN-γ)) are composed of sever¬ al different protein chains which together compose the receptor system. The function of these protein chains a 1) to bind the ligand specifically at high affinity, 2) interact with the active site of the ligand which often produces a conformational change, and 3) to transduce a signal into the intracellular compartment. Cooperation between the different proteins forming the receptor syst is considered to be a requirement for these functions. For example, high affinity binding of IL-2 requires two components (Hatakeyama M, et al (1989) , "Interleukin-2 receptor β chain gene: Generation of three receptor for by cloned human α and β chain cDNAs", Science. 244, 551- 556) . The type II interferon receptor system is compose of at least two components: a binding unit and a trans¬ ducing unit (Langer J.A. and Pestka S., (1988) "Interfer receptors". Immunology Today. 9, 393-400). It is likely that the type I IFN receptor system is similarly compose of several proteins involved in species-specific binding of IFN, interaction with IFN active site and transduction of the signal which activates transcription of specific genes.

While it is known in the art to cross-link an interferon to its cell surface receptor, this method cannot practically be used to identify transformed cells bearing the Type I receptor or to obtain sufficient recep tor to purify and characterize it, since the number of receptors per cell is low (typically, less than 5,000). Moreover, this method would not have identified the L7 protein since it does not appear to bind IFN at high affinity.

In one aspect, the present invention relates to a method of identifying a gene encoding a receptor compo¬ nent for a cytokine, particularly an interferon, and more especially a Type I interferon, which avoids the need to attempt to cross-link a labeled cytokine to a surface receptor expressed by the transformed cells and then, through detection of the bound label, distinguish cells expressing such a receptor from those which do not. Instead, the present invention relies on the ability of the cytokine to induce a detectable change in a cell surface characteristic. In a preferred embodiment, this change is an antigenic change. The change may be an increase or a decrease in one or more antigens. By way o example, with interferon one such change is an increase i MHC antigens; it is preferred to detect the increased expression of the H2 antigen, but other MHC antigens may similarly be detected, as, for example, β2 microglobulin or class II MHC antigens.

This procedure has several advantages over the method discussed above. First, it is capable of identi¬ fying cytoplasmic receptors, not merely cell surface receptors. Second, it is capable of identifying receptor which interact only indirectly with the corresponding ligand, not merely those which bind the ligand directly. Finally, it ensures that the selected clones are competen to respond to the ligand. The simplistic method can select clones expressing a molecule which binds interfero

but does not, as a result of that interaction, trigger a furthejr cellular activity.

Preferably, the cell surface antigen is recog¬ nized by an antibody, especially a monoclonal antibody. However, in certain cases, a different class of binding molecule, such as a lectin or an enzyme, may be suitable In one embodiment, the binding molecule must be labeled. Any label known in the assay art, such as a radiolabel, enzyme label or fluorescent label, may be employed; the fluorescent label is preferred. The label may be attach directly or indirectly, covalently or noncovalently. Alternatively, the binding molecule may be attached to a affinity chromatography support and cells bearing the receptor can then be affinity separated from those which do not.

Separation of the desired clones from the othe transformed cells is preferably accomplished by fluorescence-activated cell sorting (FACS) . The level o expression of the detected surface structure must of course be within the sensitivity limits of FACS, typical over 20,000 molecules/cell.

The receptor component gene must be isolated from DNA containing unrelated genes as well as sequences without coding function. The DNA to be screened may be genomic DNA or cDNA. It may be derived from the entire genome of the donor, or from a selected chromosome. It may be enriched for DNA of a particular size. The DNA m be introduced into the host cell by simple transfection DNA fragments, or the DNA fragments may be subcloned int a suitable viral or plasmid vector which is then used to transform the host cell.

The host cell used for screening is preferably one which, prior to transformation with a functional receptor gene, is incapable of responding to the inter¬ feron (or other cytokine of interest) . However, a host cell with a low level of response can be used if the genetic modification substantially enhances the response

to be detected. Since interferons are species-specific, in searching for the human IFN receptor component gene, the use of a non-human host cell is preferred.

Once the receptor component gene has been iso- lated, non-naturally occurring analogues of the receptor component are readily prepared by site-specific mutagene¬ sis of the gene, or by chemical treatment (e.g., with proteolytic enzymes) . Analogues likely to be functional can be identified by careful study of the sequence. For example, potential phosphorylation and transmembrane regions are more likely to be essential to activity. It is better to replace an amino acid with one of similar size and charge than with a radically different amino acid. The N- and C-termini may be truncated to determin which is necessary for activity.

The receptor component gene may be linked to a non-natively associated promoter, which may be constitu¬ tive or inducible. The preferred promoter is the SV40 early promoter. The codon selection of the gene may be modified to match the codon preferences of the host, to eliminate undesirable secondary structure formation, or facilitate further modifications of the gene by creating or eliminating restriction sites. The gene may be intro duced into any suitable host cell.

In an especially preferred embodiment, sheared human DNA was introduced into mouse LTK" cells together with a Herpes virus TK gene and LTK* transfected mouse cells which respond to human type I IFN's were selected fluorescence activated cell sorting (FACS) . Indirect, rather than direct, FACS selection for receptor componen was used because the number of Type I IFN receptors/cell is below that which can be detected by FACS analysis. T IFN induced increase in MHC antigens on the cell surface was chosen for use as a marker to separate responder cells.

Specifically, cells expressing biologically active receptor components, as evidenced by increased H2

expression after treatment of HAT resistant cells with human Type I IFN, were screened for. This is an approach supported by demonstration that human IFN regulates mouse H2 expression in human x mouse somatic cell hybrids expressing human IFN receptors. Cells were cultured and the FACS selection repeated. The amount of human DNA i the mouse cells was decreased by making a secondary trans fection using DNA from human IFN sensitive cloned cells from the primary transfections. Secondary transfectants expressing the receptor component were obtained by FACS screening as described above.

Genomic DNA libraries were prepared from DNA extracted from the secondary transfectants using bacterio- phage EMBL3. Human DNA restriction enzyme fragments from these phages were analyzed by Southern blot analysis to obtain unique sequences expressed in cells containing human chromosome 21. One phage isolate contained a unique 2.1 kb EcoRl restriction fragment which appeared to be from human chromosome 21 on the basis of differential hybridization to genomic DNA from mouse cells vs. human placental DNA and DNA from a human x mouse somatic cell line in which human chromosome 21 was the only human chromosome retained. It was confirmed that this was indeed a piece of human chromosome 21 by its hybridization pattern on pulse field gel blots using DNA from human x mouse and human x hamster somatic cell hybrids with and without human chromosome 21. It has been shown using Northern blot hybridization that this unique fragment hybridizes with human poly-A* RNA large enough to encode the receptor component. This piece was subcloned into plasmid vector pGEM 4 and used to screen a human placenta λgtll cDNA library purchased from Clonetech Laboratories (Palo Alto, Calif.). Seven cDNA clones were isolated. The human DNA has been subcloned into the Bluescript plasmid vector of Stratagene Cloning Systems (San Diego, Calif.) .

Example 1: Isolation of the Putative Type I IFN Receptor Component Gene

^~ Mouse LM TK" cells (lacking the enzyme thymidylate kinase, TK) were transfected by total human

DNA from the T cell leukemia Molt-4 cells using a plasmid pAGO harboring the Herpes Simplex Virus TK' cDNA as the selectable marker in Hypoxanthine-Aminopterin-Thymidine (HAT) selective medium. LM TK* cells were exposed to 200 U/ml human IFN-/31 and stained with monoclonal antibody anti-H2K ^k 12-41 (see Rosa, et al. (1986), Interferon. 7, 47-87, Academic Press, incorporated by reference) and FITC-(Fab') rabbit antimouse IgG. Using a cytofluorograp 50HH (Orthodiagnostics) , the 10% most fluorescent cells were sterilely sorted, regrown and subjected to a second round of selection. Preliminary experiments established that the twice sorted population was enriched in cells responding to human IFN-alpha or beta (but not gamma) in multiple tests such as induction of the (2*-5') oligo A synthetase, resistance to Vesicular Stomatitis Virus, and ¹²⁵ i-iFN-alpha binding. Clones with this phenotype were derived by dilution.

Polyclonal and monoclonal antibodies to mouse- human hybrids containing only human chromosome 21, which inhibit the binding and response to human (but not mouse) IFN (see Razziudin et al., cited above), blocked the H2 increase by human IFN in the positive clones indicating the presence of the human IFN surface receptor. DNA of one such primary transformant was transfected into new LM TK" cells and secondary transformants were obtained using pAGO for selection. Secondary transformants were again subjected to FACS selection as above and after three cycles, clones were derived by dilution and human IFN responders identified by the tests described above. DNA was prepared from subclone cultures which were simulta¬ neously tested for (2'-5') oligo A synthetase induction b human IFN. Subcloning or repeated FACS selection was found to be needed to maintain consistent levels of the

phenotypic response. One clone, named 011-19-26 was selected for further analysis.

Southern blot analysis with a total human DNA probe showed that the content of human DNA in some of the 5 clones was low enough to warrant attempts at molecular cloning of the transforming DNA. A genomic library was prepared from clone 011-19-26 using the EMBL-3 vector, an screened with total human DNA as probe. About 20 EMBL-3 phages containing human repetitive DNA were purified. ¹⁰ From two of the phages giving the strongest signal (15-4 and 6-40), BamHI restriction fragments of 2.1 and 1 kb containing no repetitive DNA were isolated. Southern blo analysis of genomic DNA confirmed that these fragments hybridized with human but not mouse DNA. The majority of

15 phages with human DNA from the 011-19-26 library hybrid¬ ized with the BamHI 2.1 kb fragment (probe 6-40-3) indi¬ cating that this is a major constituent of the human DNA in the 011-19-26 cells. The chromosomal origin of the 6- 40-3 probe was determined by hybridization to Southern DN

20 blots from mouse-human and hamster-human hybrids, which confirmed that the 6-40-3 probe identifies a 300 kb Notl fragment originating from chromosome 21. The 6-40-3 BamH fragment of 2.1 kb overlaps two EcoRI fragments of 2 and -5 kb, the total length of the insert in phage 6-40 being 16 kb. DNA from transformant subclones which either respond to human IFN or lack the phenotype were probed with 6-40- 3; only responders showed specific hybridization to the probe. ₀ To determine whether the 6-40-3 probe was part of a transcription unit, we screened a lambda gtll cDNA library from human placenta with this probe. Seven posi¬ tive cDNA clones were identified in the 2xl0 ⁶ phages. On clone (L7) has an EcoRI insert of 3.9 kb indicating a lon 5 transcription unit. This was confirmed on Northern blots where the same probe detected a 4.5 kb RNA, and a larger 5.1 kb band. The EcoRI insert of cDNA clone L7 of 3.9 kb was subcloned in a Bluescript vector (Stratagen Co.) and

E. coli TCI competent bacteria were transformed with this L7-BS2.vector. This transformed bacteria was deposited with the Collection Nationale de Cultures de Microorgan- ismes (C.N.C.M.), Paris, France, on November 14, 1988, under the accession number C.N.C.M. 1-816, in accordance with the Budapest Treaty. Subcultures will be made avail¬ able to the public as required by the Treaty, but no license can be inferred from the deposit.

The subcloning was carried out in two orienta- tions. T7 RNA polymerase transcripts translated in reticulocyte lysates produced a 100 kd protein in one orientation but not the other. A 62 kD protein was made by the same transcripts producing the 100 kD protein, and may be translated from an internal ATG codon.

Example 2: Characterization of the Cloned Putative IFN Type I Receptor Component

The nucleotide sequence of the L7 cDNA was established by sequencing nested deletions from both strands in the Bluescript vector. The sequence shown in Fig. 1 indicates an RNA of 3870 nucleotides followed by a poly A tail. An open reading frame starts at nucleotide 899 and ends at 3253 encoding a protein of 785 amino acids. The location of the coding region 899-3253 was confirmed by translation of deletion transcripts in reticulocyte lysates.

The length of the protein predicts a molecular weight of 86,000 which is close to the lower values calcu¬ lated by gel electrophoresis. From the electrophoresis of the translation product of the L7 cDNA in reticulocyte lysate, it appears that the protein migrates more slowly than theoretically expected, with an apparent molecular weight of 100,000. This can be explained by the abundance of acidic amino acids in the sequence.

An interesting feature of the L7 protein is a region of 12 -contiguous serines flanked by 6 glutamic acids (bases 1710-1760 in Fig. 1) . This region could be

phosphorylated. A potential transroembrane region with 1 uncharged amino acids is present from nucleotides 3056 t 3100 in Fig. 1. A hydrophobic signal peptide is observe at the N-terminus of the protein. These features are in line with the location of this protein in the cell mem¬ brane as expected for the IFN type I receptor. There is no significant homology between the L7 protein sequence and the IFN- ₇ receptor or any other known gene.

Since final proof of identity requires demon¬ stration of acquired sensitivity to human IFN in non-hum cells transfected with this cDNA clone, transient and stable transfections of the cDNA gene fused to the SV40 early promoter were analyzed in various non-human cell lines.

To test the biological activity of the protein encoded by the L7 cDNA, an expression vector pSVE3-L7 was constructed containing the entire L7 cDNA fused to the Simian Virus 40 early gene promoter. For transient expression this plasmid was transfected into mouse cells and 24 hours later the cells were treated by 500 U/ml human IFN-01 or left untreated (Benech, P., et al. (1987) Mol. Cell. Biol.. 7. pp. 4498-4504) . After another 24 hours cell extracts were prepared and (2'-5') oligo A synthetase was measured as described (Revel M. et al.

(1981), Meth. Enzvmol.. 79. p. 143). The (2'-5') oligo A synthetase activity in transfected cells treated by human IFN minus that in non-treated cells was calculated. As shown in Table 1, this increase was on the average 4.5 fold higher in the cells which had received the expressio vector containing the L7 cDNA versus the cells which had received the empty expression vector. The latter showed no significant increase over cells which had received no DNA at all (Table 1) . Stable transformants of Chinese Hamster Ovary cells (CHO) were prepared by cotransfection of the cDNA fused to the SV40 early promoter and vectors containing the DHFR gene. Clonal cell populations were isolated fro

transfection after appropriate selection and these clonal lines were screened for the presence of the cDNA and for response to human IFN-0 as shown by induction of (2'-5*) A synthetase. Clones containing L7 DNA by Southern Blot analysis and responding to 100 U/ml human IFN-0** by (2'- 5')-A synthetase induction higher than CHO-DHFR* cells, were isolated and subjected to amplification by stepwise selection with methotrexate (MTX) . In clones resistant to 50 nM or, better, 300 nM MTX, L7 RNA expression becomes detectable by Northern blots of total RNA. The response to 100 U/ml human IFN-0, is 4-5 fold higher in the 50 nM MTX clones than in CHO cells containing pSVDHFR only, which were similarly amplified (Table 2). At 300, 750 and 1000 nM MTX, the CHO-DHFR+L7 clones showed an increase in the response to human IFN-0 as compared to the highest response in CHO-DHFR control clones which were similarly treated by MTX (Fig. 4) . The amplitude of the response to human IFN-0 increases, therefore, with gene amplification by a factor of 8 fold. The response of such amplified clones to 100 U/ml human IFN was about 15% of that to 20 U/ml hamster IFN.

The present inventors have concluded that, although expression of the L7 cDNA under SV40 promoter control leads to increased response of the cells to human IFN, the amplitude of the response remains low. This may be due either to problems in expression of the protein or to the fact that the L7 protein is only one part of the IFN receptor system. The latter possibility seems more likely as significant high affinity binding of human - ⁵ 1- IFN-α to the CHO transformants was not able to be demon¬ strated. If the receptor is composed of more than one chain, binding affinity may be very low in the presence of only one of the protein chains. That the L7 protein is a component of the IFN receptor which is not directly responsible for the species specific binding of human IFN is further shown by the fact that in the amplified CHO-pSVL7 clones, the response to

hamster IFN is also increased. Fig. 5 shows that at low concentration of hamster or human IFN, which produce low stimulation of (2'-5') A synthetase activity in the con¬ trol CHO-DHFR clone (abscissa) , a CHO-pSVL7 clone (300 n MTX) gave a 5 fold higher response with either of the tw IFN species (the straight line in Fig. 5 is the expected line if both clones respond equally) . At higher IFN concentrations producing higher induction of (2'-5') A synthetase, the difference between the CHO-pSVL7 and CHO DHFR clone was less marked. The present inventors have concluded that having more L7 protein on the surface allows the cells to respond better to low concentrations of IFN. This result is in agreement with the hypothesis that L7 is one of the chains of the type I IFN receptor system, which acts in transducing the signal within the cell.

Example 3: Antibodies to L7 protein show the protein on the surface of cells

A 27 amino acid long peptide near the amino- terminus of the L7 protein (Fig. 3) , was synthesized, conjugated to KLH and used to raise rabbit antisera. Th antibodies were purified on a column of immobilized peptide and used in Western blots. In Daudi and U937 cells a protein of 130 kDa is detected in Triton X-100- deoxycholate lysates and in membrane preparations. This is probably a glycosylated form of L7 which has an N- glycosylation site near the N-terminal region. However, immunoprecipitation was not possible with the peptide antibodies.

The L7 cDNA was fused through codon 113 to the protein A gene in vector pRIT-2 (Pharmacia Fine Chemicals and expressed in E.coli. The fused protein PSN and PN2 (Fig. 3) , purified on an immunoglobulin column, were used for immunization to obtain rabbit polyclonal antibodies t the L7 protein. These antibodies immunoprecipitated from human Daudi cells labelled for 7 hours with ³⁵ -methionine

major bands of 80-85 kDa and minor bands of about 130 kDa. The 80-85 kDa bands may be forms of the protein which have lost part of the N-terminal region recognized by the above anti-peptide antibody. With the rabbit antiserum against L7 (anti-PSN and PN2) it is possible to demonstrate that the L7 protein is on the surface of human lymphoblastoid cells (Daudi) . Fig. 6 shows a fluorescence activated cell sorter (FACS) analysis demonstrating the presence of L7 on the surface of Daudi cells. The right panel shows that with Rhod- amine isothiocyanate labeled protein A alone or with normal rabbit serum a peak of fluorescence is obtained at 60 arbitrary units. However, with added anti-L7 serum the peak of fluorescence is displaced to higher values (aver¬ age 110) showing that the anti-L7 antibodies bind to the cells and hence more Rhodamine labelled Protein A is attached. The form of the curves shows that all the Daudi cells display the L7 protein on their surface.

The anti-L7 antibodies did not inhibit the binding of ¹²⁵ 1-IFN-α ₂ to human lymphoblastoid Daudi cells. Fig. 7 shows that the binding was inhibited by cold IFN-0, by antibodies against mouse-human hybrid cells containing only chromosome 21 (<chr.21) , but not by anti- L7 (<L7) , nor by normal rabbit serum (NRS) or normal mouse serum (NMS) . Therefore, the L7 protein seems not to be directly involved in the binding of IFN-α to these human cells. This would distinguish the L7 protein from the protein cross-linked to IFN-α on human cells (Razziudin et al, locus cited) . It is not excluded that the protein which binds IFN-α may be different from that which binds IFN-0. Since the L7 protein was identified by its action on the response of cells to IFN-0, there has been an attempt to see if L7 could be cross-linked to IFN-0. Since radio-iodination of active IFN-0 was not success¬ ful, cold IFN-0 was cross-linked to Daudi cells by suber- imidate treatment and the electrophoretic mobility of the L7 protein (detected by immunoblots with anti-peptide) was

analyzed to determine if it decreased by this treatment. In one experiment, a slower 150 kDa band was observed in addition to the 130 kDa L7 protein, but two other experi ments failed to show this phenomenon.

Conclusions:

Identification of the L7 protein as a protein which modulates cellular response to Type I interferon a a component of the type I IFN receptor system is based o the following:

1. Transient expression of the L7 cDNA causes mouse cel to acquire a higher response to human IFN-0.

2. Stable transformants of hamster CHO cells with the L7 cDNA show an increased response to IFN. The increase is correlated with amplification of the integrated L7 DNA. The increased response is seen with human ham¬ ster IFN and is more marked at low concentrations of IFN, suggesting L7 augments the affinity of the cells to IFN.

3. The L7 protein is on the surface of human cells.

4. The gene encoding the L7 protein is on chromosome 21 in the region (q22) assigned to the type I receptor gene(s) by somatic cell genetics.

As in the case of other cytosines (e.g., IL-2, IFN- ) , the type I IFN receptor is probably formed by several different protein chains. The L7 protein is probably one component of the type I IFN receptor system and involved in the response of cells to IFN. It is probably not the chain of the receptor system which binds human IFN in a species specific way. It is more likely t be a chain of the receptor increasing the signal produced by the binding of IFN.

The identification of the L7 protein allows one to study its structure-function relationship, to produce modified L7 protein with the aim of increasing or decreas ing the response of cells, tissues and whole organisms to

interferon of the α or 0 type, and to produce by recombi¬ nant DNA technology soluble or lipid-bound L7 protein which can act as a competitor against IFN. These inven¬ tions can be applied to pathological situations in which one wishes to either enhance or to reduce the response to these types of interferons (e.g., autoimmune processes, chronic and acute infections, tumors and leukemias, etc.) .

For example, L7 protein fragments., or analogues which interact with IFN, or antidiotypic antibodies raise against antibodies to the L7 protein, may be injected int a patient, acting as "decoys" for any interferon molecule present. Gene therapy with functional receptor genes can be used to provide more receptors and thereby increase th response to the IFN.

It should be understood that while a single specific protein, and the genetics material therefor, is described herein, the present invention is not limited thereto. The protocol for extracting genetic material coding for a protein which modulates cellular response to interferon has been shown to be effective by the experi¬ mental results described herein. Those of ordinary skill in this art will understand that if this protocol is repeated, different proteins also modulating cellular response to interferon, which proteins are likely to comprise the other chains of the Type I receptor system, may be obtained. While the protocol is specific to pro¬ teins which cause increased H2 antigen production, it is fully expected that additional proteins other than the specific one described herein will also have such proper¬ ties and the present invention is intended to generically cover all of them.

Furthermore, the process of the present inven- tion can also be repeated using other markers of interfer on activity, e.g., other changes of cell surface characteristics caused by interferon, for detection of th

interferon receptor system components. Such characteris¬ tics may include increased 02 microglobulin or class II MHC antigens. When these techniques are used, proteins which modulate cellular response to interferon will also be obtained and such purified proteins and the genetic material encoding same are also part of the present invention.

It is known that any protein obtained in accor¬ dance with the present invention modulates cellular response to interferon. This is true by definition as it is that property which guides the selection of the geneti material. While it is thought that all such proteins are part of the interferon receptor system, applicant does no wish to be limited to such theory. It is known, however, that such proteins are not merely interferon inducers in view of their activity in responding to interferon, rathe than creating it.

The foregoing description of the specific embodiments will so fully reveal the general nature of th invention that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and therefore such adaptations and modifications are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation.

TART.T: I _ {-ΛMτm-n- o V -L

Fold increase by pSVE3-L7 over pSVE3 vector alone in th response of transfected cells to human IFN-01 (500 U/ml

Fold increase of pSVE3 vector-trans erred cells over no transfected cells.

TABLE 2: (2 ^« -5') A Synthetase Induction in CHO Cells Transformed by pSVDHFR Alone (Control) or by pSVDHFR and PSVE3-L7 cDNA. Amplified bv 50 nM MTX

CHO Number of (2'-5') A synthetase activity

Transformed Clones " ³² p-(A2 ^, p) _n A, cpm by Vector Tested No IFN +HU-IFN-0 Inductio

DHFR 5 433±109 1160+433 727±51

DHFR+7CDNA 6 386±105 3673+931 3286±88

Human IFN-0 used at 100 units/ml.