Cell, Vol. 60, 225-234.
January
26, 1990, Copyright
0 1990 by Cell Press
Genetic Transfer of a Functional Human Interferon Receptor into Mouse Cells: Cloning and Expression of Its cDNA Gilles Uz6, Georges Lutfalla,’ and Ion Gresser Laboratoire d’Oncologie Virale CNRS UPR 274 7, rue Guy MBquet 94 801 Villejuif Cedex France
A cDNA coding for the human interferon a receptor has been cloned using a gene transfer approach. This consists of transferring human DNA to mouse cells and selecting for cells sensitive to human interferon a. The transfected cells expressed the human interferon a receptor, and a 5 kb human DNA was isolated from a secondary transfectant. This DNA detects an mRNA present in human cells and was used to clone a 2.7 kb cDNA from a library constructed from human Daudi cells. The sequence of this cDNA is presented. It codes for a glycoprotein of 557 amino acids with an N-terminal hydrophobic region and a single transmembranespanning segment. Mouse cells expressing the cDNA become sensitive to the antiviral activity of and express binding sites for human interferon a, demonstrating that the cloned cDNA encodes a functional human interferon a receptor. Introduction Interferon (IFN) a is a family of related proteins defined by their capacity to render cells resistant to viral infection (Isaacs and Lindenmann, 1957). Since its discovery, many more activities have been described. These multiple effects include modulation of many differentiated cellular activities, inhibition of cell growth, and antitumor effects (Gresser, 1989). The presence of IFN in several organs of normal individuals suggests that it is important in normal physiology (Tovey et al., 1987), especially in maintaining pregnancy (Roberts, 1989). Human IFNs-a are encoded by a family of at least 23 functional genes. They show a high level of sequence homology at both the amino acid and nucleotide levels and approximately 25% amino acid identity with IFN-P (Weissmann and Weber, 1986). IFNs exhibit target cell specificities. Human IFN-a subspecies show considerable variation in their specific activities on human cells and only a weak cross-reaction on mouse cells (Rehberg et al., 1982). In spite of the high structural homology between different subspecies of human IFN-a, there exists a large variation in their specificity on mouse cells. For example, the human IFN-a6 does not cross-react at all on mouse cells, whereas some IFN hybrids constructed between human * Present address: Unite de Gt?nBtique et Biochimie du DBveloppement, lnstitut Pasteur, 25, rue du Dr. Roux, 75 724 Paris Cedex 15, France.
a
IFN-aB and IFN-aD or IFN-aA and IFN-aD have significant activity on mouse cells (Streuli et al., 1981; Meister et al., 1986). Furthermore, it has been shown that single amino acid changes can render human IFN-aA active on mouse cells (Weber et al., 1987). It is now accepted that, like most polypeptide humoral factors such as lymphokines and growth factors, IFN actions are mediated through interactions with receptors present at the cell surface (Aguet, 1980; Mogensen and Bandu, 1983; Hannigan and Williams, 1986). Competitive binding of human IFNs on human cells seems to indicate that all human IFNs-a and -b share a common receptor, while IFN-y is recognized by a distinct receptor system (Aguet et al., 1984; Merlin et al., 1985). Human IFN-a receptors have been characterized in some detail from several cell lines (for reviews see Aguet and Mogensen, 1983; Mogensen et al., 1989). Affinity labeling using iodinated IFN-a and cross-linking agents or ligand blotting showed that IFN could be linked with a 95-100 kd component present in the plasma membrane (Vanden Broecke and Pfeffer, 1988; Eid et al., 1988; Uzb et al., 1988; Schwabe et al., 1988). The presence and the number of IFN receptors on the cell surface do not reflect the sensitivity of cells to IFN. In fact, IFN receptors are present in all tissues and even on the surface of most IFN-resistant cells (Aguet and Mogensen, 1983). The structure of the IFN-a receptor has not yet been elucidated, mostly because its number on the cell surface is very limited: 100 to 5000, involving difficulties in the purification of the receptor and the production of anti-receptor antibodies. As the biochemical analysis of the receptor has remained mainly unfruitful, we have chosen to clone the human IFN-a receptor gene by transfer of human DNA to mouse cells and by selection of a mouse cell expressing the human receptor. There are several reasons to suppose that such a cell would be sensitive to human IFN-a and that a functional selection might be used to isolate it. Since the specific activities of human IFNs-a on human cells as well as on murine cells (when the IFN exhibits a cross-reaction) are related to their affinities for their receptors (Aguet et al., 1984; Uzh et al., 1985; Meister et al., 1986), the mode of action of human and mouse IFN receptors must be relatively similar. It is reasonable to assume that the species specificity of human IFNs-a is due to an absence of appropriate binding contacts for the murine receptor and that a mouse cell capable of responding to mouse IFN should be sensitive to human IFN provided it expresses the human IFN receptor. In addition, somatic genetic studies indicate that mouse-human cell hybrids retaining human chromosome 21 are sensitive to human IFNs (Slate et al., 1978). Additional data support the assignment of the human IFN receptor to chromosome 21, suggesting that the human receptor can be functional in a mouse context (Epstein et al., 1982). Here we report the isolation of mouse cells transfected with human DNA and selected for expression of a functional human IFN receptor at their surface. We present the
Cell 226
molecular cloning of a cDNA containing the entire coding sequence of the human IFN-a receptor and its expression in mouse cells. Results Selection of Mouse BTG 9A Cell Transfectants Sensitive to Human IFN-aB Mouse BTG 9A cells were cotransfected with a 1 :l ratio of high molecular weight human Daudi genomic DNA and a library of full-length human cDNA cloned in a mammalian expression h phage vector containing the bacterial gene neo, which confers resistance to G418 (Okayama and Berg, 1985). This cotransfection system in which expression of the human IFN receptor can be derived from genomic human DNA and/or from human cDNA was used to increase the chances of isolating transfectants that express the receptor. Stable transfectant clones were selected in G418 medium with a frequency of 1O-z-1O-3. To detect cell clones sensitive to human IFN-a, transfected clones were treated with 0.15 uglml (30,000 U/ml) human IFN-aB and infected with vesicular stomatitis virus (VSV) as described in Experimental Procedures. At this concentration, mouse BTG 9A cells are insensitive to human IFNaB, but a transfectant clone expressing the human IFN-a receptor gene should acquire an antiviral state. To neutralize the large quantity of VSV produced by the irrelevant cell clones, which would otherwise overcome the antiviral state induced in the transfected cell clones that express the receptor, a rapid virus adsorption on cells was followed by treatment with rabbit anti-VSV antiserum to neutralize excess virus, and the cytopathic effect was left to develop in semi-solid agar medium containing antibody to VSV. Surviving cell clones were isolated individually. To avoid chronic VSV infection, cell clones were treated with mouse IFN-a/6, and the anti-VSV antiserum was maintained in G418 medium for 1 week. Cell clones were then retested for their sensitivity to human IFN-aB against VSV or encephalomyocarditis virus (EMC) infections. Most of these transfected clones were insensitive to human IFN-aB and probably represented the background of the system. However, one cell clone was clearly sensitive to human IFN-aB and it was subcloned. We refer to it as lOBH7. We have determined the sensitivity of lOBH7 cells to various human and murine IFNs and compared the behavior of these cells with that of the parental murine BTG 9A cells. Table 1 shows the activity of mouse IFN-a/6, human IFN-aB, human IFN-6, and human IFN-y assayed on parental mouse BTG 9A cells and the transfected clone lOBH7, using VSV as virus challenge. The same results were obtained using EMC virus. lOBH7 cells are as sensitive as the parental BTG 9A cells to murine IFN-a/6. On the other hand, lOBH7 cells manifest at least a 64,000-fold increase in sensitivity to human IFN-aB compared with parental cells. We have also observed an 8-fold increase in the activity of human IFN-8 on lOBH7 cells. We have not detected any antiviral activity of human IFN-y on either parental BTG 9A cells or on the transfected lOBH7 clone. The specific antiviral activity of IFN-aB (4.7 x 10” Ulmg) on lOBH7 cells is in the range of specific activities of human
Table 1. Antiviral Titers of IFN Preparations BTG 9A or lOBIf Cells BTG 9A Mouse Human Human Human
12.8 x 1Oe
January
26, 1990, Copyright
0 1990 by Cell Press
Genetic Transfer of a Functional Human Interferon Receptor into Mouse Cells: Cloning and Expression of Its cDNA Gilles Uz6, Georges Lutfalla,’ and Ion Gresser Laboratoire d’Oncologie Virale CNRS UPR 274 7, rue Guy MBquet 94 801 Villejuif Cedex France
A cDNA coding for the human interferon a receptor has been cloned using a gene transfer approach. This consists of transferring human DNA to mouse cells and selecting for cells sensitive to human interferon a. The transfected cells expressed the human interferon a receptor, and a 5 kb human DNA was isolated from a secondary transfectant. This DNA detects an mRNA present in human cells and was used to clone a 2.7 kb cDNA from a library constructed from human Daudi cells. The sequence of this cDNA is presented. It codes for a glycoprotein of 557 amino acids with an N-terminal hydrophobic region and a single transmembranespanning segment. Mouse cells expressing the cDNA become sensitive to the antiviral activity of and express binding sites for human interferon a, demonstrating that the cloned cDNA encodes a functional human interferon a receptor. Introduction Interferon (IFN) a is a family of related proteins defined by their capacity to render cells resistant to viral infection (Isaacs and Lindenmann, 1957). Since its discovery, many more activities have been described. These multiple effects include modulation of many differentiated cellular activities, inhibition of cell growth, and antitumor effects (Gresser, 1989). The presence of IFN in several organs of normal individuals suggests that it is important in normal physiology (Tovey et al., 1987), especially in maintaining pregnancy (Roberts, 1989). Human IFNs-a are encoded by a family of at least 23 functional genes. They show a high level of sequence homology at both the amino acid and nucleotide levels and approximately 25% amino acid identity with IFN-P (Weissmann and Weber, 1986). IFNs exhibit target cell specificities. Human IFN-a subspecies show considerable variation in their specific activities on human cells and only a weak cross-reaction on mouse cells (Rehberg et al., 1982). In spite of the high structural homology between different subspecies of human IFN-a, there exists a large variation in their specificity on mouse cells. For example, the human IFN-a6 does not cross-react at all on mouse cells, whereas some IFN hybrids constructed between human * Present address: Unite de Gt?nBtique et Biochimie du DBveloppement, lnstitut Pasteur, 25, rue du Dr. Roux, 75 724 Paris Cedex 15, France.
a
IFN-aB and IFN-aD or IFN-aA and IFN-aD have significant activity on mouse cells (Streuli et al., 1981; Meister et al., 1986). Furthermore, it has been shown that single amino acid changes can render human IFN-aA active on mouse cells (Weber et al., 1987). It is now accepted that, like most polypeptide humoral factors such as lymphokines and growth factors, IFN actions are mediated through interactions with receptors present at the cell surface (Aguet, 1980; Mogensen and Bandu, 1983; Hannigan and Williams, 1986). Competitive binding of human IFNs on human cells seems to indicate that all human IFNs-a and -b share a common receptor, while IFN-y is recognized by a distinct receptor system (Aguet et al., 1984; Merlin et al., 1985). Human IFN-a receptors have been characterized in some detail from several cell lines (for reviews see Aguet and Mogensen, 1983; Mogensen et al., 1989). Affinity labeling using iodinated IFN-a and cross-linking agents or ligand blotting showed that IFN could be linked with a 95-100 kd component present in the plasma membrane (Vanden Broecke and Pfeffer, 1988; Eid et al., 1988; Uzb et al., 1988; Schwabe et al., 1988). The presence and the number of IFN receptors on the cell surface do not reflect the sensitivity of cells to IFN. In fact, IFN receptors are present in all tissues and even on the surface of most IFN-resistant cells (Aguet and Mogensen, 1983). The structure of the IFN-a receptor has not yet been elucidated, mostly because its number on the cell surface is very limited: 100 to 5000, involving difficulties in the purification of the receptor and the production of anti-receptor antibodies. As the biochemical analysis of the receptor has remained mainly unfruitful, we have chosen to clone the human IFN-a receptor gene by transfer of human DNA to mouse cells and by selection of a mouse cell expressing the human receptor. There are several reasons to suppose that such a cell would be sensitive to human IFN-a and that a functional selection might be used to isolate it. Since the specific activities of human IFNs-a on human cells as well as on murine cells (when the IFN exhibits a cross-reaction) are related to their affinities for their receptors (Aguet et al., 1984; Uzh et al., 1985; Meister et al., 1986), the mode of action of human and mouse IFN receptors must be relatively similar. It is reasonable to assume that the species specificity of human IFNs-a is due to an absence of appropriate binding contacts for the murine receptor and that a mouse cell capable of responding to mouse IFN should be sensitive to human IFN provided it expresses the human IFN receptor. In addition, somatic genetic studies indicate that mouse-human cell hybrids retaining human chromosome 21 are sensitive to human IFNs (Slate et al., 1978). Additional data support the assignment of the human IFN receptor to chromosome 21, suggesting that the human receptor can be functional in a mouse context (Epstein et al., 1982). Here we report the isolation of mouse cells transfected with human DNA and selected for expression of a functional human IFN receptor at their surface. We present the
Cell 226
molecular cloning of a cDNA containing the entire coding sequence of the human IFN-a receptor and its expression in mouse cells. Results Selection of Mouse BTG 9A Cell Transfectants Sensitive to Human IFN-aB Mouse BTG 9A cells were cotransfected with a 1 :l ratio of high molecular weight human Daudi genomic DNA and a library of full-length human cDNA cloned in a mammalian expression h phage vector containing the bacterial gene neo, which confers resistance to G418 (Okayama and Berg, 1985). This cotransfection system in which expression of the human IFN receptor can be derived from genomic human DNA and/or from human cDNA was used to increase the chances of isolating transfectants that express the receptor. Stable transfectant clones were selected in G418 medium with a frequency of 1O-z-1O-3. To detect cell clones sensitive to human IFN-a, transfected clones were treated with 0.15 uglml (30,000 U/ml) human IFN-aB and infected with vesicular stomatitis virus (VSV) as described in Experimental Procedures. At this concentration, mouse BTG 9A cells are insensitive to human IFNaB, but a transfectant clone expressing the human IFN-a receptor gene should acquire an antiviral state. To neutralize the large quantity of VSV produced by the irrelevant cell clones, which would otherwise overcome the antiviral state induced in the transfected cell clones that express the receptor, a rapid virus adsorption on cells was followed by treatment with rabbit anti-VSV antiserum to neutralize excess virus, and the cytopathic effect was left to develop in semi-solid agar medium containing antibody to VSV. Surviving cell clones were isolated individually. To avoid chronic VSV infection, cell clones were treated with mouse IFN-a/6, and the anti-VSV antiserum was maintained in G418 medium for 1 week. Cell clones were then retested for their sensitivity to human IFN-aB against VSV or encephalomyocarditis virus (EMC) infections. Most of these transfected clones were insensitive to human IFN-aB and probably represented the background of the system. However, one cell clone was clearly sensitive to human IFN-aB and it was subcloned. We refer to it as lOBH7. We have determined the sensitivity of lOBH7 cells to various human and murine IFNs and compared the behavior of these cells with that of the parental murine BTG 9A cells. Table 1 shows the activity of mouse IFN-a/6, human IFN-aB, human IFN-6, and human IFN-y assayed on parental mouse BTG 9A cells and the transfected clone lOBH7, using VSV as virus challenge. The same results were obtained using EMC virus. lOBH7 cells are as sensitive as the parental BTG 9A cells to murine IFN-a/6. On the other hand, lOBH7 cells manifest at least a 64,000-fold increase in sensitivity to human IFN-aB compared with parental cells. We have also observed an 8-fold increase in the activity of human IFN-8 on lOBH7 cells. We have not detected any antiviral activity of human IFN-y on either parental BTG 9A cells or on the transfected lOBH7 clone. The specific antiviral activity of IFN-aB (4.7 x 10” Ulmg) on lOBH7 cells is in the range of specific activities of human
Table 1. Antiviral Titers of IFN Preparations BTG 9A or lOBIf Cells BTG 9A Mouse Human Human Human
12.8 x 1Oe
