key: cord-022226-qxp0gfp3
authors: Meager, Anthony
title: Interferons Alpha, Beta, and Omega
date: 2007-09-02
journal: Cytokines
DOI: 10.1016/b978-012498340-3/50026-9
sha: 
doc_id: 22226
cord_uid: qxp0gfp3

Interferon alpha (IFN-α) is a mixture of closely related proteins, termed “subtypes,” expressed from distinct chromosomal genes. Interferon β (IFN-β) is a single protein species and is molecularly related to IFN-α subtypes, although it is antigenically distinct from them. IFN omega (IFN-ω) is antigenically distinct from IFN-α and IFN-β but is molecularly related to both. The genes of three IFN subtypes are tandemly arranged on the short arm of chromosome 9. They are transiently expressed following induction by various exogenous stimuli, including viruses. They are synthesized from their respective mRNAs for relatively short periods following gene activation and are secreted to act, via specific cell surface receptors, on other cells. IFN-α subtypes are secreted proteins and as such are transcribed from mRNAs as precursor proteins, pre-IFN-α, containing N-terminal signal polypeptides of 23 hydrophobic amino acids (aa) mainly. Pre-IFN-β contains 187 aa, of which 21 comprise the N-terminal signal polypeptide and 166 comprise the mature IFN-β protein. IFN-ω contains 195 aa—the N-terminal 23 comprising the signal sequence and the remaining 172, the mature IFN-ω protein. At the C-terminus, the aa sequence of IFN-ω is six residues longer than that of IFN-α or IFN-β proteins. IFN-α, as a mixture of subtypes, and IFN-ω may be produced together following viral infection of null lymphocytes or monocytes/macrophages. The biological activities of IFNs are mostly dependent upon protein synthesis with selective subsets of proteins mediating individual activities. IFNs can also stimulate indirect antiviral and antitumor mechanisms, depending upon cellular differentiation and the induction of cytotoxic activity.

The phenomenon of viral interference was first described nearly 60 years ago when Hoskins (1935) described the protective action of a neurotropic yellow fever virus against a viserotropic strain of the same virus in monkeys. Although viral interference was further investigated in the 1940s and 1950s, the underlying mechanism was not discovered until 1957 when Isaacs and Lindemann, working at The National Institute for Medical Research (London, UK), isolated a biologically active substance from virally-infected chicken cell cultures that, on transfer to fresh chicken cell cultures, produced a protective antiviral effect (Isaacs and Lindemann, 1957) . The word Interferon (IFN) was coined for this substance. Its discovery aroused considerable scientific and medical interest since by 1957 antibiotics were widely available to control bacterial infections, but, in stark contrast, viral diseases such as influenza, measles, polio, and smallpox were virtually untreatable. Interest was further heightened by many subsequent studies that demonstrated that IFN could be produced by human cells and was active against a broad spectrum of viruses (see Schlesinger, 1959 , for an early review). At that time, IFN was being hailed by the media as a wonder drug, but it soon became clear that IFN was being produced naturally in too small quantifies for that extravagant claim to be immediately confirmed. In fact, the low production of IFN was to bedevil attempts both to characterize it molecularly and evaluate it clinically for many years following its discovery.

Although the protein nature of IFN was recognized at an early stage in its development (see Fantes, 1966 , for an early review), it was only following the introduction of large-scale production methods in the 1970s (Cantell and Hirvonen, 1977) and the simultaneous development of efficient purification procedures (Knight, 1976; Rubinstein et al., 1978) that sufficient amounts of partially pure IFN protein became available for characterization and clinical use. Gradually, it became apparent that IFN was not a single protein and that there were likely to be different types of IFN molecules. However, despite progress in the area of purification and in initial characterization by sequencing N-terminal polypeptides, IFN proteins all but defied full characterization until the advent of recombinant DNA (rDNA) technology in the late 1970s. This technology, spurred on by the pharmaceutical industry's desire to produce pharmacologically active proteins cheaply, revealed that one type of human IFN, now designated IFN-cx, was a mixture of several closely related proteins, termed subtypes, expressed from distinct chromosomal genes . Second and third types of IFN, designated IFN-I3 and IFN-y respectively, have subsequently been "cloned" (Taniguchi et al., 1980; Gray et al., 1982) but, unlike IFN-0q are single protein species. IFN-13 is molecularly related to IFN-~ subtypes but is antigenically distinct from them, whereas IFN-y is both molecularly and antigenically distinct from either IFN-cx subtypes or IFN-I3. (For this reason, IFN-y is considered separately elsewhere in this volume.) Finally, a fourth type of IFN, antigenically distinct from IFN-0~ and IFN-I3, but molecularly related to both, has more recently been cloned and characterized. Rather untypically, this new IFN type has been designated IFN omega (IFN-00) (Adolf, 1987) .

The genes for IFN-0~ subtypes, IFN-13 and IFN-03 are tandemly arranged on the short arm of chromosome 9.

They are only transiently expressed following induction by a variety of exogenous stimuli, including viruses. IFN-0~, IFN-I3 and IFN-00 proteins are synthesized from their respective mRNAs for relatively short periods following gene activation and are secreted to act, via specific cell surface receptors, on other cells. Early studies on the characterization of IFN receptors indicated that IFN-0~ and IFN-I3 were likely to share a common receptor, but it has only been comparatively recently that such receptors have been cloned (Uz~ et al., 1990; Novick et al., 1994) . IFN actions are initiated by activated receptors and cytoplasmic signal transduction pathways, which are now well characterized for the IFN-~/I3 receptor, and manifested following expression of a number of IFNspecific inducible genes. Induction of the antiviral state, which is dependent on such protein synthesis, may now be viewed as just one of the many activities attributed to IFN in general; these activities include inhibition of cell proliferation and immunomodulation (see Pestka et al., 1987 , for a review).

In the 1960s, two types of lFN were defined on the basis of the capacity of their antiviral activity to withstand acidification to pH 2. These were termed type I IFN for acid-stable IFN and type II IFN for acid-labile IFN. Type I IFN included IFN produced by virally infected leukocytes, alternatively known as leukocyte IFN, and IFN produced by virally infected human diploid fibroblasts, alternatively known as fibroblast IFN. Type II IFN, which was only produced by antigenically or mitogenically stimulated human peripheral blood mononuclear cells (PBMC), has often been referred to as immune IFN (Stewart, 1979) .

Antigenic differences were described for leukocyte and fibroblast IFN and these were put on a molecular basis when knowledge of their respective N-terminal amino acid sequences became available (Allen and Fantes, 1980; Knight et al., 1980; Levy et al., 1980; Zoon et al., 1980) . At that time, an international nomenclature committee (Stewart et al., 1980) reviewed the growing evidence for the existence of distinct molecular forms of IFN and introduced the Greek alphabetical system to apply to the then known antigenically distinct types of IFN. Leukocyte IFN was designated IFN-~, fibroblast IFN was designated IFN-13, and immune IFN became IFN-y. However, complications immediately arose when it was revealed, following the cloning of several different leukocyte IFN complementary DNAs (cDNAs) Brack et al., 1981; Goeddel et al., 1981; Streuli et al., 1980) , that leukocyte IFN was heterogeneous and contained many different, molecularly and antigenically related species, now commonly referred to as subtypes. The research group at Hoffmann-La Roche labeled the subtypes produced in E. coli RA, 0~B, 0~C, 0cD, etc. (Evinger et al., 1981; Rehberg et al., 1982) , distinguishing them from natural components of leukocyte IFN , while the Biogen group labeled these recombinant subtypes ix1, 0~2, 0c3, 0~4, etc. , regrettably without an appropriate alphabeticalnumerical correspondence: for example, ~k -~2, 0cD -0~1. However, the numerical system now prevails. When an IFN preparation is a mixture of IFN-~ subtypes, e.g., leukocyte IFN, lymphoblastoid IFN, this is often designated IFN-o~n.

The later cloning of cDNAs encoding IFN-0~-like proteins (Capon et al., 1985; Hauptmann and Swetly, 1985) initially led to the naming of this new IFN as IFN-o~ subclass II, with all of the earlier-characterized IFN-0~ subtypes being reclassified as IFN-tx subclass I. This large, unwieldy nomenclature system has been superseded by the renaming of IFN-~ II as IFN-03; this has been generally accepted with the finding that IFN-03 is antigenically distinct from IFN-~ and IFN-I] proteins and thus qualifies as a separate type of IFN (Adolf, 1987) .

Fortunately, the nomenclature for IFN-I] has remained straightforward since there is only one protein species, at least in humans (Derynck et al., 1980 Taniguchi et al., 1980) .

From the initial cloning of IFN-o~ cDNAs, there has been a plethora of reports on the cloning of new, and sometimes distinct, genomic and eDNA clones, and fairly disparate nomenclatures have arisen. Diaz and Allen (1993) therefore undertook the considerable task of compiling the IFN genes and genomic and cDNA clones from the literature and introduced an arabic-alphabetical system for naming IFN genes to enable their distinction from IFN proteins. Thus, IFN-oc genes became IFNA genes with the addition of a numeral to denote subtype, i.e., IFNA1, IFNA2, etc. (Table 25 .1). The IFN-~ gene became IFNB and, since there is only one gene in humans, it is referred to as IFNB 1. For IFN-03 genes, W has been used; hence IFNW1 (Table 25 .1). Besides genes that are capable of being expressed and translated into IFN proteins, there are a number of pseudogenes which are unable to give rise to IFN proteins, and which in this new nomenclature system are designated by a P, e.g., IFNAP22, IFNWP2, or simply IFNP1 where pseudogenes are clearly IFN-like but cannot be definitely included in any one of the IFNA, IFNB or IFNW gene families (Table 25 .1).

The IFN genomic and eDNA clones have been designated in a variety of ways, as illustrated in Table  25 .1. Pseudogenic or non-translatable clones are normally prefixed with a Greek ~. For the purposes of this chapter, and to reduce the complexity of naming IFN clones and proteins, the nomenclature system adopted by Weissmann and colleagues will be adhered to: IFN-0q, %, {x4, o~, etc.

2.1 NUMBERS, STRUCTURE, AND LOCALIZATION In humans, there are 14 nonallelic IFNA genes, one of which, IFNAP22, is a pseudogene (Table 25 .1). In addition, there are a further four nonallelic pseudogenes that possibly also belong to the IFNA gene family. Probable allelic variants of certain IFNA genes, e.g., IFNA2, are also known to exist Goeddel et al., 1981; Dworkin-Rastl et al., 1982; Emanuel and Pestka, 1993) . This extensive family of IFNA genes are tandemly arranged on the short-arm of chromosome 9 (9p23) and span a region of approximately 400 kb (Owerbach et al., 1981; Shows et al., 1982; Slate et al., 1982; Ullrich et al., 1982) . The IFNB gene and the IFNW gene/pseudogene family are also located in the same region of chromosome 9 (Meager et al., 1979a,b; Owerbach et al., 1981; Henry et al., 1984; Capon etal., 1985) .

There is a high degree of homology among the IFNA genes, but these show much less homology to either IFNB or IFNW genes. Nevertheless, all of these IFN genes share the common feature of being intron-less (Taniguchi et al., 1980; Goeddel et al., 1981; Houghton et al., 1981; Capon et al., 1985) , suggesting a very ancient origin of their common ancestral gene. It has been proposed that the primordial IFN gene arose some 500 million years ago, with the first split occurring around 400 million years ago to yield the first IFNA and IFNB genes (Wilson et al., 1983) . Since then the IFNA gene has evolved and duplicated many times to give rise to the multiple IFNA genes found in present-day animals and man (Mij~ita and Hayashida, 1982; Gillespie and Carter, 1983) . Around 100 million years ago, an IFNA gene appears to have diverged sufficiently from the main group to give rise to the IFNW gene family, which is present in most mammals except mice and dogs (Himmler et al., 1987; Roberts et al., 1992) .

It is not clear what characteristic of IFNA and IFNW genes enabled the numerous reduplication events to occur in comparison to the nonexistent or more limited (some mammals have more than one IFNB gene, e.g., bovines (Wilson et al., 1983) ) reduplication of the IFNB gene (Ohlsson et al., 1985) . It is apparent that gene conversion, as a result of mismatch repair and unequal crossover, contributed significantly to the creation of distinct, but highly homologous, nonallelic IFNA (and IFNW) genes (De Maeyer and De Maeyer-Guignard, 1988) . In most cases, both coding and noncoding regions have diverged, but in the case of IFNA13 the coding region has remained identical to that of IFNA1, although its 5' and 3' flanking regions have diverged (Todokoro et al., 1984) .

The structures of IFNA, IFNB, and IFNW genes are similar. Each gene has a 5' regulatory promoter region upstream from the transcriptional start (cap) site, a IFNA10  IFNA13  IFNA14  IFNAls  IFNA~  IFNA21  IFNAP~   IFN-oq  IFN-o h, LelF-D  IFN-oq (D)  ;~a 2, IFN-A  IFN Adapted and modified from Diaz and Allen (1993) and Allen and Diaz (1996) , which see for specific references for genomic clones and cDNAs. Reproduced with permission of Mary Ann Liebert, Inc., New York, USA.

coding region containing a nucleotide sequence encoding a signal polypeptide of 21-23 mainly hydrophobic amino acids, which is typical for secreted proteins, and consecutively the sequence encoding the mature IFN protein, followed by the 3' flanking noncoding region, which can vary in length up to 450 base pairs (bp) (Figure 25 .1) (Derynck etal., 1980 (Derynck etal., , 1981 Nagata et al., 1980; Streuli et al., 1980; Taniguchi et al., 1980; Degrave etal., 1981; Goeddel etal., 1981; Gross et al., 1981; Lawn et al., 1981a,b; Gren et al., 1984; Capon et al., 1985; Henco et al., 1985) . The 5' flanking region contains a TATA or Hogness box, which delineates the boundary of the upstream promoter, approximately 30 bp from the cap site. Farther upstream are found a number ofhexameric repeat sequences GAAANN, where N can be any base, which in their dimeric or multimeric forms act as binding sites for nuclear transcription factors and repressor molecules (Fujita et al., 1985; Ryals et al., 1985) . (This area is covered in more detail in Section 2.2, Inducers and Transcriptional Control). The 3' flanking regions vary in length and contain several polyadenylation sites and thus can give rise to mRNAs of different lengths (Mantei and Weissmann, 1982; Henco et al., 1985) . They contain above-average numbers of the sequence motifs ATTA or TTATTTAT. Such sequences are common, however, in many other cytokine genes and other genes, such as protooncogenes, that are inducibly and transiently expressed. It has been proposed that these sequences contribute to the relative instability and short half-lives of IFN and cytokine mRNAs (Caput et al., 1986; Shaw and Kamen, 1986 ).

All IFN genes are normally silent and thus require some sort of stimulus to induce expression. A wide range of inducers, including viruses, bacteria, mycoplasma, endotoxins, double-stranded polynucleotides or RNA (dsRNA), and some cytokines, have been shown to efficiently activate transcription of IFN genes (Stewart, 1979; De Maeyer and De Maeyer-Guignard, 1988 ). Such inducers have in general the potential to induce expression of all IFNA, IFNB, and IFNW genes; however, there appears to be cell-and inducer-specific selectivity that governs the type and numbers of IFN genes expressed. For instance, virally induced human diploid fibroblasts produce mainly IFN-13 and only a minor amount of IFN-~ (Havell et al., 1978) , whereas virally induced PBMC produce mainly IFN-a plus IFN-o3 and only a minor amount of IFN-~ (Cantell and Hirvonen, 1977; Adolf et al., 1990) . This has to some extent been confirmed at the mRNA level (Shuttleworth et al., 1983; Hiscott et al., 1984) . Differences have also been reported in the proportions of individual IFN-0~ subtypes produced by different cell types (Goren et al., 1986; Finter, 1991; Greenway et al., 1992) , suggesting that IFNA genes may be differentially expressed. However, the way in which such differential expression is regulated is presently not understood. Transcriptional control of IFNA genes resides in their 5' flanking region, upstream from the cap site. Nucleotide deletions outside of position-117 from the cap site have little impact on transcriptional control, but deletions farther in eliminated induction, indicating that this region,-117 to-1, contained promoter regulatory elements (Ragg and Weissmann, 1983; Weidle and Weissmann, 1983) . These have been further delineated as a purine-rich nucleotide tract between-109 and-64, containing hexameric repeats of GAAANN (GAAA G/C T/C), which appears to be necessary for inducible transcription and which has been termed the "virusregulating element" (VRE) (Ryals et al., 1985) .

Similar studies involving 5' deletions have been carried out with the IFNB gene, and it has been found that 5' sequences within-110 to -1 contain regulatory elements that are required for induction by viruses and dsRNA. The minimum VRE has been localized to-74 to-37 with respect to the cap site (Goodbourn et al., 1986; Goodbourn and Maniatis, 1988) and contains two positive virus-inducible elements, termed positive regulatory domains (PRDI,-77 to-64; PRDII,-66 to -55), and a negative regulatory domain (NRD,-57 to -37) (Figure 25 .1) (Fujita et al., 1985; Fan and Maniatis, 1989; Whittemore and Maniatis, 1990; Nourbakhsh et al., 1993) . In addition, the hexameric repeat sequences GAAANN (also present in IFNA genes) spanning from -110 to-65 contain variants of the PRDI sequence and two further regulatory elements, PRDIII (-90 to-78) and PRDW (-104 to-91) have been identified that are required for a functional VRE in IFNB gene expression in mouse L cells (Dinter and Hauser, 1987; Du and Maniatis, 1992) . PRDI and PRDIII act as binding sites for a nuclear transcription factor, designated "interferon regulatory factor-l" (IRF-1) , whose expression is transiently increased by virus infection and which appears to mediate the activation of transcription of the IFNB gene (Fujita et al., 1988; Harada et al., 1989; Xanthoudakis et al., 1989) . A second virus-inducible factor, designated "interferon regulatory factor-2" (IRF-2), also binds to PRDI but suppresses, rather than activates, transcription (Harada et al., 1989 (Harada et al., , 1990 . PRDII is a binding site for the nuclear transcription factor NFrA3 (Clark and Hay, 1989; Fujita et al., 1989a; Hiscott et al., 1989; Lenardo et al., 1989; Visvanathan and Goodbourn, 1989) , which interacts with the major groove of the DNA (Thanos and Maniatis, 1992) . Additionally, another protein, highmobility group Y/l, also binds to PRDII, interacting with the minor groove of the DNA (Thanos and Maniatis, 1992) . Both factors appear to be necessary for virus induction of the IFNB gene promoter. PRDIV contains a binding site for a protein of the cAMP response element binding protein (ATF/CREB) family of transcription factors (Du and Maniatis, 1992) .

Viral induction of the IFNB gene is thought to occur following activation of pre-existing NFtcB and by de novo synthesis of IRF-1; these nuclear transcription factors bind to the tandemly arranged PRDI and PRDII and act cooperatively to initiate/activate transcription (Leblanc et al., 1990; Lenardo et al., 1989; Visvanathan and Goodbourn, 1989; Fujita et al., 1989b; Watanabe et al., 1991) . Reporter constructs containing PRDI supported by a simian virus 40 (SV40) enhancer, or (GAAAGT)4 , which contains the functional equivalent of dimeric PRDI (N~if et al., 1991) are activated not only by virus but also by overexpression of IRF-1 MacDonald et al., 1990) . However, in most cell lines, overexpression of IRF-1 has led to poor induction of IFNB (and IFNA) genes (Harada et al., 1990; Fujita et al., 1989b) or none at all (MacDonald et al., 1990; Reis et al., 1992) . This has been attributed to the repressive effect of IRF-2, a homologue oflRF-1, which also binds to PRDI (Harada et al., 1989) . In the undifferentiated murine embryonal carcinoma (stem) cell line P19, which is refractory to viral induction oflFNB (and IFNA) genes and which expresses neither IRF-1 nor IRF-2, overexpression of an introduced IRF-1 construct leads to activation of endogenous IFN genes and to activation of reporter plasmids with the IFNB promoter (Harada et al., 1990) . In addition, cell lines permanently transformed with an antisense IRF-1 expression plasmid exhibited strongly reduced IFNB gene inducibility that nevertheless could be restored by transient transformation with an IRF-l-overproducing expression plasmid (Reis et al., 1992) . However, the role oflRF-1 in virus-induced activation of the IFNB promoter remains controversial (Whiteside et al., 1992; Pine et al., 1990) and Ruffner and colleagues (1993) have shown that in murine embryonal stem cells devoid of both IRF-1 gene alleles (IRF-1 ~ ) viral induction of IFNB was only slightly higher in control IRF-1 0/+ differentiated stem cells than that in IRF-1 ~ differentiated cells. This suggests that while IRF-1 at high levels may elicit or enhance induction of IFNB under certain circumstances, it is not essential for viral induction. In cultured mouse fibroblasts devoid of IRF-1, IFNB induction by the synthetic dsRNA molecule poly(I):poly(C) was absent, whereas induction by Newcastle disease virus (NDV) was normal (Matsuyama et al., 1993) . However, IFN induction in vivo by either virus or dsRNA has been found to be unimpaired in IRF-1 ~ mice, indicating that IRF-1 is not essential . It has also become clear recently that the PRDI site can bind factors other than IRF-1 and IRF-2, and these may be more important for regulating IFNB gene activation (Whiteside et al., 1992; Keller and Maniatis, 1991) .

In contrast, targeted disruption of the IRF-2 gene to yield mouse fibroblasts deficient in the repressor IRF-2 has been found to lead to upregulated induction of IFNB following NDV infection (Matsuyama et al., 1993) . This suggests that IRF-2 negatively regulates or represses IFNB gene induction.

The induction of the IFNA1 gene appears to be regulated differently from that of the IFNB gene. IRF-1 is bound poorly by the equivalent PRDI site in the IFNA1 promoter and this promoter also lacks an NF~cB site ( Figure 25 .1) MacDonald et al., 1990) . The IFNA1 gene VRE does contain a hexameric repeat nucleotide sequence (GAAATG)4, designated a "TG-sequence" (MacDonald et al., 1990 ), which appears to mediate virus inducibility when supported by an SV40 enhancer, but which does not respond to IRF-1 (N~if et al., 1991) . It has, however, been reported that overexpression of IRF-1 can induce IFNA genes, at least under special circumstances (Harada et al., 1990; Au et al., 1992) .

The IFNW gene, like IFNA and IFNB genes, is virus inducible and has structural features in its 5' promoter region similar to those in IFNA/B promoters ( Figure  25 .1). In particular, hexameric repeat units are present, but are organized differently from those present in IFNA/B genes (Hansen et al., 1991; Roberts et al., 1992) . However, the regulation of transcription of the IFNW gene has not been studied in detail.

The 14 IFN-(X subtypes are secreted proteins and as such are transcribed from mRNAs as precursor proteins, pre-IFN-(X, containing N-terminal signal polypeptides of 23 mainly hydrophobic amino acids (Figure 25 .2). The signal polypeptide is cleaved off before "mature" IFN-(X molecules are secreted from the cell. From their cErNA sequences, mature IFN-(X subtypes have been predicted to contain 166 amino acids (except IFN-%, 165 amino acids) Nagata et al., 1980; Goeddel et al., 1981; Lawn et al., 1981a,b; Gren et al., 1984) . The calculated molecular mass of recombinant IFN-(x subtypes is approximately 18.5 kDa, although apparent molecular masses of leukocyte-derived IFN-(X subtypes in sodium dodecyl sulphate (SDS)-polyacrylamide gels vary between 17 and 26 kDa, possibly owing to variable processing of C-terminal amino acids (Le W et aL, 1981) and post-translational modifications. The amino acid sequences of IFN-(x subtypes are highly related, with complete identity at 85 of the 166 amino acid positions (Langer and Pestka, 1985; De Maeyer and De Maeyer-Guignard, 1988) . This is illustrated in Figure 25 .2, where the amino acid sequences of the subtypes are compared to an idealized consensus sequence. Many of the positions where amino acids differ from subtype to subtype are conservative substitutions. Interestingly, IFN-(X subtypes contain four cysteine residues whose positions (1, 29, 99, and 139) are highly conserved (Figure 25 .2). These four cysteines form disulfide bridges (1-99, 29-139) which induce folding of the IFN-(X molecule (Figure 25 .3) and whose integrity is essential for biological activity (Morehead et al., 1984) . IFN-(X subtypes are predicted to contain a high proportion (-60%) of (x-helical regions and are folded to form globular proteins (Zoon and Wetzel, 1984) . It has not yet proved possible to apply x-ray crystallographic techniques to IFN-(X subtypes, or to human IFN-~3, but the three-dimensional crystal structure of recombinant mouse IFN-[3, which is approximately 60% related in amino acid sequences to its human counterpart, has been solved (Senda et al., 1992) . This has revealed that mouse IFN-[3 has a structure which consists of five (x-helices folded into a compact (x-helical bundle ( Figure 25 .4). From comparative sequence analysis it is predicted that in all mammalian IFN-(X and IFN-]3 proteins these five (x-helical domains are conserved (Korn et aL, 1994; Horisberger and Di Marco, 1995) and thus show similarities with many other cytokines, which also have (x-helical bundle structures (Bazan, 1990) .

With one exception, IFN-(X subtypes do not contain recognition sites (Asn-X-Ser/Thr) for N-linked glycosylation (Henco et aL, 1985) ; only IFN-(X14 contains two of these sites (Figure 25 .2). Nevertheless, O-linked glycosylation may be possible in other IFN-(X subtypes. For example, it has been found that natural IFN-%, purified from human leukocyte IFN, contains the disaccharide galactosyl-N-acetylgalactosamine in Olinkage to Thr-106 (Adolf et aL, 1991a) . However, since IFN-% is the only IFN-(X subtype with a theonine at position 106, it may represent the only O-glycosylated IFN-(X protein (Figure 25 comprise the mature IFN-]3 protein (Derynck et al., 1980 Taniguchi et al., 1980; Houghton et al., 1981) . Although IFN-] 3 is the same length as the majority of IFN-a subtypes, it shows only approximately 30% amino acid sequence relatedness with them ( Figure  25 .2) and is antigenically distinct. The IFN-~ protein lacks the N-terminal Cys-1 residue present in IFN-a subtypes, but contains three other cysteines at positions 17, 31, and 141, the latter two corresponding to the disulfide bond pairing 29-139 in IFN-a subtypes. Replacement of Cys-17 by serine does not result in any loss of biological activity, whereas serine substitution of Cys-141 does (Mark et al., 1981; Shepard et al., 1981) .

As mentioned previously, on the basis of the threedimensional structure of recombinant mouse IFN-~ ( Figure 25.4) , human IFN-[3 is predicted to contain five a-helices and to fold up into an a-helical bundle structure (Senda et al., 1992) . Human IFN-[3 has one potential N-glycosylation site at Asn-80 (Taniguchi et al., 1980) and N-linked oligosaccharides, primarily of the biantennary complextype, are known to be attached to this site in natural IFN-13 (Hosoi et aL, 1988) . However, these may vary considerably depending on the producer cell type (Utsumi et al., 1989) .

From cDNA sequence data, it was predicted that pre-IFN-m contains 195 amino acids, the N-terminal 23 comprising the signal sequence and the remaining 172 the mature IFN-m protein (Capon et al., 1985; Hauptmann and Swetly, 1985) . The amino acid sequence of IFN-m is therefore six residues longer at the Cterminus than IFN-R or IFN-[3 proteins. However, it has been found that natural IFN-m is heterogeneous at the N-terminus owing to variable cleavage of pre-IFN-m; about 60% of mature IFN-m molecules carry two additional N-terminal amino acids (Adolf, 1990; Shirono et al., 1990) . It is approximately 60% related to IFN-cx subtype sequences, but only 30% related to that of IFN-[3 ( Figure 25 .2), and is antigenically distinct from both IFN-~ and IFN-[3 (Adolf, 1990) . Nevertheless, the four cysteines occur in the same notional positions, 1, 29, 99, and 139, as they do in IFN-~ subtypes and it is likely that IFN-m will have a similar cx-helical bundle structure to those predicted for both IFN-cx and IFN-[3 proteins (Senda et al., 1992) . IFN-m has one potential site at Asn-78 for N-linked glycosylation and natural IFN-m has been demonstrated to be a glycoprotein with biantennary complex oligosaccharides (containing neuraminic acid) attached at this site (Adolf, 1990; Adolf et al., 1991 b) .

Type I IFNs (IFN-0t,-[3, and-m) are produced by a variety of normal cell types responding to extracellular or intracellular stimuli (Stewart, 1979) . IFN-~, as a mixture of subtypes, and IFN-m may be produced together following viral infection of null lymphocytes or monocytes/macrophages (Cantell and Hirvonen, 1977; Adolf, 1990) . The proportions of IFN-0t subtypes may vary according to the type of virus used as inducer (Hiscott, 1984; Finter, 1991) . However, production of IFN-[3 is usually restricted to double-stranded polynucleotide, e.g., poly-inosinic, poly-cytidylic acid, or

It is well established that the biological activities of IFNs are mostly dependent upon protein synthesis with selective subsets of proteins mediating individual activities. Antiviral, antiproliferative and immunomodulatory activities have been ascribed to IFN-~/[3/c0 (reviewed in Pestka et al., 1987; De Maeyer and De Maeyer-Guignard, 1988) . The proteins and mechanisms involved in these activities are described below. virally-induced normal fibroblasts and other tissue cell types, e.g., epithelial cells (Meager et al., 1979; Stewart, 1979) . In all the above cases, the amount of IFN secreted is dependent on the dose of the inducer. Besides normal cells, a range of transformed and tumor-derived cell lines are known IFN producers, e.g., MG63 human osteosarcoma cell line (Meager et al., 1982) , and the Namalwa B-lymphoblastoid cell line (Phillips et al., 1986) . Generally speaking, adherent fibroblastic cell lines produce mainly IFN-[3 and only a minor quantity of IFN-~ (Havell et al., 1978) , whereas nonadherent myeloid or lymphoid cell lines produce mainly IFN-c~ and only a small amount of IFN-I3 (Cantell and Hirvonen, 1977; Shuttleworth et al., 1983; Zoon et al., 1992) .

Actual production of IFNs lasts only a matter of a few hours following induction. This is due to IFN mRNA instability and the rapid shut-off of IFN gene transcription (Caput et al., 1986; Shaw and Kamen, 1986) . Under conditions where IFN mRNA stability is increased, e.g., by blocking protein and RNA synthesis following induction, IFN production has been shown to be "superinduced" once the block on protein synthesis is removed (Meager et al., 1979; Stewart, 1979) .

Despite there being vast numbers of viruses with different replication strategies, it appears that many viruses can be countered by relatively few IFN-inducible "antiviral" proteins (Samuel, 1987) . One of the bestcharacterized of these is a family of enzymes collectively known as "2-5A synthetase" which, in the presence of dsRNA (often an intermediate of viral RNA synthesis) catalyses the formation of an unusual oligonucleotide, ppp (A2'p) nA (2-5A), which in turn activates an IFNinduced latent endonuclease, RNase L (Williams and Kerr, 1978; Wreschner et al., 1981; Ghosh et al., 1991; Hovanessian, 1991; Lengyel, 1982; Zhou et al., 1993) . When activated, the RNase L degrades viral (and cellular mRNA) and therefore inhibits viral protein synthesis. Small RNA viruses (picornaviridae), e.g., Mengo virus and murine encephalomyocarditis virus (EMCV), whose replication is cytoplasmic are most inhibited by the induction of2-5A synthetase (Lengyel, 1982; Rice et al., 1985; Chebath et al., 1987; Kumar et al., 1988) . A further important IFN-induced "antiviral" protein is a dsRNA-dependent protein kinase, now designated PKR, which in the active form phosphorylates the peptide initiation factor, eIF2, involved in polyribosomal translation of mRNA (Miyamoto and Samuel, 1980; Gupta et al., 1982; Samuel, 1987) . Phosphorylated eIF2 is inactive and thus viral protein synthesis is inhibited. This inhibition has been associated with the loss of replicating capacity of reoviruses and rhabdoviruses such as vesicular stomatitis virus (VSV).

The 2-5A synthetase and PKR antiviral mechanisms are rather general and potentially could affect a wide range of viruses. However, IFN can induce certain proteins that inhibit specifically one class of virus. For example, the IFN-inducible Mx proteins block the replication of influenza virus, probably by inhibiting the nuclear phase of viral transcription (mouse cells) or later cytoplasmic phases (human cells), without affecting the replication of many other viruses (Staeheli, 1990; Mel6n et al., 1992; Ronni et al., 1993) .

In addition, since IFNs can impair various steps of viral replication, including penetration, uncoating and assembly of progeny virions as well as transcription and translation, there are likely to be several other antiviral proteins and mechanisms (De Maeyer and De Maeyer-Guignard, 1988) . For instance, some viruses, e.g., herpes virus and certain retroviruses, appear to be inhibited at the relatively late stage of virus particle maturation and budding (Aboud and Hassan, 1983) .

In the course of evolution, many viruses have developed countermechanisms by which they can disrupt the antiviral mechanisms induced by IFN.

Such countermechanisms often point to the significance of particular "antiviral" proteins. One of the main "targets" for several different viruses is the IFN-inducible PKR. The action of this kinase is overcome in adenovirus or Epstein-Barr virus (a member of the herpes virus family) by the production of small viral RNA molecules, VAIand EBER-RNAs, respectively, which bind to PKR and block its activation by dsRNA (Clarke et al., 1991; Ghadge et al., 1991; Mathews and Shenk, 1991) . Reoviruses and vaccinia virus (a member of the pox virus family) produce viral proteins (sigma3 and SKI, respectively), that bind to dsRNA and thus reduce activation of PKR (Sen and Lengyel, 1992) . Interestingly, if IFN-treated VSV-infected cells are coinfected by vaccinia virus, VSV replication is rescued, presumably partly by the inhibitory effect of SKI on PKR (Whitaker-Dowling and Youngner, 1983) . Vaccinia virus also produces a nonfunctional protein analog of elF2 which competes with the real eIF2 for phosphorylation by PKR and thus dilutes out the antiviral effect of activated PKR (Beattie et al., 1991) . Other viruses, such as influenza, may activate latent cellular inhibitors of PKR activity, e.g., a 58 kDa protein (p58) (Lee et al., 1992) .

The 2-5A synthetase-RNase L system can also be subverted. For example, EMCV, a picornavirus, can inactivate RNase L in several cell lines, but this inactivation is usually blocked by IFN treatment (Lengyel, 1982) . Herpes viruses, in contrast, appear to inhibit RNase L activation by producing competing analogs of2-5A (Cayley et al., 1984) .

Some viruses even .have the ability to block the transcription of IFN-inducible genes. The "early" Ela regulatory proteins of adenoviruses prevent the activation of ISGF3 by IFN, probably by inhibiting the transcription of the ISGF37 subunit Gutch and Reich, 1991; Kalvakolanu et at., 1991; Nevins, 1991) . In the case of hepatitis B virus-infected cells, the so-called virus-specified "terminal protein" inhibits IFN-inducible gene expression .

The antiviral mechanisms induced by IFN are mediated by enzymes, e.g., 2-5A synthetase and PICR, whose activities have broad implications for cell growth and proliferation. Viral replication may be regarded as a form of pathological growth of a foreign, "cell-like", entity at the expense of a living cell. In the presence of IFN, enzymes are activated which curtail protein synthesis in general, but because viral protein synthesis is normally rapid, the inhibitory effect on viral replication appears more dramatic than on the slower and more complex cellular growth. Possibly, the "IFN system" was evolved more as a part of a complex, interactive network of intercellular mediators of cell growth and proliferation than as one for antiviral mechanisms. Recent investigations tend to support the role of IFNs in regulating cell growth. For example, if a mutant form of PKR that is unable to phosphorylate eIF2 is introduced into cells, they undergo neoplastic transformation (Koromilas et al., 1992; Lengyel, 1993; Meurs et al., 1993) . This suggests that PKR normally acts as a "tumorsuppressor"' gene product. Therefore, one of the mechanisms by which IFN inhibits cell proliferation could be through its capacity to induce enhanced expression/activity of PKR. IFN-~ has also been reported to inhibit cyclin-dependent CDK-2 kinase, which is responsible for phosphorylation of retinoblastoma (RB) protein, and this could contribute to antiproliferative activity (Kumar and Atlas, 1992; Resnitzky et al., 1992; Zhang and Kumar, 1994) .

The 2-5A synthetase-RNase L system may also have antiproliferative and tumor suppressor activities. For instance, the levels of these two enzymes are high in growth-arrested cells: introduction of 2-5A-like oligoadenylates into proliferating cells also causes growth impairment (Sen and Lengyel, 1992; Lengyel, 1993; Zhou et al., 1993) .

The IFN-stimulated increases in synthesis of PKR and 2-5A synthetase are dependent on IFN-inducible transcription factors, such as IRF-1 (ISGF2) Pine et al., 1990; Williams, 1991; Reis et al., 1992) . The latter has a short half-life and thus transcription of IFN-inducible genes is rapidly repressed by the longer-lasting, inhibitory IRF-2 (Harada et al., 1989) . If IRF-2 is overexpressed, cells become transformed as even low level constitutive production of PKR and 2-5A synthetase, which can regulate normal cell growth, is abrogated. This transformation was reversed by overexpressing IRF-1 (Harada et al., 1993) , indicating that IRF-1 can be viewed as a pivotal player in the growth, regulatory, and tumor suppressor machinery. A variety of other IFN-induced mechanisms, including suppression of oncogenes (Contente et al., 1990) , depletion of essential metabolites (Sekar et al., 1983) , and increased cell rigidity (E. Wang et al., 1981) , could also contribute to its antiproliferative activity.

The antiproliferative effects of IFNs in different tumor cell lines cultured in vitro is highly variable. Besides tumor cell lines, IFN-0t/[3 have antiproliferative activity in hematopoietic precursor cells, e.g., of the myeloid lineage (Rigby et al., 1985; De Maeyer and De Maeyer-Guignard, 1988 , for review). They are also potent inhibitors of angiogenesis, the process whereby blood capillaries are formed to envasculate tissues (Sidky and Borden, 1987) .

Besides activating intracellular processes, IFNs can also activate intercellular activities, especially within the immune system, which are an essential part of host defense against infectious and invasive diseases. Thus, IFNs can stimulate indirect antiviral and antitumor mechanisms, which in the main rest upon cellular differentiation and the induction of cytotoxic activity. For example, in the presence of antigen-specific antibodies, macrophages can effect cell-mediated cytotoxicity. Such antibody-dependent cell-mediated cytotoxicity (ADCC) is enhanced by IFN, possibly through an augmentation of immunoglobulin G (IgG)-Fc receptor (FcR) expression (Hokland and Berg, 1981; Vogel et al., 1983; De Maeyer and De Maeyer-Guignard, 1988 ). In addition, another category of leukocytes comprising large granular lymphocytes and known as natural killer (NK) cells are activated, by unknown mechanisms, to kill virally-infected or tumor cell targets independently of major histocompatibility complex (MHC) antigen expression (Trinchieri and Perussia, 1984; Rager-Zisman and Bloom, 1985; De Maeyer and De Maeyer-Guignard, 1988) .

IFNs can stimulate increased expression of class I MHC antigens, i.e., HLA-A, -B, -C, which are crucial for recognition of foreign antigen by cytotoxic T lymphocytes (CTL, CD8+); recognition of virally infected cells by CTL depends on class I MHC antigen presentation of viral antigens at the cell membrane (Heron et al., 1978; Fellous et al., 1979) . IFN-(x/[3 have sometimes been observed to increase class II MHC antigen expression, which is necessary to trigger both humoral and cell-mediated immunity, but probably play a lesser role than IFN-y, which is the major class II MHC antigen inducer (Baldini et al., 1986; Rhodes et al., 1986; De Maeyer and De Maeyer-Guignard, 1988 ).

All of the biological activities so far described (see above) for IFNs have followed from in vitro experimentation.

Here it is possible to pick and choose conditions that favor particular outcomes, e.g., the antiviral response, by adjusting doses of IFN, times of incubation, levels of virus challenge, and so on. Such experiments illustrate the range of biological activities of IFNs but cannot define their physiological roles. That IFNs have the potential for inducing antiviral and antitumor activity suggests their main role in vivo is to act as regulators of host defense mechanisms, and to prevent pathophysiological events occurring. Investigations in experimental animals have supported this likelihood. For example, injection of mice with anti-IFN-~/]3 antibody has been shown to increase their susceptibility to a range of virus infections (Virelizier and Gresser, 1978; Gresser, 1984) . The earliest evidence for an antitumor effect of IFN-tx/13 came from inoculation of murine L1210 cells into mice. L1210 cells are sensitive to the antiproliferative action of IFN-~/13 in vitro and in vivo, IFN~/[3 prevented tumor growth by these cells. However, when a clone of L1210 was isolated that was resistant to the antiproliferative action of lFN-~/[3, there occurred a similar retardation of tumor growth upon IFN-c~/]3 treatment to that observed with "sensitive" L1210 cells, suggesting that IFN was acting indirectly in vivo by a host-mediated mechanism (Gresser et al., 1970 (Gresser et al., , 1972 . A similar conclusion was reached more recently using IFN-resistant B-cell lymphoma cells . In the intervening years, many studies have been conducted confirming that IFN can act directly (e.g., human IFN-a 2 against a range of human tumor xenografts in nude mice where human IFN-0~ 2 has no activity on the murine immune system) and indirectly (reviewed by Balkwill, 1989) . Although antitumor activity has been clearly demonstrated by the application of exogenous IFNs, it is not certain that endogenously produced IFNs are involved in countering tumor growth. However, some experimental evidence that endogenous IFN could play a role in host resistance to cancer or its spread has been obtained by treating mice with anti-IFN antibodies. Under these conditions, the intraperitoneal transplantability of six different experimental murine tumors was observed (Gresser, 1984) .

IFN-a/~ can inhibit the growth of hematopoietic progenitor cells in vitro (Rigby et al., 1985) and this is also likely to occur in vivo. Such an occurrence is undesirable in most instances, but suppression of proliferation of bone marrow hematopoietic cell precursors has been turned to advantage in protecting tumor-bearing mice against the cytotoxicity of chemotherapeutic agents such as 5-fluorouracil (5-FU) (Stolfi et al., 1983) .

IFNs exercise their actions in cells via IFN-specific cell surface receptors. These receptors bind IFNs with high affinity (Aguet, 1980) and transduce the signal occasioned by ligand (IFN) binding across the cell membrane into the cytoplasm. IFN-cq IFN-[3 and IFN-co share the same binding sites (Aguet et al., 1984; Flores et al., 1991) , but IFN-y binds to different sites (Branca and Baglioni, 1981; Aguet et al., 1988) . The binding of IFN-c~ and IFN-~ to lymphoid cells and fibroblasts has been studied extensively and has been reviewed (Rubinstein and Orchansky, 1986; Branca, 1988; Langer and Pestka, 1988; Grossberg et al., 1989) , and results have generally demonstrated the presence of up to a few thousand complex, high-affinity receptors per cell. Chemical cross-linking studies with ~2SI-labeled IFN-~ or IFN-[3 to receptor-bearing cells have led to the identification of various IFN-receptor complexes, their molecular masses ranging from 80 to 300 kDa (Joshi et al., 1982; Eid and Mogensen, 1983; Faltynek et al., 1983; Raziuddin and Gupta, 1985; Thompson et al., 1985; Hannigan et al., 1986; Vanden Broecke and Pfeffer, 1988; Colamonici et al., 1992) . Such results have suggested that there are either multiple binding sites for IFN-cx, IFN-[3, and IFN-m or that there are complex multichain receptors (Colamonici et al., 1992; Hu et al., 1993) .

Although on human cells all the IFN-c~, IFN-~, and IFN-m proteins compete for common binding sites, individual IFN-c~ subtypes show different levels of activities on cells (Streuli et al., 1981; Week et al., 1981; Rehberg et al., 1982) which appear to correlate with their binding behavior to the cell surface (Uzd et al., 1985) . In particular, IFN-% shows a much lower binding for human membrane receptors than either "IFN-0~ 2 or IFN-% (Uz6 et al., 1988) . Interestingly, this differential binding of human IFN-~ subtypes is not manifested in bovine cells, and all of the subtypes exhibit high specific activities (Yonehara et al., 1983; Shafferman et al., 1987) . IFN-I3 and IFN-m are also active in bovine cells (Capon et al., 1985; Adolf et al., 1990) , but this cross-reactivity does not extend to mouse cells, a feature that has provided experimental systems in which to characterize IFN-receptors. Thus, somatic cell genetic studies with human • rodent hybrid cells containing various combinations of human chromosomes have provided evidence that the presence of human chromosome 21 confers sensitivity of such hybrid "rodent" cells to human IFN-~, IFN-I3 and IFN-m (Tan et al., 1973; Slate et al., 1978; Epstein et al., 1982; Raziuddin et al., 1984) . Further, it was demonstrated that antibodies raised against human chromosome 21encoded cell surface proteins were able to block the binding and action of human IFN-~ to human cells, indicating that this chromosome contained a gene(s) specifying the human IFN cell surface receptor .

The elucidation of the full complement of components of the IFN-~/[3/m receptor has long been sought. One methodology used to isolate receptor cDNAs involves transfecting mouse cells with total human DNA and then selecting for cells sensitive to human IFN-cz. After several attempts, this approach led successfully to the isolation of a 2.7 kb eDNA from a library constructed from human lymphoblastoid (Daudi) cells, which encoded an IFN-cz binding protein (Uzd et al., 1990) containing 557 amino acids (molecular mass 63485 Da) including a signal sequence of 27 mainly hydrophobic amino acids. This protein has a structure typical of a transmembrane glycoprotein: a large N-terminal extracellular domain, which potentially could be highly glycosylated owing to a preponderance of N-linked glycosylation sites, a short hydrophobic transmembrane domain, and an intracellular or cytoplasmic tail (Figure 25.5) . Its amino acid sequence shows little homology with any currently available sequences of proteins, including the sequence of the human IFN-7 receptor (Aguet et al., 1988) ; however, the extracellular domain has been predicted to show structural similarities with the latter receptor and to a lesser extent with the so-called hematopoietin receptor supergroup (Bazan, 1990a,b) . The gene coding for this putative IFN-cz receptor has been mapped to chromosome 21.q22 , in confirmation of the earlier rodent x human hybrid cell data (Tan et al., 1973; Slate et al., 1978; Epstein et M., 1982; Raziuddin et al., 1984) .

Although the cloned "IFN-cz receptor" could be shown to confer sensitivity to human IFN-% in transfected mouse cells (Uzd et al., 1990) , such cells were relatively insensitive to human IFN-% and human IFN-[3. These findings, together with those from anti-IFN-cz receptor antibody blocking studies (Colamonici et al., 1990; Revel et al., 1991; Uz~ et al., 1991) and affinity cross-linking studies with IFN-% (Colamonici et al., 1992) , have suggested that a second IFN-cz receptor exists or another component is required besides the cloned IFN-cz 8 binding protein, to complete the receptor complex. This hypothesis is further supported by a study in which introduction of a yeast artificial chromosome (YAC) containing a segment of human chromosome 21 into chinese hamster ovary (CHO) cells conferred a greatly increased response to both IFN-0~ 2 and IFN-0~8, as well as an increased response to IFN-I3 and IFN-m, whereas the expression of the IFN-~ 8 binding protein alone did not confer sensitivity (Revel et al., 1991) . However, these increased responses can be "knocked out" by disruption of the IFN-~ 8 binding protein gene in the YAC, and then reconstituted by expression of the cDNA encoding the IFN-% binding protein (Cleary et al., 1994) , suggesting that cell surface expression of this protein is required for a fully functional receptor (see also Hertzog et al., 1994; Constantinescu et al., 1994) .

A second human IFN-~/I3 receptor, which is probably the additional component of the receptor complex referred to above, has been cloned (Novick et al., 1994) . The 1.5 kb cDNA encodes a 331-amino-acid protein, including a signal sequence, which has the predicted structure of a transmembrane glycoprotein. The Nterminal ectodomain (217 amino acids) corresponds in sequence to a soluble 40 kDa IFN-(x/[3 binding protein, p40, isolated from urine. This domain is linked to a transmembrane segment (21 amino acids) and a relatively small cytoplasmic domain of 67 amino acids ( Figure 25 .5)i Overall, the primary sequence shows little IFN-a/]3 binding protein bind IFN-a 2 but are insensitive to its effects, suggesting that an accessory protein, possibly the cloned IFN-c~ 8 binding protein, is required for signaling (Novick et al., 1994; Constantinescu et al., 1994) . The findings that anti-p40 antiserum and a particular monoclonal antibody to the IFN-cx 8 binding protein (Benoit et al., 1993) both block the biological activity of IFN-0q3co indicate that the IFNhomology with that of the previously cloned IFN-a 8 . c~/13 and IFN-a 8 binding proteins are in close proximity, binding protein (Uzd et al., 1990) , but when the extracellular domains are compared, 23.4% relatedness is found (Novick et al., 1994) , suggesting that both of these IFN binding proteins belong to the same so-called class II cytokine receptor family (Uz6 et al., 1995) . Two classes of cytokine receptor (class I and class II) have been proposed by Bazan (1990a,b) , these being distinguished by the positions of cysteine pairs in the extracellular domain. The latter is comprised of fibronectin type III-like units containing around 200 amino acids and designated D200 (Uzd et al., 1995) . A schematic drawing of both IFN-a/[5 receptor chains is shown in Figure 25 .5. Subsequently, it has been found that alternative splicing of the IFN-a/13 receptor gene can produce a transcript encoding a long form of the receptor protein containing a larger cytoplasmic domain of 251 amino acids .

Mouse cells transfected with the cDNA encoding the and thus probably interact to form a high-affinity IFNa/]3/co receptor complex. The most likely scenario on present evidence is for a two-chain IFN-c~/13/co receptor, comprising the cloned IFN-c~ 8 binding protein (Uz6 et al., 1990) and the long form of the IFN-c~/[3 binding protein Lutfalla et al., 1995) , each of which binds to some extent particular IFN types or IFN-(x subtypes but which together more strongly bind all IFN-~, -13, and -co species and function to transmit signals across the cell membrane. However, it is not completely ruled out that other cell surface components, e.g., membrane glycosphingolipids, are required for fully functional receptors (Colamonici et al., 1992; Platanias et al., 1994; Ghislain et al., 1995; Uz4 et al., 1995) or, possibly, that there are alternative IFN receptors, e.g., the Epstein-Barr virus/complement C3d receptor as an IFN-c~ receptor on B-lymphocytes (Delcayre et al., 1991) . Vaccinia virus and other orthopoxviruses contain a gene B 18R encoding a soluble type I IFN receptor which, unlike the class II cytokine type receptors, belongs to the immunoglobulin superfamily (Symons et al., 1995; Colamonici et al., 1995) .

7.1

The intracellular domains of the two cloned IFN-binding proteins are unrelated to the tyrosine kinase class of receptors, e.g., epidermal growth factor receptor (EGF-R) and platelet-derived growth factor-receptor (PDGF-R), and are not predicted to have kinase activity of any sort (Uzd et al., 1990; Novick et al., 1994) . However, it appears that the cytoplasmic domain of the IFN-0~ 8 and a/13 binding proteins associate with nonreceptor tyrosine kinases TYK2 and Janus kinase 1 (JAK1), respectively, known to be involved in the signal transduction pathway of IFN-a/[3 and other cytokines (Novick et al., 1994; Ghislain et al., 1995; Ihle, 1995; Ihle and Kerr, 1995; Velasquez et al., 1995) . The current understanding of this pathway is as follows. After binding of IFN-a/[3/c0 to their cognate receptors, the intracellular domains are phosphorylated by TYK2 and JAK1. These phosphorylated domains act as docking sites for the cytoplasmic STAT (signal transducers and activators of transcrilStion ) proteins p84/p91 (STATla/b) and p113 (STAT 2) (Ihle, 1996) . The latter undergo tyrosine phosphorylation mediated by receptor-associated TYK2/JAK1, dimerize, translocate to the nucleus, and combine with a DNA binding protein, p48, to form the IFN-stimulated gene factor-3 (ISGF3) transcription factor complex (Schindler et al., 1992; Velasquez et al., 1992; Mtiller et al., 1993; Platanias et al., 1994; Shuai et al., 1994; Gupta et al., 1996; Yan et al., 1996) . Both TYK2 and JAK1 need to be reciprocally activated for signal transduction to occur, since cell mutants lacking either TYK2 or JAK1 are unresponsive to IFN-0~ (Ihle, 1995; Ihle and Kerr, 1995) . ISGF3 binds to cis-acting IFN-stimulated response elements (ISRE), present in the promoter regions of IFN-inducible genes, to initiate their transcription (Williams, 1991) . Targeted disruption of the STAT 1 gene in mice has shown that STAT 1 has an obligatory role in IFN-cx and IFN-y signaling (Durbin et al., 1996; Meraz et al., 1996) . The JAK1/TYK2-ISGF3 pathway may not be the only "receptor-to-cell nucleus" signaling mechanism activated in IFN-stimulated cells. There has been some evidence to implicate protein kinase C (PKC) pathways as well (Reich and Pfeffer, 1990; Pfeffer et al., 1991; C. Wang et al., 1993) . However, this remains controversial owing to the lack of specificity of kinase inhibitors used (Kessler and Levy, 1991; James et al., 1992) .

IFN-inducible genes have a common regulatory nucleotide sequence (G/A)GGAAAN(N)GAAACT in their 5' flanking region and this type of sequence, which resembles the VRE sequences (Ryals et al., 1985; present in IFN genes, is designated interferonstimulated response element (ISRE) (Williams, 1991) . The resemblance between ISRE and VRE sequences probably accounts for the finding that many IFNinducible genes are transcriptionally activated by virus infection or dsRNA, which also activate the transcription of IFN genes (Hug et al., 1988; Wathelet et al., 1988) . As mentioned previously (see Section 5.1.3), IFN-receptor occupation activates cytoplasmic ISGF-3 and this complex is translocated to the nucleus and binds to ISRE of IFNinducible genes as a transcriptional activator. In addition, a second factor, ISGF2, forms complexes with ISRE in IFN-stimulated cells. ISGF2 is a single, inducible phosphoprotein that has been shown to be identical to IRF-1 Pine et al., 1990; Williams, 1991; Reis et al., 1992) . The role of a third transcription factor, ISGF1, which is constitutively produced and requires only the central 9 bp core of ISRE for binding, remains to be fully defined (Kessler et al., 1988 ) . A number of other negative regulatory factors, including IRF2 (Harada et al., 1989) and the ISGF2 (IRF1)/ISGF3yrelated "human interferon consensus sequence binding protein" (ICSBP) (Weisz et al., 1992; Bovolenta et al., 1994) , which also bind to ISRE, are also probably involved in the regulation of transcription of IFN-inducible genes.

It is clear that the regulation of expression of IFNinducible genes is complex and that the mechanisms that control their selective expression are not fully understood (see Taylor and Grossberg, 1990, for review) . IFNinducible proteins, whose number probably exceeds 20, include both those proteins induced early after IFN stimulation and those proteins that may be produced at later times, often in response to the actions of "early" IFN-inducible proteins (Sen and Lengyel, 1992) . The full set of IFN-inducible proteins is probably not known, but several have been identified and characterized. Table  25 .2 shows an incomplete list of IFN-inducible proteins together with their likely functions. Some of these proteins are not exclusively induced by IFN-0~/[3/c0; IFN-y and certain other cytokines, e.g., tumor necrosis factor-~ (TNF-~) often induce spectra of proteins that overlap with the set induced by IFN-~/13/~ (Revel and Chebath, 1986; Rubin et al., 1988; Wathelet et al., 1992) . It should be noted that IFN-inducible proteins tend also to be cell type-specific and thus not all proteins listed in Table 25 .2 will be expressed in all cell types. In some cases, IFN-inducible proteins are completely absent from a cell before IFN stimulation, but in other cases Revel and Chebath, 1986; Sen and Lengyel, 1992; Samuel, 1987; Staeheli, 1990. Revel and Chebath, 1986; Lengyel, 1992. De Maeyer and De Maeyer Guignard, 1988; Heron et aL, 1978 . Schwemmle and Staeheli, 1994 . Ronni et aL, 1993 . Revel and Chebath, 1986 . C. Wang et al., 1993 . Kumar and Atlas, 1992 . Loeb and Haas, 1992 . AIIdridge et aL, 1989 . Fellous et aL, 1982 . Sen and Lengyel, 1992 . Sen and Lengyel, 1992 . Bange et al., 1994 . Chebath et aL, 1983 . Choubey and Lengyel, 1992 . Lawn et aL, 1981a Constantoulakis et aL, 1993 . Porter and Itzhaki, 1993 . Hokland and Berg, 1981 . Thomas and Linch, 1991 . Revel and Chebath, 1986 they are being constitutively produced, their synthesis augmented by IFN.

The genes for mouse IFN-~ subtypes and mouse IFN-[3 (no functional mouse IFN-c0 gene has been found) are located on mouse chromosome 4 (Dandoy et al., 1984 (Dandoy et al., , 1985 De Maeyer and De Maeyer-Guignard, 1988, for review) . These genes are, like their human counterparts, intronless and of comparable structure. Twelve mouse IFN-~ genes or pseudogenes have been identified, of which the cDNAs for 10 different genes have been cloned and expressed (Langer and Pestka, 1985; De Maeyer and De Maeyer-Guignard, 1988) . Mouse IFN-~ subtype proteins contain 166 or 167 amino acids, or exceptionally 162 (mouse IFN-~8) , and the four cysteines at positions (Langer and Pestka, 1985) . There is only a single-copy mouse IFN-[3 gene and this encodes the 16I-amino-acid mature mouse IFN-[3 protein (Higashi et al., 1983; De Maeyer and De Maeyer-Guignard, 1988) . Mouse IFN-13 contains only one cysteine and thus cannot form intramolecular disulfide bonds. It has three potential N-linked glycosylation sites and is heavily glycosylated when secreted from mouse fibroblasts; the molecular mass of the native glycoprotein is approximately 34 kDa compared to the predicted 17 kDa for the nonglycosylated counterpart (De Maeyer and De Maeyer-Guignard, 1988 ). The amino acid sequence of mouse IFN-I3 is about 4:8% related to that of human IFN-I3. The threedimcnsional structure of mouse IFN-I3 has been solved (Senda et aL, 1992 ) (see Section 3.1.1 ) and the protein has been shown to comprise five s-helices folded into a compact s-helical bundle (Figure 25.4) .

Induction of transcription of mouse IFN-0~ subtype genes and the mouse IFN-I3 gene is probably regulated by transcription factor-binding nucleotide sequences present in the 5' noncoding promoter region, in a similar way to that of human IFN-~ and IFN-[3 genes (see Section 2.2). For. example, repeated GAAA-rich sequences are present in the 5' flanking regions of most mouse IFN-cz subtype genes and these are likely to be important for virus-inducible transcription (Shaw et M., 1983; Zwarthoff et al., 1985) . Inducers of mouse IFN-~ and IFN-I3 synthesis, which include a number of viruses and double-stranded polynucleotides, are similar to those which induce human IFN-c~, IFN-[3, and IFN-60 production (Stewart, 1979; De Maeyer and De Maeyer-Guignard, 1988) . Similarly, the type of IFN produced follows the pattern found among different human cell types: fibroblastic and epithelial cell lines produce mainly IFN-~, whereas leukocytes produce mainly IFN-0~ subtypes (De Maeyer and De Maeyer-Guignard, 1988) .

The biological properties of mouse IFN-0~ and IFN-~ are similar to those of human IFN-0~, IFN-J3 and IFN-03 (see Section 5). Since mouse and human IFN-~ subtypes are only 40% homologous, there is considerable species preference in biological activity, i.e., mouse IFN-cx is weakly active in human cells and vice versa. Mouse IFN-13 is also not active in human cells (Stewart, 1979) .

Rather less is known regarding receptors for mouse IFN-~ and IFN-13, than for the human counterparts, but it is probable that they comprise two or more chains, as is the case for the human IFN-~/J3/00 receptors (Uzd et al., 1995) . The mouse equivalent receptor chain to the IFN-0~ 8 binding protein (Uzd et al., 1990) has been cloned (Uz~ et al., 1992) . The gene for this mouse IFN-~/13 receptor has been located to mouse chromosome 16 (Cheng et al., 1993) . The mouse IFN-~/13 receptor is 564 amino acids long and is divided into a large Nterminal extracellular domain (403 amino acids), a short hydrophobic transmembrane segment (20 amino acids), and a cytoplasmic domain (141 amino acids) (Uz~ et al., 1992) . The extracellular domain contains eight potential N-linked glycosylation sites and is predicted to exhibit the two-D200 domain structure of the human IFN-~ 8 binding protein extracellular domain (Figure 25 .5) (Uzd et al., 1995) . Further mouse IFN-0~/]3 rceptors or components thereof await identification and characterization. Signal transduction via mouse IFN-~/]3 receptors is expected to involve the JAK1/TYK2-ISGF3 (STAT 1/2) pathway as outlined for human IFN-0~/]3/60 receptors (see Section 7.1). In STAT 1 genedeleted mice there are no overt developmental abnormalities, but they display a complete lack of responsiveness to mouse IFN-0~ and IFN-~, (Durbin et al., 1996; Meraz et al., 1996) . As a consequence, STAT 1 -/-mice are highly susceptible to infection by viruses and microbial pathogens. STAT 1 is therefore an obligatory mediator in the signal transduction pathway triggered by IFNs. Targeted disruption of the cloned mouse IFN-0~/]3 receptor gave rise to a knockout with a similar phenotype (Miiller et al., 1994) . Such mice, lacking the IFN-cx/13 receptor, developed normally but were unable to respond to mouse IFN-0~/]3 and thus unable to cope with viral infections.

The potent antiviral activity of IFN-~/]3/c0 together with their potential antitumor actions provided the impetus for large-scale manufacture of IFNs for the purpose of clinical evaluation in a variety of viral and malignant diseases. In the early 1970s, IFN production depended on pooled, human buffy coats (leukocytes) and thus only limited quantities could be made (Cantell and Hirvonen, 1977) . Later in that decade, human lymphoblastoid cells (e.g., Namalwa), which could be grown to large culture volumes, became available for IFN production. By the 1980s, following the cloning of IFN-~ and IFN-J3, these IFN species were massproduced by recombinant rDNA technology, leading to abundant availability of certain IFN-0~ subtypes, e.g., IFN-0~ 2 and "stabilized" IFN-]3 ser 17. There followed production of IFN-7 and IFN-c0 by this means. Clinical usage of IFN-~ preparations far exceeds that of IFN-[3 and IFN-c0 because of early production difficulties with the latter types, though these are now solved.

At the beginning of the 1980s there was tremendous enthusiasm, both from manufacturers of IFNs and from clinicians, to evaluate the therapeutic potential of IFNs. However, early clinical trials had been poorly devised, were not "blinded", and often yielded only anecdotal evidence of success. It was only after many controlled, randomized studies had been conducted that it became apparent that IFNs in general, administered as a single agent, were not beneficial for the treatment of the majority of malignant diseases, including the major cancers (lung, breast, colon) of the developed world. The initial optimism all but vanished and was replaced in the mid-to-late 1980s by a more sober and realistic appreciation of the potential therapeutic value ofIFNs. A number of general conclusions have been drawn, as follows. (i) IFN-0~ and IFN-]3, and to a lesser extent IFN-3', have antitumor activity in a small number of cancers, particularly in those that are relatively slow-growing and well-differentiated. (ii) There is no indication that heterogeneous IFN-~ preparations containing mixtures of IFN-0~ subtypes (e.g., leukocyte IFN, lymphoblastoid IFN) have different clinical effects from those of homogenous, recombinant IFN-0~ subtype or IFN-]3 preparations. (iii) Continuous or intermittent high dosing appears to be required for antitumor efficacy. (iv) IFNs probably work best in patients with a minimal tumor burden (Balkwill, 1989) .

A major concern that has emerged from clinical studies is that IFNs all generate a considerable number of undesirable, clinically observable, side-effects, including fever, chills, malaise, myalgia, headache, fatigue, and weight loss, and in certain cases these have been severe enough for treatment to be halted (Bottomly and Toy, 1985; Rohatiner et al., 1985; Goldstein and Laszlo, 1986) . In addition, a variable proportion (1-40%) of patients treated with IFN-0~ or IFN-]3, especially recombinant IFN-~ 2 and recombinant IFN-]3 ser 17, develop neutralizing antibodies to the IFN species used (Rinehart et al., 1986; Antonelli et al., 1991) , that in some instances have been associated with clinical "resistance" to IFN (Steis et al., 1988; Oberg et al., 1989; Freund et al., 1989; Fossa et al., 1992) . A further important, but generally unrecognized, side-effect of IFN-~ treatment is the possible induction of certain types of autoimmune disease (Feldmann et al., 1989; Gutterman, 1994) , probably mediated via IFN-induced upregulation of MHC antigen expression and generalized immunosuppression.

The most responsive cancer to IFN-a therapy is a very rare form of B-cell leukemia, known as "hairy cell" leukemia (HCL), in which a response rate up to 80% has been reported (Gutterman, 1994; Baron et al., 1991; Vedantham et al., 1992) . In HCL patients, the "hairy cells" invade the spleen and bone marrow and the disease takes an indolent course. It has been shown convincingly that IFN-~ therapy continued over several months leads to a clearance of "hairy cells" and in some patients a long-term remission is achieved. IFN-~ ser 17 or IFN-y were less effective against HCL (Saven and Piro, 1992) . The IFN-~-induced mechanisms whereby clearance of "hairy cells" is achieved are not fully understood, but it is believed that a direct action of IFN-~ leading to differentiation of "hairy cells" to a nonproliferating phenotype is involved (Vedantham et al., 1992; Gutterman, 1994) . Not all patients benefit greatly from IFN-~ treatment and some develop neutralizing antibodies, particularly when IFN-0~ 2 is used (Steis et al., 1988) . When such neutralizing antibodies cause resistance to further IFN-~ 2 treatment, clinical responses can be "rescued" by switching to a heterogeneous IFN-0~ preparation, e.g., leukocyte IFN-c~ (von Wussow et al., 1991) . However, on the whole, IFN-~ therapy of HCL appears at least as effective and durable as chemotherapy with the drug pentostatin (2-deoxycoformycin) (Saven and Piro, 1992) . IFN-0~ therapy has also been shown to slow down the progression of chronic myelogenous leukemia (CML) (Baron et al., 1991; Gutterman, 1994) . In this malignant disease, leukemic cells grow slowly in the initial chronic, but benign, phase and persist for 2-4 years, but there follows a dramatic "blast crisis" producing rapidly proliferating myeloid leukemia cells and a fatal outcome. CML patients treated with IFN-~ in the chronic phase often achieve durable remissions, associated with the elimination of leukemic cells bearing the so-called "Philadephia chromosome", sometimes lasting up to 8 years.

Other malignancies in which IFN-a therapy seems to work, although with generally a lower percentage of patients responding than in HCL and CML, include lowgrade non-Hodgkin lymphoma, cutaneous T cell lymphoma, carcinoid tumors, renal cell carcinoma, squamous epithelial tumors of the head and neck, multiple myeloma, and malignant melanoma. In most of these cancers, complete responses are low compared to partial responses, but IFN-~ may help with maintenance therapy of diseases in some cases, e.g., multiple myeloma (Mandelli et al., 1990; Johnson and Selby, 1994) .

The neovascularization of primary tumors is a crucial step in their development and thus the anti-angiogenic activity of IFN-~/I3 (Sidky and Borden, 1987) may have therapeutic value in certain early malignancies, e.g., primary melanoma (Gutterman, 1994) . Kaposi sarcoma, often found in AIDS patients, has been regarded as an angiogenic tumor or angioproliferative disease, which may explain why IFN-c~ treatment can lead to regression of lesions in up to 40% of patients with this condition (De Wit et al., 1988; Groopman and Scadden, 1989) .

In preclinical systems, the combination oflFN therapy and conventional chemotherapy has appeared to offer greater chances of producing effective treatment of many cancers, but in clinical trials this strategy has produced mostly disappointing results (see Wadler and Schwartz, 1990, for review) . This may be due to (i) the inability of preclinical models accurately to predict the clinical situation; (ii) the lack of understanding of the biochemical interactions and biological consequences of combining IFNs and chemotoxic agents; (iii) a failure to incorporate information on dose, scheduling, and sequence of administration of IFNs and chemotoxic agents into clinical trials.

Despite having proven antiviral activity in vitro, IFNs have not proved the hoped-for panacea for most common viral infections in man. IFN-0~/~ prevent the replication of common cold viruses (rhinoviruses and coronaviruses) in the test tube and when administered to volunteers in the form of a nasal spray, but cannot "cure" colds once they are established (Scott r al., 1982; R.M. Douglas et al., 1986; Turner et al., 1986) . IFN-ix is only partially effective in preventing influenza virus infections (Treanor et al., 1987) . Topical applications of IFN-~/~ in the form of creams or ointments to herpes virus lesions, e.g., in herpes zoster (chickenpox), and genital warts (Condyloma acuminatum) caused by papilloma viruses have been investigated, but have given limited beneficial effects. However, when administered parenterally, i.e., by intramuscular or intravenous injection, greater beneficial effects of IFN-(x/6 on virally caused lesions and warts have been found, although not to an extent that IFN therapy has become the treatment of choice (Schneider et al., 1987; J.M. Douglas et al., 1990; Baron et al., 1991; Gutterman, 1994) . Another wart-like disease, juvenile laryngeal papilloma (JLP), which can severely obstruct the airways of young children, caused by the same papilloma virus types (6 and 11) as cause genital warts, has also been found to respond beneficially to IFN-~ therapy. Disappointingly, IFN-cz therapy appears neither curative nor of substantial value as an adjunctive agent in the long-term management ofJLP (Healy et al., 1988) .

Probably the most successful application of IFN-~ therapy to viral disease is in the treatment of chronic active hepatitis, caused by either hepatitis B or C viruses (Baron et al., 1991; Gutterman, 1994) . Up to about 40% of chronic active hepatitis B patients respond to IFN-~ therapy; viral infectivity markers disappear and seroconversion and cure follow. It is interesting in the case of hepatitis B virus that viral activity is responsible for inhibiting the endogenous IFN system , and thus the administration of exogenous IFN-cz constitutes a replacement therapy. In hepatitis C virus infection, some serotypes of the virus are apparently more sensitive to IFN-~ therapy than others and prolonged treatment may be necessary (>6 months) to prevent relapses occurring (Gutterman, 1994) .

Both IFN-~ and IFN-[3 have been shown to inhibit human immunodeficiency virus-1 (HIV-1) replication in vitro (Hartshorn et al., 1987) . However, in vivo, there is little evidence showing that IFN-c~ therapy has any longterm beneficial effect in asymptomatic HIV-l-positive individuals or MDS patients (Friedland et al., 1988; Lane et al., 1990) , except for limited regressions in Kaposi sarcoma lesions (De Wit et al., 1988; Groopman and Scadden, 1989) . Combination therapies for HIV-1infected individuals involving IFN-~ and antiviral drugs such as zidovudine (AZT) have also proved to be ineffective (Berglund, 1991 ) .

As mentioned earlier, IFN-0~/~ inhibits hematopoiesis and therefore induces leukopenia in patients. This effect has generally been thought to be undesirable and it can lead to immunosuppression; however, it has proved useful for the treatment of diseases in which there is uncontrolled leukocytosis, e.g., thrombocytosis (markedly elevated platelet numbers), associated with various myeloproliferative diseases (Gisslinger et al., 1989) . Resistance to IFN-~ 2 therapy has occurred in such patients when neutralizing antibodies to IFN-~ 2 have developed, but successful retreatment with a heterogeneous IFN-~ preparation, lymphoblastoid IFN (IFN-0~N1) has been reported (Brand et al., 1993) . The findings that production of IFN-~ and IFN-7 was deficient in multiple sclerosis (MS) patients (Neighbor and Bloom, 1979) stimulated clinical trials to evaluate IFNs in MS. Rather unexpectedly, it has repeatedly been found that IFN-]3, either natural fibroblast-derived or the later recombinant IFN-J3 ser 17 (IFN-J3-1b), injected intrathecally, subcutaneously, or intramuscularly in patients with relapsing/remitting disease leads to a reduced rate of exacerbations of the disease and thus is possibly of clinical benefit in some patients (Jacobs et al., 1981 (Jacobs et al., , 1987 (Jacobs et al., , 1993 The IFNB Multiple Sclerosis Study Group, 1993; Paty et al., 1993) . The IFN-[3-induced mechanisms that contribute to this beneficial outcome are not known, but probably immunomodulatory actions are involved, e.g., suppression of growth and activity of autoreactive T lymphocytes in the central nervous system (Goodkin, 1994) . It is unclear whether IFN-c~ would have a similar effect. However, the results with IFN-[3 treatment have been encouraging so far, although more follow-up of patients will be necessary to monitor any effects on the clinical progression of MS (Ebers, 1994) .

Accumulation and breakdown of RNA-deficient intracellular virus particles in interferon-treated NIH 3T3 cells chronically producing Moloney murine leukaemia virus

Inhibition of the cellular response to interferons by products of the adenovirus type 5 EIA oncogene

Antigenic structure of human interferon 0~1 (interferon alpha II): comparison with other human interferons

Monoclonal antibodies and enzyme immunoassays specific for human interferon (IFN) c01: evidence that IFN-c01 is a component of human leukocyte IFN

Purification and characterisation of natural interferon c01: two alternative cleavage sites for the signal peptidase

Natural human interferon-~2 is O-glycosylated

Human interferon c01: isolation of the gene, expression in Chinese hamster ovary cells and characterization of the recombinant protein

High affinity binding of 12SI-labelled mouse interferon to a specific cell surface receptor

Various human interferon ~ subclasses cross-react with common receptors: their binding activities correlate with their specific biological activities

Molecular cloning and expression of the human interferon-7 receptor

Interferon [3 increases expression of vimentin at the mRNA and protein levels in differentiated embryonal carcinoma (PSMB) cells

Nomenclature of the human interferon proteins

A family of structural genes for human lymphoblastoid (leukocyte-type) interferon

Neutralizing antibodies to interferon-cx: relative frequency in patients treated with different interferon preparations

Distinct activation of murine interferon-~ promoter region by IRF-1/ISGF-2 and virus infection

Human recombinant interferon c~-2C enhances the expression of class II HLA antigens on hairy cells

Cytokines in Cancer Therapy

IFP35 is an interferon-induced leucine zipper protein that undergoes interferon-regulated cellular redistribution

The interferons: mechanisms of action and clinical applications

Shared architecture of the hormone binding domains in type I and II interferon receptors

Haemopoietic receptors and helical cytokines

Vaccinia virus-encoded eIF-2 alpha homolog abrogates the antiviral effect of interferon

A monoclonal antibody to recombinant human IFN-c~ receptor inhibits biologic activity of several species of human IFN-c~, IFN-[3 and IFN-m

Combined treatment of symptomatic human immunodeficiency virus type 1 infection with native interferon c~ and zidovudine

Clinical side effects and toxicities of interferon

Molecular interactions between interferon consensus sequence binding protein and members of the interferon regulatory factor family

Molecular analysis of the human interferon-cx gene family

The interferon receptors

Evidence that types I and II interferons have different receptors

Successful retreatment of an anti-interferon resistant polycythaemic vera patient with lymphoblastoid interferon-czN1 and in vitro studies on the specificity of the antibodies

Preparation of human leukocyte interferon for clinical use

Two distinct families of human and bovine interferon-c, genes are coordinately expressed and encode functional polypeptides

Identification of a common nucleotide sequence in the 3'-untranslated region of mRNA molecules specifying inflammatory mediators

Activation of the ppp(A2'p)nA system in interferon-treated, herpes simplex virus-infected cells and evidence for novel inhibitors of the ppp(A2'p)nA-dependent RNase

Interferon-induced 56,000 M r protein and its mRNA in human cells: molecular cloning and partial sequence of the eDNA

Constitutive expression of (2'-5') oligo A synthetase confers resistance to picornavirus infection

GART, SON, IFNAR, and CRF-24 genes cluster on human chromosome 21 and mouse chromosome 16

Interferon action: nucleolar and nucleoplasmic localization of the interferoninducible 72-kD protein that is encoded by the If: 204 gene from the gene 200 cluster

Sequence requirement for specific interaction of an enhancer binding protein (EBP1) with DNA

Binding of Epstein-Barr virus small RNA EBER-1 to the double-stranded RNA-activated protein kinase DAI

Knockout and reconstitution of a functional human type I interferon receptor complex

Characterization of three monoclonal antibodies that recognize the IFNa2 receptor

Multichain structure of the IFN-cx receptor on hematopoietic cells

Vaccinia virus B18R gene encodes a type I interferon-binding protein that blocks interferon transmembrane signaling

Role of interferon cz/[3 receptor chain 1 in the structure and transmembrane signalling of the interferon ~/[3 receptor complex

Inhibition of Revmediated HIV-1 expression by an RNA binding protein encoded by the interferon-inducible 9-27 gene

Expression of gene rrg is associated with reversion of NIH 3T3 transformed by LTR-c-H-ras

Linkage analysis of the murine interferon-0~ locus on chromosome 4

Segregation of restriction fragment length polymorphism in an interspecies cross of laboratory and wild mice indicates tight linkage of the murine IFN-[3 gene to the murine IFN-cx genes

Nucleotide sequence of the chromosomal gene for human fibroblast (131) interferon and of the flanking regions

Epstein Barr virus/complement C3d receptor is an interferon c~ receptor

Interferons and Other Regulatory Cytokines

Isolation and structure of a human fibroblast interferon gene

Isolation and characterisation of a human fibroblast interferon gene and its expression in Escherichia coli

Clinical and virological effects of high-close recombinant interferon alpha in disseminated AIDS-related Kaposi's sarcoma

Nomenclature of the human interferon genes

Cooperative interaction of multiple DNA elements in the interferon-J3 promoter

Cloning and expression of a long form of the 13 subunit of the interferon 0q3 receptor that is required for signalling

A randomized trial of combination therapy with intralesional interferon alpha 2b and podophyUin versus podophyllin alone for the therapy of anogenital warts

Prophylactic efficacy of intranasal alpha-2 interferon against rhinovirus infections in the family setting

An ATF/CREB binding site protein is required for virus induction of the human interferon ]3 gene

Targeted disruption of the mouse Stat 1 gene results in compromised innate immunity to viral disease

Molecular cloning of human alpha and beta genes from Namalwa cells

Treatment of multiple sclerosis

Isolated interferon alpha-receptor complexes stabilised in vitro

Human interferon-~A, -~2 and -~2 (Arg) genes in genomic DNA

Direct evidence that the gene product of the human chromosome 21 locus, IFRC, is the interferon-cx receptor

Recombinant human leukocyte interferon produced in bacteria has antiproliferative activity

Characterization of an interferon receptor on human lymphoblastoid cells

Two different virus-inducible elements are required for human ]3-interferon gene regulation

Purification, concentration and physico-chemical properties of interferons

Enhanced expression of HLA antigens and ~2-microglobulin on interferon-treated human lymphoid cells

Modulation of tubulin mRNA levels by interferon in human lymphoblastoid cells

Why are there so many subtypes of alpha-interferons

Human interferon omega (co) binds to the ~/~ receptor

Recombinant interferon c~-2A combined with prednisone in metastatic renal-cell carcinoma: treatment results, serum interferon levels and the development of antibodies

Expression of the terminal protein region of hepatitis B virus inhibits cellular responses to interferons c~ and 3' and double-stranded RNA

Recombinant human interferon (IFN) alpha-2b in chronic myelogenous leukaemia: dose dependency of response and frequency of neutralizing anti-interferon antibodies

A randomized placebo-controlled trial of recombinant human interferon alpha 2a in patients with AIDS

Delimitation and properties of DNA sequences required for the regulated expression of human interferon

Evidence for a nuclear factor(s), IRF-1, mediating induction and silencing properties to human IFN-J] gene regulatory elements

Involvement of a cis-element that binds an H2TF-1/NF-~cB like factor(s) in virus-induced interferon-[3 expression

Induction of the transcription factor IRF-1 and interferon-J3 mRNAs by cytokines and activators of second messenger pathways

Induction of endogenous IFN-~ and IFN-J3 genes by a regulatory transcription factor, IRF-1

Binding of the adenovirus VAI RNA to the interferon-induced 68-kDa protein kinase correlates with function

Configuration of the interferon-cx/[3 receptor complex determines the context of the biological response

Cloning, sequencing and expression of two murine 2'-5'-oligoadenylate synthetases

Concerted evolution of human interferon ~ genes

Long-term interferon therapy for thrombocytosis in myeloproliferative diseases

Human leukocyte interferon produced by E. coli is biologically active

The structure of eight distinct cloned human leukocyte cDNAs

Interferon therapy in cancer: from imagination to interferon

Overlapping positive and negative regulatory domains of the human [3-interferon gene

The human ~-interferon gene enhancer is under negative control

Interferon beta-lb

Human monocytes and lymphocytes produce different mixtures of s-interferon subtypes

Expression of human immune interferon eDNA in E. coli and monkey cells

Selective production of interferon-alpha subtypes by cultured peripheral blood mononuclear cells and lymphoblastoid cell lines

Novel human leukocyte interferon subtype and structural comparison of ~ interferon genes

Role of interferon in resistance to viral infection in vivo

Interferon and cell division. I. Inhibition of the multiplication of mouse leukaemia L1210 cells in vitro by interferon preparations

Mechanism of antitumour effect of interferon in mice

Injection of mice with antibody to interferon enhances the growth of transplantable murine tumours

Interferon therapy for Kaposi's sarcoma associated with the acquired immunodeficiency syndrome (AIDS)

The structure of a thirty-six kilobase region of the human chromosome including the fibroblast interferon gene IFN-~

Interferon receptors and their role in interferon action

Interferon action against reovirus: activation of interferon-induced protein kinase in mouse L929 cells upon reovirus infection

The SH2 domains of Stat 1 and Stat 2 mediate multiple interactions in the transduction of IFN~ signals

Regression of the interferon signal transduction pathway by the adenovirus E1A oncogene

Cytokine therapeutics: lessons from interferon ~. Proc. Natl Acad. Sci. USA 91

Differential human interferon alpha receptor expression on proliferating and non-proliferating cells

The genes for the trophoblast interferons and the related interferon ~II possess distinct 5'-promoter and 3'-flanking sequences

Structurally similar, but functionally distinct factors, IRF-1 and IRF-2, bind to the same regulatory elements of IFN and IFN-inducible genes

Absence of type I IFN system in EC cells: transcriptional activator (IRF-1) and repressor (IRF-2) are developmentally regulated

Anti-oncogenic and oncogenic potentials of interferon regulatory factors 1 and 2

Activity of interferons alpha, beta, and gamma against human immunodeficiency virus replication in vitro

A novel class of human type I interferons

Synthesis of two distinct interferons by human fibroblasts

Treatment of recurrent respiratory papillomatosis with human leukocyte interferon

Structural relationship of human interferon-~ genes and pseudogenes

The gene for human fibroblast interferon (IFB) maps to 9p21

Enhanced expression of 132-microglobulin and HLA antigens on human lymphoid cells by interferon

A gene on human chromosome 21 located in the region 21q22.2 to 21q22.3 encodes a factor necessary for signal transduction and antiviral response to type I interferons

Structure and expression of a cloned cDNA for mouse interferon-[5

Structure and expression in Escherichia coli of canine alpha-interferon genes

Differential expression of human interferon genes

Induction of human interferon gene expression is associated with a nuclear-factor that interacts with the NF~cB site of the human immunodeficiency virus enhancer

Interferon enhances the antibody-dependent cellular cytotoxicity (ADCC) of human polymorphonuclear leukocytes

Interferon-alpha hybrids

A protective action of neurotropic against viserotropic yellow fever virus in Macacus rhesus

Structural characterization of fibroblast human interferon-beta

The absence of introns within a human fibroblast interferon gene

Interferon-induced and double-stranded RNA-activated enzymes: a specific protein kinase and 2',5'-oligoadenylate synthetases

Evidence for multiple binding sites for several components of human lymphoblastoid interferon-cx

Interferon betal b is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial

Cytokine receptor signalling

STATs: signal transducers and activators of transcription

JAKs and STATs in signaling by the cytokine receptor superfamily

Virus interference. I. The interferon

Intrathecal interferon reduces exacerbations of multiple sclerosis

Intrathecally administered natural human fibroblast interferon reduces exacerbations of multiple sclerosis: results of a multicenter, double-blind study

A phase III trial of intramuscular recombinant beta interferon as treatment for multiple sclerosis: current status

Role of protein kinase C in induction of gene expression and inhibition of cell proliferation by interferon cx

The treatment of multiple myelomaman important MRC trial

Interferon receptors. Crosslinking of human leukocyte alpha-2 to its receptor on human cells

Inhibition of interferon-inducible gene expression by adenovirus E1A proteins: block in transcriptional complex formation

Identification and characterisation of a novel repressor of interferon-~ expression

Protein kinase activity required for an early step in interferon-~ signalling

Two interferon-induced nuclear factors bind a single promoter element in interferon-stimulated genes

Interferon: purification and initial characterisation from diploid cells

Human fibroblast interferon: amino acid analysis and amino terminal amino acid sequence

Three-dimensional model of a human interferon alpha consensus sequence

Malignant transformation by a mutant of the IFN-inducible dsRNA-dependent protein kinase

Studies on the role of the 2'-5'-oligoadenylate synthetase-RNase L pathway in l-interferon-mediated inhibition of encephalomyocarditis virus replication

Interferon c~ induces the expression of retinoblastoma gene product in human Burkitt lymphoma Daudi cells: role in growth regulation

Interferon alpha in patients with asymptomatic human immunodeficiency virus (HIV) infection: a randomized, placebo-controlled trial

Structure of interferons

Interferon receptors

DNA sequence of two closely linked human leukocyte interferon genes

DNA sequence of a major human leukocyte interferon gene

Synergism between distinct enhanson domains in viral induction of human beta interferon gene

Characterization and regulation of the 58,000-dalton cellular inhibition of the interferon-induced, dsRNA-activated protein kinase

The involvement of NF-~cB in 13-interferon gene regulation reveals its role as a widely inducible mediator of signal transducfion

Biochemistry of interferons and their actions

Tumor-suppressor genes: news about the interferon connection

Amino-terminal amino acid sequence of human leukocyte interferon

Amino acid sequence of a human leukocyte interferon

Molecular analysis of a human interferoninducible gene family

The interferon-inducible 15-kDa ubiquitin homolog conjugates to intracellular proteins

The structure of the human interferon c~/[3 receptor gene

Mutant U5A cells are complemented by an interferon-a[3 receptor subunit generated by alternative processing of a new member of a cytokine receptor gene cluster

Different pathways mediate virus inducibility of the human IFN-~I and IFN-~ genes

Maintenance treatment with recombinant interferon alpha 2b in patients with multiple myeloma responding to conventional induction chemotherapy

Controlled transcription of a human m-interferon gene introduced into mouse L-cells

The nucleotide sequence of a cloned human leukocyte interferon cDNA

Site-specific mutagenesis of the human fibroblast interferon gene

Adenovirus virus-associated RNA and translational control

Targeted disruption of IRF-1 or IRF-2 results in abnormal type I IFN gene induction and aberrant lymphocyte development

Involvement of a gene on chromosome 9 in interferon production

Interferon production: variation in yields from human cell lines

Somatic cell genetics of human interferon production in human-rodent cell lines

The effect of hypertonic salt on interferon and interferon mRNA synthesis in human MG63 cells

Interferon-induced Mx proteins form oligomers and contain a putative leucine zipper

Targeted disruption of the Stat 1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signaling pathway

Tumor suppressor function of the interferon-induced double-stranded RNA-activated protein kinase

Recent divergence from a common ancestor of human IFN-~ genes

Mechanism of interferon action. Interferon mediated inhibition of reovirus mRNA translation in the absence of detectable mRNA degradation but in the presence of protein phosphorylation

Regulated expression of a gene encoding a nuclear factor, IRF-1, that specifically binds to IFN-[3 gene regulatory elements

Roles of the 29-138 disulfide bond of subtype A of human interferon in its antiviral activity and conformational stability

The protein tyrosine kinase JAK1 complements defects in the IFN-a/[3 and -~, signal transduction

Functional role of type I and type II interferons in antiviral defense

Multimerization of AAGTGA and GAAAGT generates sequences that mediate virus inducibility by mimicking an interferon promoter element

The structure of one of the eight or more distinct chromosomal genes for human interferon-c~

Absence of virus-induced lymphocyte suppression and interferon production in multiple sclerosis

Transcriptional activation by viral regulatory proteins

Interferon-J3 promoters contain a DNA element that acts as a position-independent silencer on the NF-~cB site

The human interferon ~/J3 receptor: characterisation and molecular cloning

Treatment of malignant carcinoid tumors with recombinant interferon alfa-2b: development of neutralizing interferon antibodies and possible loss ofantitumor activity

Close linkage of ~ and ~ interferons and infrequent duplication of ]3 interferon in humans

Leukocyte and fibroblast interferon genes are located on human chromosome 9

Interferon beta-lb is effective in relapsing-remitting multiple sclerosis. II. MRI analysis results of a multicenter, randomized, double-blind, placebo-controlled trial

Interferons and their actions

Transmembrane signalling by interferon ~ involves diacylglycerol production and activation of ~ isoform of protein kinase C in Daudi cells

Large scale production of human interferon from lymphoblastoid cells

Purification and cloning of interferon-stimulated gene factor 2 (ISGF2): ISGF2 (IRF-1) can bind to the promoters of both beta interferon and interferon-stimulated genes, but is not a primary transcriptional activator of either

Tyrosine phosphorylation of the ~ and ~ subunits of the c~ and [3 subunits of the type I interferon receptor

Gene targeting in human somatic cells: complete inactivation of an interferoninducible gene

Interferons and natural killer cells

Not more than 117 base pairs of 5'-flanking sequence are required for inducible expression of a human IFN-~ gene

Receptors for human interferon-alpha: two forms of interferon-receptor complexes identified by chemical cross-linking

Receptors for human cx and [3 interferon but not for 7 interferon are specified by human chromosome 21

Specific molecular activities of recombinant and hybrid leukocyte interferons

Evidence for involvement of protein kinase C in the cellular response to interferon ~

A single DNA response element can confer inducibility by both ~-and 7-interferons

Enhanced in vivo therapeutic response to interferon in mice with an in vitro interferon-resistant B-cell lymphoma

Mice devoid of interferon regulatory factor (IRF-1) show normal expression of type I interferon genes

Critical role of a common transcription factor, IRF-1, in the regulation of IFN-~ and IFN-inducible genes

Interferons and interleukin-6 suppress phosphorylation of the retinoblastoma protein in growth-sensitive hematopoietic cells

Interferon-activated genes

Components of the human type

Antigen presentation by human monocytes: effects of modifying major histocompatibility complex class II antigen expression and interleukin 1 production by using recombinant interferons and corticosteroids

Double-stranded RNA-dependent protein kinase and 2-5A system are both activated in interferon-treated, encephalomyocarditis virus-infected HeLa cells

The effect of recombinant-DNA-derived interferons on the growth of myeloid progenitor cells

Phase I/II trial of human recombinant ~-interferon serine in patients with renal cell carcinoma

Interferons as hormones of pregnancy

Central nervous system toxicity of interferons

Control of IFN-inducible MxA gene expression in human cells

Tumour necrosis factor and IFN induce a common set of proteins

The interferon receptors

Human leukocyte interferon purified to homogeneity

Human leukocyte interferon: isolation and characterisation of several molecular forms

Induction of type I interferon genes and interferon-inducible genes in embryonal stem cells devoid of interferon regulatory factor I

A 46-nucleotide promoter segment from an IFN-cx gene renders an unrelated promoter inducible by virus

Interferon induction of the antiviral state proteins induced by interferons and their possible roles in the antiviral mechanisms of action

Treatment of hairy cell leukaemia

Interferon-dependent tyrosine phosphorylation of a latent cytoplasmic transcription factor

Interference between animal viruses

Interferon treatment of human genital papilloma virus infection: importance of viral type

The interferoninduced 67-kDa guanylate-binding protein (hGBP1) is a GTPase that converts GTP to GMP

Prevention of rhinovirus colds by human interferon alpha-2 from E. coli

Inhibition of ornithine decarboxylase in human fibroblast cells by type I and type II interferons

The interferon system: a bird's eye view of the biochemistry

Three-dimensional crystal structure of recombinant murine interferon-[3

Specific residues within an amino-terminal domain of 35 residues of interferon-~ are responsible for recognition of the human interferon-a cell receptor and for triggering biological effects

Structure and expression of cloned murine IFN-a genes

A conserved AU sequence from the 3' untranslated region of GM-CSF mRNA mediators selective mRNA-degradation

A single amino acid change in IFN-[31 abolishes its antiviral activity

Existence and unique N-terminal sequence of alpha II (omega) interferon in natural leukocyte interferon preparation

Clustering of leukocyte and fibroblast genes on human chromosome 9

Interferon activation of the transcription factor Star91 involves dimerization through SH2-phosphotyrosyl peptide interactions

Antibodies to chromosome 21 coded cell surface components block binding of human ~ interferon but not 7 interferon to human cells

Expression of interferon-~ and interferon-J3 genes in human lymphoblastoid (Namalwa) cells

Inhibition ofangiogenesis by interferons: effects on tumour-and lymphocyte-induced vascular responses

Presence of human chromosome 21 alone is sufficient for hybrid cell sensitivity to human interferon

Chromosomal location of a human cx interferon gene family

Expression of a functional human type I interferon receptor in hamster cells: application of functional yeast artificial chromosome (YAC) screening

Interferon-induced proteins and the antiviral state

Resistance to recombinant interferon alfa-2a in hairy-cell leukaemia associated with neutralizing anti-interferon antibodies

The Interferon System

Interferon nomenclature

Modulation of 5-fluorouracil-induced toxicity in mice with interferon or with the interferon inducer, polyinosinic-polycytidylic acid

At least three human type a interferons: structure of a 2

Target specificity of two species of human interferon-a produced in Escherichia coli and of hybrid molecules derived from them

Vaccinia virus encodes a soluble type I interferon receptor of novel structure and broad species specificity

The linkage of genes for the human interferon-induced antiviral protein and indophenol oxidase-B traits to chromosome G-21

Human leukocyte and fibroblast interferons are structurally related

Recent progress in interferon research: molecular mechanisms of regulation, action, and virus circumvention

The high mobility group protein HMGI (Y) is required for NF-~cB-dependent virus induction of the human IFN-13 gene

An intracellular 50kDa Fc~,-binding protein is induced in human cells by 0~-IFN

Characterization of human beta-interferon-binding sites on human cells

Two non-allelic human interferon a genes with identical coding regions

Intranasally administered interferon as prophylaxis against experimentally induced influenza A virus infection in humans

Human natural killer cells: biologic and pathologic aspects

Prevention of experimental coronavirus colds with intranasal alpha-2b interferon

Nucleotide sequence of a portion of human chromosome 9 containing a leukocyte interferon gene cluster

Characterization of four different mammalian-cell-derived recombinant human interferon-[31s

Receptor dynamics of closely related ligands: "fast" and "slow" interferons

Electrostatic interactions in the cellular dynamics of the interferon-receptor complex

Genetic transfer of a functional human interferon 0~ receptor into mouse cells: cloning and expression of its cDNA

Murine tumor cells expressing the gene for the human interferon 0q3 receptor elicit antibodies in syngeneic mice to the active form of the receptor

Behaviour of a cloned murine interferon alpha beta receptor expressed in homospecific or heterospecific background

0~ and ]3 interferons and their receptor and their friends and relations

Characterization of interferon-alpha binding sites on human cell lines

Mechanism of interferon action in hairy cell leukaemia: a model of effective cancer biotherapy

A protein tyrosine kinase in the interferon a/J5 signalling pathway

Distinct domains of the protein tyrosine kinase tyk2 required for binding of interferon-a/J3 and for signal transduction

Role of interferon in the pathogenesis of viral diseases of mice as demonstrated by the use of anti-interferon serum. V. Protective role in mouse hepatitis virus type 3 infection of susceptible and resistant strains of mice

Double-stranded RNA activates binding of NF-~cB to an inducible element in human [3-interferon promoter

Interferon-induced enhancement of macrophage Fc receptor expression: [3-interferon treatment of C3H/HeJ macrophages results in increased numbers and density of Fc receptors

Effective natural interferon-~ therapy in recombinant interferon-m-resistant patients with hairy cell leukaemia

Antineoplastic activity of the combination of interferon and cytotoxic agents against experimental and human malignancies: a review

Interferon c~ induces a protein kinase C-e (PKC-e) gene expression and a 4.7kb PKC-8 related transcript

Interferon increases the abundance of submembranous microfilaments in HeLa-S3 cells in suspension culture

Activation of IFN-13 element by IRF-1 requires a translational event in addition to IRF-1 synthesis

Regulation of two interferon-inducible human genes by interferon, poly(rI):poly(rC) and viruses

Regulation of gene expression by cytokines and virus in human cells lacking the type-I interferon locus

Antiviral activities of hybrids of two major human leukocyte interferons

The 5'-flanking region of a human IFN-c~ gene mediates viral induction of transcription

Human interferon consensus sequence binding protein is a negative regulator of enhancer elements common to interferon-inducible genes

Vaccinia rescue of VSV from interferon-induced resistance: reversal of translation block and inhibition of protein kinase activity

Identification of novel factors that bind to the PRD1 region of the human [3-interferon promoter

Postinduction repression of the 13-interferon gene is mediated through two positive regulatory domains

Treatment of malignant carcinoid tumors with recombinant interferon alfa-2b: development of neutralizing interferon antibodies and possible loss of antitumour activity

Transcriptional regulation of interferon-stimulated genes

Inhibition of protein synthesis by 2'-5' linked adenine oligonucleotides in intact cells

A comparison of vertebrate interferon gene families detected by hybridisation with human interferon DNA

Ribosomal RNA cleavage, nuclease activation and 2-5A (ppp(A2'p)nA) in interferon treated cells

Multiple protein-DNA interactions with the human interferon-[3 regulatory element

Phosphorylated interferon-~ receptor 1 subunit (IFNaR1) acts as a docking site for the latent form of the 113 kDa STAT 2 protein

Different binding of human interferon-c~l and-c~2 to common receptors on human and bovine cells. Studies with recombinant interferon produced in Escherichia coli

Interferon-0~ inhibits cyclin Eand cyclin Dl-dependent CDK-2 kinase activity associated with RB protein and E2F in Daudi cells

Expressing cloning of 2-5A-dependent RNAase: a uniquely regulated mediator of interferon action

Comparative structures of mammalian interferons

Amino-terminal sequence of the major component of human lymphoblastoid interferon

Purification and characterization of multiple components of human lymphoblastoid interferon-c~

Organization, structure and expression of murine interferon c~ genes