WO2000077205A1 - HUMAN NUCLEAR FACTORS ASSOCIATED WITH dsRNA (NFAR) - Google Patents

Abstract

Nucleotide and amino acid sequences of human Nuclear Factors Associated with dsRNA (NFAR-1 and NFAR-2) are disclosed. Specific binding molecules (e.g., complementary primers or probes, antibodies) are disclosed that identify the genes and proteins, and can distinguish the related NFAR-1 and NFAR-2 forms. Recombinant polynucleotides and polypeptides, vectors and expression constructs, and genetically engineered cells and non-human organisms can be made from the native genes and proteins. Arrays of different polynucleotides, polypeptides, specific binding molecules, or combinations thereof can be made on artificial substrates. Kits comprising polynucleotides, polypeptides, specific binding molecules, instructions for their use, or combinations thereof are also disclosed. Moreover, the structure of DNA, RNA, and protein; the amount of DNA, RNA, and protein; and the localization of DNA, RNA, and protein can be determined or manipulated. Thus, NFAR-1 and NFAR-2 activity can be increased or decreased, or expressed in places or times selected in accordance with the invention. Candidate chemical agents can be screened for the ability to potentiate or to inhibit activity of an NFAR gene or protein, or to potentiate or to inhibit association of PKR with NFAR-1 or NFAR-2. Furthermore, processes of making and using the aforementioned products are disclosed.

Description

HUMAN NUCLEAR FACTORS ASSOCIATED WITH dsRNA (NFAR)

GOVERNMENT LICENSE RIGHTS The U.S. government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license others on reasonable terms as provided by the terms of CA84247-01 awarded by the National Institutes of Health.

BACKGROUND OF THE INVENTION

1. Field of Invention The present invention relates to human genes for Nuclear Factors Associated with dsRNA (NFAR), and their protein products.

2. Description of the Related Art

The interferons (IFNs) are pleiotropic cytokines that are considered important in host defense and manifest their biological properties by inducing a number of IFN- responsive genes. One key inducible gene is the serine/threonine, dsRNA-dependent protein kinase (PKR) which has a molecular weight of about 68 kDa in human cells. PKR is a member of a previously described family of dsRNA-binding proteins that includes E. coli RNAse III, D. melanogaster Staufen, and X. laevis 4F.1 and 4F.2 (St. Johnston et al., 1992; Bass et al., 1994). Interaction with dsRNA structures of greater than 35 basepairs (bp) causes PKR to autophosphorylate and subsequently to catalyze the phosphorylation of substrate targets, the most well-characterized being the alpha subunit of eukaryotic initiation factor 2 (elF2 ). Phosphorylated elF2α effectively sequesters elF2B, a rate-limiting component of translation, and thereby inhibits protein synthesis in the cell. Most of the studies relating to PKR's function have focused on its participation in

IFN's retaliatory response to virus infection. For example, PKR can be activated by a number of viral-specific RNA. Since activation of PKR is highly detrimental to the virus life cycle, numerous viruses have evolved mechanisms to neutralize the function of this kinase. One of the most well-characterized PKR repressors, adenovirus VAI RNA, prevents PKR autophosphorylation by interfering with the binding of dsRNA activators. Similarly, Espstein-Barr virus may utilize the EBER RNA in a comparable strategy. Vaccinia virus reportedly encodes two proteins capable of inhibiting PKR. The first of these is referred to as E3L and contains a dsRNA-binding motif similar to those found in PKR. E3L, which is primarily detected in the nucleus, competes for dsRNA activators and may even bind to and inhibit the kinase through dsRNA bridging. A second vaccinia protein K3L shares homology to the PKR substrate elF2α and is thought to function by competitively sequestering the kinase. In addition, HIV-1 may utilize the Tat gene product to suppress PKR, while hepatitis C has been reported to encode PKR interacting proteins referred as NS5a and E2, respectively.

Recent data have also revealed that PKR almost certainly plays a role in mediating apoptosis. For example, recombinant vaccinia viruses overexpressing wild-type PKR have been shown to induce the rapid apoptosis of human HeLa cells. Similarly, vaccinia viruses lacking E3L become more sensitive to cell-mediated apoptosis presu- mably since they are less effective at preventing PKR activation. Mouse fibroblasts deficient for PKR activity were also reported to be more resistant to dsRNA and TNF- induced apoptosis. Significantly, our own laboratory (Balachandran et al., 1998) has recently demonstrated that cells inducibly overexpressing functional PKR, but not a dominant-negative PKR variant, are highly sensitive to dsRNA-mediated and viral induced apoptosis. PKR-dependent apoptosis was accompanied by the induction of Fas, a member of the tumor necrosis factor receptor family and was mediated through the FADD death signaling pathway. Conversely, cells expressing dominant-negative PKR variants became transformed and contain no detectable Fas, FADD, FLICE, Bax, Bad, or TNFR-1 mRNA or protein; this signifies a block at the transcriptional level. Although the mechanisms of PKR-induced apoptosis remain to be fully clarified, it is likely that viruses target PKR, not only to avoid a shutdown of protein synthesis rates, but also to avoid triggering programmed cell death.

These observations support the conclusion that PKR is a receptor or inducer in signaling pathways regulated by dsRNA, but PKR's role and its precise mechanism of action remain to be fully resolved. For example, it has been reported that the inhibitory subunit of NF-κB, referred to as IkB, can be phosphorylated in vitro by PKR. Mouse fibroblasts lacking PKR activity have also been shown to exhibit diminished responses to dsRNA signaling as indicated by reduced activity of the dsRNA-regulated genes IRF- 1 and NF-κB. Collectively, these data imply that PKR substrates other than elF2α are likely prevalent in the cell.

Therefore, the identification of PKR substrates and putative regulators of PKR- induced signaling pathways was a problem that needed to be addressed. Accordingly, we have isolated proteins that associate with PKR using a yeast two-hybrid system with PKR as bait to address this need. N.B. This association is called "binding" between NFAR and its partner hereinafter although other components may be present in the binding complex (e.g., dsRNA).

Herein, we report the isolation and characterization of two highly conserved, dsRNA-binding nuclear proteins NFAR-1 and -2 (for Nuclear Factor Associated with dsRNA), that are generated from a differentially spliced mRNA transcribed from the same gene. Aside from demonstrating that both NFAR bind to PKR in yeast, we show that they share homology with elF2α and are substrates for PKR. In addition, both NFAR colocalize with PKR in the cell and co-immunoprecipitate with PKR in vitro and in vivo. Importantly, NFAR-1 and NFAR-2 were found to augment the transcription of genes controlled by the IFNβ or Fas promoter, an effect that could be blocked by the vaccinia virus encoded PKR inhibitory protein E3L. Such results show NFAR-1 and NFAR-2 are positive regulators of transcription whose evolutionary conservation is indicative of their importance. Therefore, we disclose that NFAR-1 and NFAR-2 function as proximal targets for

PKR-induced signaling, and may regulate gene expression in apoptosis and tumori- genesis, activation of the host defense response, and other dsRNA-regulated signaling events. It is also an objective of the invention to provide a locus at chromosome band 19p13.1-13.2 for mapping and for correlating rearrangement with tumorigenesis at the level of an NFAR gene, as well as a cell marker for a ubiquitously expressed, evolution- arily conserved, nuclear phosphoprotein and an affinity reagent for PKR at the level of an NFAR protein. These and other objects of the present invention will be evident to a person skilled in the art from the description below.

SUMMARY OF THE INVENTION

We provide polynucleotides and polypeptides for a human Nuclear Factor Associated with dsRNA (NFAR), as well as binding molecules specific for such. At least two different products, NFAR-1 and NFAR-2, result from alternative splicing. Thus, polynu- cleotides, polypeptides, and specific binding molecules that distinguish between NFAR- 1 and NFAR-2 are useful to identify common or diverse functions of the two isoforms.

Also provided are processes to detect NFAR polynucleotide or polypeptide expression; to potentiate or to inhibit NFAR polynucleotide or polypeptide expression; to identify, to isolate, or to detect an NFAR polynucleotide or polypeptide; to produce native NFAR polynucleotide or polypeptide, NFAR recombinant polynucleotide, or NFAR fusion polypeptide; to make a binding molecule specific for NFAR polynucleotide or polypeptide; to identify, to isolate, or to detect the specific binding molecule; and to make a vector, an expression construct, a transfected cell, or a non-human transgenic animal. In particular, the structure of DNA, RNA, and protein; the amount of DNA,

RNA, and protein; and the localization of DNA, RNA, and protein can be determined or manipulated. Thus, NFAR-1 and NFAR-2 activity can be increased or decreased, or expressed in selected places or at selected times.

Further processes are provided to screen candidate chemical agents for the ability to modulate expression of an NFAR gene and/or activity of an NFAR protein. Candidate chemical agents that modulate association of PKR with NFAR may also be identified or isolated. Reporter constructs and expression systems for screening candidate chemical agents are provided.

Probes or primers complementary to a NFAR polynucleotide may be used for monitoring gene expression and functional studies. From its nucleotide sequence, specific binding molecules (e.g., antisense polynucleotides, ribozymes, triple helix- forming polynucleotides) can be used to inhibit NFAR expression. Its amino acid sequence may be used for preparation of specific binding molecules (e.g., polyclonal or monoclonal antibodies, antibody fragments, humanized antibodies, single chain anti- bodies, phage hybrid proteins, or other members of a combinatorial library) for monitoring protein expression and functional studies. Specific binding molecules made against an NFAR amino acid sequence may also be used to inhibit NFAR expression.

Uses of NFAR polynucleotide, polypeptide, and specific binding molecule are further described below. Kits comprising the aforementioned products are also provided to practice the described processes; such kits would further comprise instructions for performing the processes, standards to calibrate quantitiative assays, positive or negative controls suitable to perform such processes, other reagents to perform the processes, and combinations thereof.

These and other aspects of the present invention will be apparent to a person of skill in the art from the following description.

DESCRIPTION OF TABLES AND FIGURE Tables 1-2 show nucleotide and amino acid sequences of human NFAR-1 (SEQ ID NOS:1-2, respectively) deposited under GenBank Accession No. AF167569. SEQ ID NO:1 shows only the sense strand so a double-stranded human NFAR-1 cDNA would include SEQ ID NO:1 and its complement.

Tables 3-4 show nucleotide and amino acid sequences of human NFAR-2 (SEQ ID NOS:3-4, respectively) deposited under GenBank Accession No. AF167570. SEQ ID NO:2 shows only the sense strand so a double-stranded human NFAR-2 cDNA would include SEQ ID NO:2 and its complement.

Table 5 shows an alignment of members of this dsRNA-binding protein family using CLUSTAL algorithm and PAM250 weight table: NFAR-1 (SEQ ID NO:2), NFAR-2 (SEQ ID NO:4), MPP4 2043 bp (SEQ ID NO:5), MPP4 1523 bp (SEQ ID NO:6), NF90 (SEQ ID NO:7), mlLF3 (SEQ ID NO:8), XI4F.1 (SEQ ID N0.9), and XI4F.2 (SEQ ID NO: 10). Residues that are identical to either NFAR-1 or NFAR-2 are boxed.

Table 6 shows exon and intron sequences (SEQ ID NOS:11-51) of NFAR-1 and NFAR-2 represented by upper case and lower case letters, respectively. The adenine base of the ATG start codon is present in exon 2. Figure 1 shows the intron/exon organization of a human NFAR gene. Exons are boxed and introns are shown as connecting lines. Arrows and numbers indicate the corresponding nucleotide positions of two different NFAR cDNA with position 1 corresponding to the adenine base of the ATG start codon for both, and position 2109 for NFAR-1 and position 2685 for NFAR-2 corresponding to the last base of the stop codon. The 5'-UTR exons are shown as open boxes and coding sequences as solid boxes.

DESCRIPTION OF THE INVENTION "NFAR" refers to Nuclear Factors Associated with dsRNA, including the NFAR-1 and NFAR-2 isoforms, other native (i.e., derived from nature) genes and proteins, and derivative forms thereof (e.g., mutants and analogs not found in nature). The chemical structure of NFAR may be a polymer of natural or non-natural nucleotides connected by natural or non-natural covalent linkages (i.e., polynucleotide) or a polymer of natural or non-natural amino acids connected by natural or non-natural covalent linkages (i.e., polypeptide). See Tables 1-4 of WIPO Standard ST.25 (1998) and M.P.E.P. § 2422 for a non-limiting list of natural and non-natural nucleotides and amino acids.

The natural linkage for polynucleotides is a phosphodiester bond made between the 3' hydroxyl group of one nucleotide and the 5' phosphate group of the succeeding nucleotide. Non-natural backbones that include a phosphorus heteroatom are phosphorothioates, chiral phosphorothioates, phosphorodithioates, phosphotriesters, aminoalkylphosphotriesters, methyl and other alkyl phosphonates including 3'-alkylene phosphonates and chiral phosphonates, phosphinates, phosphoramidates including 3'- amino phosphoramidate and aminoalkylphosphoramidates, thionophosphoramidates, thionoalkylphosphonates, thionoalkylphosphotriesters, and boranophosphates having normal 3'-5' linkages, 2'-5' linked analogs of these, and those having inverted polarity wherein the adjacent pairs of nucleoside units are linked 3'-5' to 5'-3' or 2'-5' to 5'-2'. Non-natural backbones that do not include a phosphorus heteroatom are formed by short chain alkyl or cycloalkyl internucleoside linkages, mixed heteroatom and alkyl or cycloalkyl internucleoside linkages, or one or more short chain heteroatomic or heterocyclic internucleoside linkages. These include those having morpholino linkages (formed in part from the sugar portion of a nucleoside); siloxane backbones; sulfide, sulfoxide, and sulfone backbones; formacetyl and thioformacetyl backbones; methylene formacetyl and thioformacetyl backbones; alkene containing backbones; sulfamate backbones; methyleneimino and methylenehydrazino backbones; sulfonate and sulfonamide backbones; amide backbones; and others having mixed N, O, S and CH₂ groups. Post-transcriptional modifications include polyadenylation and splicing.

Various modifications to the polynucleotides can be introduced to increase stability and half-life. Possible modifications include, but are not limited to, the addition of flanking sequences of ribo- or deoxy- nucleotides to the 5' and/or 3' ends, blocking or cyclization of the 5' and/or 3' ends, or the use of phosphorothioate or 2'-0-methyl rather than phosphodiesterase linkages within the backbone.

The natural linkage for polypeptides is an amide bond made between the α- carboxyl group of the N-terminal residue and the α-amino group of the C-terminal residue. Backbones may be modified by using D-amino acids, modified amino acids, and peptidomimetics. Non-peptide linkages include, for example, -CH₂NH- -CH₂S- -CH2-CH₂- -CH=CH- (cis and trans), -COCH₂- -CH(OH)CH₂- -CH₂SO- and others having mixed P, O, Si, S and CH₂ groups. Post-translational modifications include acetylation, acylation, amidation, disulfide bonding, formylation, glycosylation, hydroxylation of γ-carboxyglutamic acid, methylation, phosphorylation, proteolysis, and sulfatation.

"Mutants" are polynucleotides and polypeptides having at least one function that is more active or less active, an existing function that is changed or absent, a novel function that is not naturally present, or combinations thereof. "Analogs" are polynucleotides and polypeptides with different chemical structure, but substantially equivalent function as compared to the native gene or protein. NFAR functions are described in detail herein. Mutants and analogs can be made by genetic engineering or chemical synthesis, but the latter is preferred for non-natural nucleotides or amino acids or linkages.

"Oligonucleotides" and "oligopeptides" are short versions of polynucleotides and polypeptides (e.g., less than 50 nucleotides or amino acids). Generally, they can be made by chemical synthesis, but cleavage of longer polynucleotides or polypeptides can also be used. Electrophoresis and/or reverse phase high-performance liquid chromatography (HPLC) are suitable biochemical techniques to purify short products. "Human NFAR" means an NFAR derived from a human and includes mutants and polymorphisms thereof, but excludes similar genes and proteins derived from other organisms. In particular, other dsRNA-binding proteins described in the literature such as E. coli RNAse III, D. melanogaster Staufen, human TRPB and mouse Spnr, mouse ILF3, human MPP4 and NF90, and X. laevis 4F.1 and 4F.2, are excluded from the invention. Human NFAR genes can be isolated using stringent hybridization: suitable conditions for oligonucleotides could be 400 mM NaCI, 40 mM PIPES pH 6.4, 1 mM EDTA, 50°C or 70°C; suitable conditions for polynucleotides of 50 bases or longer could be 500 mM NaHPO₄ pH 7.2, 7% sodium dodecyl sulfate (SDS), 1% bovine serum albumin (BSA, fraction V), 1 mM EDTA, 45°C or 65°C. Human NFAR proteins can be isolated using immunoblotting: suitable conditions could be 50 mM Tris-HCI pH 7.4, 500 mM NaCI, 0.05% TWEEN 20 surfactant, 1% bovine serum albumin (BSA, fraction V), room temperature. Blocking and washing conditions can be varied to obtain a hybridization or immunoblotting signal that is target specific and reduces the background.

An "isolated" product is at least partially purified from its cell of origin (e.g., human, other mammal, bacterium, yeast). For example, as compared to a lysate of the cell of origin, the isolated product is at least 50%, 75%, 90%, 95% or 98% purified from other chemically-similar solutes (e.g., nucleic acids for polynucleotides, nucleoproteins for polypeptides). For a chemically-synthesized polymer of nucleotides or amino acids, purity is determined by comparison to prematurely terminated or blocked products and may, as a practical matter, be considered isolated without purification. Purification may be accomplished by biochemical techniques such as, for example, cell fractionation, centrifugation, chromatography, and electrophoresis. Generally, solvent (e.g., water) and functionally inert chemicals like buffers and salts are disregarded when calculating purity. Cloning is often used to isolate the desired product. The meaning of "heterologous" depends on context. For example, ligation of heterologous nucleotide regions to form a chimera means that the regions are not found colinear in humans (e.g., a human-derived NFAR polynucleotide and identical to the native sequence ligated to a human non-NFAR transcriptional regulatory region). Another example is that fusion of heterologous amino acid domains means that the domains are not found colinear in humans (e.g., a human-derived NFAR polypeptide and identical to the native sequence joined to a human non-NFAR dimerization domain). Ligation of nucleotide regions or joining of amino acid domains, one derived from human and another derived from a non-human, are heterologous because they are derived from different species. In a further example, transfection of a vector or expression construct into a heterologous host cell or transgenesis of a heterologous non-human organism means that the vector or expression construct is not found in the cell's or organism's genome in nature. A "recombinant" product is the result of ligating heterologous regions for a recombinant polynucleotide or fusing heterologous domains for a recombinant polynucleotide. Recombination may be genetically engineered in vitro with purified enzymes or in vivo in a cultured cell, or by natural phenomena like gene rearrangement and chromosomal translocation.

According to one aspect of invention, polynucleotides (e.g., DNA or RNA, single- or double-stranded) that specifically hybridize to NFAR genes and transcripts can be used as probes or primers. Such polynucleotides could be full length covering the entire gene or transcribed message (e.g., a recombinant clone in a phagemid, plasmid, bacteriophage, cosmid, shuttle vector, yeast artificial chromosome or YAC, bacterial artificial chromosome or BAC, or other vector), a particular coding region, or a shorter length sequence which is unique to NFAR genes or transcripts but contains only a portion of same. A probe stably binds its target to produce a hybridization signal specific for an NFAR polynucleotide or polypeptide, while a primer may bind its target less stably because repetitive cycles of polymerization or ligation will also produce a specific amplification signal. The polynucleotide may be at least 15, 30, 45, 60, 90, 120, 240, 360, 480, 600, 720, 1200, 2400, 5000, 10K, 20K, 40K, 100K, 250K, or 500K nucleotides long. For example, a probe or primer having a nucleotide sequence comprising bases 2062 to 2106 may be used to distinguish between NFAR-1 (SEQ ID NO:1) and NFAR-2 (SEQ ID NO:3) at the level of the transcript (e.g., RNA or cDNA). A recombinant clone or expression construct containing an NFAR nucleotide sequence is a preferred form of the invention; such clone or construct could be single- or double-stranded, and comprised of DNA, RNA, natural and non-natural nucleotides, natural and non-natural linkages, or combinations thereof. The expression construct further comprises a regulatory region for gene expression (e.g., promoter, enhancer, silencer, splice donor and acceptor sites, polyadenylation signal, cellular localization sequence) and, optionally, an origin of replication that allows chromosomal or episomal replication in a selected host cell. The expression construct may be based on a vector with region(s) from a mammalian gene (e.g., actin, glucocorticoid receptor, histone, immunoglobulin heavy or light chain, metallothionein) or a virus (e.g., adenovirus, adeno-associated virus, baculovirus, cytomegalovirus, herpes simplex virus, Moloney leukemia virus, mouse mammary tumor virus, Rous sarcoma virus, SV40 virus), as well as regions suitable for gene manipulation (e.g., selectable marker, linker with multiple recognition sites for restriction endonucleases, promoter for in vitro transcription, primer annealing sites for in vitro replication). Preferred are transcriptional regulation by tetracyline or dimerized macrolides.

Modulation of gene expression may be effected by affecting transcriptional initiation, transcript stability, translation of the transcript into protein product, protein stability, or a combination thereof. Quantitative effects can be measured by techniques such as in vitro transcription, in vitro translation, northern hybridization, nucleic acid hybridization, reverse transcription-polymerase chain reaction (RT-PCR), run-on transcription, solution hybridization, southern hybridization, cell surface protein labeling, metabolic protein labeling, immunoprecipitation (IP), enzyme linked immunosorbent assay (ELISA), electrophoretic mobility shift assay (EMSA), radioimmunoassay (RIA), immunostaining, and fluorescence activated cell analysis (FACS).

Gene expression can be assayed by use of a reporter or selectable marker gene whose protein product is easily assayed. Reporter genes include, for example, alkaline phosphatase, β-galactosidase (LacZ), chloramphenicol acetyltransferase (CAT), β- glucoronidase (GUS), green fluorescent protein (GFP), horseradish peroxidase (HRP), β-lactamase, luciferase (LUC), and derivatives thereof. Such reporter genes would use cognate substrates that are preferably assayed by a chromogen, fluorescent, or lumi- nescent signal. Alternatively, assay product may be tagged with a heterologous epitope (e.g., FLAG, MYC, SV40 T antigen, glutathione S transferase, hexahistidine, maltose binding protein) for which cognate antibodies or affinity resins are available. Examples of drugs for which selectable marker genes exist are ampicillin, geneticin (G418)/kanamycin/neomycin, hygromycin, puromycin, and tetracycline. A metabolic enzyme (e.g., dihydrofolate reductase, HSV-1 thymidine kinase) may be used as a selectable marker in sensitive host cells or auxotrophs. For example, methotrexate can increase the copy number of a polynucleotide linked to a DHFR selectable marker and gancyclovir can negatively select for a viral thymidine kinase selectable marker.

Another aspect of the invention is a NFAR recombinant like a transcriptional chimera or a translational fusion. In transcriptional chimeras, at least a transcriptional regulatory region of a heterologous gene is ligated to an NFAR polynucleotide or, alternatively, a transcriptional regulatory region of an NFAR gene is ligated to at least a heterologous polynucleotide. The reading frames of an NFAR polypeptide and at least a heterologous amino acid domain are joined in register for a translational fusion. If a reporter or selectable marker is used as the heterologous region or domain, then the effect of mutating NFAR nucleotide or amino acid sequences on NFAR function may be readily assayed. In particular, a transcriptional chimera may be used to localize a regulated promoter of a NFAR gene and a translational fusion may be used to localize NFAR protein in the cell. Sequence specificity may be changed or conferred by joining a NFAR polypeptide to a heterologous DNA-binding domain (DBD) of known sequence specificity. For translational fusions, a protease recognition site (e.g., for enterokinase, Factor Xa, or thrombin) may be included.

According to yet another aspect of the invention, an NFAR DNA is transcribed to produce an NFAR RNA transcript, the NFAR RNA is translated to produce an NFAR nascent chain which is folded, and there might be post-translational processing (e.g., acetylation, acylation, amidation, disulfide bonding, glycosylation, phosphorylation, hydroxylation of γ-carboxyglutamic acid, methylation, phosphorylation, proteolysis, and sulfatation). Nascent chain, folded NFAR, and post-translationally processed NFAR nascent chain which is folded, and there might be post-translational processing (e.g., acetylation, acylation, amidation, disulfide bonding, glycosylation, phosphorylation, hydroxylation of γ-carboxyglutamic acid, methylation, phosphorylation, proteolysis, and sulfatation). Nascent chain, folded NFAR, and post-translationally processed NFAR are known generically as polypeptide. A native human NFAR polypeptide has a relative mobility of about 90 kDa or about 110 kDa in denaturing PAGE-SDS (uncertainty in molecular weight determined from relative mobility can be ± 10%).

Typically, a nucleotide sequence may show as little as 85% sequence identity, and more preferably at least 90% sequence identity compared to SEQ ID NO:1 or 3, excluding any deletions or additions which may be present, and still be considered related. Nucleotide sequence identity may be at least 95% and, more preferably, nucleotide sequence identity is at least 98%. Amino acid sequences are considered to be related with as little as 90% sequence identity compared to SEQ ID NO:2 or 4. But 95% or greater sequence identity is preferred and 98% or greater sequence identity is more preferred.

Use of complex mathematical algorithms is not required if amino acid sequences can be aligned without introducing many gaps. But such algorithms are known in the art, and implemented using default parameters in commercial software packages provided by DNASTAR, Genetics Computer Group, Hitachi Genetics Systems, and Oxford Molecular Group (formerly Intelligenetics). See Doolittle, Of URFS and ORFS, University Science Books, 1986; Gribskov and Devereux, Sequence Analysis Primer, Stockton Press, 1991 ; and references cited therein. Percentage identity between a pair of sequences may be calculated by the algorithm implemented in the BESTFIT computer program (Smith and Waterman, J. Mol. Biol., 147:195-197, 1981 ; Pearson, Genomics, 11:635-650, 1991). Another algorithm that calculates sequence divergence has been adapted for rapid database searching and implemented in the BLAST computer program (Altschul et al., Nucl. Acids Res., 25:3389-3402, 1997).

Conservative amino acid substitutions, such as Glu/Asp, Val/lle, Ser Thr, Arg/Lys and Gln/Asn, may also be considered when making comparisons because the chemical similarity of these pairs of amino acid residues would be expected to result in functional equivalency. Amino acid substitutions that are expected to conserve the biological function of the polypeptide would conserve chemical attributes of the substituted amino acid residues such as hydrophobicity, hydrophilicity, side-chain charge, or size. Functional equivalency or conservation of biological function may be evaluated by methods for structural determination and bioassay as disclosed herein. Thus, amino acid sequences are considered to be related with as little as 90% sequence similarity between the two polypeptides; however, 95% or greater sequence similarity is preferred and 98% or greater sequence similarity is most preferred.

The codons used in the native nucleotide sequences may be adapted for translation in a heterologous host by adopting the codon preferences of the host. This would accommodate the translational machinery of the heterologous host without a substantial change in the chemical structure of the polypeptide. NFAR polypeptide and its variants (i.e., deletion, domain shuffling or duplication, insertion, substitution, and combinations thereof) are useful for determining structure- function relationships (e.g., alanine scanning, conservative or non-conservative amino acid substitution). See Wells (Bio/Technology, 13:647-651 , 1995) and US Patent 5,534,617. Directed evolution by random mutagenesis or gene shuffling (see US Patent 5,811 ,238, WO 97/35966, WO 98/27230, WO 98/31837, WO 98/42832, WO 99/23236, WO 99/29902, and WO 99/62847) using NFAR and other members of the family of dsRNA-binding proteins. Mutant and analog NFAR polypeptides are encoded by suitable mutant and analog NFAR polynucleotides.

Structure-activity relationships of NFAR may be studied (i.e., SAR studies) using variant polypeptides produced by an expression construct transfected in a host cell with or without endogenous NFAR. Thus, mutations in discrete domains of the NFAR polypeptide may be associated with decreasing or even increasing activity in the protein's function. As described in the Examples, a variety of mutations have been made to define domains critical for the function of NFAR-1 and NFAR-2. NFAR function involves regulation of gene expression, at the level of transcription or post-transcription, and include RNA processing. NFAR-1 and NFAR-2 differ in their ability to regulate gene expression and may differ in the genes they regulate (e.g., NFAR-1 may be a more potent regulator of PKC-Θ gene transcription). NFAR-1 and NFAR-2 may also interact with each other and, for example, regulate the other's activity. Mutants have also been used to map PKR phosphorylation to the N-terminal 250 residues of NFAR. Agents which bind NFAR (i.e., drug screening) may also be useful for potentiating or inhibiting NFAR function in the cell. The action of an agent can be be determined by comparing the agent's effects on the two isoforms and mutant versions thereof. Mutations that are functionally significant and genetic polymorphisms in NFAR nucleotide and amino acid sequences are also aspects of the invention. Mutations may be in regulatory and/or coding regions of the gene. They are also useful to establish structure-function relationships in NFAR. Such mutations may also be useful to detect cancer cells or to monitor tumor evolution (e.g., solid tumors, leukemias, lymphomas). Polymorphism in an NFAR gene may be used in linkage mapping and to study a possible role for NFAR in genetic disease. Identification of mutations by molecular genetic or cytogenetic techniques may also determine how NFAR expression is activated in a cancer cell. An NFAR nucleotide sequence can be used to produce a fusion polypeptide with at least one heterologous peptide domain (e.g., an affinity or epitope tag). Oligopeptide is useful for producing specific antibody and epitope mapping of NFAR-specific antibody. A polypeptide may be at least 15, 25, 50, 100, 250, 500, 750 or 1000 amino acids long. Oligopeptide may be conjugated to one affinity tag of a specific binding pair (e.g., antibody-digoxygenin/hapten/peptide, biotin-avidin/streptavidin, glutathione S transferase-glutathione, maltose binding protein-maltose, polyhistidine-nickel, protein A or G/immunoglobulin). Either a full-length NFAR polypeptide (SEQ ID NO:2 or 4), a C- terminal fragment thereof (e.g., starting from residue 595 of NFAR-1 or NFAR-2), a shorter fragment unique to the NFAR amino acid sequence (e.g., residues 612 to 702 of NFAR-1 , residues 612 to 894 of NFAR-2, residues 703-894 of NFAR-2), or an oligopeptide unique to an NFAR isoform (e.g., residues 688 to 702 of NFAR-1 or NFAR-2) can be produced; optionally including a heterologous peptide domain. The latter oligopeptides have been used to raise antiserum.

NFAR polypeptide may be synthesized by chemical means, purified from natural sources, synthesized in transfected host cells, or combinations thereof. Polypeptides synthesized in a transfected bacterium from an expression construct will be devoid of eukaryotic post-translational modifications but, if such are desired, an expression construct may be transfected into a suitable eukaryotic cell (e.g., yeast, other fungus or mold, insect, human or other mammal, hamster, mouse, rat, somatic or stem) or organism (e.g., insect, fish, bird, plant, non-human mammal, hamster, mouse, rat, rabbit, goat, sheep, cattle). NFAR polynucleotides perse or an expression construct comprising an NFAR polynucleotide may be introduced into the host cell or organism by a technique such as chemical transfection (e.g., calcium phosphate, DEAE-dextran), Hpofection, electroporation, naked DNA transfection, biolistics, infection by recombinant virus, or microinjection; preferably, the introduced polynucleotide is an expression construct and the expression construct integrates into the eukaryotic genome of the host cell or organism. Alternatively, a homologous region from the NFAR gene can be used to target a heterologous regulatory region to the NFAR locus in the eukaryotic genome and activate endogenous NFAR transcription controlled by the heterologous regulatory region (i.e., gene activation of endogenous NFAR). NFAR polypeptide may be produced in vitro by culturing transfected cells, produced in vivo by transgenesis, or cells may be transfected ex vivo and then transplanted into an organism. The NFAR nucleotide sequence or a portion thereof can be used as a probe to monitor NFAR rearrangement and/or expression, especially in tumor cells from tissue culture or patient biopsies. The invention also provides hybridization probes and amplification primers (e.g., polymerase chain reaction or PCR, ligation chain reaction or LCR, other isothermal amplification reactions). A pair of such primers has been used for RT-PCR assays to quantitate NFAR transcript abundance within cells. Amplification primers may be between 15 and 30 nucleotides long (preferably about 25 nucleotides), anneal to either sense or antisense strand (preferably the pair will be complementary to each strand), and terminate at the 3' end anywhere within SEQ ID NOS:1 and 3 or their complements (preferably within -10 to +10 nucleotides of an intron/exon boundary). Therefore, this invention will be useful for development and utilization of NFAR primers and other oligonucleotides to quantitate cognate RNA and DNA within cells. This information may then be used for determination of gene rearrangement and expression profiling of NFAR in tumorigenesis; NFAR expression and function may be disregulated in tumor cells. Expression profiling of NFAR may also be performed in virally-infected cells, IFN-treated cells, and other cells in which there is activation of host defense. A host cell may be transfected with an expression construct comprised of an NFAR polynucleotide. Also provided are NFAR transgenic non-human organisms and mutants (e.g., site-directed mutations and gene knock outs) thereof, and NFAR mutants of human somatic cells. One NFAR allele has been knocked out in an ES cell line and is being used to generate a mouse with the null allele.

According to another aspect of invention, an antibody raised against NFAR in rabbits can be used as a marker to detect the expression of NFAR in human tumor tissues by immunohistochemical and cytological techniques, or immunoblotting. Other polyclonal and monoclonal antibodies may be prepared by immunizing animals (e.g., chicken, hamster, mouse, rat, rabbit, goat, horse) with NFAR antigen. Antibody fragments may be prepared by proteolytic cleavage or genetic engineering; humanized antibody and single chain antibody may be prepared by transplanting sequences from the antigen binding domains of antibodies to framework molecules. In general, specific binding molecules may be prepared by screening a combinatorial library for a member which specifically binds NFAR antigen (e.g., phage display library). NFAR antigen may be full-length protein or fragment(s) thereof. See US Patents 5,403,484, 5,723,286, 5,733,743, 5,747,334, and 5,871 ,974. Antisense polynucleotides are presumed to directly block translation of NFAR by hybridizing to complementary target RNA and thereby preventing its translation. They are generally oligonucleotides composed of deoxyribonucleotides or ribonucleotides, especially those with methyl phosphonate and phosphorothioate backbones, and complementary to the translation initiation site (e.g., between the -10 and +10 regions of the target nucleotide sequence).

Ribozymes catalyze specific cleavage of an NFAR transcript. The mechanism of ribozyme action involves sequence-specific hybridization to complementary target RNA, followed by endonucleolytic cleavage. The composition of the ribozyme includes one or more sequences complementary to the target RNA and catalytic sequences responsible for mRNA cleavage (e.g., hammerhead and hairpin motifs). For example, specific ribozyme cleavage sites within a potential RNA target are initially identified by scanning an NFAR transcript for ribozyme cleavage sites which include the following sequences: GUA, GUU and GUC. Once identified, an oligonucleotide of between about 15 and about 20 ribonucleotides corresponding to the region of the RNA target containing the cleavage site can be evaluated for predicted structural features, such as secondary structure, that can render the oligonucleotide sequence unsuitable. The suitability of candidate sequences can then be evaluated by their ability to hybridize and cleave an NFAR transcript using ribonuclease protection assays.

Specific binding molecules used in triplex helix formation for the inhibition of transcription should be single stranded and composed of deoxyribonucleotides. The base composition of these oligonucleotides must be designed to promote triple helix formation via Hoogsteen base pairing rules, which generally require sizeable stretches of either purines or pyrimidines to be present on one strand of a duplex. Nucleotide sequences can be pyrimidine-based and result in TAT and CGC triplets across the three associated strands of the resulting triple helix. The pyrimidine-rich molecules provide base complementarity to a purine-rich region of a single strand of the duplex in a parallel orientation to that strand. In addition, specific binding molecules can be chosen that are purine-rich (e.g., containing a stretch of guanines). These molecules will form a triple helix with a DNA duplex that is rich in GC pairs, in which the majority of the purines are located on a single strand of the targeted duplex, resulting in GGC triplets across the three strands in the triplex. Alternatively, the potential sequences that can be targeted for triple helix formation can be increased by creating a switchback molecule. Switchback molecules are synthesized in an alternating 5'-3', 3'-5' manner, such that they base pair with first one strand of a duplex and then the other, eliminating the necessity for a sizeable stretch of either purines or pyrimidines to be present on one strand of a duplex.

According to another aspect of invention, oligonucleotides can be selected from an NFAR nucleotide sequence. This set of primers is specific for amplification of an NFAR gene and can be used in pairs for PCR and RT-PCR amplification of DNA and RNA, respectively. A single oligonucleotide can be used for specific hybridization to a NFAR nucleotide sequence.

An NFAR polynucleotide may be ligated to a linker oligonucleotide or conjugated to one affinity tag of a specific binding pair (e.g., antibody-digoxygenin/hapten/peptide epitope, biotin-avidin/streptavidin, glutathione S transferase-glutathione, maltose binding protein-maltose, polyhistidine-nickel, protein A/G-immunoglobulin). The NFAR polynucleotide may be conjugated by ligation of a nucleotide sequence encoding the affinity tag or by direct chemical linkage to a reactive moiety on the affinity tag by cross- linking.

Polynucleotides and polypeptides may be used as an affinity tag to identify, to isolate, and to detect interacting proteins that bind an NFAR gene or an NFAR protein. Optionally, bound complexes may be identified, isolated, and detected indirectly though a specific binding molecule for the NFAR gene, the NFAR protein, or the interacting protein. Such interacting proteins may regulate NFAR gene expression (e.g., affinity chromatography of DNA-binding proteins, electrophoretic mobility shift assay, one- hybrid system) or form protein complexes with regulate the cellular function of NFAR (e.g., membrane protein cross-linking, screening a phage display library, two-hybrid system). The invention is not limited to such protein agents but may also be used to identify, to isolate, and to detect other chemical agents which may regulate NFAR gene expression or NFAR protein function by screening, for example, a combinatorial or natural product library for agents which potentiate or inhibit the IFN signaling pathway. Binding of polynucleotides, polypeptides, or polynucleotide/polypeptide may take place in solution or on a substrate. Attachment of NFAR polynucleotide or polypeptide, interacting protein, or specific binding molecule to a substrate before, after, or during binding results in capture of an unattached species. See US Patents 5,143,854 and 5,412,087. A set of oligopeptides which define all possible linear epitopes of NFAR may be arranged on a substrate to map the epitope specifically bound by a binding molecule (e.g., polyclonal or monoclonal antibody). Once a reactive epitope is defined, it may be used to isolate the specific binding molecule or to inhibit binding between NFAR and the specific binding molecule. See US Patent 5,194,352.

Thus, NFAR polynucleotide, NFAR polypeptide, or specific binding molecule may be optionally attached to a substrate. The substrate may be solid or porous and it may be formed as a sheet, bead, or fiber. The substrate may be made of cotton, silk, or wool; cellulose, nitrocellulose, nylon, or positively-charged nylon; natural rubber, butyl rubber, silicone rubber, or styrenebutadiene rubber; agarose or polyacrylamide; silicon or silicone; crystalline, amorphous, or impure silica (e.g., quartz) or silicate (e.g., glass); polyacrylonitrile, polycarbonate, polyethylene, polymethyl methacrylate, polymethylpentene, polypropylene, polystyrene, polysulfone, polytetrafluoroethylene, polyvinylidenefluoride, polyvinyl acetate, polyvinyl chloride, or polyvinyl pyrrolidone; or combinations thereof. Optically-transparent materials are preferred so that binding can be monitored and signal transmitted by light. Such reagents would allow capture of a molecule in solution by a specific interaction between the cognate molecules and then could immobilize the molecule on the substrate. Monitoring NFAR expression is facilitated by using a microarray.

Polynucleotide, polypeptide, or specific binding molecule may be synthesized in situ by solid-phase chemistry or photolithography to directly attach the nucleotides or amino acids to the substrate. Attachment of the polynucleotide, polypeptide, or specific binding molecule to the substrate may be through a reactive group as, for example, a carboxy, amino, or hydroxy radical; attachment may also be accomplished after contact printing, spotting with a pin, pipetting with a pen, or spraying with a nozzle directly onto a substrate. Alternatively, the polynucleotide, polypeptide, or specific binding molecule may be reversibly attached to the substrate by interaction of a specific binding pair (e.g., antibody-digoxygenin/hapten/peptide, biotin-avidin/streptavidin, glutathione S transferase-glutathione, maltose binding protein-maltose, polyhistidine-nickel, protein A or G/immunoglobulin); cross-linking may be used if irreversible attachment is desired. By synthesizing the polynucleotide, polypeptide, or specific binding molecule in situ or otherwise attaching it to a substrate at a predetermined position or to a coded substrate, the identity of the polynucleotide, polypeptide, or specific binding molecule can be determined without sequence analysis. For example, a polynucleotide, polypeptide, or specific binding molecule of known sequence can be determined by its position (e.g., cartesian or polar coordinates) or decoding a signal (e.g., combinatorial tag, electromagnetic radiation) on the substrate. A nucleotide or amino acid sequence will be correlated with each position or signal of the substrate(s). A substrate may have a pattern of different polynucleotides, polypeptides, or specific binding molecules (e.g., at least 1 ,000, 10,000, 100,000 or 1 ,000,000 distinguishable positions) at high density (e.g., at least 1 ,000, 10,000, 100,000 or 1 ,000,000 distinguishable positions per cm²). The number of sequences that can be differentiated by the signal is only limited by the physical uncertainties (e.g., number of combinations, interference between a property of electromagnetic radiation like wavelenth, frequency, energy, polarization, etc.). Multiplex analysis may be used to monitor NFAR expression together with expression of oncogenes, tumor suppressor genes, transcription factors, cell cycle regulators, IFN receptors, kinases, phosphatases, other downstream effectors of the dsRNA signaling pathway, or combinations thereof. Such multiplex analysis may be performed using different polynucleotides, polypeptides, or specific binding molecules arranged in high density on a substrate (i.e., a microarray). However, simultaneous solution methods such as multi-probe ribonuclease protection assay or multi-primer pair amplification associate each transcript with a different length of detected product which is resolved by separation on the basis of molecular weight.

A screening method may comprise administering a candidate chemical agent to an organism, or incubating a candidate chemical agent with a cell or tissue, and determining whether or not NFAR gene or protein activity is modulated. Modulation may be an increase or decrease in activity (i.e., potentiation or inhibition for NFAR function, respectively). Optionally, PKR or dsRNA may be added in assays. NFAR gene or protein activity may be increased at the level of rate of transcriptional initiation, rate of transcriptional elongation, stability of transcript, translation of transcript, rate of translational initiation, rate of translational elongation, stability of protein, rate of protein folding, proportion of protein in active conformation, functional efficiency of protein (e.g., activation or repression of transcription), or combinations thereof. See US Patents 5,071 ,773 and 5,262,300. High-throughput screening assays are preferred. The screening method may comprise incubating a candidate chemical agent with a cell containing a reporter construct, the reporter construct comprising an NFAR- responsive transcription regulatory region covalently linked in a cis configuration to a downstream gene encoding an assayable product; and measuring production of the assayable product. A candidate chemical agent which increases production of the assayable product would be identified as an agent which activates gene expression from the NFAR-responsive region, and a candidate chemical agent which decreases production of the assayable product would be identified as an agent which inhibits gene expression from the NFAR-responsive region. See US Patents 5,849,493 and 5,863,733.

The screening method may comprise measuring in vitro transcription from a reporter construct incubated with NFAR (or fragment thereof) in the presence or absence of a candidate chemical agent, the reporter construct comprising a transcription regulatory region which is responsive to NFAR (or fragment thereof); and determining whether transcription is altered by the presence of the candidate chemical agent. In vitro transcription is preferably assayed using a cell-free extract (more preferably, a nuclear extract); partially purified fractions of the cell-free extract; purified transcription factors or RNA polymerase; or combinations thereof. See US Patents 5,453,362, 5,534,410, 5,563,036, 5,637,686, 5,708,158, and 5,710,025.

The screening method may comprise incubating NFAR (or fragment thereof) with a candidate chemical agent and PKR (or fragment thereof); and determining the amount of the PKR (or fragment thereof) which is associated with NFAR (or fragment thereof), the desired chemical agent being one which increases or decreases specific interaction between PKR and NFAR. Preferably, at least one of the NFAR (or fragment thereof) and the PKR (or fragment thereof) is immobilized to a substrate to facilitate separation of bound from unbound complexes.

Methods for measuring transcriptional or translational activity in vivo can be any which are known. For example, a nuclear run-on assay may be employed to measure transcription of a reporter gene. The translation of the reporter gene may be measured by determining the activity of the translation product. Methods for measuring the activity of an assayable product of certain reporter genes are well known. Candidate chemical agents can also be screened for use in regulating gene expression in apoptosis or tumorigenesis, activation of the host defense response, and other dsRNA-regulated signaling events by their ability to modulate the activity of NFAR. The ability of a candidate chemical agent to modulate the transcriptional activity of NFAR may be assessed by measuring transcription from an NFAR-responsive regulatory region.

A transcription reaction comprises a regulatory region responsive to NFAR and a reporter gene. The reporter gene operably linked to the regulatory region in a reporter construct could be any gene known in the art. In a preferred embodiment, the length of the promoter region to be assayed is less than 200 bp and no more than 1000 bp. For example, such regulatory regions may comprise upstream sequences of the Fas or IFNβ genes.

Suitable methods for measuring in vitro transcription are known. In vitro transcription may be carried out by incubating a reporter construct, labeled nucleotides (e.g., [³²P]-ATP), transcriptionally active cell-free extract, nucleotides, and buffer reagents in the presence and absence of a candidate chemical agent. The procedures for producing cell-free extracts and partially purified fractions are well-described in the art; the conditions for in vitro transcription are also well known. The labeled transcript can be separated by slab or capillary gel electrophoresis, detected by autoradiography, and quantitated by any technique known in the art. Optionally, in vitro transcription can be carried out in the presence of NFAR and/or PKR.

A candidate chemical agent which increases production of an assayable product in the cell or the amount of in vitro transcription indicates its ability to enhance NFAR activity. In contrast, a candidate chemical agent which decreases production of assay- able product in the cell or the amount of in vitro transcription indicates its ability to reduce NFAR activity. These agents can potentially be administered to a human.

Candidate chemical agents regulating the binding between PKR and NFAR may be identified and isolated. NFAR can be attached to a support. A candidate chemical agent is incubated with the immobilized NFAR in the presence of PKR. Alternatively, PKR can be immobilized on a support and a candidate chemical agent can be incubated with the immobilized PKR in the presence of NFAR. After incubation, non- binding components can be washed away, leaving PKR bound to NFAR/support or NFAR bound to PKR/support, respectively. Washing may be facilitated by forming the solid support into a bilious strip, a well of a 96-well plate, a bead or end of an optical fiber, a chromatography column, or a porous membrane. Solution transfer may be accomplished by fluid channels, magnetic particles, or robotics.

The amount of NFAR or PKR can be quantified by any means known in the art. For example, it can be determined using a binding assay detected by autoradiography, enzyme colorimetry, excitation energy transfer, fluorescence polarization, fluorescence quenching, liquid scintillation, or surface plasmon resonance. The amount of bound NFAR or PKR may be compared with and without the candidate chemical agent. A desirable agent is one which increases or decreases the binding of NFAR to PKR.

Bound complex may be visualized by X-ray crystallography or nuclear magnetic resonance spectroscopy to identify contact points between subunits of the oligomer. Small molecule mimetics can be designed to increase or decrease formation of oligomers. See US Patents 5,790,421 and 5,835,382.

The identification of other genes and proteins whose expression or activity is NFAR-dependent will provide additional targets for drug development. Gene expression profiles may be compared prior to and after induction of NFAR transcription or transcriptional activation by NFAR. Transcription of NFAR-dependent genes may be activated by introducing the NFAR gene under the control of an inducible promoter into a host cell that lacks endogenous NFAR activity.

NFAR-dependent genes may be identified by techniques detecting differential expression such as a subtractive cDNA library screened with post-induction transcripts minus pre-induction transcripts, or by differential screening of cDNA or genomic clone libraries. Differential message display (US Patents 5,459,037, 5,599,672, 5,665,544, 5,707,807, 5,807,680, 5,814,445, 5,851 ,805, and 5,876,932); subtractive hybridization (US Patents 5,316,925, 5,643,761 , 5,804,382, 5,830,662, 5,837,468, 5,846,721 , and 5,853,991); computer-assisted comparison with an electronic database (e.g., US

Patent 5,840,484); differential screening of arrayed cDNA clones or libraries (e.g., US Patents 4,981 ,783, 5,206,152, and 5,624,801); reciprocal subtraction differential display (US Patent 5,882,874); serial analysis of gene expression (US Patent 5,866,330); and other proprietary techniques (e.g., US Patents 5,871 ,697, 5,972,693, and 6,013,445) may be used to identify NFAR-dependent genes.

The NFAR-dependent gene transcripts will be translated into NFAR-dependent proteins, such proteins may be identified by comparing the pattern of proteins expressed prior to and after induction of NFAR (with or without dsRNA). For example, pre- and post-induction cultures of the host cells may be [³⁵S]-pulsed, protein extracts may be made from whole cell lysates or subcellular fractions, and NFAR-dependent proteins will be identified by their increased or decreased signal intensity in two- dimensional gels of [³⁵S]-labeled proteins from pre- and post-induction cultures. Proteins of interest (i.e., labeled proteins which increase or decrease in abundance) may be isolated, N-terminal or internal peptide amino acid sequence may be determined, and the NFAR-dependent genes of interest identified by cloning with degenerate polynucleotides whose sequences are predicted according to the determined amino acid sequence. NFAR-dependent genes may also be identified by promoter trapping. NFAR may be induced in cells after introducing the NFAR gene under the control of an inducible promoter into a host cell that lacks endogenous NFAR activity. A clone library of gene fragments inserted into a promoter vector can be constructed to operably link the gene fragment with a reporter gene, such that a promoter contained in the gene fragment may direct the transcription of the indicator gene. A suitable indicator gene will be transcribed and produce a detectable indicator product under appropriate assay conditions. Individual clones of the library may be introduced into the host cell, and colonies replica plated under conditions of NFAR induction or non-induction. Gene fragments will be isolated from colonies which produce indicator product only when NFAR activity is induced because they could contain NFAR-dependent promoters.

Alternatively, a construct containing the indicator gene but no operably linked promoter may be randomly integrated into the chromosome of a cell. Clones which contain integrations near NFAR-dependent promoters may be identified after induction of NFAR activity by screening for the indicator product. Those integration sites could mark the sites of NFAR-dependent promoters and isolating the NFAR-dependent genes associated with such promoters may also identify NFAR-dependent genes.

Differentially expressed genes may be isolated and cloned through differential message display, RNA fingerprinting, representational difference analysis (RDA), subtractive hybridization, substraction between electronic databases, differential screening of arrayed cDNA clones or libraries, reciprocal subtraction differential display, serial analysis of gene expression, and generation of expressed sequence tags (EST). See Soares (Curr. Opin. Biotechnol., 8:542-546, 1997) and references cited therein.

All publications, patent applications, and patents cited in this specification are incorporated by reference in their entirety where they are cited. Such references are also cited as indicative of the skill in the art.

The following examples are meant to be illustrative of the present invention, however practice of the invention is not limited or restricted in any way by them.

EXAMPLES Plasmids:

The construction of PKR K296R in pET11a (Novagen) has been described by Barber et al. (1991). To create fusion GAL4 DNA-binding domain/PKR polypeptides, the EcoRI site in pGBT9 (Clontech) was mutagenized to create an Ndel site and thereby generate pGBTIO. Ndel/BamHI-digested PKR K296R encoding all 551 amino acids of the PKR gene was then inserted into similarly treated pGBTIO, thereby generating pGBT DM. Construction of expression constructs encoding the amino terminus of PKR

(residues 1-296) and the carboxyl terminus of PKR (residues 244-551) has been described by Katze et al. (1991). Ndel sites were inserted at the initiating ATG codon of PKR 1-296 and at codon 244 (Banl site) of PKR 244-551 , respectively (Barber et al., 1995). Inserts were retrieved using Ndel/BamHI and inserted into pGBTIO to create pGBT 1-296 and pGBT 244-551. Yeast vectors expressing vaccinia virus E3L have been described by Sharp et al. (1998).

Full-length cDNA of NFAR-1 and NFAR-2 were made by subcloning C-terminal Pstl-Bglll NFAR fragments from yeast two-hybrid pGAD10 vectors into pGAD424 (Clontech). The N-terminus of NFAR was obtained from cDNA characterized from HeLa cells and cloned into Ndel-Pstl digested pGAD424-(Ndel) vectors containing the C-terminus of NFAR 1 or NFAR 2. For bacterial expression, NFAR-1 and NFAR-2 were subcloned into pET14b (Novagen) to generate expression constructs. For mammalian cell expression, NFAR-1 and NFAR-2 cDNA were subcloned into EcoRV/ BamHl digested pcDNA3.1 (Invitrogen) to generate expression constructs. The cDNA of vaccinia virus proteins E3L and K3L were similarly cloned into Hindi ll/BamHI cut pcDNA3.1.

NFAR-2 deletion mutations M5 (Δ7-173), M7 (Δ174-232), and M8 (Δ238-384) were created in pET14b (bacterial) and pcDNA3.1 (mammalian) using the QUICK- CHANGE site-directed mutagenesis kit (Stratagene). All variants were confirmed by sequencing. The NFAR-M9 variant (residues 1-417) was created by cloning the NFAR Ndel/Pstl fragment into pGAD424 and subsequently into pET11 and pcDNA3.1 expression vectors. Vaccinia virus E3L was FLAG-tagged by cloning into pCMV-2B. The Fas promoter (Wada et al., 1995) and IFNβ promoter (Thanos and Maniatis, 1995) were cloned into pGL3 (Promega).

Yeast Strains and Two-Hybrid Screen:

To detect proteins that interacted with PKR, the yeast strain S. cerevisiae Y190 was used in a two-hybrid assay. Y190 transfected with pGBT DM were grown overnight in synthetic media lacking tryptophan. Cultures were transfected with a Jurkat cell cDNA library fused to the activation domain of GAL4 in vector pGADIO (Clontech).

Transfectants were plated onto synthetic media lacking tryptophan, leucine, and histidine and containing 25 mM 3-aminotriazole (3-AT). Colonies were allowed to develop for up to seven days at 30°C. Colonies were lifted with nitrocellulose filters and blue colonies identified by galactosidase assay (Chien et al., 1991).

Colonies were grown to mid-late log phase and assayed for β-gal activity to quantitate the interaction between PKR bait and human polypeptides fused to GAL4.

Three individual co-transformants were assayed in each case and the average values of β-gal units determined. The interaction between pVA3 (SV40) and pTD1 (p53) was used as a positive control.

Hexahistidine-Fusion Protein Expression and Purification:

PKR was cut out of pET11a using Ndel/BamHI and inserted into similarily cut pET14b such that the amino terminus of the kinase was fused in-frame with six histidine residues (Novagen). The resultant pET14/PKR expression construct was used to transfect BL21[DE3]pLysS bacteria (Novagen) and cultures grown overnight in the absence of IPTG (Barber et al., 1991). At late log phase, bacteria were treated with 0.5 mM IPTG to induce expression of PKR. Cells were lysed by sonication in high salt buffer (400 mM NaCI, 5 mM imidizole, 20 mM Tris-HCI pH 7.5), 100 units/ml aprotinin, 0.2 mM PMSF, and 0.1% NP-40 surfactant. After centrifugation, the cleared lysate was applied to a nickel affinity column (Novagen). After washing the column, proteins were eluted using 100 mM imidazole in the same buffer and dialyzed against 20 mM Tris- HCI pH 7.5, 100 units/ml aprotinin, 0.2 mM PMSF, 50 mM KCI, 0.5 mM EDTA, 0.5 mM DTT, and 10% glycerol. His-tagged proteins were further purified by SUPEROSE 12 gel filtration (Pharmacia), examined by COOMASSIE dye-stained gels after PAGE- SDS, and quantitated by spectrophotometer or by immunoblotting using anti-PKR monoclonal antibody (Laurent et al., 1985).

NFAR-1 and NFAR-2 cDNA in pET14b were expressed in bacteria as described above. The bacteria were harvested and lysed in 2 mM imidazole, 0.5 mM NaCI, 20 mM Tris-HCI pH 7.9, 0.2 mM PMSF, and 100 units/ml aprotinin (lysis buffer). Inclusion bodies were resuspended in lysis buffer containing 6 M urea and applied to a nickel affinity column (Novagen). The column was washed in lysis buffer containing 6 M urea and 60 mM NaCI. Proteins were eluted with 1 M imidazole, 0.5 M NaCI, 20 mMTris pH 7.5, 0.2 mM PMSF, 100 units/ml aprotinin, and 6 M urea, and then dialyzed in urea- containing buffer (20 mM Tris pH 8.0, 50 mM NaCI, 0.1 mM EDTA, 0.2 mM PMSF, 100 units/ml aprotinin, 10% glycerol, and 1 M urea). Urea was removed by dialyzing in the same buffer lacking the chaotropic agent. Purified His-tagged NFAR proteins were examined by COOMASSIE dye-stained gels after PAGE-SDS and detected by immunoblotting using a monoclonal antibody to the hexahistidine tag (R&D Systems).

For in vitro expression of [³⁵S]-labeled E3L, the TNT transcription/translation system was used (Promega).

dsRNA-Binding Assays:

The dsRNA-binding activity of NFAR was measured by incubating a purified His- tagged NFAR protein or HeLa cell lysate with poly l:C agarose. HeLa cells were lysed in binding buffer (20 mM Tris pH 7.5, 50 mM KCI, 500 mM NaCI, 1% NP-40 surfactant, 1 mM EDTA, 1 mM DTT, 100 mM PMSF, and 1 mg/ml of aprotinin). Lysates were adjusted to 200, 500 and 1000 mM NaCI concentrations and incubated with 0.1 ml of poly l:C agarose slurry. Agarose was incubated at 4°C for 3 hr and then washed five times with 1 ml of the binding buffer at the indicated NaCI concentrations. Competition assays were done by adding excess dsRNA (poly l:C from Pharmacia), ssRNA (polyA from Sigma), and dsDNA (poly dA:dT from Sigma). After the final wash, poly l:C agarose was resuspended in protein loading dye, boiled, and separated by PAGE- SDS. Proteins were transferred to nitrocellulose membrane and probed with anti- NFAR antiserum or with anti-hexahistidine monoclonal antibody.

Confocal Microscopy:

HeLa or Cos-7 cells were fixed at room temperature in freshly prepared 1 % paraformaldehyde in PBS, washed with PBS, and permeabilized for 20 min on ice with 0.2% TRITON X-100 surfactant. Antigen localization was determined after incubation of permeabilized cells with mouse monoclonal antibody against PKR, rabbit antiserum against NFAR, anti-tubulin mouse mAb, anti-FLAG antibody (M2, Stratagene), or with DAPI to stain nuclear DNA, in PBS for 1 hr at room temperature. Cells were double labeled with the respective secondary antibodies conjugated with TEXAS RED dye or FITC. For exponentially growing cells, an AXIOVERT confocal scanning microscope (Zeiss) was used. For analysis for cells in mitosis, a BX40 immunoflourescence microscope (Olympus) was utilized.

PKR Kinase Assays: For in vitro assays, approximately stoichiometric amounts of purified PKR was incubated with His-tagged NFAR-1 or NFAR-2 in kinase buffer (20 mM Tris-HCI pH 7.5, 0.01 mM EDTA, 50 mM KCI, 10 μg/ml aprotinin, 0.3 mg/ml BSA, 2 mM MgCI₂, 2 mM MnCI₂, 1.25 μM [γ-³²P]-ATP, 0.1 mM PMSF, and 5% glycerol). Activators (poly l:C or heparin from Sigma) were then added in various concentrations and incubated for 15 min at 30°C. Reactions were stopped using an equal volume of 2 X protein loading buffer (5% SDS, 5% β-mercaptoethanol, 150 mM Tris-HCI pH 6.8, and 20% glycerol), boiled, and separated by 10% PAGE-SDS; results were visualized by autoradiography.

Immunoblots and Antibodies: PKR protein expression in yeast was detected by growing cultures overnight in synthetic medium. The yeast were grown in YPD medium for about 3-5 hr, rinsed in PBS, and incubated in the presence of zymolase in EDTA buffer. Cells were washed in PBS and lysed in protein loading buffer as above. Following separation by 10% PAGE-SDS, the proteins were transferred to nitrocellulose and incubated with anti- human PKR monoclonal antibody (Laurent et al., 1985). Blots were subsequently washed and incubated with anti-mouse antibody conjugate (Gibco BRL) and specific proteins visualized using a chemiluminescence substrate (Pierce Chemicals). To analyze protein expression in mammalian cells, the same procedure was carried out as above, except that the cells were first disrupted in a lysis buffer (10 mM Tris-HCI pH 7.5, 50 mM KCI, 1 mM DTT, 2mM EDTA, 0.2 mM PMSF, 100 units/ml aprotinin, and 1% TRITON X-100 surfactant). An equal volume of protein disruption buffer was added to the lysate and boiled before separation by PAGE-SDS. For NFAR analysis, a rabbit antiserum raised against E. co//-produced NFAR-1 was used. For NFAR protein analysis in human tissue, extracts were prepared from whole-tissue homogenates (Clontech) and analyzed using the anti-NFAR rabbit antiserum.

Northern Blot Analysis: Adult and fetal human multiple tissue northern blots (Clontech) were hybridized with radiolabeled cDNA probe representing NFAR-1. Blots were washed and autoradiographed prior to being re-probed with radiolabeled actin cDNA as a control.

Co-lmmunoprecipitation Assays: For in vitro analysis, equal amounts of His-tagged PKR and NFAR-1 were incubated in binding buffer (0.2% NP-40 surfactant, 80 mM NaCI, 1 mM EDTA, 100 units/ml aprotinin, 0.2 mM PMSF, 1 mM DTT, and 20 mM Tris-HCI pH 7.5) on ice for 10 min. To immunoprecipitate PKR, anti-PKR monoclonal antibody or pre-immune mouse IgG was added to the mix and rotated at 4°C for 2 hr. Protein G agarose was added for 1 hr. Proteins were washed four times in cold incubation buffer and boiled in protein disruption buffer as above. Proteins were separated by PAGE-SDS and analyzed by immunoblotting using anti-NFAR rabbit antiserum. To immunoprecipitate NFAR, rabbit antiserum raised to NFAR-1 or control pre-immune rabbit antiserum was added to the proteins as above. After washing, the proteins were electrophoresed as described and immunoblotted using anti-PKR monoclonal antibody.

For in vivo co-immunoprecipitation analysis, HeLa cells (3 x 10⁶ per incubation) were lysed in RIPA buffer (1% deoxycholate, 0.2% SDS, 1% NP-40 surfactant, 20 mM Tris pH 7.5, 100 mM NaCI, and protease inhibitors as above). Anti-PKR, anti-NFAR, or pre-immune antibodies were added to total cell extracts and incubated for 2 hr. Protein G was added to the mix and after 1 hr the cells were washed with the same lysis buffer four times. The proteins were examined by immunoblotting using anti-PKR or anti- NFAR antibodies as described above.

In Vivo Labeling of Cells:

HeLa cells were treated for 4 hr with 10 μg/ml poly l:C (Sigma) in the presence of [³²P]-orthophosphate. The cells were disrupted in lysis buffer, as described above for co-immunoprecipitation assays and incubated with pre-immune serum and/or protein G to pre-clear. After pre-clearing for 1 hr by centrifugation, poly l:C agarose (Pharmacia) were added to the incubations for 2 hr. Protein G was added to the mix. The extracts were then washed four times prior to electrophoresis and autoradiography.

Transient Transfection Assays:

Cos-7 cells were cultured in Dulbecco's modified Eagles' medium supplemented with 10% fetal bovine serum. Sixty thousand cells were plated in 12-well dishes for luciferase assays: 0.7 μg of pcDNA3.1 -based expression constructs and 0.3 μg of the pGL3-based expression constructs were co-transfected using the LIPOFECTAMINE PLUS lipid reagents (Gibco BRL). Cells were lysed 48 hr post-transfection in 40 μl of lysis buffer (20 mM Tris-HCI pH 7.5, 50 mM KCI, 100 mM NaCI, 1% NP-40 surfactant, 1 mM EDTA, 1 mM DTT, and 25 mM PMSF). Luciferase assays were read on a TD 20/20 luminometer (Promega). All assays were performed independently, a minimum of three times, in duplicate.

RT-PCR:

Cos-7 cells were transfected as above, except that 10 x 10⁴ cells were seeded into 6-well dishes and transfected with 1 μg of pcDNA3.1 -based expression constructs and 0.4 μg of the reporter expression vector. Total RNA was isolated using TRIZOL reagent (Gibco BRL) according to the manufacturer's protocol. After quantitation, the samples were treated with DNAsel (Gibco BRL). RT-PCR was performed using the PROSTAR ULTRA HF RT-PCR system (Stratagene). First strand cDNA synthesis was performed at 37°C for 30 min. As a control for equal mRNA concentrations, β-actin was reversed transcribed using the following primers: 5'-TGACGGGGTCACCCACAC- TGTGCCCATCTA-3' (SEQ ID NO:52) and 5' CTAGAAGCATTTGCGGTGGACGATG- GAGGG-3' (SEQ ID NO:53). Luciferase mRNA was reverse transcribed using 5'-CAG- CGAGACGGAGTATCTTGACGG-3' (SEQ ID NO:54) and 5'-GAAGGAATCTCTCCCC- TCGCGGTG-3' (SEQ ID NO:55). PCR products were visualized on 1% agarose gels and quantitated using the QUANTITY ONE computer program (BioRad).

Isolation of NFAR-1 and NFAR-2:

To identify new proteins that interact with PKR, we used a yeast two-hybrid system to screen for proteins that bind to full-length kinase (Fields and Song, 1989). Wild-type PKR is toxic to yeast, however, since it becomes activated in vivo, phosphorylates the yeast elF2α homolog SUI2, and blocks translation. To circumvent this problem, a catalytically-inactive PKR variant with an amino acid substitution in catalytic domain II (K296R) was fused to the DNA-binding domain of GAL4 transcrip- tional activator (pGBT Dll) and then used as bait. Yeast expressing PKR was subsequently transfected with a human T-cell cDNA expression library (Jurkat) fused to the GAL4-activation domain and eight positive clones were obtained from screening about 4 x 10⁶ transformants. The specificity of interaction was determined by retrieving the interacting, Jurkat library-derived plasmids from yeast and co-transforming them with control heterologous baits.

Clones GR-1 to GR-8, isolated by a yeast two-hybrid system, were found to contain overlapping 3' co-terminal cDNA sequences, fused to the GAL4-activation domain, in the same reading frame. But further analysis indicated that two related proteins had in fact been isolated since the cDNA sequences representing these clones diverged towards the carboxyl terminal end of the putative open reading frames. Since later studies indicated that the GR clones encoded proteins that bound dsRNA, we named the two isoforms NFAR-1 and NFAR-2 for Nuclear Factors Associating with dsRNA. Their authenticity was confirmed by sequencing analysis of (1) human NFAR cDNA clones isolated from two-hybrid yeast screenings with PKR, (2) additional human NFAR cDNA clones isolated after screening of T-cell cDNA libraries by hybridization, and (3) an entire human NFAR gene from PAC clones isolated by hybridization of a large-insert genomic library. These data confirmed that we had isolated two NFAR cDNAs generated from a differentially spliced mRNA encoded from a single NFAR gene. The two NFAR cDNAs represented open reading frames encoding proteins containing 702 amino acids for NFAR-1 and 894 amino acids for NFAR-2, respectively.

Database searches revealed that residues 401-468 and residues 524-591 of NFAR-1 and NFAR-2 contain amino acid sequences with homology to dsRNA-binding motifs. Therefore, NFAR-1 and NFAR-2 are considered to be members of the dsRNA- binding family of proteins that includes the Drosophila maternal effect protein Staufen and the vaccinia virus protein E3L. Database comparative analysis further revealed that NFAR-1 and NFAR-2 are about 75% indentical at the amino acid level with the X. laevis dsRNA-binding proteins 4F.1 and 4F.2 whose function is unknown. Significantly, both NFAR proteins were also found to share a homologous domain of residues 166- 235 as compared with the PKR substrate elF2α (23% identity or 49% conservation over 69 amino acids). It is notable that this homologous domain of elF2α contains serine 51 , the single target site for PKR phosphorylation. More extensive domains of homology exist from residues 222 to 235 and include a conserved VI RV box. This domain also entirely overlaps with the vaccinia virus encoded PKR inhibitor K3L, an elF2α homolog. It has been reported that K3L binds 400-fold more efficiently than elF2α to PKR and efficiently suppresses the kinase's function.

To further investigate the interaction of PKR with NFAR-1 and NFAR-2, we defined the PKR domains responsible for interacting with human NFAR in yeast. This study revealed that PKR domains containing dsRNA-binding motifs (PKR 1-296 and 1- 551) interacted most efficiently with the carboxyl-terminal domains of both NFAR-1 and NFAR-2 that also contained two dsRNA-binding motifs (i.e., residues 378 to 702 of NFAR-1 and residues 397 to 894 of NFAR-2). Conversely, PKR domains containing the catalytic domains of the kinase (PKR 266-551) and lacking dsRNA-binding motifs did not bind to human NFAR in yeast. It is interesting that full-length NFAR-1 and NFAR-2 did not interact efficiently with full-length PKR or even with PKR 1-296, probably because the conformation of these proteins is affected by being fused to portions of the GAL4 transcription factor. It is also noteworthy that E3L, the vaccinia virus encoded PKR inhibitor that binds dsRNA, also interacted efficiently with carboxyl- terminal domains of NFAR-1 and NFAR-2 (i.e., residues 378 to 702 of NFAR-1 and residues 397 to 894 of NFAR-2). These data indicate that the binding of PKR and NFAR in yeast may be enhanced through dsRNA bridging. In Vivo Expression and Tissue Distribution of NFAR-1 and NFAR-2:

To characterize the in vivo protein expression profile of NFAR isoforms, selected cell lines were examined: equal amounts of whole-cell extracts from Jurkat, HeLa, Cos- 7, 293T, and 3T3-L1 cells were separated by 8.5% PAGE-SDS and immunoblotted with rabbit antiserum raised to recombinant purified His-tagged NFAR-1 from E. coli. This shows that human NFAR exists as at least two major species of relative molecular weight about 90 kDa and about 110 kDa in the cells examined.

To determine whether the isolated NFAR cDNA clones encoded similar-sized proteins, NFAR-1 and NFAR-2 cDNAs were His-tagged and expressed transiently in mammalian 293T cells using pcDNA3.1 -based expression constructs or pcDNA3.1 alone. After 48 hours, cells were lysed and equal amounts of the cell extracts were precipitated using nickel resin. Precipitates were washed, separated by 8.5% PAGE- SDS, and immunoblotted using antiserum to purified NFAR-1. As a control, total lysate from untransfected 293T cells was electrophoresed on the same gel and analyzed by immunoblotting with NFAR antiserum. Immunoblot analysis indicates that recombinant NFAR-1 migrated like the 90 kDa protein observed in the cell lines, while recombinant NFAR-2 migrated like the 110 kDa protein.

Taken together, these observations indicate that both isolated NFAR cDNA are full-length representations of the major NFAR species of 90 kDa and 110 kDa proteins that exist in human cells.

To extend our analysis of the expression profile of the NFAR genes, transcripts and proteins were examined in selected human tissues. Northern blot analysis indicated that NFAR mRNA were synthesized in nearly all human tissues examined. Moreover, we found that a number of mRNA species were apparent (between about 4 and about 8 kb), indicating that a variety of spliced variants of the NFAR gene could exist in the cell. The major species was about 8.0 kb and a minor species was present at about 3.5 kb (uncertainty in molecular weight determined from relative mobility in gel electrophoresis can be ± 10%).

To check the translational status of these transcripts, representative proteins from tissue analyzed by northern blot were examined by immunoblot analysis using anti-NFAR antibody. Equal amounts of protein extract were separated by 8.5% PAGE- SDS and immunoblotted against rabbit antiserum raised against recombinant NFAR-1 or PKR. Actin served as a control. In the majority of tissues examined, the two major species of NFAR, namely the 90 and 110 kDa proteins, were visible, with the 110 kDa protein being more predominant. Interestingly, lesser amounts of NFAR were seen in the liver and spleen and to a certain extent kidney, indicating that the regulation of these proteins may be governed to some extent at the translational or post-translational level. In addition, larger protein species of about 200 kDa were seen in the thymus and ovary, indicating that other NFAR-like proteins may exist in these tissues.

Since human NFAR showed extensive homology to X. laevis 4F.1 and 4F.2, we next examined whether antiserum raised against a human NFAR could cross-react with Xenopus proteins. X. laevis oocytes and embryos at various stages of development were lysed, precipitated with poly l:C agarose, precipitates were washed and then separated by 8.5% PAGE-SDS, and immunoblotted using rabbit antiserum raised against recombinant NFAR-1. S. frugiperda insect cells (Sf9) were similarly analyzed. The antiserum detected two poly l:C binding proteins of relative molecular weight about 95 kDa and about 115 kDa in all Xenopus extracts examined. Furthermore, similar- sized dsRNA-binding proteins were observed in insect cells (i.e., about 90 kDa and about 110 kDa). These data indicate that the proteins detected in Xenopus and insect cell extracts are homologs of NFAR-1 and NFAR-2, and that both proteins have been conserved through evolution.

NFAR-1 and NFAR-2 Bind dsRNA:

To confirm that human NFAR proteins were able to interact with dsRNA, poly l:C agarose or agarose alone were used to precipitate endogenous NFAR-1 and NFAR-2 from HeLa whole-cell extract. Complexes were washed under various salt conditions (i.e., 0 to 1 M NaCI), spearated on 8.5% PAGE-SDS, and immunoblotted using rabbit antiserum raised against recombinant NFAR-1. E. co//-expressed His-tagged NFAR-1 and NFAR-2 were also incubated with poly l:C agarose or agarose alone, and similarly treated. Both NFAR-1 and NFAR-2 bound efficiently to poly l:C agarose, but not to agarose alone. Increasing the salt concentration to 1 M NaCI prevented the binding of the NFAR proteins to poly l:C agarose. Furthermore, recombinant NFAR-1 and NFAR- 2 could also bind to poly l:C agarose, but not to agarose alone.

Whole-cell extracts of HeLa cells were incubated with poly l:C agarose in the presence of increasing amounts of competitor dsRNA (poly l:C from 0.05 to 1 mg/ml), DNA (poly dA:dT from 0.05 to 1 mg/ml), and ssRNA (poly A from 0.05 to 1 mg/ml). Beads were washed and then eluted proteins were separated by 8.5% PAGE-SDS, transferred to nitrocellulose, and immunoblotted with rabbit antiserum raised against recombinant NFAR-1. The binding of human NFAR proteins to poly l:C agarose could be inhibited by preincubating the lysate with dsRNA, but not with competitor ssRNA or dsDNA.

These data confirm that NFAR-1 and NFAR-2 preferentially bind to dsRNA. To confirm the identification of NFAR domains responsible for binding to dsRNA, a number of NFAR variants were constructed and expressed as His-tagged proteins in E. coli: NFAR 2-M5 [Δ 7-173], NFAR 2-M7 [Δ174-232], NFAR 2-M8 [Δ238-384], and NFAR-M9 [Δ 418-702]. Equal amounts of purified His-tagged NFAR proteins were incubated with poly l:C agarose, washed with high salt buffer, separated by 8.5% PAGE-SDS, and developed by immunoblotting with rabbit antiserum raised against recombinant NFAR-1. Immunoblotting analysis with anti-NFAR antibody indicated that all NFAR variants containing the dsRNA-binding motifs were able to bind dsRNA. The only exception was NFAR-M9, which contains residues 1-417 but lacks both dsRNA- binding motifs. Thus, the carboxyl-terminal domain of this human NFAR contains both dsRNA-binding motifs and are essential for interaction with dsRNA.

PKR and NFAR-1 or NFAR-2 Co-lmmunoprecipitate In Vitro and In Vivo: To further analyze the binding of human NFAR with PKR, we determined whether NFAR-1 and NFAR-2 could co-immunoprecipitate with PKR, in vitro and in vivo. Thus, E. co//-expressed and affinity-purified His-tagged PKR and NFAR-1 or NFAR-2 were incubated in vitro at 4°C for 1 hr, and then co-immunoprecipitated using an anti-PKR monoclonal or control antibody. Precipitates were washed and separated by 8.5% PAGE-SDS. Immunoblotting with the NFAR antiserum revealed that both NFAR-1 and NFAR-2 bind to PKR when the anti-PKR antibody, but not the control antibody, was used in the immunoprecipitation reactions. Similarly, PKR was found to reciprocally co-immunoprecipitate with NFAR-1 and NFAR-2 when NFAR antiserum was used, but not with pre-immune serum. Inputs represent about 10% of the total amount of PKR and NFAR proteins used in the initial incubation reaction.

To demonstrate the in vivo interaction of NFAR and PKR, HeLa cell extracts were incubated with the NFAR antiserum. Immunoprecipitates were washed, separated by 8.5% PAGE-SDS, and then immunoblotted. Endogenous PKR co-immuno- precipitated with endogenous NFAR-1 and NFAR-2 from HeLa cells as confirmed using anti-PKR monoclonal antibody in immunoblot analysis. PKR was not detected when pre-immune rabbit serum was used in immunoprecipitation. Likewise, both NFAR-1 and NFAR-2 co-immunoprecipitated with PKR when the anti-PKR monoclonal antibody was used in immunoprecipitation, but not when control antibody of similar heavy chain subtype was used. This indicates that PKR binds to NFAR-1 and NFAR-2 not only in vitro, but also in vivo. Inputs represent about 10% of the HeLa cell extract used in the assays.

PKR Co-Localizes with NFAR in the Nucleus of the Cell:

PKR is predominantly found in the cytoplasm, where it is closely associated with ribosomes. Electron micrographs, however, have also indicated that some PKR may be localized to the nucleus. To determine if PKR co-localizes with the NFAR in the cell, we employed confocal scanning microscopy, in double labeling studies, using rabbit polyclonal or mouse monoclonal antibody directed to NFAR or PKR, respectively.

NFAR antibody was detected with FITC-conjugated anti-rabbit IgG and PKR antibody with TEXAS RED dye-labeled anti-murine IgG. Merged filter analysis reveals co- localization of NFAR and PKR in the nucleus in areas of overlap (i.e., a yellow image). As a control, tubulin antibody was used under similar conditions. HeLa cells incubated with both antibodies revealed that the distribution of NFAR is predominantly in the nucleus. Though the majority of PKR was found in the cytoplasm, a fraction of about 20-25% was also found to be distributed in the nucleus. Importantly, merged filter analysis revealed that both NFAR and PKR co-localized in the nucleus of the cell. HeLa cells similarly treated with a monoclonal antibody to tubulin, a cytoplasmic protein, were also probed with the same TEXAS RED dye- conjugated secondary antibody used in the PKR localization analysis, to demonstrate the specificity of the primary antibodies. These data indicate that NFAR and PKR can co-localize in the interphase cell.

To complement our analysis, we analyzed the cellular distribution of NFAR and PKR during cellular mitosis. We observed that neither NFAR nor PKR was associated with cellular DNA (visualized by DAPI staining) during the mitotic phase of HeLa cells. Similarly, human NFAR and PKR did not appear to be associated with mitotic spindle formation as seen with an anti-tubulin antibody. The functions of human NFAR and PKR in mitosis and control of the cell cycle, if any, has been further characterized.

NFAR-1 and NFAR-2 are Nuclear Phosphoproteins and Substrates for PKR:

Comparison of sequences revealed that these human NFAR proteins exhibited homology with the PKR substrate elF2α. To determine if NFAR proteins are indeed substrates for PKR, NFAR-1 and NFAR-2 were expressed as His-tagged proteins in E. coli. The recombinant material was then purified over nickel-affinity columns to about 90% purity as estimated by COOMASSIE dye-stained gels after PAGE-SDS. PKR was also expressed as a His-tagged protein and underwent a similar purification protocol followed by gel filtration. Equal amounts of the purified, recombinant NFAR proteins were subsequently incubated in the presence of increasing amounts of dsRNA or its absence.

NFAR-1 or NFAR-2 alone showed no indication of autophosphorylation when incubated in the presence of dsRNA and [γ³²P]ATP. In contrast, PKR efficiently auto- phosphorylated when incubated with dsRNA as has been previously described. At high amounts of dsRNA, PKR autophosphorylation is typically inhibited, arguably since inter- molecular PKR-PKR interactions are prevented by each PKR molecule individually binding to dsRNA (Thomis and Samuel, 1995; Barber et al., 1995; Wu and Kaufman, 1997). Significantly, PKR was able to phosphorylate stoichiometric amounts of NFAR- 1 as well as NFAR-2 when incubated together, although NFAR-2 appeared to be marginally less stable during the kinase reaction (possibly because of proteolysis). Phosphorylation of NFAR-1 and NFAR-2 at higher concentrations of dsRNA by PKR was inhibited, similar to the effect seen with PKR alone.

To examine the role of the dsRNA-binding motifs in interactions between PKR and NFAR, an NFAR variant lacking both dsRNA-binding motifs (NFAR-M9) was also examined as a substrate for the kinase. This variant was purified by nickel affinity chromatography and incubated with PKR in the presence of [γ³²P]ATP and increasing amounts of dsRNA. NFAR-M9 served as an efficient substrate for PKR and, thus, the dsRNA-binding motifs of NFAR are not essential for the interaction with PKR or for phosphorylation by this kinase. These data indicate that although PKR and human NFAR can bind through dsRNA bridging, as demonstrated in our yeast studies, dsRNA is not essential for the interaction between these proteins.

To evaluate the in vivo phosphorylation status of human NFAR, including whether these proteins can be phosphorylated by PKR, mouse 3T3 L1 cells inducibly overexpressing PKR (WT-PKR) or control cells transfected with vector alone (VEC) were treated with poly l:C in the presence of [³²P]-orthophosphate for 4 hr. Cells were lysed and extracts were incubated with poly l:C agarose. After washing, precipitates were separated by PAGE-SDS and autoradiographed. Mouse NFAR has a relative molecular weight of about 90 kDa and 110 kDa. This reveals that a proportion of the NFAR exist as phosphoproteins in vivo. But 3T3 L1 cells overexpressing heterologous PKR (WT-PKR) several-fold higher than the endogenous mouse PKR in control cells (VEC) contained three- to four-fold higher amounts of phosphorylated NFAR-1 and NFAR-2. The identities of NFAR-1 and NFAR-2 were confirmed by immunoblotting equal amounts of the poly l:C precipitated protein complexes using rabbit antiserum raised against recombinant NFAR. This analysis confirmed that the amounts of NFAR proteins were equal in both the control and PKR overexpressing cells, demonstrating that the increase in NFAR phosphorylation was not a reflection of elevated NFAR protein amounts. These data indicate that human NFAR proteins exist in vivo as nuclear phosphoproteins and are substrates for PKR.

NFAR-1 and NFAR-2 Augment Gene Expression:

A number of ubiquitous, evolutionarily conserved, nuclear phosphoproteins have been shown to function by linking different signal-responsive transcriptional activators to the basal transcription apparatus (Chen et al., 1997). To examine a role for NFAR in the regulation of gene expression, NFAR-1 and NFAR-2 were co-transfected into Cos-7 cells with luciferase expression constructs driven by various promoters, including those controlling SV40, Fas and IFNβ expression. After 48 hr, co-transfected cells were lysed and luciferase activity determined. Measurements were done in duplicate, three times. Both Fas and IFNβ gene expression can be induced by treating cells with dsRNA. This reveals that luciferase gene expression driven by either a Fas or an IFNβ promoter was augmented in the presence of NFAR-1 , about two-fold higher, and NFAR-2, about five-fold higher, indicating that both function as positive regulators of gene expression.

To determine whether the enhancement of gene expression was governed at the transcriptional or translational level, Cos-7 cells were co-transfected with NFAR-1 or NFAR-2 expression constructs and a luciferase reporter gene driven by the IFNβ promoter. Total RNA was isolated after 48 hr, and luciferase mRNA measured using RT-PCR. Actin mRNA was similarly measured to ensure that equal amounts of RNA were analyzed. This shows significant increases in the amount of reporter gene transcripts in cells transfected with NFAR-1 or NFAR-2, but not vector alone. These results indicate that human NFAR protein functions to positively regulate transcription. NFAR-2 appeared to be the more potent inducer of gene expression. Although it has not yet been determined whether NFAR functions as a co-activator of transcription or in elongation, stabilization or transportation of cellular mRNA (i.e., post-transcriptional regulation), it appears highly purified NFAR proteins do not bind directly to IFNβ or SV40 promoter sequences as determined by electrophoretic mobility shift assays.

To examine the importance of the dsRNA-binding motifs in the function of these proteins, His-tagged NFAR-2 variants were generated and examined for their ability to stimulate the expression of IFNβ or Fas promoter-driven luciferase reporter genes. NFAR-1, NFAR-2, and NFAR variants (M5, M7, M8 and M9) were transfected into Cos- 7 cells. After 48 hr, NFAR protein expression was monitored by lysing the cells and immunoblotting an aliquot of extracts using a anti-hexahistidine antibody to ensure that all NFAR variants were expressed equally. This indicated that removal of the dsRNA- binding motifs (NFAR-M9) greatly reduced the ability of these proteins to stimulate gene expression, indicating a critical role for these domains in NFAR function. Removal of the elF2α domain of homology (NFAR2-M7) also moderately reduced the ability of NFAR to simulate the expression of Fas and IFNβ driven reporter gene constructs. The NFAR variant NFAR2-M5, lacking the first 173 residues of the amino terminus as well as NFAR2-M8 (Δ238-384) similarly showed markedly less activity than wild-type NFAR-2. This indicates that the amino terminus as well as the dsRNA- binding motifs are essential for the functioning of these proteins in the cell.

Our yeast two-hybrid analysis had revealed that the vaccinia virus encoded inhibitor E3L could also interact with NFAR. To determine the effect of E3L on NFAR function in vivo, we transfected vaccinia virus K3L or E3L in pcDNA3.1 along with NFAR-1 or NFAR-2 expression constructs in Cos-7 cells expressing an SV40 promoter- driven luciferase reporter gene. After 48 hr, cells were lysed and luciferase activity determined. Measurements were done in duplicate, three times. Our results show that K3L and E3L alone moderately increased the synthesis of luciferase protein. But E3L, and to a much lesser extent K3L, abrogated the ability of both NFAR-1 and NFAR-2 to stimulate expression of an SV40 promoter-driven luciferase reporter gene. Similar effects were observed in co-transfection studies using the IFNβ promoter driving the luciferese gene. Conceivably, E3L may prevent NFAR activity by binding directly to PKR, as has been previously shown, or by directly binding to NFAR, as our yeast data indicate.

To examine this possibility further, in vitro translated [³⁵S]-labeled E3L was incubated with purified His-tagged recombinant NFAR-1 or NFAR-2. The mixtures were incubated with rabbit antiserum raised against recombinant NFAR or pre-immune serum from the same animal, and then precipitated with protein G agarose. After washing, bound proteins were separated by 12.5% PAGE-SDS and autoradiographed. Co-immunoprecipitation studies revealed that NFAR-1 and NFAR-2 were precipitated with E3L only when the NFAR antiserum was added.

Further evidence indicating that E3L could potentially interact with human NFAR in vivo was demonstrated by using confocal microscopy to show that human NFAR and E3L could co-localize in the nucleus of the cell. Cos-7 cells transfected with FLAG- tagged E3L were incubated with rabbit antiserum raised against recombinant NFAR and an mouse anti-FLAG monoclonal antibody. FITC-conjugated anti-rabbit secondary antibody was used to detect endogenous NFAR, and TEXAS RED dye-conjugated anti- mouse antibody was used to detect E3L. Merged filters revealed colocalization of NFAR and E3L in the nucleus of cells. We conclude that E3L binds to and inhibits NFAR function.

Genomic Organization of a Human NFAR Gene:

To determine the genomic organization of a human NFAR gene, we screened large-insert PAC genomic libraries (loannou et al., 1994) using a human NFAR cDNA as a probe. After identification, a single isolated colony was prepared and the insert was sequenced to determine the structure of this human NFAR gene. Additional DNA sequencing was used to determine the majority of the entire NFAR genomic region. Exon coding regions in the genomic clone were confirmed using sequence data derived from NFAR-1 and NFAR-2 cDNA clones isolated from our two-hybrid screening of Jurkat cDNA libraries, as well as from clones isolated by hybridization screening of other T-cell cDNA libraries. Additional proteins interacting with NFAR-1 and/or NFAR-2 can be identified and isolated using the disclosure of US Patent 6,057,101. Fig. 1 shows the intron-exon organization of this NFAR gene in relation to mRNA representing NFAR-2 (panel A) and NFAR-1 (panel A) proteins. NFAR spanned 13.5 kb in a genomic clone and contained 22 exons, with the initiating ATG of both NFAR-1 and NFAR-2 being present in exon 2. NFAR-1 and NFAR-2 are identical for the first 17 exons, up to amino acid 687. NFAR-1 then contains two extra exons, 18 and 19, the first of which contains several termination codons. Thus, translation of NFAR-1 stops at amino acid 702. NFAR-2 was found to lack exons 18 and 19, but to contain three additional coding exons to generate an extended product of 894 amino acids. The extra domain of 207 amino acids of NFAR-2 is rich in glycine residues. Table 6 shows the sizes of the introns and the intron-exon splice junction sequences in these regions. All of the splice donor and acceptor sequences agree with the GT/AG consensus sequence. The size of introns ranged from 0.88 kb to 1.3 kb. The intron splice phasing is type 0 (intron located between codons) for introns 1 , 2, 4, 5, 9, 12, 14 and 15; type 1 (the intron interrupts the first and second bases of the codon) for introns 8, 10, 11, 16, 17, 18 and 19; and type 2 (the intron interrupts the second and third bases of the codon) for introns 3, 6 and 7. The two dsRNA-binding motifs reside in exons 12-13 and exons 14-15, respectively. Inspection revealed that the 5'-UTR of this NFAR gene lacks a TATA box and CCAAT sequence near the start site, and further examination will be required to identify the initiation of transcription. Interestingly, our sequence differs from the sequence of the NF90 clone that was also isolated from Jurkat cells (GenBank Accession No. U 10324) and is in general agreement with the sequence of partial cDNA clones MPP4 of 1523 nucleotides and 2043 nucleotides (GenBank Accession No. U07156). We believe that differences in the information representing NF90 may have arisen through sequencing discrepancies (Kao et al., 1994). In addition, NF90 contains an extra cytosine and thymidine base at position 1780 (CT) that is absent from our genomic clones, cDNA clones, and from the MPP4 clones. These extraneous nucleotides, which may have been inadvertently introduced as errors during PCR, changes the reading frame of NF90 at amino acid residue 594 and prematurely terminates the protein at amino acid residue 671. The 1523 bp MPP4 lacks bases 1320 to 2109 of NFAR-1 and bases 1320 to 2685 of

NFAR-2; the 2043 bp MPP4 lacks bases 1834 to 2109 of NFAR-1 and bases 1834 to 2685 of NFAR-2. See Table 5. Termination should occur at amino acid residue 702 for NFAR-1 or amino acid residue 894 for NFAR-2, to generate predicted NFAR proteins of about 90 kDa and about 110 kDa, which was confirmed by PAGE-SDS.

Both NFAR-1 and NFAR-2 contain two dsRNA-binding motifs located towards the carboxyl terminus (i.e., residues 419 to 464 and residues 535 to 604) and are members of the same dsRNA-binding protein family that includes PKR. Although the mouse homolog of NFAR has not yet been identified, it is likely that the X. laevis dsRNA-binding proteins 4F.1 and 4F.2 are NFAR homologs with the proteins sharing about 75% identity at the amino acid level. Thus, this human NFAR gene and its encoded products represent a newly identified family of evolutionarily conserved dsRNA-binding proteins.

Chromosomal Localization of a Human NFAR Gene:

As additional confirmation that NFAR-1 and NFAR-2 are derived from the same gene, their chromosomal location was determined by fluorescent in situ hybridization (Stokke et al., 1995). Digoxigenin dUTP-labeled probes were hybridized to normal metaphase chromosomes derived from PHA-stimulated peripheral blood lymphocytes. Initial experiments resulted in a single specific labeling of the middle of the short arm of chromosome 19. Subsequent measurements revealed that NFAR is located at a position which is 59% of the distance from the centromere to the telomere of chromosome arm 19p, an area that corresponds to band 19p13.1-13.2. A total of 80 metaphase cells were analyzed with 72 exhibiting specific labeling.

Interestingly, chromosome 19p13 was commonly rearranged in malignant ovarian tumors and acute lymphocytic leukemias (ALL). Abnormality of chromosome band 19p13 has also been reported to recurrently occur in malignant fibrous histiocytoma (MFH), gastric cancer, and T-cell lymphomas. Since PKR is a putative tumor suppressor gene that may exert some of its effects through the phosphorylation of NFAR, it is plausible that inactivation of NFAR could conceivably neutralize PKR function, possibly contributing to tumorigenesis. Primer pairs to amplify exons or introns can provide sequence tagged sites

(STS) at chromosome 19p13.1-13.2 to map a breakpoint within or outside this human NFAR gene. Translocation might serve, for example, to disregulate NFAR expression, favor one isoform over the other, produce a fusion protein, or combinations thereof as expected for a role for NFAR as a oncogene or tumor suppressor gene.

Discussion:

Extensive studies with PKR has revealed that this kinase requires at least 35 bp of dsRNA for minimal binding and subsequent activation, and 85 bp or greater for maximum kinase activity (Manche et al., 1992). Once bound to dsRNA, PKR is believed to undergo a conformational change and to expose sites required for trans- phosphorylation and activation of target substrates (Wu and Kaufman, 1997). Evidence also suggests that two molecules of PKR may bind to the same dsRNA structure so that they may activate one another (Barber et al., 1995; Wu and Kaufman, 1996). Catalytically inactive PKR dominant-negative mutants have been shown to inhibit PKR function by dimerizing with functional kinase to form inactive heterodimers (Barber et al., 1995; Romano et al., 1995). Expression of such PKR variants in immortalized mouse 3T3 fibroblasts results in cellular transformation with concomitant transcriptional repression of a number of pro-apoptotic genes such as FADD, TRADD, and caspase-8 (Koromilas et al., 1992; Meurs et al., 1993; Barber et al., 1995; Balachandran et al., 1998). Although the mechanism of PKR variant-induced tumorigenesis is presently unknown, it may involve perturbation of the elF2 pathway. For example, PKR variants may compete with the endogenous kinase and associate with elF2α, leading to stimulation of translation in the cell (Donze et al., 1995). But it is also likely that PKR variants may contribute towards malignant transformation by perturbing alternative cellular signaling pathways, such as those involving NFAR or other proteins. The importance of the dsRNA-binding motifs in NFAR function was underscored by demonstrating that variants lacking these domains lost their ability to stimulate gene expression in the cell.

Other members of this family of dsRNA-binding proteins, including X. laevis 4F.1 , D. melanogaster Staufen, human TRBP, mouse Spnr, and human NF90, exhibit about 75%, 12%, 16%, 55% and 64% identity at the amino acid level with NFAR, respectively. A partial human cDNA clone with unknown function referred to as MPP4 exhibited the largest degree of homology with NFAR (i.e., about 69%). MPP4 was found to react with an antibody that recognizes phosphorylation sites associated with M-phase proteins (Matsumoto-Taniura et al., 1996). During progression into M phase, many proteins, including mitotic regulators and interphase structural proteins, become hyperphosphorylated by activated M-phase kinases such as the M phase promoting factor (MPF). These M-phase phosphoproteins are considered to permit the disassembly of interphase structures as well as stimulate M-phase enzymatic activity. A mouse cDNA with similarity to NFAR-1 , but unknown function, has also recently been reported (Zhong et al., 1999). Significantly, however, none of these dsRNA-binding proteins have been reported to be substrates for PKR.

Although the VIRV box is conserved, the serine 51 site for PKR phosphorylation of elF2α is not identically preserved in NFAR. Interestingly, the removal of this domain does not appear to prevent the phosphorylation of NFAR by PKR. But it is possible that PKR may have multiple targets within NFAR and further experiments are ongoing to identify those residues that are specifically targeted by the kinase. This domain of NFAR also shared homology with the K3L product, a putative elF2α pseudo-substrate that is encoded by vaccinia virus with the purpose of binding to and inhibiting PKR. Although the importance of the elF2α domain of homology in NFAR-1 and NFAR-2 remains to be clarified, removal of these amino acids reduced the ability of the NFARs to function in vivo, indicating the likely importance of this domain to NFAR function. Removal of other amino-terminal domains of the NFAR proteins also reduced their ability to activate gene expression. These data, taken together, indicate that the integrity of much of the NFAR protein is required for efficient functioning in the cell. Importantly, our data indicate that both NFAR-1 and NFAR-2 are substrates for

PKR. Phosphorylation of NFAR-1 and NFAR-2 was enhanced by low concentrations of dsRNA but inhibited at higher dsRNA concentrations, perhaps because the dsRNA- binding motifs of PKR and NFAR might each accommodate different dsRNA at high concentrations. This might prevent PKR from associating with substrates such as NFAR. Data obtained from the yeast two-hybrid system also suggest that dsRNA may facilitate the interaction of NFAR with PKR. But NFAR variants lacking the dsRNA- binding motifs retained their ability to serve as substrates for PKR, indicating that the domains are not necessary for this interaction. Taken together, PKR can become activated and will phosphorylate NFAR-1 and NFAR-2 in the presence of dsRNA of either cellular or viral origin. dsRNA may also facilitate the interaction by acting as a bridge between NFAR and PKR. It is possible that once bound to dsRNA, PKR may preferentially activate nearby targets that also contain dsRNA-binding motifs, such as NFAR-1 and NFAR-2. It appears that NFAR proteins are able to associate with one another in yeast. This indicates that they may also form complexes with one another in the cell. It is unclear, however, why there are two similar NFAR proteins in the cell. Both NFAR appear to be expressed concomitantly in nearly all cells and tissues analyzed, arguing against one spliced form being preferentially translated in certain types of tissue, in the absence of the other. It is possible that the regulation of NFAR-2 may be governed by factors not associated with NFAR-1 since the former protein has an extra 192 amino acids at the carboxyl terminus. Nevertheless, the fact that there are two NFAR proteins present in the cell is almost certainly biologically significant as highlighted by their conservation throughout evolution.

Studies with X. laevis 4F.1 and 4F.2 proteins indicated that they were able to associate with RNA:DNA hybrids, suggesting that they could be involved in enhancing transcriptional elongation (Bass et al., 1994). Other possible NFAR functions include the modulation of mRNA stability of normally rapidly turned over mRNA representing cytokines, pro-apoptotic genes, and proto-oncogenes (Wang et al., 1999; Otero et al., 1999).

Efficient cellular transcription, processing, and transport may preferentially take place in regions of the nucleus harboring concentrations of transcription factors and spliceosome assembly proteins that promote gene expression (Ascoli and Maul, 1991). Presently, a number of factors have now been identified to colocalize to such areas, referred to as ND domains (Maul et al., 1996). Some of these genes are known to be growth suppressive and, significantly, many are induced by interferon (Korioth et al., 1995; Dyck et al., 1994). It is noteworthy that these nuclear regions have been shown to be preferentially localized sites of replication by at least three DNA virus families, including cytomegalovirus (CMV) and herpes simplex type 1 (HSV-1) (Ishov et al.,

1997; Maul et al., 1996). Although it remains to be clarified whether PKR and/or NFAR are associated with such complexes, it is possible that these domains are protected from being utilized by viruses by the interferon system. The suppression of NFAR and PKR activity by E3L might assist viral replication. Thus, the activation of PKR by viral RNA may modulate the activity of the NFAR proteins and constitute an important component of host defense in the cell.

Further yeast two-hybrid screens with a carboxyl-terminal polypeptide of NFAR-2 isolated two more interacting proteins: Translocated in LipoSarcoma (TLS also known as FUS) and Survival of Motor Neurons (SMN). Thus, NFAR may regulate the activity of these two proteins. TLS is an RNA-binding protein that is also involved in chromosomal translocations in liposarcomas and myeloid leukemias. It is a component of hnRNP complexes and may also act as a transcription factor. Reduced levels of or mutations in SMN results in spinal muscular atrophy. SMN interacts with spliceosomal snRNP proteins and is critical for their assembly.

Expression profiling after transfection of NFAR-1 into 293T cells showed that there was a dramatic increase in transcription of protein kinase C theta (PKC θ), which is involved in T-cell activation. Thus, NFAR may regulate PKC activity and NF-κB- mediated signaling pathways as a transcription factor.

REFERENCES

Ascoli and Maul (1991) J. Cell Biol., 112:785-795.

Balachandran et al. (1998) EMBO J., 17:6888-6902. Barber et al. (1991) Biochemistry, 30:10356-10361.

Barber et al. (1995) J. Biol. Chem., 270:17423-17428.

Bass et al. (1994) Curr. Biol., 4:301-314.

Chen et al. (1997) Cell, 90:569-580.

Chien et al. (1991) Proc. Natl. Acad. Sci. USA, 88:9578-9582. Donze et al. (1995) EMBO J., 14:3828-3834.

Dyck et al. (1994) Cell, 76:333-343.

Fields and Song (1989) Nature, 340:245-246. loannou et al. (1994) Nature Genet, 6:84-89.

Ishov et al. (1997) J. Cell Biol., 138:5-16. Kao et al. (1994) J. Biol. Chem., 269:20691-20699.

Katze et al. (1991) Mol. Cell. Biol., 11:5497-5505.

Korioth et al. (1995) J. Cell Biol., 130:1-13.

Koromilas et al. (1992) Science, 257:1685-1689.

Laurent et al. (1985) Proc. Natl. Acad. Sci. USA, 82:4341-4345. Manche et al. (1992) Mol. Cell. Biol., 12:5238-5248.

Matsumoto-Taniura et al. (1996) Mol. Biol. Cell., 7:1455-1469.

Maul et al. (1996) Virology, 217:67-75.

Meurs et al. (1993) Proc. Natl. Acad. Sci. USA, 90:232-236. Otero et al. (1999) Mol. Cell. 3:109-118. Romano et al. (1995) Mol. Cell Biol., 15:365-378. St. Johnston et al. (1992) Proc. Natl. Acad. Sci. USA, 89:10979-10983. Sharp et al. (1998) Virology, 250:302-315. Stokke et al. (1995) Genomics, 26: 134-137. Thanos and Maniatis (1995) Cell, 83:1091-1100. Thomis and Samuel (1995) J. Virology, 69:5195-5198. Wada et al. (1995) J. Biol. Chem., 270:18007-18012. Wang et al. (1999) Mol. Cell. Biol., 19:4552-2560. Wu and Kaufman (1996) J. Biol. Chem., 271 :1756-1763. Wu and Kaufman (1997) J. Biol. Chem., 272:1291-1296. Zhong et al. (1999) Nature Genet., 22:171-174.

While our invention has been described in connection with what is presently considered to be practical and preferred embodiments, it should be understood that it is not to be limited or restricted to the disclosed embodiments but, on the contrary, is intended to cover various modifications, substitutions, and combinations within the scope of the appended claims. In this respect, one should note that the protection conferred by the claims is determined after their issuance in view of later technical developments and would extend to all legal equivalents.

Therefore, it is to be understood that variations in our invention that are not described herein will be obvious to a person skilled in the art and could be practiced without departing from the invention's novel and non-obvious elements with the proviso that the prior art is excluded.

Table 1: Human NFAR-1 nucleotide sequence (SEQ ID NO:1)

atgcgtccaa tgcgaatttt tgtgaatgat gaccgccatg tgatggcaaa gcattcttcc gtttatccaa cacaagagga gctggaggca gtccagaaca tggtgtccca cacggagcgg gcgctcaaag ctgtgtccga ctggatagac gagcaggaaa agggtagcag cgagcaggca gagtccgata acatggatgt gcccccagag gacgacagta aagaaggggc tggggaacag aagacggagc acatgaccag aaccctgcgg ggagtgatgc gggtgggcct ggtggcaaag tgcctcctac tcaaggggga cttggatctg gagctggtgc tgctgtgtaa ggagaagccc acaaccgccc tcctggacaa ggtggccgac aacctggcca tccagcttgc tgctgtaaca gaagacaagt acgaaatact gcaatctgtc gacgatgctg cgattgtgat aaaaaacaca aaagagcctc cattgtccct gaccatccac ctgacatccc ctgttgtcag agaagaaatg gagaaagtat tagctggaga aacgctatca gtcaacgacc ccccggacgt tctggacagg cagaaatgcc ttgctgcctt ggcgtccctc cgacacgcca agtggttcca ggccagagcc aacgggctga agtcttgtgt cattgtgatc cgggtcttga gggacctgtg cactcgcgtg cccacctggg gtcccctccg aggctggcct ctcgagctcc tgtgtgagaa atccattggc acggccaaca gaccgatggg tgctggcgag gccctgcgga gagtgctgga gtgcctggcg tcgggcatcg tgatgccaga tggttctggc atttatgacc cttgtgaaaa agaagccact gatgctattg ggcatctaga cagacagcaa cgggaagata tcacacagag tgcgcagcac gcactgcggc tcgctgcctt cggccagctc cataaagtcc taggcatgga ccctctgcct tccaagatgc ccaagaaacc aaagaatgaa aacccagtgg actacaccgt tcagatccca ccaagcacca cctatgccat tacgcccatg aaacgcccaa tggaggagga cggggaggag aagtcgccca gcaaaaagaa gaagaagatt cagaagaaag aggagaaggc agagcccccc caggctatga atgccctgat gcggttgaac cagctgaagc cagggctgca gtacaagctg gtgtcccaga ctgggcccgt ccatgccccc atctttacca tgtctgtgga ggttgatggc aattcattcg aggcctctgg gccctccaaa aagacggcca agctgcacgt ggccgttaag gtgttacagg acatgggctt gccgacgggt gctgaaggca gggactcgag caagggggag gactcggctg aggagaccga ggcgaagcca gcagtggtgg cccctgcccc agtggtagaa gctgtctcca cccctagtgc ggcctttccc tcagatgcca ctgccgagca ggggccgatc ctgacaaagc acggcaagaa cccagtcatg gagctgaacg agaagaggcg tgggctcaag tacgagctca tctccgagac cgggggcagc cacgacaagc gcttcgtcat ggaggtcgaa gtggatggac agaagttcca aggtgctggt tccaacaaaa aggtggcgaa ggcctacgct gctcttgctg ccctagaaaa gcttttccct gacacccctc tcgcccttga tgccaacaaa aagaagagag ccccagtacc cgtcagaggg ggaccgaaat ttgctgctaa gccacataac cctggcttcg gcatgggagg ccccatgcac aacgaagtgc ccccaccccc caaccttcga gggcggggaa gaggcgggag catccgggga cgagggcgcg ggcgaggatt tggtggcgcc aaccatggag gctacatgaa tgccggtgct gggtatggaa gctatgggta cggaggcaac tctgcgacag caggctacag tgactttttc acagactgct acggctatca tgattttggg tcttcctaga gcgtctaaaa gtattgcaca caaaatcaac tttttactcc aatttcctcc aactccaaaa cccaaagtgt ccgtgctgtg tccctgtgct tcactgggtt tctcaaccgt ggcttttcac cgcagcttgt ctgaaactct tagcctgcag aatttaagac aatggcagtt tttatcgtga tttgcctttg aacttggtcc tattgaagtt cacaataagt ggaaaacaat tttttcagag aatgtatttt tgtgcagaat tgcacagaat tctagagaca gcgttgttcg gcatcaaggc aaaagcccac ctttgctttt tatggaaagc attactttat ttaaagagac agacaatgac gcattttaat ctacctttgt cttaatttac agcaggtttt gtatgaattt ttaacctttt aacaaactcc caaatctggt tgatgccttt gacagtgatg aaaacgattt caccacatct gaatccagag aaaccggctt tttttcttat tgcgagcatg ttaaaacgtt gggaacatgt ggggaattgt atattgcgct gaattaactt ctcccgcctc ttgtaatgct ctggtgggtt cttgtttggg aatgcgatat tttgtggctg gtttagctag agagtgaact ctcaaaggta tcaaaactgt gcttccatta ttagtgcaag aaacagacag gctttaaggg gtagatgacg tgaaattttg caagtcttaa ttacagctgc agatgcatgg gattctggat ttttttgttg ctttttagtt taatgggact ttaaaagtaa ttgaggagaa agaaccgtga tgttccctgt ttctccagta aaggactggc ttttgcttgg gcagaggtgg tgctgctggg tgtgcagctg ccacagactc caaaggcgta gaagtttgtg ccaacacacg gagtcattct ggctctctgc tgaggcccct gttttctggc aggtgccctc cttggaaact ggttttggct ctgatcagcg gttctttttg cagcaaagcc tgcatctgtg ttgacttgca agattttgcg tttattcagg caaaaactgg tcaaaatggt tactacatga tttgttccca gaggtttgaa acattcagtg aaacttttta aaactttgat tgcatgatgt attttttttt tagaaagtta ttgtttgaga ataatgtctt tttataccag gaaaatagtt atcctgaatg acgttgaaaa ctccccctcc cctttatttt tttttaatca atacatgtga aagtaaaaaa aaaaaaaaaa aaaaaa

Table 2: Human NFAR-1 amino acid sequence (SEQ ID NO:2)

MRP RIFVNDDRHVMAKHSSVYPTQEELEAVQN VSHTERALKAVSD IDEQEKGSSEQAESDN MDVPPEDDSKEGAGEQKTEHMTRTLRGVMRVGLVAKC LLKGDLDLELVLLCKEKPTTALLDKV ADNLAIQLAAVTEDKYEILQSVDDAAIVIKNTKEPPLSLTIHLTSPWREEMEKVLAGETLSVN DPPDVLDRQKC AALASLRHAKWFQARANGLKSCVIVIRVLRD CTRVPTWGP RGWPLELLCE KSIGTANRPMGAGEALRRVLECLASGIVMPDGSGIYDPCEKEATDAIGH DRQQREDITQSAQH ALRLAAFGQLHKVLGMDPLPSK PKKPKNENPλDYTVQIPPSTTYAITP KRPMEEDGEEKSPS KKKKKIQKKEEKAEPPQAMNA MRLNQLKPG QYKLVSQTGPVHAPIFT SVEVDGNSFEASGP SKKTAKLHVAVKVLQD GLPTGAEGRDSSKGEDSAEETEAKPAWAPAPWEAVSTPSAAFPSD ATAEQGPILTKHGKNPVMELNEKRRGLKYELISETGGSHDKRFVMEVEVDGQKFQGAGSNKKVA -^YAALAA EK FPDTPLALDANKKKRAPVPVRGGPKFAAKPHNPGFG GGPMHNEVPPPPNLR GRGRGGSIRGRGRGRGFGGA HGGYMNAGAGYGSYGYGGNSATAGYSDFFTDCYGYHDFGSS

Table 3: Human NFAR-2 nucleotide sequence (SEQ ID NO:3)

atgcgtccaa tgcgaatttt tgtgaatgat gaccgccatg tgatggcaaa gcattcttcc gtttatccaa cacaagagga gctggaggca gtccagaaca tggtgtccca cacggagcgg gcgctcaaag ctgtgtccga ctggatagac gagcaggaaa agggtagcag cgagcaggca gagtccgata acatggatgt gcccccagag gacgacagta aagaaggggc tggggaacag aagacggagc acatgaccag aaccctgcgg ggagtgatgc gggtgggcct ggtggcaaag tgcctcctac tcaaggggga cttggatctg gagctggtgc tgctgtgtaa ggagaagccc acaaccgccc tcctggacaa ggtggccgac aacctggcca tccagcttgc tgctgtaaca gaagacaagt acgaaatact gcaatctgtc gacgatgctg cgattgtgat aaaaaacaca aaagagcctc cattgtccct gaccatccac ctgacatccc ctgttgtcag agaagaaatg gagaaagtat tagctggaga aacgctatca gtcaacgacc ccccggacgt tctggacagg cagaaatgcc ttgctgcctt ggcgtccctc cgacacgcca agtggttcca ggccagagcc aacgggctga agtcttgtgt cattgtgatc cgggtcttga gggacctgtg cactcgcgtg cccacctggg gtcccctccg aggctggcct ctcgagctcc tgtgtgagaa atccattggc acggccaaca gaccgatggg tgctggcgag gccctgcgga gagtgctgga gtgcctggcg tcgggcatcg tgatgccaga tggttctggc atttatgacc cttgtgaaaa agaagccact gatgctattg ggcatctaga cagacagcaa cgggaagata tcacacagag tgcgcagcac gcactgcggc tcgctgcctt cggccagctc cataaagtcc taggcatgga ccctctgcct tccaagatgc ccaagaaacc aaagaatgaa aacccagtgg actacaccgt tcagatccca ccaagcacca cctatgccat tacgcccatg aaacgcccaa tggaggagga cggggaggag aagtcgccca gcaaaaagaa gaagaagatt cagaagaaag aggagaaggc agagcccccc caggctatga atgccctgat gcggttgaac cagctgaagc cagggctgca gtacaagctg gtgtcccaga ctgggcccgt ccatgccccc atctttacca tgtctgtgga ggttgatggc aattcattcg aggcctctgg gccctccaaa aagacggcca agctgcacgt ggccgttaag gtgttacagg acatgggctt gccgacgggt gctgaaggca gggactcgag caagggggag gactcggctg aggagaccga ggcgaagcca gcagtggtgg cccctgcccc agtggtagaa gctgtctcca cccctagtgc ggcctttccc tcagatgcca ctgccgagca ggggccgatc ctgacaaagc acggcaagaa cccagtcatg gagctgaacg agaagaggcg tgggctcaag tacgagctca tctccgagac cgggggcagc cacgacaagc gcttcgtcat ggaggtcgaa gtggatggac agaagttcca aggtgctggt tccaacaaaa aggtggcgaa ggcctacgct gctcttgctg ccctagaaaa gcttttccct gacacccctc tcgcccttga tgccaacaaa aagaagagag ccccagtacc cgtcagaggg ggaccgaaat ttgctgctaa gccacataac cctggcttcg gcatgggagg ccccatgcac aacgaagtgc ccccaccccc caaccttcga gggcggggaa gaggcgggag catccgggga cgagggcgcg ggcgaggatt tggtggcgcc aaccatggag gctacatgaa tgccggtgct gggtatggaa gctatgggta cggaggcaac tctgcgacag caggctacag tcagttctac agcaacggag ggcattctgg gaatgccagt ggcggtggcg gcgggggcgg tggtggctcc tccggctatg gctcctacta ccaaggtgac aactacaact caccggtgcc cccaaaacac gctgggaaga agcagccgca cgggggccag cagaagccct cctacggctc gggctaccag tcccaccagg gccagcagca gtcctacaac cagagcccct acagcaacta tggccctcca cagggcaagc agaaaggcta taaccatgga caaggcagct actcctactc gaactcctac aactctcccg ggggcggggg cggatccgac tacaactacg agagcaaatt caactacagt ggtagtggag gccgaagcgg cgggaacagc tacggctcag gcggggcatc ctacaaccca gggtcacacg ggggctacgg cggaggttct gggggcggct cctcatacca aggcaaacaa ggaggctact cacagtcgaa ctacaactcc ccggggtccg gccagaacta cagtggccct cccagctcct accagtcctc acaaggcggc tatggcagaa acgcagacca cagcatgaac taccagtaca gataagcccc cgcggggcgg agatttctac cttctgcact tactccccat cagaagatcg agttttatgc atcacagtta acatgtcagc tggccctcca ggcccccgcc cccatcccgt ccacgttgct gtgtcgtgag gtgcagcggg tcaccctgtg gcccgtcctg tgacccatat ttagccgtgt ttgggactcc gtgtcttcaa tggtttgtta gttgccatta caactttgtc tgggtagagt ttttgagttt ttgcagttca gtatccctct gtctattcac acttcgtgtt agtggtaact cagtttgtct ttaaatagtt acagaaggga tacgtcattt gttaatgctt ttgtgaagtg agttaaacga gctttctgta ttttaatgct ttagtgtttc agtttttata agtgaagatt ttattttaaa aaccagtggg aaagagtggg gggtttcttt ttatgtctgg gtcattcagg cagtacatct gaattaagct gaatgtagac aaataaagaa aaacaaaact aaaaaaaaaa aaaaaaaaaa aa

Table 4: Human NFAR-2 amino acid sequence (SEQ ID NO:4)

MRPMRIFλ DDRHVMAKHSSλTΪ-PTQEELEAVQNMVSHTERALKAVSD IDEQEKGSSEQAESDN MDVPPEDDSKEGAGEQKTEH TRTLRGλΛyiRVGLVAKCLL KGDLDLELVLLCKEKPTTALLDKV ADNLAIQLAAVTEDKYEILQSVDDAAIVIKNTKEPPLS TIHLTSPWREEMEKVLAGETLSVN DPPDV DRQKCLAALASLRHAK FQARANGLKSCVIVIRV RD CTRVPT GPLRG P ELLCE KSIGTANRPMGAGEALRRVLECLASGIVMPDGSGIYDPCEKEATDAIGHLDRQQREDITQSAQH -ALRLAAFGQLHKVLGMDP PSKMPKKPKNENPVDYTVQIPPSTTYAITPMKRPMEEDGEEKSPS KKKKKIQKKEEKAEPPQA NALMRLNQLKPGLQYKLVSQTGPVHAPIFTMSVEVDGNSFEASGP SKKTAKLHVAVKVLQDMGLPTGAEGRDSSKGEDSAEETEAKPAWAPAPWEAVSTPSAAFPSD ATAEQGPILTKHGKNPVME NEKRRG KYELISETGGSHDKRFVMEVEVDGQKFQGAGSNKKVA KAYAALAALEKLFPDTPLALDANKKKRAPVPVRGGPKFAAKPHNPGFGMGGPMHNEVPPPPNLR GRGRGGSIRGRGRGRGFGGA HGGYMNAGAGYGSYGYGGNSATAGYSQFYSNGGHSGNASGGGG GGGGGSSGYGSYYQGDNYNSPVPPKHAGKKQPHGGQQKPSYGSGYQSHQGQQQSYNQSPYSNYG PPQGKQKGYNHGQGSYSYSNSYNSPGGGGGSDYNYESKFNYSGSGGRSGGNSYGSGGASYNPGS HGGYGGGSGGGSSYQGKQGGYSQSNYNSPGSGQNYSGPPSSYQSSQGGYGRNADHSMNYQYR

NFAR-1 R P M R I F V N D D R H V M A K 17

NFAR-2 R P M R I F V N D D R H V M A K 17

MPP4 2043bp R P M R I F V N D D R H V M A K 17

MPP4 1523bp R P M R I F V N D D R H V M A K 17

NF90 R P M R I F V N D D R H V M A K 17 mI F3 A L Y H H H F I T R R R R R P M R I F V N D D R H V M A K 30

X14F.1 R P M R I F N D D R H V M A K 17

X14F.2 0

NFAR-1 H S S V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 47

NFAR-2 H S S V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 47

MPP4 2043bp H S S V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 47 C

^■

MPP4 1523bp H S S V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 47

NF90 H S S V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 47 en mlLF3 H S S V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 60

X14F.1 H S V V Y P T Q E E L E A V Q N M V S H T E R A L K A V S D 47

X14F.2 0

NFAR-1 I D E Q E K G S S E Q A E S D N M D V P P E D D S K E G A 77

NFAR-2 I D E Q E K G S S E Q A E S D N M D V P P E D D S K E G A 77

MPP4 2043bp W I D E Q E K G S S E Q A E S D N M D V P P E D D S K E G A 77

MPP4 1523bp W I D E Q E K G S S E Q A E S D N M D V P P E D D S K E G A 77

NF90 I ^"HI E Q E K G S S E Q A E S D N M D V P P E D D S K E G A 77 mI F3 I D E Q E K G N S E L L R Q K I T H P Q T M R A K E G A 90

X14F.1 W I D Q Q E K D C S G E Q E Q P M A E E T E T T E E G K D S 77

X14F.2 0

NFAR-1 G E Q K T E H M T R T L R G V M R V G L V A K C L L L K G 106

NFAR-2 G E Q K T E H M T R T L R G V M R V G L V A K C L L L K G 106

MPP4 2043bp G E Q K T E H M T R T L R G V M R V G L V A K G L L L K G 106

MPP4 1523bp G E Q K T E H M T R T L R G V M R V G L V A K G L L L K G 106

NF90 G E Q K T E H M T R T C R G V M R A G P G G Q S A S Y R G 107 mILF3 G E Q K A E H M T R T L R G V M R V G L V A K G L L L K G 119

X14F.1 E M K T G E N P T R T L R G V M R V G L V A K G L L L K G 106

X14F.2 A K G L L L K G

NFAR-1

NFAR-2

MPP4 2043bp

MPP4 1523bp

NF90 n mILF3 X14F.1

X14F.2

NFAR-1 A A V T E D K Y E I L Q S V D D A A I V I K N T K E P P L S 166

NFAR-2 A A V T E D K Y E I L Q S V D D A A I V I K N T K E P P L S 166

MPP4 2043bp A A V T E D K Y E I L Q S V D D A A I V I K N T K E P P L S 166

MPP4 1523bp A A V T E D K Y E I L Q S V D D A A I V I K N T K E P P L S 166

NF90 A A V T E D K Y E I L Q S V D D A A I V I K N T K E P P L S 167 ILF3 T T V T E D K Y E I L Q S V D D A A I V I K N T K E P P L S 179

X14F.1 E T V S E D K Y E V I Q N I R E A L I V V K S T K E P P L T 166

X14F.2 K T V S E D K Y E V V Q N I R E A S I V I K N T K E P P L T 68

NFAR-1 L T I H L T s P V V R E E M E K V L A G E T L S V N D P P D 196 NFAR-2 L T I H L T s P V V R E E M E K V L A G E T L S V N D P P D 196 MPP4 2043bp L T I H L T s P V V R E E M E K V L A G E T L S V N D P P D 196 MPP4 1523bp L T I H L T s P V V R E E M E K V L A G E T L S V N D P P D 196 NF90 L T I H L T s P V V R E E M E K V L A G E T L S V N D P P D 197 mILF3 L T I H L T s P V V R E E M E K V L A G E T L S V N D P P D 209 X14F.1 L N I R L T s P L V R E E M E K L S A G E T L T V S D P P D 196 X14F.2 L H I R L T s P L V R E E V E K L s A G E T L T V S D P P D 98

NFAR-1 V I V I R V L R D L C T R V P T W G P L R G P L E L L C E 256 NFAR-2 V I V I R V L R D L C T R V P T W G P L R G W P L E L L C E 256 MPP4 2043bp V I V I R V L R D L c T R V P T W G P L R G P L E L L C E 256 MPP4 1523bp V I V I R V L R D L c T R V P T W G P L R G W P L E L L C E 256 NF90 V I V I R V L R D L c T R V P T W G P L R G P L E L L C E 257 ILF3 V I V I R V L R D L c T R V P T G P L R G P L E L L C E 269 X14F.1 V I V I R V L R D L c T R V P T W E P L R G P L E L L C E 256 X14F.2 V I V I R V L R D L c T R V P T W E P L R G W P L E L L C E 158

NFAR-1 K s I G T A N R P M G A G E A L R R V L E C L A S G I V M P 286 NFAR-2 K s I G T A N R P M G A G E A L R R V L E C L A S G I V M P 286 MPP4 2043bp K s I V T N R P M G A G E A L R R V L E c L A S G I V M P 286 MPP4 1523bp K s I G T A R P M G A G E A L R R V L E c L A S G I V M P 286 NF90 K s I G T A N R P M G A G E A L R R V L E c L A S G I V M P 287 mILF3 K s I G T A N R P M G A G E A L R R V L E c L A S G I V M P 299 X14F.1 K A I G T A N R P M G A G E A L R R V L E c L s S G I L M P 286 X14F.2 K A I G T A N R P M G A G E A L R R V L E c L S S G I L M P 188

NFAR-1 P K N E N P V - D Y T V Q I P P s T T Y A I T P M K R P M E 375 NFAR-2 P K N E N P V - D Y T V Q I P P s T T Y A I T P M K R P M E 375 MPP4 2043bp P K N E N P V - D Y T V Q I P P s T T Y A I T P M K R P M E 375 MPP4 1523bp P K N E N P V - D Y T V Q I P P s T T Y A I T P M K R P M E 375 NF90 P K N E N P V - D Y T V Q I P P s T T Y A I T P M K R P M E 376 mILF3 P K N E N P V - D Y T V Q I P P s T T Y A I _ T_. P M K R P M E 388 X14F.1 T K V E T P V I D Y T V Q I P P s T T Y A M P JP_ L K R P I E 376 X14F.2 T K I E I P I I D Y T V Q I P P s T T Y A M P A L K R P I E 278

NFAR-1 E D G E E K S P S K K K K K I Q K K E E K A E P P Q A M N A 405 NFAR-2 E D G E E K S P S K K K K K I Q K K E E K A E P P Q A M N A 405 MPP4 2043bp E D G E E K S P S K K K K K I Q K K E E K A E P P Q A M N A 405 MPP4 1523bp E D G E E K S P S K K K K K I Q K K E E K A E P P Q A M N A 405 NF90 E D G E E K S P S K K K K K I Q K K E E K A E P P Q A M N A 406 en en mILF3 E D G E E K S P S K K K K K I Q K K E E K A D P P Q A M N A 418 X14F.1 E D G D D K S P S K K K K K I Q K K D E K S E P P Q V M N A 406 X14F.2 E D G E D K S P S K K K K K I Q K K D E K S E P P Q A M N A 308

NFAR-1 L M R L N Q L K P G L Q K L V S Q G P V H A P I F T M S 435 NFAR-2 L M R L N Q L K P G L Q K L V S Q G P V H A P I F T M S 435 MPP4 2043bp L M R L N Q L K P G L Q Y K L V S Q T G P V H A P I F T M S 435 MPP4 1523bp L M R L N Q L K P G L Q Y K L V S Q G P V H A P I F T M S 435 NF90 L M R L N Q L K P G L Q K L V S Q G P V H A P I F T M S 436 mILF3 L M R L N Q L K P G L Q Y K L I S Q G P V H A P I F T M S 448 X14F.1 L M R L N Q L K P G L Q Y K L I S Q T G P V H A P V F T M S 436 X14F.2 L M R L N Q L K P G L Q Y K L I S Q T G P V H A P I F T M S 338

NFAR-1 V E V D G N S F E A S G P S K K T A K L H V A V K V L Q D M 465

NFAR-2 V E V D G N S F E A S G P S K K T A K L H V A V K V L Q D M 465

MPP4 2043bp V E V D G N S F E A S G P S K K T A K L H V A V K V L Q D M 465

MPP4 1523bp V E V D G 440

NF90 V E V D G N S F E A S G P S K K T A K L H V A V K V L Q D M 466 mILF3 V E V D G S N F E A S G P S K K T A K L H V A V K V L Q D M 478

X14F.1 V E V D D K T F E A S G P S K K T A K L H V A V K V L Q D M 466

X14F.2 V E V D D K T F E A S G P S K K T A K L H V A V K V L Q D M 368

NFAR-1 G L P T G A E G R D S S K G E D S A E E T E A K P A V V A P 495 NFAR-2 G L P T G A E G R D S S K G E D S A E E T E A K P A V V A P 495 MPP4 2043bp G L P T G A E G R D S S K G E D S A E E T E A K P A V V A P 495 MPP4 1523bp 440 en σ. NF90 G L P T G A E G R D S S K G E D S A E E T E A K P A V V A P 496 mILF3 G L P T G A E G R D S S K G E D S A E E S D G K P A I V A P 508 X14F.1 G L P T G M E E K - - - - - E E G T D E S E Q K P V V Q T P 491 X14F.2 G L P T G I D E K - - - - - E E[S]V D E S E E K P V L Q T P 393

NFAR-1 A P V V E A V S T P S A A F P S D A T A E Q G P I L 521 NFAR-2 A P V V E A V S T P S A A F P S D A T A E Q G P I L 521 MPP4 2043bp A P V V E A V S T P S A A F P S D A T A E Q G P I L 521

NFAR-1 T K H G K N P V M E L N E K R R G L K Y E L I s E G G S H 551 NFAR-2 T K H G K N P V M E L N E K R R G L K Y E L I s E T G G S H 551 MPP4 2043bp T K H G K N P V M E L N E K R R G L K Y E L I s E T G G S H 551 MPP4 1523bp 440 NF90 T K H G K N P V M E L N E K R R G L K Y E L I S E T G G s H 556 mILF3 T K H G K N P V M E L N E K R R G L K Y E L I s E T G G S H 564 X14F.1 T K H G K N P V M E L N E K R R G L K Y E M I s E T G G S H 548 X14F.2 T R H G K N P V M E L N E K R R G L K Y E L I s E T G G S H 450

NFAR-1 D K R F V M E V E V D G Q K F Q G A G S N K K V A K A Y A A 581 NFAR-2 D K R F V M E V E V D G Q K F Q G A G S N K K V A K A Y A A 581 MPP4 2043bp D K R F V M E V E V D G Q K F Q G A G S N K K V A K A Y A A 581 MPP4 1523bp 440 NF90 586 mILF3 594 en

^■ l X14F.1 578 X14F.2

480

NFAR-1 L A A L E K L F P D T P L A L D A N K K K R A P V P V R G G 611

NFAR-2 L A A L E K L F P D T P L A L D A N K K K R A P V P V R G G 611

MPP4 2 043bp L A A L E K L F P D T P L A L D A N K K K R A P V P V R G G 611

MPP4 1523bp 440

NF90 T, 7\ T- T. TT V T. T. 'D Γ. τ> r> T. q 600 mILF3 L A A L E K L F P D T P L A L 624

X14F.1 L s A L E K L F P D Y T M Y T 608

X14F.2 L s A L E K L F P D Y T T Y T

510

NFAR-1 P K F A A K P H N P G F G M G G P M H N E V P P P P N L R G 641 NFAR-2 P K F A A K P H N P G F G M G G P M H N E V P P P P N L R G 641

MPP4 2043bp 611 MPP4 1523bp 440 NF90 P L M P _τ]κ R R E P Q Y 612 ILF3 P K F A A K P H N P G F G M G G P M H N E V P P P P N I R G 654

X14F.1 P K F A G K - H N Q G F G M - - - M Y S E V P P P Q A M R G 634

X14F.2 P K F A G K H N Q G F G M M Y N E V P P P Q V M R G 536

NFAR-1 R G R G G S I R G R G R G R G - F G G A N H G G - Y M N A G 669 NFAR-2 R G R G G S I R G R G R G R G F G G A N H G G Y M N A G 669

MPP4 2043bp 611 MPP4 1523bp 440 NF90 612 en oo ILF3 R G R G G N I R G R G R G R G F G G A N H G G G Y M N A G 683

X14F.1 R G R G G M N R G R G R G R G G F G G G Y M N S G 659

X14F.2 R G R G G M N R G R G R G R G G F G G G N Y G G - Y M N S G 565

NFAR-1 A G Y G S Y G Y G G N S A T A G Y S D F F T D - - 692

NFAR-2 A G Y G S Y G Y G G N S A T A G Y S 9 F Y S N G G 694

MPP4 2043bp 611 MPP4 1523bp 440

Table 5 (continued): Alignment of members of dsRNA-binding protein family

r~- o »_D

PH

M r-l r-1

Table 5 (continued): Alignment of members of dsRNA-binding protein family

CN rH rH o 01 i

[ [ >-0

rH CM rH CM

1 1 m

Di Pi O PH tt PH σ, P ^P M¹

P_H P_H PH H rH rH

2 2

2 e X X Table 5 (continued): Alignment of members of dsRNA-binding protein family

-H. -ςfi

H rH ΓH

Table 6: Exon-intron junctions (SEQ ID NOS:11-51) of NFAR-1 and NFAR-2

Exon Intron Exon

o. Size (base) Junction No. Size (kb) Junction No.

174 GACAAG gtgctagctg 1 1 .012 gatcttatag AGTTGA 2

93 AAAATG gtaagtttgt 2 0, .365 acatccctag CGTCCA 3

230 GGCTGG gtaagtgagg 3 0. .154 tgtccctcag GGAACA 4

181 CTTGCT gtaagtgtgg 4 0, .886 ttatttttag GCTGTA 5

147 CTGGAG gtagggaggc 5 1, .306 ccatgaatag AAACGC 6

92 TTCCAG gttccaggtt 6 0 .397 tttcccctag GCCAGA 7

96 GGCTGG gtaaggcatg 7 0. .624 tccctttcag CCTCTC 8

113 TGCCAG gttggggcct 8 0, .418 ctgccttcag ATGGTT 9

98 GCGCAG gtatagtcat 9 0 .572 ccttctccag CACGCA 10

112 ACACCG gtaagcctgc 10 0, .088 ctccttgcag TTCAGA 11

111 AGAAAG gtactggccg 11 0, .666 ctctctgtag AGGAGA 12

200 GTTAAG gtgagtgtgg 12 0. .349 cctcttgcag GTGTTA 13

168 GCCGAG gtctccaccc 13 0, .363 ttcccttgag AACGTA 14

138 ATGGAG gtgcccttct 14 0, .101 cactttccag GTCGAA 15

177 GCTAAG gtgagcagtg 15 0. .093 tctcttctag CCACAT 16

154 ATGCCG gtaagggccc 16 0, .131 cccttttcag GTGCTG 17

56 TACAGT gtaagtgtgc 17 0, ,444 ccttttgtag GACTTT 18

1346 GTAACA gtcttcattc 18 1. .374 ccctgtttag CAGTTC 19

361 AATTCA gtgagttggc 19 0. .842 tcccccagag ACTACA 20

129 AACAAG gtgggcctgg 20 0. .497 tctctcctag GAGGCT

670 GAGGCT (polyA4- signal for NFAR-1)

Claims

WE CLAIM:

1. An isolated human Nuclear Factor Associated with dsRNA (NFAR) polypeptide and having a relative molecular weight of about 90 kilodaltons or about 110 kilodaltons.

2. The human NFAR polypeptide of claim 1 having an amino acid sequence comprised of SEQ ID NO:2 or SEQ ID NO:4.

3. The human NFAR polypeptide of claim 2 having an amino acid sequence comprised of residues 612 to 702 of SEQ ID NO:2.

4. The human NFAR polypeptide of claim 2 having an amino acid sequence comprised of residues 612 to 894 of SEQ ID NO:4.

5. An isolated human NFAR polypeptide having an amino acid sequence comprised of at least 25 contiguous amino acids from the human NFAR polypeptide of claim 1 excluding amino acid sequences encoded by cDNA clones NF90 and MPP4.

6. A fusion polypeptide having an amino acid sequence comprised of (a) at least 25 contiguous amino acids from the human NFAR polypeptide of claims 1-5 and (b) at least one heterologous amino acid sequence.

7. An isolated complex comprised of the human NFAR polypeptide of claims 1-5 and dsRNA-dependent protein kinase (PKR).

8. An artificial substrate on which the polypeptide of claims 1-6 is one of at least 1000 different polypeptides attached to the substrate.

9. An isolated polynucleotide encoding the human NFAR polypeptide of claim 1 in its entirety, or the complement thereof.

10. The polynucleotide of claim 9 encoding NFAR-1 and having a nucleotide sequence comprised of SEQ ID NO:1 , or the complement thereof.

11. The polynucleotide of claim 9 encoding NFAR-2 and having a nucleotide sequence comprised of SEQ ID NO:3, or the complement thereof.

12. An isolated polynucleotide encoding the polypeptide of claim 5, or the complement thereof.

13. A chimeric polynucleotide having a nucleotide sequence comprised of (a) at least 30 contiguous nucleotides from the polypeptide of claims 1-5 and (b) at least one heterologous nucleotide sequence, or the complement thereof.

14. An expression construct comprised of (a) at least the polynucleotide of claims 9- 13 or the complement thereof and operably linked to (b) at least one transcriptional regulatory region.

15. The expression contruct of claim 14, wherein it is double stranded RNA or DNA.

16. An artificial substrate on which the polynucleotide of claims 9-13 is one of at least 1000 different polynucleotides attached to the substrate.

17. A transfected cell or transgenic non-human organism containing the chimeric polynucleotide of claim 13 or the expression construct of claim 14, wherein the chimeric polynucleotide is comprised of a heterologous nucleotide sequence or the expression construct is comprised of a transcriptional regulatory region that does not originate from the cell or non-human organism.

18. A binding molecule specific for the polypeptide of claims 1-4.

19. A process of using the expression construct of claims 14-15 to produce a human NFAR protein.

20. A process of using the expression construct of claims 14-15 to transfect a cell with the expression construct being integrated into the cell's genome or to produce a non-human transgenic organism with the expression construct being integrated into the transgenic organism's genome.

21. A process of identifying or isolating or detecting the human NFAR polypeptide of claims 1-4.

22. A process of detecting or potentiating or inhibiting expression of the human NFAR polypeptide of claims 1-4.

23. A process of identifying or isolating or detecting a human NFAR gene encoding the human NFAR polypeptide of claims 1-4.

24. A process of detecting or potentiating or inhibiting expression of a human NFAR gene encoding the human NFAR polypeptide of claims 1-4.

25. A process of identifying or isolating or detecting a binding molecule specific for the human NFAR polypeptide of claims 1-4.

26. A process of identifying or isolating or detecting an agent which specifically binds to the human NFAR polypeptide of claims 1-4.

27. A process of using the binding molecule of claim 18 to identify or to isolate or to detect a human NFAR protein.

28. A process of using the binding molecule of claim 18 to inhibit expression of a human NFAR protein.

29. A process of using the substrate of claim 8 to identify or to isolate or to detect a human NFAR-specific binding molecule.

30. A process of using the substrate of claim 16 to identify or to isolate or to detect a human NFAR polypeptide.

31. A chemical agent that potentiates or inhibits at least one activity of the human NFAR polypeptide of claims 1-4.