CA1337281C - Cloning of cdnas for interferon-induced genes and their use for monitoring the response to interferon - Google Patents

Abstract

Human DNA encoding enzymes having (2'-5') oligo A syn-thetase has been sequenced. The amino acid sequences of the enzymes have been deduced. Antigenic peptides have been prepared which may be used to raise antibod-ies which recognize and immunoprecipitate (2'-5') oligo A synthetase. Methods of monitoring interferon activ-ity in a subject are presented.

Description

Background of the Invention Throughout this application, various publications are referenced by the name of the author and date of publication within parentheses. Full citations for these references may be found at the end of the specification listed in alphabetical order immediately preceding the claims.

Many of the biological effects of interferon (IFN) appear to be mediated by the induction of new mRNAs and proteins in cells exposed to IFNs (for review: Revel, 1984; Lebleu and Content, 1982; Baglioni and Nilsen, 1983).
Among these IFN-induced proteins, two groups appear particularly important: 1) translation regulatory enzymes (ds RNA dependent protein kinase and (2'-5') oligo A snythetase, (2'-5') oligo A-activated nu-clease, 2-phosphodiesterase); and 2) cell surface antigens (HLA-A, B, C, B2-microglobulin, HLA-DR).
Other cellular and excreted proteins probably play important roles as well (Weil et al., 1983; Chebath et -al., 1983; Wallach et al., 1983). With the exception of the HLA genes (Malissen et al., 1982; Schamboeck et al., 1983), the structure and sometimes the function of the IFN-induced proteins is unknown and so is the mechanism by which IFNs activate specifically these genes. To address these questions, several cDNAs from IFN-induced genes have been recently cloned (Chebath et al., 1983; Merlin et al., 1983; Friedman et al., 1984;
Samanta et al., 1984). We have, in particular, studied the cDNA and gene coding for the human (2'-5') oligo A
synthetase, a ds RNA-activated enzyme that converts ATP
into ppp(A2'pA)n oligomers (Rerr and Brown, 1978) which in turn bind to and activate the latent RNAse F
(Schmidt et al., 1978). The (2'-5') oligo A synthetase is strongly induced in cells by all three types of human IFNs, and its increase is a good marker of IFN
activity (Wallach et al., 1982). The enzyme is induced during differentiation of hematopoietic cells, and denotes an autocrine secretion of IFN-beta (Yarden et al., 1984). The enzyme is similarly induced late in the S phase of synchronized embryo fibroblasts (Wells and Mallucci, 1985). The enzyme activity drops when cell growth starts (Etienne-Smekens et al., 1983;
Creasey et al., 1983) and appears to be involved in the antigrowth effect of IFN (Rimchi et al., 1981). Defi-ciency in the (2'-5') oligo A synthetase or in the (2'-5') oligo A-activated RNAse F has also been correlated with partial loss of the antiviral effects of IFNs (Salzberg et al., 1983; Epstein et al., 1981), although this is probably not the only mechanism by which IFN
inhibits virus growth (Lebleu and Content, 1982). The (2'-5') oligo A nucleotides have been detected in many eucaryotic cells and even in bacteria (Laurence et al., 1984) and the synthetase is likely to be a wide-spread enzyme. The enzyme has been purified from mouse . ! ' ' ~ .

_ 3 _ 1 33728 1 (Dougherty et al., 1980) and human cells (Yand et al., 1981); Revel et al., 1981); a larqe and a small form of the enzyme have been observed (Revel et al., 1982; St.
Laurent et al., 1983) but their structures were not elucidated.

The (2'-5') oligo A synthetase, induced in cells ex-posed to IFNs (~ovanessian et al., 1977; Zilberstein et al., 1978) has a number of unusual properties. Its main activity is the synthesis from ATP of 5'triphos-phorylated short oligo A chains (of up to 15 A, withmainly dimers to pentamers), but in contrast to other RNA polymerases, it adds adenylate or one other nucle-otide specifically to the 2'OH of adenylate in oligo A
(Rerr and Brown, 1978; Samanta et al., 1980), or to other (oligo) nucleotides with a free 2'0~ adenylate such as NAD (Ball, 1980) or even tRNA (Ferbus et al., 1981). ~o be active, the enzyme has to bind to double-stranded RNA stretches of minimum 50 bp (Minks et al., 1979), and must therefore possess several binding sites: for nucleotide triphosphates, for 2'OH adenosine polynucleotides and for double stranded RNA. The en-zyme binds to 2', 5' ADP-Sepharose (Johnston et al., 1980), to poly (rI)(rC)-Agarose (Hovanessian et al., 1977) and to Cibacron Blue-Sepharose (Revel et al., 1981). In different cells, the (2'-5') oligo A synthe-tase activity is in the cytosol (Revel et al., 1981) or in ribosomal salt washes (Dougherty et al., 1980), as well as in the nuclear sap (Nilsen et al., 1982b) and even in large amounts in the nuclear matrix. It is notable that cellular RNAs can replace poly (rI)(rC) for activation of the enzyme (Revel et al., 1980) and the synthetase may even have a role in Hn RNA process-ing (Nilsen et al., 1982a). Some (2'-5') oligo A syn-thetase is bound to plasma membranes and can be incor-,~i ..~

porated in budding virions (Wallach and Revel, 1980).
These complex interactions may ensure a localized action of the (2'-5') oligo A system (Nilsen and Baglioni, 1983) and explain its multiple suggested roles in normal and virus-infected cells. The synthetase amounts to less than 0.1~ of the proteins in IFN-treated cells, and its structure could not be determined directly.

Brief Description of the Figures Figure 1 depicts the structure and sequence of (2'-5') oligo A synthetase E1 cDNA clone 174-3:

Figure lA depicts the restriction map of El cDNA
clone 174-3. The insert base pairs are numbered in the same direction as pBR322 DNA. The pBR Eco Rl site is on the right. Both strands of the insert (dotted lines) were sequenced (Maxam &
Gilbert, 1980) from the restriction sites indicated by the vertical lines. The coding strand is 5' to 3' from right to left. Following the right Pstl site there were 17G and 72T, followed by the dinucleotide GA and the 3T of the sequence shown in (B) which are therefore not part of the tails. At the 3' end, tails of 45A
and lOC preceded the left Pstl site.

Figure lB depicts the nucleotide sequence having the longest coding frame. The first T is nucleotide 92 following the tails of the insert (right end in A). The Sau 3A1 site and the Eco Rl of the insert are at positions 129 and 480 respectively of the sequence shown.

Figure 2 depicts the size and induction of El specific mRNAs in SV80 and Namalva cells:

Figure 2A depicts the hybridization of nick-translated [32P]-cDNA of clone El to electrophoretic blots of denatured poly A+-RNA
from SV80 cells. The RNAs were prepared at the indicated hour after IFN-beta-l addition. The apparent size of the RNA is indicated on the autoradiography. Left lane, rRNA markers.

Figure 2B is the same as 2A with RNA from Namalva cells treated with IFN-alpha for the indicated time. Left lane: rRNA markers.

Figure 3 depicts the characterization by hybridization to RNA blots of recombinant plasmid clone C56, harbouring cDNA for an IFN-induced mRNA Poly(A)+- RNA
from IFN-treated SV80 cells (I) or from non-treated cells (C), 7 micrograms were electrophoresed on agarose gels and after blotting to nitrocellulose were hybridized to nick-translated [32P]-plasmid DNA of either the C56 clone, a human HLA cDNA clone or a rat tubulin cDNA clone. Exposure was for 48 h. Position of radioactive 18S ribosomal RNA marker is indicated.

Figure 4 depicts the partial restriction map and nucleotide sequence of the C56 450 bp insert. The C56 plasmid was digested with Hind 3, end-labeled with alpha [32p] -dCTP by the DNA polymerase I-large fragment (Klenow enzyme, Boehringer) and the Hind 3 - Pst 1 fragments were separated on a 1~ agarose gel. In order to sequence the complementary strand, the plasmid was 5'-labeled at the Bql 2 site with gamma [32P]-ATP by the T4-Polynucleotide kinase (Biolabs) and the Bgl 2 Pstl fragments were isolated. Sequencing was made by the Maxam and Gilbert technique. Sequence of coding strand A (right to left) is shown in the lower panel. The two first thymidylic residues of the sequence of strand A probably correspond to the AT
tail as indicated in the upper diagram.

Figure 5 depicts the time course of the induction of C56 mRNA by IFN:

Figure 5A depicts Poly (A)+ RNA,7 micrograms from Namalva cells treated with IFN-alpha 1000 U/ml for the indicated times were electrophoresed on agarose gels and, after blotting, were hybridized with nick-translated [32P]-C56 plasmid DNA.

Figure 5B depicts Poly (A)+RNA,7 micrograms from SV80 cells treated with 200 U/ml IFN-beta for the indicated times. The asterisk indicates an RNA
sample from cells treated with IFN-beta-1 purified on monoclonal antibody column (2x103 U/mg).

Figure 5C depicts Poly (A)+RNA, 1 microgram, from SV80 cells treated as in (5B) was hybridized in liquid with 3' end-labeled fragment I of C56 DNA
(see Fig. 4). The hybrids were treated with S1--nuclease and analyzed on denaturing gels. The mRNA-hybridized probe ( ~ ) is shorter than the self-reassociated probe (~

Figure 6 depicts the restriction map of cDNAs for the 1.6 and 1.8 kb (2'-5') oligo A synthetase mRNAs.

. . ~_ Figure 6A depicts the map of the 1.6 kb cDNA.
The position of the E1 cDNA (Merlin et al., 1983) and of the lambda gt 10 cDNAs is shown. pA is the polyadenylation site. The exon limits are shown by vertical dotted lines. The size of the genomic DNA fragments carrying each exon are given in parentheses. The vertical arrow shows the position of the additional splice site in the 1.8 kb RNA. The strategy for sequencing the 9-21 and 5-21 cDNAs is indicated. The sequence from the 3' EcoR1 site (E) to the Pstl site (P) was determined in the E1 cDNA (Merlin et al., 1983).

Figure 6B depicts a map of the 1.8 kb cDNA. The lambda gtlO clone 48-1 was isolated using the Pstl-Pstl genomic fragment containing exon 8 of the 1.8 kb RNA (Fig. 9). Exons are numbered as for the 1.6 kb E cDNA. The truncated exon 7 is designated 7a.

Figures 7A and B depict the nucleotide sequences of the two (2'-5') oligo A synthetase cDNAs. The nucleotides of the 1.8 kb cDNA clone 48-1 are numbered as for the 1.6 kb cDNA clone 9-21. Amino acid numbering is given in parentheses. Translation starts at the first or second codon of the ATGATG sequence.
Limits between exons are shown by vertical bars.
(Glycos.) indicates a possible glycosylation site in E18. Single base variations, possibly allelic differences, were detected between clones or genomic DNA in the 1.6 kb sequence at 376 (T for C), 525 (G
for A), 807 (G for C), 811 (A for G); in the 1.8 kb sequence at 1087 (G for A), 1115 (G for C).

_ 8 Figure 8 depicts the hydropathy plot of the C-termini of the E16 and E18 (2'-5') oligo A synthetases. The computer program of Kyte and Doolittle (1982) was used. Hydrophobic regions are over the midline. The acidic region in E18 corresponds to amino acids 353 to 358 in Figure 7.

Figure 9 depicts the restriction map of the human (2'-5') oligo A synthetase gene. A map constructed from three overlapping genomic clones is shown with the position of the 7 exons of the 1.6 kb RNA and the additional 8th exon of the 1.8 kb RNA (black bars).

Figure 10 depicts the promoter region of the human (2'-5') oligo A synthetase gene. A restriction map of the Sphl-Sphl 0.85 kb fragment from the 4.2 kb EcoR1 genomic DNA segment in Figure 9 is shown. The 5' end of the mRNAs is marked as cap.

Figure 11 depicts the sequence of the human (2'-5') oligo A synthetase promoter region. The sequence of the Sau3a-Hpal segment of Figure 10 shown, aligned for comparison with the promoter region of the human IFN--beta-1 gene (Degrave et al., 1981). Numbering is from the presumed cap site. A purine-rich transcription-regulatory sequence around -75 in the IFN-beta-1 pro-moter (Zinn et al., 1983), repeated at -10, is under-lined. The TATA box is doubly underlined.

SummarY of the Invention The present invention concerns human DNA encoding anenzyme having (2'-5') oligo A synthetase activity.
One form of the DNA has the nucleotide sequence set forth in Figure 7A. Another form of the DNA has the sequence of nucleotides 1-1322 set forth in Figure 7A
which overlaps with the sequence of nucleotides 901-1590 set forth in Figure 7B.

An enzyme having (2'-5') oligo A synthetase activity has the amino acid sequence set forth in Figure 7A.
Another enzyme having (2'-5') oligo A synthetase acti-vity has the sequence of amino acids 1-364 set forth in Figure 7A which overlaps with the sequence of amino acids 290-400 set forth in Figure 7B.

A 1.6 kb and 1.8 kb RNA having nucleotide sequences complementary to the nucleotide sequences in Figures 7A and 7B have been isolated.

A method of monitoring the response of a patient to an interferon comprises measuring the concentration of (2'-5') oligo A synthetase mRNA in cells or body fluids of the patient by hybridizing to the mRNA DNA
complementary thereto.

Antigenic peptides of the present invention have an amino acid sequence contained within the amino acid sequences set forth in Figures 7A and 7B. Antibodies raised against these antigenic peptides recognize and immunoprecipitate (2'-5') oligo A synthetase.

A method of monitoring interferon activity in a subject comprises measuring the amount of (2'-5') oligo A synthetase in a cell or body fluid of the subject at predetermined time intervals, determining the differences in the amount of said synthetase in the cell or body fluid of the subject within the different time intervals, and determining therefrom the amount of synthetase in the cell or body fluid of the subject and thereby the interferon activity of the - subject. The synthetase may be measured by contacting the synthetase with an antibody of the present invention so as to form a complex therewith and determining the amount of complex so formed.

1 33728 l Det~ile~ De~cri~ion o~ the Invention The present invention concerns human DNA encoding an enzyme having (2'-S') oliqo A synthetase activity and having the nucleotide sequence set forth in Figure 7A.
The DNA may also comprise the sequence of nucleotides 1-1322 set forth in Figure 7A and the overlapping se-quence of nucleotides 901-1590 set forth in Figure 7B.
The DNA of the present invention has the restriction enzyme sites set forth in Figure 9.

An enzyme having l2'-5') oligo A synthetase activity has the amino acid sequence set forth in Figure 7A.
This enzyme comprises about 364 amino acids and has a molecular weight of about 41,500 daltons. Another enzyme having (2'-5') oligo A synthetase activity com-prises the sequence of amino acids 1-364 set forth in Figure 7A and the sequence of amino acids 290-400 set forth in Figure 7B. This enzyme comprises about 400 amino acids and has a molecular weight of about 46,000 daltons-The present invention provides a 1.6 kb ~NA having anucleotide sequence complementary to the nucleotide sequence set forth in Figure 7A. Also provided is a 1.8 kb RNA co~prising a nucleotide sequence comple-mentary to the sequence of nucleotides 1-1322 set forth in Figure 7A and the sequence of nucleotides 901-lS90 set forth in Figure 7B.

.

A method of monitoring the response of a patient to an interferon comprises measuring the concentration of (2'-5') oligo A synthetase mRNA in cells or body fluids of the patient by hybridizing to the mRNA DNA comple-mentary thereto. The mRNA may be the 1.6 kb or 1.8 kb RNA of the present invention.

A method for evaluating the response of cells and tis-sues to interferon comprises hybridizing RNA from cells or tissues exposed to interferon with cDNA complementa-ry to the RNA, and determining the extent of hybrid-ization. The RNA is extracted from cells or tissues which have been exposed to interferon, immobilized on a membrane filter and hybridized to labeled cDNA sp-ecific for interferon-induced mRNAs. The method may also ,comprise L~ situ hybridization of labeled cDNA to slic-es of tissues and then evaluating by microscopic exami-nation autoradiography, or fluorescence. The cells ortissues analyzed may be of human or other animal ori-gin.

A kit for carrying out a method for evaluating the response of cells and tissues to interferon contains a cDNA complementary to a sequence set forth in-Figure 7A
or 7B, reagents to carry out the hybridization tests for nick-translation with deoxy ribonuclease I and [32P]-gamma -dCTP, reagents for hybridization on nitro-cellulose membranes, and reagents for RNA extractionfrom cells.

Also provided are antigenic peptides having amino acid sequences contained within the amino acid sequences set forth in Figure 7A and Figure 7B.

An antigenic peptide of the present invention has the amino acid sequence comprising the 17 C-terminal amino acids of the amino acid sequence set forth in Figure 7A
and having the amino acid sequence: ARG-PRO-PRO-ALA-SER-SER-LEU-PRO-PHE-ILE-PRO-ALA-PRO-LEU-HIS-GLU-ALA.
Another antigenic peptide has the amino acid sequence:
GLU-LYS-TYR-LEU-ARG-ARG-GLN-LEU-THR-LYS-PRO-ARG-PRO-VAL-ILE-LEU-ASP-PRO-ALA-ASP.

Antibodys raised against the antigenic peptides of the present invention recognize and immunoprecipitate (~'-5') oligo A synthetase.

A method of monitoring interferon activity in a subject comprises measuring the amount of (2'-5') oligo A syn-thetase in a cell or body fluid of the subject at pre-determined time intervals, determining the differences in the amount of said synthetase in the cell or body fluid of the subject within the different time inter-vals, and determining therefrom the amount of synthe-tase in the cell or body fluid of the subject and thereby the interferon activity of the subject. The amount of synthetase may be measured by contacting the synthetase with an antibody of the present invention so as to form a complex therewith and determining the amount of complex so formed.

A method of monitoring interferon activity may further comprise, extracting (2'-5') oligo A synthetase from a cell or body fluid which has been exposed to interfer-on, labeling the extracted synthetase with an identi-fiable marker to form a labeled synthetase, contacting the labeled synthetase with an antibody of the present invention under suitable conditions so as to form a .

labeled-synthetase-antibody complex, and detecting the marker in the complex, thereby detecting the syntheta-se. The marker may be 35S-methionine.

A kit for carrying out the method of monitoring inter-feron activity comprises an antibody of the present invention, materials for extracting the synthetase, materials for labeling the synthetase, and materials for detecting the marker and determining the amount of synthetase.
The present invention also provides cloned DNA that specifically hybridizes to messen~er RNAs which appear in human cells after exposure to interferon. The cloned cDNA may be specific for the (2'-5') oligo A
synthetase mRNAs of 3.6, 1.8 and 1.6 kilobase. A
cloned DNA of the present invention is specific for the mRNA of a 56,000 Mr-protein, which mRNA is 2 kilobase and which has the sequence defined in Figure 1.

A partial cDNA clone (El) for the (2'-5') oligo A syn-thetase mRNA from human SV80 cells, was first obtained through its ability to select by hybridization an mRNA
producing (2'-5') oligo A synthetase activity upon translation in Xenopus l~evis oocytes (Merlin et al., 1983). The El cDNA insert (675 bp) hybridizes to 3 RNA
species of 1.6, 1.8 and 3.6 kb which are coinduced by IFN in SV80 cells, accumulate for 12 hours and are found in the cytoplasmic polysomal fraction (Benech et al, 1985). Two other early transcripts (2.7 and 4 kb) appear in lesser amounts. Analysis of various types of human cells has shown that these RNAs are differential-ly expressed in a cell specific manner. In B lympho-blastoid cells (Namalva, Daudi) only the 1.8 kb RNA
accumulates, while in amniotic WISH cells, in histiocy-1 ~3~8~

tic lymphoma U937 cells and in HeLa cells, the 1.6 kbRNA is predominantly induced by IFN with some 3.6 kb RNA, but little 1.8 kb RNA. ~n diploid fibroplasts FSll, in SV80 fibroplastoid cells and in the T cell line CEMT, all 3 stable RNAs are expressed (Benech et al., 1985). The type of (2'-5') oligo A synthetase RNA
expressed does not depend on the species of IFN used (alpha, beta, or gamma) but rather seems developmental-ly regulated in the cell.

The different (2'-5') oligo A synthetase transcripts appear to originate from a single gene (Benech et al., 1985) . Restriction mapping showed: 1) that the El cDNA corresponds to the 3' end of the 1.6 kb RNA; 2) that the 1.8 kb RNA has a different 3' end than the 1.6 kb RNA and contains an additional downstream exon; and 3) that the 3.6 kb RNA has the same 3' end as the 1.8 kb RNA but is incompletely spliced. Hybridization-translation experiments using specific genomic DNA
fragments also demonstrate that both the 1.8 and 1.6 kb RNAs actively code for (2'-5') oligo A synthetase (Benech et al., 1985).

cDNA clones for the 1.6 and 1.8 kb RNAs have been iso-lated and sequenced, which enabled the deduction of the amino acid sequences of two forms of the IFN-induced (2'-5') oligo A synthetase in human cells. The two proteins differ in their C-termini, which is hydropho-bic in the 1.6 kb RNA product (E16) and acidic in the 1.8 kb RNA product (E18). A complete mapping of the (2'-5') oligo A synthetase gene shows that the 1.6 kb RNA is coded by 7 exons and the 1.8 kb RNA by 8 exons.
The sequence of the presumed transcription initiation site and promoter region of the IFN-activated human (2'-5') oligo A synthetase gene shows a striking homol-..
., ogy to the promoter region of the human IFN-beta-l gene.

ExAMpr~E ]

S MeAsure of (2'-5'~ oligo A synthet~se ~RNA b~y cDNA
clones.

A) Isol~tion of E-cDNA c]one~

Total RNA was prepared from 109 SV80 cells (S~40-trans-formed human fibroblasts) treated for 12 hours with 200 units per ml IFN-beta. The RNA was extracted by 3M
LiCl - 6M urea and purified by passage on oligo dT-cellulose. The 0.4 mg poly A+-RNA obtained were frac-15 tionated in a preparation of gel electrophoresis appa-ratus in 1.5~ agarose/6M urea 25rnM sodium citrate pH
3.5. The 17-18S RNA fraction, was used to prepare cDNA
as follows: 2 micrograms RNA were heated for 1 min at 90C with 2 micrograms oligo (dT)12-18 in 6 20 ters water, cooled at 0C, supplemented with salts to a final concentration of SOmM Tris-HCl pH 8.3, 10mM
MgC12, 75mM RC1 and incubated 5 min at 42C before adding lmM dithiothreitol, lmM each dATP, dTGP, 0.SmM
dCTP, 20 micro-Ci 32P-dCTP (300 (Ci/mmol)), 4mM Na-25 pyrophosphate and 20 units reverse transcriptase in afinal volume of 0.1ml. The mixture was incubated for 45 minutes at 42C, the reaction was stopped with 10mM
EDTA, 0.2~ Na-dodecyl sulfate and the cDNA extracted with phenol-chloroform, treated with 0.3N NaOH for 2 30 hours at 52C and neutralized. The cDNA was filtered on Sephadex G-75, ethanol precipitated, and tailed by dATP with terminal transferase. The synthesis of the second cDNA strand was primed with ol igo (dT) and car-ried out as for the first strand for 2 hours at 42C

but without radioactive nucleotide and without pyro-phosphate. To insure blunt ends the ds cDNA was incu-bated with ~. ~nli DNA polymerase I large fragment first in 20mM Tris-HCl pH 8, 75mM KCl, SmM MgCll, lmM
dithiothreitol for S minutes at 37C (trimming reac-tion) and then under the conditions of filling-in with ATP. The ds cDNA ws fractionated by sedimentation on a 5-20% sucrose gradient and the heaviest fractions were tailed with dCTP and annealed with equimolar amounts of P~tl-cut pBR322 plasmid DNA tailed with dCTP. About 7 ng DNA were mixed with 100 microliters of frozen, Cacl2-treated~ E. coli MM294. After 30 minutes at 0C, and 5 minutes at 37C, the bacteria were grown in 2ml of LB-broth for 2 hours at 37C, and plated on L~-agar plates with 10 micrograms/ml tetracycline. About 1.4xlO5 tetracycline-resistant, ampicillin-sensitive colonies were obtained per microgram recombinant plas-mid DNA.

To identify the cDNA clone of the (2'-5') oligo A syn-thetase mRNA, a total of 3,000 plasmid DNA clones were screened by hybridization to RNA of IFN-treated SV80 cells, and the DNA-selected RNA was tested by injection into Xenopu~ l~evis oocytes and a measure of the (2'-5') oligo A synthetase activity formed according to the method of Shulman and Revel (1980). Pools of plasmid DNA from 12 individual clones (3 micrograms DNA each;
cut with ~Q Rl) were applied onto a 0.4cm diameter nitrocellulose filter and prehybridized for 2 hours at 37C in 50% formamide, 2mM Pipes buffer p~ 6.4, 0.75M
NaCl, lmM EDTA (buffer A). Three filters with pBR322 DNA and thirty filters of recombinant DNA pools were incubated together with 300 micrograms poly A+-RNA
[calculated to have a 10-fold excess of each insert cDNA over the presumed amount of (2'-5') oligo A syn-1 33728l thetase mRNA, 0.09 micrograms or .~1 in lml buffer A
for 20 hours at 37C. The filters were washed twice at 37C with buffer A, 4 times in 20mM Tris-HCl pH 7.5, 0.15M NaCl, lmM EDTA, 0.5~ Na dodecyl sulfate (once at 37C and 3 times at 52C) and then 4 times with 10mM
Tris-HCl pH 7.5, lmM EDTA (buffer C) at 52C. Each filter was next washed individually in buffer C at 52C
and the RNA was eluted by heating 2 min at 96C in 0.3ml buffer C with 40 micrograms rabbit liver tRNA per ml. After quick cooling the etnanol precipitation, the RNA was dissolved in 2 microliters water. Ten X~noDu~
l~evi~ oocytes were microinjected with 0.7 microliters RNA and after 18 hours at 19C, the oocytes were homog-enized in their incubation medium (Shulman and Revel, 1980) and 0.15 ml of homogenate was mixed with poly (rI) (rC)-agarose beads. The beads were incubated for 16 hours at 30C with 2.5 mM [32P]-alpha-ATP (0.3 Ci/mmol), 10 mM dithiothreitol, and 10 microliters of the liquid phase were incubated with 0.35 units bacte-rial alkaline phosphatase in 30 mM Tris-base for 60 minutes at 37C. The digest was submitted to paper electrophoresis on Whatman 3MM paper at 3,000 V for 4 hours, and the spots corresponding to (2'-S') ApA and (2'-5') ApApA were cut and counted by scintillation.
From the DNA-selected RNA, 1 microliter was used for i~
vitro translation in a reticulocyte lysate as described in Weissenbach et al. (1979), to measure the total mRNA
activity of the sample.

The ratio of (2'-5') oligo A synthetase activity over total mRNA activity in the DNA-selected RNA samples was calculated for each filter. One filter with pool 174, out of the 250 pools of 12 individual plasmids, gave consistently a ratio about 10 times higher as other pools or as pBR322 DNA. The plasmid DNA of each indi-, ., vidual clone of pool 174 was tested on separate filters and clone 174-3 was found to give consistently a 35-100 fold enrichment of the (2'-5') oligo A synthetase mRNA over total mRNA as compared to total RNA
or pBR322 DNA-selected RNA (Table 1). Clone 174-3 was identified as the (2'-5') oligo A synthetase cDNA and designated E-cDNA. The structure and sequence of this cDNA is shown in Fig. 1. The E-cDNA
clone contains the sequence for the 100 carboxy terminal amino acids of the enzyme and a 192 nucleotide-long untranslated region preceding the poly A-tail.

IDENrIFICATION BY HYBRIDIZATION-TRANSLATION OF THE CLONE
OF (2'-S') OLIGO A SYNTHElASE cDNA
E mRNA ~ctivitv measured bv oocvte injection (2) of induccd RNA
E%pt 1 E~pt 2 E~pt.3 (2'-S') (2-S ) (2'-S') oligo A(specificoligo A (spccificoligo A(spccific cpm activity)-cpm activity)- cpm activity)-Total poly A~-RNA 40S0 (0.007) 3440(0.004) 4900 (Q01) RNA xlected on:350 S70 (0.03) 610 (0.02) pBR filtcrs 625 (0.05) S95 725 plasmid pool 174 2320 -- --othcr pools 230-- +60 ~lonffo~pool 174 2 625 9S0 (0.04) 82S

3 6460 (4.78) 39,800 (2.7)11.23S (0.7S) 4 600 1,820 (0.48) 1,030 (0.03) S 745 S00 (0.19) S30 7 1~70 (0.1) 365 605 8 395 800 (0.02) 600 9 490 100 1.030 (0.03) 1465 (0 14) 290 1.155 (0.06) I l 1860 (0.27) 365 735 l~o RNA 195 185 S90 * Specific activity ratio of (2'-5') oligo A synthesis in mRNA-infected oocytes to translation of same RNA in reticulocyte lysates.
**Average of 28 pools B) Measure of interferon-induced (2'-5') oliqo A synthetase mRNA
by hybridization of E-cDNA.
Plasmid DNA of clone 174-3 (E-cDNA) can be used to detect the complementary RNA on electrophoretic blots of total cell RNA. Poly A~-RNA is prepared ~rom SV80 .~L~ . ~
~"r _ ..
~.

cells treated various times by 2000/ml interferon-beta and 7 micrograms RNA are denatured in 50% formamide, 6%
formaldehyde, electrophoresed in 1.3% agarose with 6%
formaldehyde and blotted onto nitrocellulose according to the procedures of Thomas (1980) and Fellous et al.
(1982). The E-cDNA plasmid labeled by nick-translation with [32P]-qamma-dCTP according to Merlin et al.
(1983), is hybridized to the nitrocellulose blot.

In SV80 cells, three RNA species which are all coordi-nately induced by the interferon treatment, hybridize with E-CDNA (Fig. 2), a large RNA species of 3.6 kilo-bases and 2 smaller species of 1.85 and 1.65 kilobase.
In non-treated SV80 cells, no E-specific RNA is found.
The 3 RNA species appear at 4 hours, are maximum at 12 hours and decrease slowly thereafter. The RNAs are still clearly detected at 24 hours after interferon.
Additional RNA species seen only at 4 hours are most probably precursors of the more stable species. The same 3 RNA species are seen in human diploid fibro-blasts treated by interferon. However, in cells of thehemopoietic lineage such as lymphoblastoid Namalva cells, only one main RNA species hybridizes to E-cDNA
(Fig. 2) and corresponds to the 1.85 kilobase RNA spe-cies. The same RNA pattern is seen in other lympho-blastoid cells, in erythroid HL-60 and in promonocyte U937 cells.

The different E-specific RNA pattern in fibroblasts and lymphoid cells corresponds to different forms of the (2'-5') oligo A synthetase in these cells. Lymphoid cells contain an enzyme of molecular weight 30,000 daltons, while fibroblasts contain two forms of the enzyme of molecular weight 80,000 and 30,000 daltons, as reported by Revel et al. (1982). The small 1.85 - ' t;

kilobase mRNA is sufficiently long t-o code for the 30,000 Mr enzyme but not for the larger form, while the 3.6 kilobase E-mRNA codes for the 80,000 Mr form of the enzyme. All three E-specific RNA species hybridize to a single clone of human genomic DNA, and probably orig-inate from a single gene, the 3.6 kilobase RNA havingan additional interferon exon as compared to the 1.85 kilobase RNA.

Leucocyte interferon-alpha induces E-specific RNA as well as does fibroblast interferon-beta. The multi-plicity of ~NA species revealed by hybridization to E-cDNA suggests that different interferon species, which all induce (2'-5') oligo A synthetase, could induce different forms of the RNA and of the enzyme. Differ-ent interferon species can also vary in their efficacyfor inducing E-mRNA.

RNA to be treated in the above hybridization assay to E-cDNA may be prepared from various cells in culture or from tissues taken from patients receiving interferon therapy or suffering from viral diseases or from dis-ease in which an elevated (2'-5') oligo A synthetase was observed (Schattner et al., 1981). RNA may also be prepared from blood cells, such as leukocytes, ob-tained from peripheral blood. The electrophoretic blotcan be replaced by a dot-hybridization method, in which RNA samples are directly applied to nitrocellulose in circles or rectangles of defined area, and the radioac-tive cDNA is hybridized to the nitrocellulose sheet.
The radioactivity of each circle of rectange is then measured by direct counting or by autoradiography fol-lowed by screening of the autoradiographic film.

An alternative method is to perform hybridization Ln Situ on tissue slices obtained from biopsies of tissues exposed to interferon. This can be preferentially applied to brain biopsies in patients receiving inter-feron for a brain viral disease or tumor, in order to measure whether the brain has been exposed to interfer-on when the drug is given either by intrathecal injec-tion or by systemic injection. The method may be ap-plied to skin biopsies when the interferon treatment is given locally as an ointment for skin lesions. It is obvious that many other application are possible. The tissue slices may be fixed and hybridized in ~i~ to radioactive DNA, followed by an autoradiography with a sensitive photographic emulsion. The cDNA may also be labeled by fluorescent nucleotides or by modified nu-cleotides which can bind fluorescent molecules, and thehybridization to the tissue slice can be monitored by fluorescent microscopy.

An increase in hybridization of the E-cDNA was compared to a proper control cell RNA or tissue sample, indicat-ing that the cell or tissue has been exposed to inter-feron. The rapidity (4-24 hours) and sensitivity (1-200 units of interferon per ml) of the method makes it very useful to follow a treatment by external interfer-on, or formation of endogenous interferon in blood andtissue of patients.

~ pr~ 2 Cloned cDNA for the interferon-induced 56,000 Mr pro-tein A) Isol~tion of cloned C56-cDNA

The cloned cDNA was isolated from the library of recom-binant plasmids described in Example 1. The principle of the method used was differential hybridization. Two duplicate sets of the 3,000 bacterial clones grown on nitrocellulose filters were hybridized either to [32p]_ cDNA from 17S-18S poly A+-RNA of SV80 cells treated by interferon-beta(200 U per ml), or to [32P]-cDNA from total poly A+-RNA of non-treated SV80 cells. The ra-dioactive cDNA were reverse transcribed from mRNA as inExample 1. About 40% of the bacterial clones hybrid-ized strongly to the "interferon-treated" cDNA probe and 8% gave a clear differential signal, hybridizing preferentially or uniquely to the n interferon-treated"
cDNA as compared to the "non-treated" cDNA. The lat-ter group.of clones was then screened by hybridizing the plasmid DNA from each clone, labeled radioactively by nick-translation, to electrophoretic blots of RNA
from interferon-treated SV80 cells and from non-treated cells. By this criterion, 1-2% of the original 3,000 bacterial clones were found to contain a plasmid cDNA
clone corresponding to an interferon-induced mRNA. One of these plasmid cDNA clones, designated C56, showed a particularly strong differential hybridization. This C56 DNA hybridizes to an 18S RNA present in interferon-treated cells but completely absent from control cell RNA (Fig. 3). rn comparison, HLA-A,B,C mRNA which is increased 5-fold in SV80 cells after interferon-treat-ment (Fellous et al., 1982), appears much less induced ` - 24 - 1 3372 8 1 than C56 mRNA and under the experimental conditions of Fig. 3, gives a clear signal also with "non-treated~
RNA.

The mRNA selected by hybridization to C56 cDNA immobi-lized on nitrocellulose filters, followed by elution from the films (as in Example 1) was translated in a reticulocyte lysate cell-free system and the [35S]-methionine-labeled translation products were analyzed by polyacrylamide gel electrophoresis in Na-dodecyl sulfate according to the method described in Weissenbach et al. tl979) adapted from Laemle (1970).
The CS6 cDNA-selected RNA is translated into a 56,000-Mr protein. The sequence of the C56 cDNA permits one skilled in the art to deduce 65 amino acids of the carboxy terminal sequence of the 56,000 Mr protein (Fig. 4).

Hybridization of the C56 cDNA to RNA extracted from SV80 cells treated various times by interferon-beta (200 U per ml), shows that the C56 mRNA starts to ap-pear at 1 hour after interferon addition (Fig. S). The CS6 RNA reaches its maximum after 4 hours, but is still detectable, although reduced, at 24 hours. Induction of CS6 mRNA was also demonstrated in diploid fibro-blasts, and in lymphoblastoid cells. Induction was proportional to the concentration of interferon be-tween 10 and 200 units per ml. CS6 mRNA was also in-duced by interferons alpha and gamma, although the latter was less efficient. The absence of this mRNA in non-treated cells and its strong and rapid increase after interferon addition make the C56 cDNA an excel-lent probe to evaluate the response of cells to inter-feron. The techniques described for E-cDNA in Example 1, can be similarly applied to the C56 cDNA.

`- - 25 - 1 33728 1 The availability of a number of cDNA corresponding to mRNA induced by interferon offers new perspectives. In particular, interferon- is needed at 100-fold lower concentrations to induce HLA-A,B,C mRNA than to induced E-mRNA or C56 mRNA (Wallach ~ ~1. (1982); on the other hand, some subspecies of interferon-alpha, such as alpha-d can induce E-mRNA when a concentration 100 times lower than those needed to induce HLA-A,B,C mRNA.
A comparison of the hybridization of different cloned cDNAs to the same RNA sample, can indicate what type of interferon is involved. Thus, more information can be derived from the comparison of different cDNA than from the use of only one cDNA probe.

ExAMprlE 3 Kit For The Measure Of Interferon-Induced mRN~

The Rit would provide the cloned cDNA specific for the mRNA of the (2'-5') oligo A synthetase and for the mRNA
of the 56,000 Mr protein, described herein, as well as reagents to carry out the hybridization tests: com-prising reagents for nick-translation with deoxy-ribo-nuclease I and [32P]-gamma-dCTP, reagents for hybrid-ization on nitrocellulose membranes, and reagents forRNA extraction from the cells.

ExAMprJE 4 Sequence of cDNA for the 1 6 kb (2'-5'~ oligo ~ synthe-tase mRNA

The partial El cDNA clone (Merlin et al., 1983), shown to be the 3' end of the 1.6 kb (2'-5') oligo A synthe-tase by IFN in human cells (Benech et al, 1985) was used to screen a lambda gtlO cDNA library from SV80 cell RNA (Wolf and Rotter, 1985). By restriction map-ping, clone lambda gtlO 9-2 was found to contain the El cDNA at the 3' end of a 1.32 kb ~QRl insert (Fig. 6A) which was subcloned in pBR (9-21 cDNA). Sequencing was carried out as outlined in Fig. 6A and confirmed that the the 9-21 cDNA contains the C-terminus and 3' un-translated sequence previously reported for the El cDNA
(Merlin et al., 1983). The 9-21 cDNA sequence (Fig 7) predicts an open reading frame of 364 amino acids starting at an ATGATG sequence. A computer program based on the 3-base periodicity of protein-coding se-quences (Trifonov, 1984) indicated that the only com-patible reading frame is the one starting from this ATGATG. It is possible that translation initiates at the second ATG in this site, since it is the only one preceded by an A at -3 and having homology with the concensus translation initiation sequence (Rozak, 1984).

The enzyme thus coded by the 1.6 kb (2'-5') oligo A
synthetase RNA has a molecular weight of about 41,700 daltons, based on the deduced amino acid sequence, which is in good agreement with the apparent 38,000 Mr protein seen by SDS-polyacrylamide gel electrophoresis of the in Yi tro translation product of RNA hybridized to El cDNA (Merlin et al., 1983). The C-terminal hep-tadecapeptide predicted by the open reading frame, was synthesized chemically and used to immunize rabbits.
The antiserum obtained (C in Fig. 12) precipitates specifically a protein migrating at 38,000-Mr in SDS
gel electrophoresis from 35S-methionine labeled ex-tracts of cells treated by IFN which is absent from untreated cells. Two experiments confirmed that this ~ 337281 protein has (2'-S') oligo A synthetase activity: 1) it was removed from the extracts by passsage through a poly (rI)(rC) agarose column; and 2) the supernatant remaining after immunoprecipitation was depleted of a large part of the enzymatic activity.
s ExAMpr~E 5 Seeuence of cDNA for the 1.~ kb (2'-5') oligo A synthe-tase m~NA

A genomic DNA fragment corresponding to the additional exon of the 1.8 kb RNA (Benech et al., 1985; see Fig.
9) was used as probe to isolate an E18 cDNA clone, 48-1, from the same lambda gtlO cDNA library of SV80 RNA.
The restriction map of the E18 cDNA clone (Fig. 6B) confirmed that its 5' end is part of the E16 cDNA but that its 3' end differs. Sequencing (Fig. 7) revealed that the junction is at nucleotide 1071 of the E16 9-21 cDNA clone, the last 247 nucleotide of E16 being re-placed by a 515 nucleotide-long sequence terminated by a different polyadenylation site. This difference accounts for the 0.2 kb difference in size between the two mRNAs seen on Northern blots. The 5' portion of the E18 cDNA shows no base change from the sequence of the E16 cDNA, but is incomplete. The gene mapping described below, indicates that both 1.6 and 1.8 kb mRNAs have the same 5' end.

The 3' region of the E18 cDNA which diverges from the E16 sequence, contains an open reading frame ending after 54 codons. This reading frame, which leaves a 350 nucleotide-long untranslated region, was confirmed by the computer program based on the 3 base periodicity of protein-coding sequences (Trifonov, 1984). An al-ternate longer open reading frame would not be in the same computed phrase as the S' portion common with the E16 cDNA. A hydropathy plot (Ryte and Doolittle, 1982) on the prprediced C-termini of the 1.6 and 1.8 kb mRNA
protein products, indicates a striking difference be-S tween the two forms of the (2'-5') oligo A synthetase (Fig. 8). The C-terminus of the E16 protein is very hydrophobic, while that of the E18 protein is hydrophi-lic and contains two acidic regions (Asp-Asp-Glu-Thr-Asp-Asp and Glu-Glu-Asp) (Fig. 7). Furthermore, a possible glycosylation site is present in the C-termi-nus of the E18 product (Fig. 7).

_ - 29 - 1337281 ExA~pr~ 6 Or~nizAtion of the hum~n t~'-5') oligo A synthet~se gene Three overlapping genomic clones were isolated using the El cDNA as probe (Benech et al., 1985), one from a library of partial E~QRl digest of human blood cell DNA
(Mory et al., 1981) and two from a library of partial ~1~1 and Hae3 digest of embryonic human DNA (Maniatis et al., 1978). The genomic clones represent about 29 kb of human DNA and no evidence for more than one E
gene was found while screening the libraries. Southern blots of genomic DNA are consistent with the existence of a single gene (Fig. 9). By Northern blot analysis using genomic DNA fragments as probes, by Sl nuclease mapping and by sequencing, the E16 cDNA 9-21 was shown to correspond to five exons on the gene (Fig. 9). The ATGATG sequence is found in exon 3, while the termina-tion codon and 3' untranslated region with the polyade-nylation site of the 1.6 kb RNA are found in exon 7.The structure of the more 5' exons 1 and 2 is described below. The sequences of the intron-exon boundaries were determined (Table 2) and follow the CAG and GT
rule for the splice acceptor and donor sites (Breathnach and Chambon, 1981).

A sequence CTGAC/T is commonly found not far from the splice acceptor, as reviewed recently by Reller (1984).
It is notable that the CTGAC/T region shows base com-plementarity to the sequence of the intron/exon 3'boundary (acceptor site; Table 2), in addition to the complementarity of the intron donor site with the CTGAC
sequence pointed out by Keller (1984) as playing a role in the lariot model.

.~

The sequences of the 5 exons containing the coding region of the (2'-5') oligo A synthetase produced by the 1.6 kb mRNA, indicates that the enzyme is composed of domains with differing amino acid compositions (Ta-ble 3). The first exonic domain (60 amino acids) is rich in aspartic acid, in the second (amino acids 61 to 156) arginine is predominant, the next two exons (amino acids 157 to 21~ and 219 to 295) are lysine rich, and the C-terminus of the E16 product (296 to 364) is very rich in proline and alanine.

3s -, EXON-INTRON BOUNDARIES IN THE HUMAN
(2'-5') OLIGO A ~YN~ ASE GENE

(~.2) -S0 ..CC~Il~IGAGGAAACGAAACCAACAG C_ TCCAAG....
e~on 3 (4.2) ..AAG.GTG.GTA.AAG GTGAGCGG...1.3kb (4.2)..GGTTTGCCIlACTAAG 214 CATCAATTATTA~ CAG GGT.GGC.TCCTCA
e~on 4 (4.2) ..GAT.GCCCTG.G GTGAGAGCTC...2.3kb (3.3)..GA~GA~GCTGAC 503 CCTAAGTTGTAAGTTTTACCCAGACAG .G~AG.TTG.ACT..
e~on S (33) ..TGG.TAC.CAAAAT GTATGGTTT...5.3kb (3.1)..TGAGCAAACCAA 668 GAl~ lccTcTTcTcAG TGT.AAG.AAG.AAG
e~on 6 (3-1) ..ACG.AAA.CCCAG GTATGCTATCCCACATGGCTTG.Ø9kb (3.1) -~t 1- 917 TAc~l~l~l~lAAATGcTGcTcTGcAG GCCT.GTGATC..
e~on 7a (3-1) ~oRI-(0.7) ..TGG.ATT.CTG.CTG GTGAGACCT....GAATTCATTCCCCTAAG

ATTGA~I~l-lGCTTCGGGCTC..1.4kb (wholc intron = 1.6kb) (6.8) BamHI-. ..GGATCCAG
ATGGCATGTCACAGTATACTAAATGCTCAC
T 'ATccAGcTGcAATGcAGGAAGAcTccc- 1072 e~on 8(1.8k~RNA) 1585 C~ TGTGATcATGTGTcTcAcc~lll~AG GCT.GAAAGC...AATAAAAATAAAGCAAATACCATTTATTGGGTG.

For exon numbering see Fig. 7 and 9. The self-complementary regions between the CTGAT/C, or CTTAC, CTGTC (Reller, 1984) and splice acceptor CAG are underlined. The polyadenylation sites with a conserved undecanucleotide of the 1.6 and 1.8 kb RNAs (see Fig. 7) are underscored by dots. The numbers in parentheses are the size of the EcoRl genomic fragments carrying the introns or exons (see Fig. 9).
The start and end of each exon is numbered as in the 9-21 E cDNA of Fig. 7.

, (2'-5') OLIGO A SYNTHETASES

E 16 C-tcrm.E 18 C-tcrm 1-60 61-lS6 157-218 219-295 296-346 347-364 347 100 AA(60) (96) (62) (71) (51) (18) (54) ALA2 (3.3)7 (7.3)0 (0.0)3 (3 9t4 (7.8)3 (16 7)4 (7 4) ARG4 (67)10 (104)3 (4.8)S (6.5)1 (20)1 (S.6) 2 (3.7) ASNI (1.7)2 (2.1)2 (3.2~3 (3.9)3 (5.9)0 (0.0) 1 (1.9) ASP6 (100)S (5.2)2 (3.2)1 (1.3)4 (7.8)0 (0.0) S (9 3) CYS4 (6.7)1 (1.0)2 (3.2)2 (2.6)1 (2.0)0 (0 0) 1 (1.9) GLNI (1.7)7 (7.3)6 (9.7)S (6.S)2 (3.9)0 (0 0) 3 (5 6) GLU2 (3.3)7 (7 3)S (8.1)4 (S.2)2 (3.9)1 (5.6) S (9.3) GLY2 (3.3)9 (9.4)3 (4.8)3 (3.9)6 ~11.8)0 (0 0)2 (3.7) HISI (1.7) ()l (1.6) 1 (1.3~ 0 (00) 1 (5.6~ 3 (S 6) ILES (8 3)2 (2.1)3 (4.8)4 (5 2)2 (3 9)1 (5 6~ 2 (3 7) LEl,' S(8.3)13(13.5) 8(129)10(1301 6(118) 2(11 1) 2(3.7) LYSS (8.3)2 (2.1)7 (11.3)9 (11.7)2 (3 9)0 (0.0)1 ( I 9) MET3 (5.0)0 (0.0)0 (0.0)1 ~1.3)0 (0 0)0 (0.0) 0 (0.0) PHE4 (6.7)7 (7 3)3 (4.8)3 (3 9)l (2.0)1 (S.6) 1 (1-9) PRO3 (S.0)4 (4.2)3 (4.8)4 (S.2)6 (11.8)S (27.8)4 (7.4) SER4 (6.7)8 (8.3)3 (4.8)1 (1-3)3 (S.9)2 (11.1) S (9.3) THR2 (3.3)4 (4.2)S (8.1)6 (7.8)1 (2.0)0 (0.0)8 (14.8) TRP0 (0.0)1 (1 0~1 (1.6)2 (2.6)4 (7.8)0 (Q0) 1 (I 9) TYR2 (3.3)0 (0.0)3 (4.8)7 (9 1)1 (2.0)0 (Q0) 4 (7.4) VAL4 (6.7)7 (7.3)2 (3.2)3 (3.9)2 (3.9)1 (S.6) 0 (0.0) -~ 33 ~ 1 3 3 72 8 1 Although the E18 cDNA 48-1 is incomplete, we found that exons 1-6 (Fig. 9) hybridize to the 1.8 kb mRNA as well as to the 1.6 kb mRNA on Northern blots. The structure of the two RNAs is most likely identical up to exon 7.
The additional splicing from the middle of exon 7 to exon 8 characterizing the E18 cDNA, was confirmed by sequencing these intron-exon boundaries in the genomic DNA clone (Table 2). The truncated exon 7a present in the E18 cDNA is followed by a 1.6 kb intron containing the polyadenylation site of the 1.6 kb RNA. Exon 8 begins 98 bp downstream from the unique ~m~l site of the gene (Table 2, Fig. 9). The genomic exon 8 ends by the polyadenylation site of the 1.8 kb RNA, character-ized by a tandem repeat of the AATAAA signal. Although exon 7 and 8 have no homology, a conserved undecanucle-otide ACCATTTATTG, in which the third cytidine is poly-adenylated, is present at the end of both exons (Table 2). As pointed out previously (Benech et al., 1985), a hairpin-loop structure can be formed in both cases between this conserved and undecanucelotide and the AATAAA region; such structures may participate in the cell-specific mechanism which determines whether cleav-age and polyadenylation of the transcripts occur at the end of exon 7 or at the end of exon 8.

Based on the above gene mapping, the enzyme coded for by the 1.8 kb mRNA should be identical to the E16 prod-uct in the first 346 amino acids, which are followed by a specific 54 amino acid-long region, rich in aspartic acid, glutamic acid and threonine. The 400 amino acid-long E-18 enzyme would have a molecular weiqht of 46,000.

~ ~ 34 ~ 1 3 3 7 2 8 1 ~X~prE 7 Two for~s of the h~lm~n (2'-5'~ oligo A ~ynthetAse ~ro-duced by ~Itern~tive ~licing of the s~me ~ene S Northern blot analysis of SV80 RNAs revealed that 3 species of RNA (1.6, 1.8 and 3.6 kb) hybridizing to El cDNA accumulate in cells up to 12 hours after exposure to IFN (Merlin et al., l9a3). Additional unstable transcripts were also seen. The relationship between these RNAs was investigated by transcript mapping on genomic DNA clones. In two human genomic libraries, the El cDNA identified only one series of overlapping genomic DNA clones which represent 29 kb of human DNA
(Fig. 9A) and were found to contain an apparently unique (2'-5') oligo A synthetase gene (Benech et al., 1985a). By Sl nuclease analysis and partial gene se-quencing, the 9-21 (El) cDNA was found to correspond to 5 exons (numbered 3-7 on Fig. 9A and in the sequence of Fig. 7). The 3' end and polyadenylation site of this cDNA was identified at the end of exon 7 (Fig. 9).
~owever, hybridization of further downstream genomic DNA fragments to Northern blots of SV80 RNA, revealed (Benech et al., 1985) that only the 1.6 kb RNA ended at the polyadenylation site in exon 7, while both the 1.8 and 3.6 kb RNAs hybridized to an additional exon lo-cated 1.6 kb downstream and which ends also by a poly-adenylation site (exon 8, Fig. 9). Thus the 9-21 (El) cDNA represents the 1.6 kb RNA and was renamed E16 cDNA. It was further found that the 3' half of exon 7 does not hybridize to the 1.8 kb RNA indicating that the transcript is formed by a splicing event from the middle of exon 7 to exon 8. All the 5' upstream exons hybridized to both 1.6 and 1.8 kb RNAs, indicating that the 2 RNAs differ only in their 3' ends. This was - _ 35 _ 1 33 728 1 confirmed by the isolation from the SV80 lambda-gtlO
cDNA library of a cDNA clone for the 1.8 kb RNA (clone 48-1 or E18 cDNA, Fig. 7B), which demonstrated the differential splicing and ended at the polyadenylation site of exon 8 (Fig. 9). A similar cDNA clone was found in a Daudi cDNA library by Saunders and Williams (1984). The E18 sequence locks the last 247 nucleo-tides of E16 which are replaced by 515 nucleotides accounting for the difference in size between the 1.6 and 1.8 kb RNAs.
The 1.8 kb RNA would thus code for a 46,000 Mr protein (E18) which differs from the E16 protein in its C-ter-minus. Like the E16 protein, the E18 product has ds RNA binding and (2'-5') oligo A synthetase activity as shown by translation of mRNA selected by hybridization to E18-specific DNA fragments (Benech et al., 1985).
This suggests that the first 346 aminoacids common to the 2 proteins contain the catalytic sites. Examining the exon composition this common part appears composed of a N-terminal acidic domain, followed by three basic regions. The last 18 residues of the E16 protein form a very hydrophobic domain, which is replaced in E18 by a longer hydrophilic and acidic region which also con-tains a potential glycosylation site. This difference between the 2 enzymes may determine their ability to dimerize, or interact with other proteins and cellular structures. For example, E16 may bind to membranes while E18 may interact with basic proteins in ribosomes or in the nucleus.
Two forms of the (2'-5') oligo A synthetase were found by gel filtration in extracts of IFN-treated human cells (Revel et al., 1982): a 30-40 kd enzyme which could correspond to a monomeric form of the E16 or E18 proteins, and a 60-80 kd enzyme which remains to be identified. The 3.6 kb RNA does not seem to code for a large enzyme since transcript mapping showed that this RNA contains intronic regions (e.g. between exon 7 and 8) which are removed from the 1.8 kb RNA and have no 5 open reading frame. We also failed to see large E mRNA
in oocyte translations. An 80 kd protein in SDS was reported in purified human (HeLa) synthetase (Yand et al., 1981) but is enzymatic activity was not demon-strated. In enzyme purified from Namalava and CML
cells (Revel et al., 1981b) we could detect a 40 kd band in SDS. Thus it remains possible that the 60-80 kd enzyme form is a dimer of the 40 kd protein. The human synthetase may differ from that in mouse cells where a large 3.8 kb RNA was seen under denaturing 15 conditions which codes for a 80 kd enzyme (mainly cyto-plasmic), in addition to a 1.5 kb RNA coding for a 30 kd enzyme (mainly nuclear) (St. Laurent et al., 1983).
The human E cDNA detects a 3.8-4 kb and a 1.6-1.7 kb RNA in mouse cells, the large RNA hybridizing to E18-20 specific DNA (Mallucci et al., 1985). It is possiblethat in human cells the large RNA is further processed - into 1.8 kb RNA, which has not been seen in mouse cells. Shulman et al. (1984) have used the fact that the bulk of the (2'-5') oligo A synthetase in human 25 cells behaves as a smaller protein than in mouse cells to map the human synthetase gene to chromosome 11 in human rodent-hybrid cells. Antisera specific to E16 and E18 will help to elucidate the relationship between these proteins and the two forms of the native enzyme 30 seen in human cells.

, ~ XAr~pr~ ~ 8 Cell ~eci~ic e~re~ion of the two 12'-5') oliqo A
synthetA ~e tnRNA~

S RNA from a number of human cell lines have been exam-ined in Northern blots with the E cDNA probe (Merlin et al., 1983; Benech et al., 1985). Table 4 shows that human cells can be qrouped in 3 classes according to the predominant E mRNA species induced by IFN. Lympho-blastoid B cell lines from Burkitt lymphomas have main-ly the 1.8 kb RNA. Instead, several cell ines have the 1.6 and 3.6 kb RNA but little 1.8 kb ~NA. If the 3.6 kb RNA is a partially spliced precursor of the 1.8 kb RNA, these cells may have an inhibitlon in the process-ing of the 3.6 kb RNA. -T-lymphocyte lines (CEMT from an acute leukemia and Gash from hairy cell leukemia) contain like fibroplastic cells, all 3 E RNA species.
The E18 polyadenylation (pA) site seems, therefore, to be used in all human cells to produce either 3.6 or 1.8 kb RNA. The E16 pA site seems not to be used in B
lymphoblastoid cells. A conserved undecanucleotide present in both E16 and E18 pA sites can form a hairpin-loop with the AATAAA siqnal and could have a role in site selection (Benech et al., 1985). E18 has a tandem repeat of the AATAAA s ignal and could be a stronger pA site. Transcripts ending at the E18 pA site accumulate earlier after IFN addition than the 1.6 kb RN~ (Benech et al., 1985).

TARr~E 4 PREDOMINANT (2'-5') OLIGO A SYNTHETASE RNA SPECIES

3.6 kb 3.6 kb 1.8 kb 1.8 kb 1.6 kb1.6 kb B lymphoblastoid HistiocyticBurkitt lymphoma: Fibroblastic:
lymphoma U937 - Daudi -SV80 - Namalva -FSll Amniotic Wish - Raji Cervix Ca HeLa T cells:
Raji x HeLa hybrids -CEMT
Hairy cell-leuk.:
-Gash The type of synthetase predominantly made may vary in different human cells. We found no correlation between the cytoplasmic or nuclear localization of the synthe-tase and the type of RNA present in the cells. Howev-er, Namalava cells seemed to have mainly the 30-40 kd enzyme upon gel filtration while HeLa and SV80 cells had also the 60-80 kd form (Revel et al., 1982).

EXAMpr~ g Promoter region of the (2'-5') oligo ~ synthet~se gene The ~hl-~hl fragment of 0.85 kb (Fig. 10) from the genomic 4.2 kb EcoRl fragment (Fig. 9) which contains part of exon 3 of the E16 cDNA 9-21 clone, hybridized in Northern blots with the 1.6, 1.8, 2.7 and 3.6 kb RNAs. However, upstream regions did not. Several experiments allowed to localize the RNA transcriptional ..

~ 33~28 1 start in this fragment. Sl nuclease analysis first showed that exon 3 starts about 50 nucleotides upstream of the end of the 9-21 cDNA. A primer extension exper-iment using an oligonucleotide from the end of the 9-21 cDNA, indicated that the 5' end of the mRNA is about 230 nucleotides from the 5' end of this cDNA. RNA
hybridization with riboprobes produced in SP6 (Green et al., 1983) and RNAse digestion indicated two exons of 70 and 110 nucleotides preceding exon 3. By Sl nucle-ase analysis using a probe labeled at the unique ~aal site (Fig. 9), the 5' end of the mRNA was finally loca-ted 17 nucleotides upstream from the ~al site. The sequence of this region is shown in Fig. 11. The loca-tion of the transcription initiation site 17 residues before the ~1 site, is supported by the presence of a TATAA box at position -30 A striking feature of the upstream sequences, is the high purine content (69.6~) mostly adenine (58.9%). Run of a homolo~y matrix with other known promoter upstream sequences revealed a surprising homology with the human IFN promoters in particular with the sequence of the IFN-beta-l gene promoter (Degrave et al., 1981). The purine-rich re-gion from -75 to -85 of the IFN-beta-l promoter, which contains the essential transcription signal described by Zinn et al., (1983), shows 90~ homology with the region of the presumed promoter of the (2'-5') oligo A
synthetase just upstream of the TATAA box (-40 to -50)(Fig. 11). This purine-rich signal is repeated in the IFN-beta-l promoter in the segment between the TATAA box and the cap site; in this region, which may also have regulatory functions (Nir et al., 1984) the homology between the I~N-beta-l gene and the (2'-5') - oligo A synthetase gene is high. In contrast, search for homology with promoters of other genes, such as ~LA
genes (Malissen et al., 1982; Schamboeck et al., 1983) and the metallothionein II gene (Rarin and Richards, 1982) which are activated by IFNs (Fellous et al., 1982; Rosa et al., 1983b; Friedman et al., 1984) showed no apparent sequence relationship in this region of the (2'-5') oligo A synthetase gene promoter. Also, no 5 significant homology was seen with the body of the IFN-beta-l gene.

The 5' untranslated leader of the (2'-5') oligo A syn-thetase mRNA (exon 1, 2 and part of exon 3) contains 10 two short introns whose positions were tentatively determined by Sl analysis as shown in Fig. 11. The entire human (2'-5') oligo A synthetase gene is about 13 kb (Fig. 9) and the sum of the exons agrees with the observed sizes of the mRNAs.

~XAMPr.E 10 r~mhd~ GTIO cDNA clones of the (2'-5') oliqo A synthe-A lambda-gtlO cDNA library prepared from poly A+ RNA of human SV80 cells (Wolf and Rotter, 1985) was screened using as probe the Pstl~ l insert of the El cDNA
plasmid described previously (Merlin et al., 1983).
25 The insert corresponding to the 3' end of the 1.6 kb E
RNA (Benech et al., 1985), was purified by agarose gel electrophoresis and nick-translated (Rigby et al., 1977). Plaques were repeatedly picked from 9cm plates (105 phages), and small scale lambda-DNA preparations 30 were analyzed by restriction mapping by routine proce-dures (Maniatis et al., 1982) . Fifteen lambda-gtlO
cDNA clones containing the El cDNA fragment were iso-lated and phages 9-2 and 5-2 with the longest inserts were cut by Ec~-Rl and the inserts sublconed into pBR322 to obtain E16 cDNA clones 9-21 and 5-21 of Fig. 6A.
The same library was rescreened with a human genomic ;l 0.9 kb fragment from phage lambda-chEl (Benech et al., 1985), a fragment which specifically hybridizes to the 1.8 kb RNA. We thereby isolated lambda-gtlO cDNA clone 48-1 of Fig. 6B, along with another cDNA clone representing a partially spliced E
RNA. Sequencing was carried out according to Maxam and Gilbert (1980). Restriction enzymes were from New Engl`and Biolabs and Boehringer. ~omology matrix and 10 hydropathy plot computer programs of Pustell and Rafatos (1982a,b) were run on an IBM PC. Three base periodicity to locate protein coding frames was comput-ed according to Trifonov (1984) .

~ P~

Genomic DNA clones contAinin~ the (2'-5'~ oligo A syn-thetA~e gene 20 Three overlapping genomic clones were isolated as pre-viously described (Benech et al., 1985): lambda-chEl from a partial EcoRl-cut DNA library (Mory et al., 1981) and lambda-chE2 and E3 from a partial ~L~l/Hae 3 DNA library (Maniatis et al., 1978) . The genomic EcoRl 25 fragments of these phages were subcloned in pBR322.
Exon mapping was done: 1) by Southern blot hybridiza-tion of restriction digests from subcloned genomic fragments to various cDNA probes; 2) by hybridization of genomic DNA restriction fragments to Northern blots 30 of poly A+ RNA from IFN-treated and untreated cells as described (Benech et al., 1985); and 3) by sequencing of intron-exon boundaries in comparison to cDNA.

The internal ~hl-~hl 0.87 kb segment of the genomic 4.2 kb ~QRl fragment containing the 5' end of the mRNA, was subcloned in the ~hl site of pBR322 before sequencing. Primer extensions using synthetic oligode-oxyribonucleotides of 18-20 bases complementary to the mRNA (gift of Dr. D. Segev, InterYeda) were done as before (Rosa et al., 1983a). Riboprobe synthesis after subcloning in the SP6 vector was carried out according to instructions of Promega Biotec. DNA from Daudi lymphoblastoid cells and from FSll foreskin fibroblasts was prepared according to Wigler et al. (1979) and Southern blot analysis was done on Gene-Screen Plus*
nylon fiber sheets using hybridization procedure B
recommended by the manufacturer (New England Nuclear).

* trade mark SUPPLEMENTARY DISCLOSURE

Further aspects of the pre6ent invention a6 described in the Principal Disclo6ure a~e now set out. One of these iB a transfer vector comprising lambda-gt 11-E16 DWA and the lac Z gene, the DNA
being fused in phase with the lac Z gene BO aB to enable expression of the DNA in suitable host cell. A microorganism may be transformed by the transfer vector. Escherichia Coli i8 a suitable microorganism for the transformation.
Another aspect is a detailed rapid assay of (2~-5') oligo A
synthetase RNAs in human peripheral white blood cells, which may be used in the assay methods described in the Principal Di~closure.
In the drawings:
Figure 12 depicts the expression of E16 cDNA in E. coli.
Extracts of E. coli lysogen Agtll-E16 induced by IPTG at 42C were assayed on poly (rI)(rC) agarose beads for (2~-5') oligo A
synthesis. Cont = extracts of E. coli with E16 cDNA in opposite orientation to lac Z gene. Nam ~ extracts of IFN-treated Namalva cells. Electrophoresis at pH 3.5 of alkaline phosphata~ed 32p-a-ATP labelled products are shown:
Figure 13 depicts the rapid method for assay of (2~-5') oligo A
synthetase RNAs in human peripheral white blood cells:
Figure 14 depicts the quick cell blot for (2'-5') oligo A
synthetase E RNAs in human PBMC according to the method of Figure 14. Indicated number of cells and IFN (16 H treatment) were used.
Autoradiography with 32p-cDNA.
Example 12 The 9-21 cDNA of Example 4 was subcloned in lambda-gtll 80 as to fuse the coding frame in phase with the lac Z gene. Extracts of the E. coli lysogen containing this phase, showed clearly (2'-5') oligo A synthetase activity afte~ binding to poly(rI)(rC) agarose, while no activity wa6 found when the 9-21 cDNA had been fused in the opposite orientation (Pig. 13). Thi~ expres6ion in E. coli demonstrates that the cDNA indeed corresponds to the 6tructural gene coding for the ds RNA activated (2'-5') oligo A 6yntheta6e and that the protein of about 40 kd coded by the IFN induced RNA is the enzyme itself, and not a regulatory factor. This protein does not seem to require post-translational modifications to exhibit enzymatic activity.
The transformed cell containing the 9-21 cDNA has been designated Escherichia coli lambda-gtll-E16 and deposited under Accession No. I496 with the Collection National Cultures de Micro-organismes, Institut Pasteur, 25 rue du Docteur Roux, 75724-Paris-Cedex 15, France. This deposit was made pursuant to the Budapest Treaty On the International Recognition Of The Deposit Of Micro-organisms For The Purposes Of Patent Procedure.
ExamPle 13 Ouick cell blot assaY of (2~-5~) oliqo A sYnthetase RNAs for the clinical monitorinq of IFN action The usefulness of measuring the (2~-5~) oligo A synthetase has been shown in human peripheral blood mononuclear cells (PBMC) to monitor the response of patients to IFN-beta (Schattner et al., 1981a) and the IFN-beta i.m. in3ections (Schoenfeld et al., 1984).
Since the enzyme level of PBMC in normal healthy individuals is rather constant, this assay has allowed the diagnosis of viral infections evidenced by an increase in the enzyme in the PBMC and granulocytes (Schattner et al., 1981b, 1984 Schoenfeld et al., 1985). Decrease in the enzyme characterize acute leukemias with numerous blast cells ~Wallach et al., 1982: Schattner et al., 1982). This technique has also been pioneered by Williams et al., (1981) and is now in wide use.
Synthetase E is strongly induced in cells treated by all three types of IFNs, alpha, beta and gamma, and its increase is a good marker of IFN activity (Wallach et al., 1982). It is therefore possible to use measurements of E levels to determine whether cells in vitro or in vivo have been exposed to IFN and respond to it.
This measurement may be used as an assay for IFN in unknown solutions, by exposing cells to said solutions and detecmining the increase in E levels (Revel et al., U.S. Patent No. 4,302,533).
The measurement may al~o be used to establish whether IFN is pcoduced in increased amounts in whole organi6ms including man.
The (2~-5~) oligo A 6ynthetase increa6es during differentiation of hematopoietic cell6 a6 a result of autocrine secretion of IFN-beta (Yarden et al., 1984). Another important application of E
measurements is in the monitoring of patient6 under IFN therapy.
Besides clinical change6, it i6 po66ible to establish that the patients respond to IFN by measuring the PBMC E level which increases 5-10 fold during systemic IFN-alpha as well as beta treatment (Schattner et al., 1981a Schoenfeld et al., 1984). It is clear that assay of other IFN-induced activities or molecules may be used as well as the assay of the E enzyme, but thi~ method has been the most widely used tWilliams et al., Borden).
Now the assay of E RNA in human PBMC is used for the same purpose. A quick cell blot (Cheley and Anderson, 1984) using the 9-21 E cDNA as probe was developed for PBMC (Fig. 13). Oligo-nucleotides derived from the E cDNA may also be used as probes.
The effect of 10 U/ml IFN can easily be detected by this method (Fig. 14). Positive signals were obtained in a patient treated by units/day of IFN-alpha-c.

_ - 46 - 1 33728 1 Ref~rence~

1. Baglioni, C. and Nilsen, T.W. (1983), in Gresser, I.(ed) Interferon 5, Acad. Press, New York, pp.
23-42.

2. Ball, L.A. (1980), Ann. N.Y. Acad. Sci. ~Q: 486-496.

3. Benech, P., Merlin, G., Revel, M. and Chebath, J.
(1985), Nucl. Acids Res. 1~: 1267-1281.

4. Breathnach, R. a-d Chambon, P. (1981), Ann. Rev.
Biochem. 50: 349-383.

5. Chebath, J., Merlin, G., Metz, R., Benech, P. and Revel, M. (1983), Nucl. Acids Res. 11: 1213-1226.

6. Creasey, A., Eppstein, D.A., Marsh, Y.V., Rhan, Z.
and Merigan, T.C. (1983), Mol. Cell. Biol. 3: 780-786.

7. Degrave, W., Derynck, R., Tavernier, J., Haegemen, G. and Fiers, W. (1981), Gene 1~: 137-143.

8. Dougherty, J., Samanta, H., Farrell, P. and - Lengyel, P. (1980), J. Biol. Chem. 2~: 3813-3816.

9. Epstein, D.A., Czarniecki, W., Jacobson, H., Friedman, R.M. and Panet, A. (1981), Eur. J. Bio-chem. lL~: 9-15.

10. Etienne-Smekens, M., Goldstein, J., Ooms, H. and Dumont, J. (1983), Eur. J. Biochem. l~Q: 269-273.

11. Fellous, M., Nir, U., Wallach, D., Merlin, G., Rubinstein, M. and Revel, M. ~1982), Proc. Natl.
Acad. Sci. USA ~: 3082-3086.

12. Ferbus, D., Justesen, J., Besancon, F. and Thang, M.N. (1981), Biochem. Biophys. Res. Commun. m Q:
-847-856.

13. Friedman, R.L., Manly, S.P., McMahon, M. Rerr, I.M. and Stark, G.R. (1984), Cell 38: 745-755.

14. Green, M., Maniatis, T. and Melton, D.A. (1983), Cell ~2: 681-694.

15. Hovanessian, A.G., Brown, R.E. and Rerr, I.M.
(1977), Nature 268: 537-540.

16. Johnston, M., Friedman, R.M. and Torrence, P.F.
(1980), Biochemistry L~: 5580-5585.

17. Rarin, M. and Richards, R.I. (1982), Nature 797-802.

18. Reller, W. ~1984), Cell 39: 423-425.

19. Rerr, I.M. and Brown, R.E. (1978), Proc. Natl.
Acad. Sci. USA 75: 256-260.

20. Rimchi, A., Shure, H., Lapidot, Y., Rapoport, S., Panet, A. and Revel, M. (1981), FEBS Lett.

21. Ritamura, N., Semler, B.L., Rothberg, P.G., Larsen, G.R., Adler, C.J., Dorner, A.J., Emini, E.A., Hanecak, R., Lee, J.J., van der Werf, S., _ 48 -Anderson, C.W. and Wimmer, E. (1981), Nature 291:
547-553.

22. Rozak, M. (1984), Nature 308: 241-246.

23. Rrishnan, I. and Baglioni, C. (1980), Proc. Natl.
Acad. Sci. USA 77: 6506-6510.

24. Ryte, J. and Doolittle, R.F. (1982), J. Mol. Biol.
157: 105-132.

25. Laemmli, W.R., (1970), Nature ~2~: 680-685.

26. Laurence, L., Marti, J., Roux, D. and Cailla, H.
(1984), Proc. Natl. Acad. Sci. USA 81: 2322-2326.

27. Lebleu, B. and Content, J. (1982) in Gresser, I.
(ed) Interferon 4, Acad. Press, New York, pp. 47-48.

28. Lengyel, P. (1982), Ann. Rev. Biochem. 51: 251-282. -29. Malissen, M., Malissen, B. and Jordan, B. (1982), Proc. Natl. Acad. Sci. USA ~: 893-897.

30. Maniatis, T., Hardison, R., Lacy, E., Lauer, J., O'Connell, C., Quon, D., Ge~ Ree, S. and Efstradiatis, A. (1978), Cell 1~: 687-701.

31. Maniatis, T., Fritsch, P. and Sambrook, J. (1982), MoleculAr Cloning: A l~bor~tory ~Anu~l, Cold Spring Harbor Laboratory, Cold String ~arbor, N.Y.

_ 49 _ 7337281 32. Maxam, A.M. and Gilbert, W. (1980), Meth. Enzymol.
65: 499-560.

33. Merlin, G., Chebath, J., Benech, P., Metz, P. and Revel, M. (1983), Proc. Natl. Acad. Sci. USA 80:
4904-4908.

34. Minks, M.A., West, D.R., Benvin, S. and Baglioni, C. tl979), J. Biol. Chem. 2~: 10180-10183.

85. Mory, Y., Cherna~ovsky, Y., Feinstein, S.L., Chen, L., Nir, U., Weissenbach, J., Malpiece, Y., Tiollais, r. ~ Marks, D., Ladner, M., Colby, C. and Revel, M. (1981), Eur. J. Biochem. 12Q: 192-202.

36. Nilsen, T.M., Maroney, P.A., Robertson, ~.D. and Baglioni, C. (1982a), Mol. Cell. Biol. ~: 154-160.

37. Nilsen, T.M., Wood, D.L. and Baglioni, C. (1982b), J. Biol. Chem. 2~1: 1602-1605.

38. Nir, U., Cohen, B., Chen, L. and Revel, M. (1984), Nucl. Acids. Res. 12: 6979-6993.

39. Pustell, J. and Rafatos, F.C. (1982a), Nucl. Acids Res. lQ: 4765-4782.

40. Pustell, J. and Rafatos, F.C. (1982b), Nucl. Acids Res. lQ: 51-59.

41. Ra~, N.B.R. and Pitha, P.N. tl981), Proc. Natl.
Acad. Sci. USA ~: 7426-7430.

42. Revel, M. (1984), in Becker, Y. (ed), Antivir~l Drt-g~ ~n~ Interferons: The Molecul~r b~si~ of ~heir ~ctivity, Martinus Nijhoff, Boston, pp. 357-433.

43. Revel, M., Rimchi, A., Shulman, L. Fradin, A., Shuster, R., Yakobson, E., Chernajovsky, Y., Schmidt, A., Shure, R. and Bendori, R. (1980), Ann. N.Y. Acad. Sci. ~5Q: 459-472.

44. Revel, M., Wallach, D., Merlin, G., Schattner, A., Schmidt, A., Wolf, D., Shulman, L. and ~imchi, A.
(1981), Meth. Enzymol. 79: 149-161.

45. Revel, M., ~imchi, A., Friedman, M., Wolf, D., Merlin, G., Panet, A., Rapoport, S. and Lapidot, Y. (1982) Texas Rep. Biol. Med. ~1: 452-462.

46. Rigby, P. Dieckman, M., Rhodes, C. and Berg, P.
tl977), J. Mol. Biol. 11~: 237-251.

47. Rosa, F., Berissi, ~., Weissenbach, J., Maroteaux, L., Fellous, M. and Revel, M. (1983a), EMBO J. 2:
239-243.

48. Rosa, F., Hatat, D., Adadie, A., Wallach, D., Revel, M. and Fellous, M. (1983b), EMBO J. ~, 1585-1589.

49. Salzberg, S., Wreschner, D.H., Oberman, F., Panet, A. and Bakhanaschvili, M. (1983), Mol. Cell. Biol.
~, 1759-1765.

50. Samanta, H., Dougherty, J.P. and Lengyel, P.
(1980), J. Biol. Chem. 2~: 9807-9813.

,, , -Sl- 1337281 5i. Samanta, H., Pravtcheva, D.D., Ruddle, F.H. and Lengyel, P. (1984), J. Interferon Res. 4, 295-300.

52. Saunders, M.E. and Williams, B.R.G. (1984), Anti-viral Res. ABSTR No. 3, p. 60.

53. Schamboeck, A., Rorman, A.J., Ramb, A. and Strominger, J.L. (1983), Nucl. Acids Res. 11:
8663-867 5 .

54. Schattner, A., Wallach, D., Merlin, G., Hahn, T., Levin, S. and Revel, M. (1981), Lancet iL: 497-500 .

55 . Schmidt, A., Zilberstein, A., Shulman, L., Federman, P., Berissi, H. and Revel, M. (1978), FEBS Lett. ~: 257-26 4 .

56. Shulman, L.M. Barker, P.E., Hart, J.T., Messer Peter, P.G. and Ruddle, F.R. (1984), Somatic Cell Mol. Genet. lQ: 247-257.

57 . Shulma~, L. and Revel, M. (1980), Nature ~:Z: 98-100 .

58. Stark, G., Dower, W.J., Schimke, R.T., Brown, R.E.
and Rerr, I.M. (1979) Nature 278: 471-473.

59. St. Laurent, G., Yoshie, O., Floyd-Smith, G., Samanta, H., Sehgal, P. and Lengyel, P. (1983), Cell 33, 95-102.

60-. Strachan, T., Sodoyer, R. Damotte, M. and Jordan, B.R. (1984), EMBO J. 3: 887-894.

, ~ .

61. Thomas (1980), Proc. Natl. Acad. Sci. USA 11:
5201-5205.

62. Trifonov, E.N. (1984), Codata Bulletin 56: 21-26.

63. Wallach, D. and Revel, M. (1980), Nature 2~1: 68-70.

64. Wallach, D., Fellous, M. and Revel, M. (1982), Nature ~2: 833-836.

65. Wallach, D., Schattner, A., Merlin, G., Rimchi, A., Fellous, M. and Revel, M. (1982), in Meringan, T.C., Friedman, R.M. (eds) Interferons, UCLA Sym-posia Vol. XXV, Acad. Press. New York, pp. 449-463.

66. Wallach, D., Aderka, D., Budilovsky, S. and ~ahn, T. (1983), in DeMaeyer, E., Schellekens, H. (eds) The Biology sf the Inte~feron System, Elsevier Sci. Publ., Amsterdam, pp. 293-303.

67. Weil, J., Epstein, C.J., Epstein, L.B., Sednak, J.J, Sabran, J.L., Grossberg, S.E. (1983), Nature ;~1: 437-439.

68. Weissenbach, J., Zeevi, M., Landau, T. and Revel, M. (1979), Eur. J. Biochem ~: 1-8.

69. Wells, V. and Mallucci, L. (1985), J. Cell Biol., in press-70. Wigler, M., Sweet, R., Sim, G.R., Wold, B., Pellicer, A., Lacy, E., Maniatis, T., Silverstein, S. and Axel, R. (1979), Cell l~i: 777-785.

71. Wolf, D. and Rotter, V. (1985), Proc. Natl. Acad.
Sci. USA ~: 790-794.

72. Yang, K., Samanta, H., Dougherty, J., Jayaram, B., Broeze, R. and Lengyel, P. (1981), J. Biol. Chem.
2~: 9324-9328.

73. Yarden, A., Shure-Gottlieb, R., Chebath, J., Rev-el, M. and Rimchi, A. (1984), EMBO J. 3, 969-973.

74. Zilberstein, A., Rimchi, A., Schmidt, A. and Rev-el, M. (1978), Proc. Natl. Acad. Sci. USA 75:
4734-4738.

75. Zinn, R., DiMaio, D. and Maniatis, T. (1983), Cell 34: 865-879.

'.~ ,

Claims (23)

1 A DNA molecule coding for an enzyme having (2'-5') oligo A synthetase activity and having the amino acid sequence set forth in Fig. 7A.

2. A DNA molecule coding for an enzyme having (2'-5') oligo A synthetase activity and having the amino acid sequence 1 to 346 set forth in Fig. 7A and 347 to 400 set forth in Fig. 7B.

3. A DNA molecule according to claim 1, showing the restriction map set forth in Fig. 9.

4. A DNA molecule according to claim 1, comprising the nucleotide sequence set forth in Fig. 7A.

5. A DNA molecule according to claim 2, comprising the nucleotide sequence 1 to 1071 set forth in Fig. 7A and 1072 to 1590 set forth in Fig. 7B.

6. A DNA molecule coding for a C-terminal heptadecapeptide of the enzyme encoded by the DNA molecule according to claim 1 and comprising nucleotides 1075 to 1124 set forth in Fig.
7A.

7. A DNA molecule according to any one of claims 1, 2, 3, 4, 5 or 6 which comprises the promoter having the DNA
sequence set forth in Fig. 11.

8. A DNA expression vector comprising an inserted DNA
sequence consisting essentially of the DNA sequence according to claim 4.

9. A DNA expression vector comprising an inserted DNA
sequence consisting essentially of the DNA sequence according to claim 5.

10. A microorganism transformed by an expression vector according to claims 8 or 9.

11. A microorganism according to claim 10, which is Escherichia coli.

12. An amino acid sequence corresponding to a DNA molecule coding for an enzyme having (2'-5') oligo A synthetase activity and being depicted in Fig. 7A.

13. An amino acid sequence corresponding to a DNA molecule coding for an enzyme having (2'-5') oligo A synthetase activity and being composed of amino acids 1 to 346 set forth in Fig. 7A and amino acids 347 to 400 set forth in Fig. 7B.

14. An antigenic peptide comprising the amino acid residues 348 to 364 set forth in Fig. 7A and having the amino acid sequence:

ARG-PRO-PRO-ALA-SER-SER-LEU-PRO-PHE-ILE-PRO-ALA-PRO-LEU-HIS-GLU-ALA.

15. An antigenic peptide comprising the amino acid residues 284 to 303 common to clones E16 and E18 of Figs. 7A and 7B
and having the amino acid sequence:

GLU-LYS-TYR-LEU-ARG-ARG-GLN-LEU-THR-LYS-PRO-ARG-PRO-VAL-ILE-LEU-ASP-PRO-ALA-ASP.

16. A method of monitoring interferon activity in a subject which comprises measuring the amount of (2'-5') oligo A
synthetase in a cell or body fluid of the subject at predetermined time intervals, determining the differences in the amount of said synthetase in the cell or body fluid of the subject within the different time intervals, by contacting the synthetase with an antibody being obtained by immunizing an animal with the peptide of claim 15, so as to form a complex of said (2'-5') oligo A synthestase with said antibody, and determining the amount of complex so formed and determining therefrom the amount of synthetase in the cell or body fluid of the subject and thereby the interferon activity of the subject.

17. The method of claim 16 further comprising the extraction of (2'-5') oligo A synthetase from a cell or body fluid which has been exposed to interferon, labelling the extracted synthetase with an identifiable marker to form a labelled synthetase contacting the labelled synthetase with the antibody under suitable conditions so as to form a labelled synthetase-antibody complex, and detecting the marker in the complex, thereby detecting the synthetase.

18. The method of claim 17 wherein the marker is 35 S-methionine.

CLAIMS SUPPORTED BY SUPPLEMENTARY DISCLOSURE

19. A transfer vector comprising the DNA of claim 1.

20. A transfer vector comprising the DNA of claim 2.

21. The transfer vector of claim 19 which comprises lambda-gt 11-E16 DNA encoding an enzyme having (2'-5') oligo A synthetase activity and having the nucleotide sequence set forth in Figure 7A, and the lac Z gene, the DNA being fused in phase with the lac Z gene so as to enable expression of the DNA in a suitable host cell.

22. A microorganism transformed by the transfer vector of claim 21.

23. Escherichia coli according to claim 22.