Recombinant human interleukin-1.alpha.

Abstract

The invention relates to the cloning of the human IL-1 gene, its engineering into suitable expression vectors, transformation of host organisms with such expression vectors and production of biologically active recombinant human IL-1 by culture of such transformed cells. Additionally, the invention relates to the isolation and use of the resulting recombinant human IL-1 polypeptides.

Description

BACKGROUND OF THE INVENTION

Interleukin-1 (IL-1) is a protein synthesized and secreted by activated macrophages. As part of the body's defense mechanism against infection and other forms of injury, this polypeptide hormone stimulates the proliferation and/or differentiation of a broad spectrum of cell types (including T and B lymphocytes, liver cells, bone marrow cells, connective tissue elements, skeletal muscle, brain cells, etc.). Through its actions on these diverse cell populations, IL-1 modulates immune function, fever, liver cell function (increased synthesis and secretion of acute phase reactants; increased uptake of amino acids, iron and zinc), production and release of neutrophils from the bone marrow, skeletal muscle proteolysis, changes in connective tissue, etc. IL-1 has also been described in the prior art as lymphocyte activating factor (LAF), leukocyte endogeneous mediator (LEM), endogeneous pyrogen (EP), and mononuclear cell factor (MCF). Until recently all studies of IL-1 were conducted with partially purified protein preparations, therefore, it has not been certain whether all the activities associated with IL-1 are contained within one molecule, or whether fragments of IL-1 or other macrophage proteins are responsible for certain of the functions outlined above.

Since it has been difficult to prepare sufficient amounts of human IL-1 for structural and activity studies, the biochemical nature of this molecule is poorly understood. IL-1 preparations show evidence of size and charge heterogeneity. IL-1 activity is associated with single polypeptide chains with molecular weights anywhere in the range between 12,000 and 19,000.

Recently, the gene coding for mouse IL-1 was cloned, sequenced, and expressed in Escherichia Coli. See in this regard Lomedico et al., Nature 312, pp. 458-462 (Nov. 29, 1984). In conjunction with the sequencing studies on purified "natural" mouse IL-1, it is possible to understand how this hormone is synthesized to yield the population of molecules possessing size and charge heterogeneity. When purified natural mouse IL-1 is electrophoresed on SDS-polyacrylamide gels, one finds multiple polypeptides with molecular weights between 12,000 and 19,000, all of which are biologically active. These polypeptides have different amino-terminal sequences and demonstrate charge heterogeneity on Tris-glycinate polyacrylamide gels. Sequencing of the cloned mouse IL-1 cDNA and in vitro translation experiments proved that IL-1 is initially synthesized as a precursor polypeptide of 270 amino acids. Biologically active IL-1 can be obtained from E. coli by expressing the carboxy-terminal 156 amino acids of this precursor. Hence, IL-1 activity is proteolytically released from the carboxy-terminus of the 270 amino acid precursor protein. Multiple points of protease attack will generate a population of molecules with "ragged" amino-termini, thus providing an explanation of the size and charge heterogeneity observed in purified "natural" IL-1.

The cloning of a putative gene for human IL-1 was described by Auron et al., Proc. Natl. Acad. Sci. USA, 81, 7907 (1984) which was published in February 1985. The DNA and protein sequences described therein are only partially homologous to the sequences described below.

The purification of natural human IL-1 to homogeneity has been reported by Lachman, Fed. Proc. 42, No. 3, 2639-2645 (June 1983). The method used molecular weight fractionation, isoelectric focusing and preparative polyacrylamide gel electrophoresis. Due to the use of sodium dodecyl sulfate in the last step, the product was denatured and exhibited only a trace of its original biological activity. See also Schmidt. J. Exp. Med. 160, 772-787 (September 1984) for a purification scheme using HPLC methods to produce a single charged species of human IL-1 and Kronheim et al., J. Exp. Med. 161, 490-502 (March 1985).

It has also been known in the art to produce antibodies directed against murine IL-1. See Mizel et al. J. Immun. 131, 1834 (1983). These antibodies, which were raised in goat, were utilized to develop an assay for IL-1 and also in the production of an anti-IL-1 immunoabsorbent column which in turn is useful for further purification of either natural or recombinant murine IL-1. The anti murine IL-1 antibody crossreacts poorly with human IL-1.

SUMMARY OF THE INVENTION

The present invention relates to the cloning of the human IL-1 gene, its engineering into suitable expression vectors, transformation of host organisms with such expression vectors and production of biologically active recombinant human IL-1 by culture of such transformed cells. Additionally, the present invention relates to the isolation and use of the resulting recombinant human IL-1 polypeptide.

Thus, the present invention utilizes recombinant DNA technology as the means to discover the DNA sequence and the deduced amino acid sequence for human interleukin-1 and to its production and to its use.

More particularly, the present invention relates to the isolation and identification of DNA sequences coding for biologically active forms of human interleukin-1. This was accomplished by employing a mouse IL-1 cDNA clone to isolate a partial human IL-1 genomic clone. This genomic clone was used in turn to isolate a human IL-1 cDNA clone. The sequence of this cDNA revealed the structure of the human IL-1 precursor protein. Expression of the carboxy-terminal 154 amino acids of this precursor in E. coli resulted in the production of biologically active IL-1 protein.

Thus, more particularly, the present invention relates to the isolation and identification of DNA sequences encoding the human IL-1 precursor and biologically active molecules contained therein, and to the construction of recombinant DNA expression vehicles containing such DNA sequences operatively linked to expression--effecting promoter sequences and to the expression vehicles so constructed. In another aspect, the present invention relates to host culture systems, such as various microorganism and vertebrate cell cultures transformed with such expression vehicles and thus directed in the expression of the DNA sequences referred to above. In other aspects, this invention relates to the means and methods of converting the end products of such expression to novel entities, such as pharmaceutical compositions, useful for the prophylactic or therapeutic treatment of humans or in diagnostic assay systems. In preferred embodiments, this invention provides particular expression vehicles that are constructed properly such that human interleukin-1 is produced in the host cell in mature form.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be more readily understood when considered in connection with the accompanying drawings wherein:

FIG. 1 shows the nucleotide sequence and predicted amino acid sequence of mouse IL-1 precursor cDNA.

FIG. 2A shows the nucleotide sequence and predicted amino acid sequence of the carboxy-terminal region of human IL-1 precursor (from phil #7) with the partial sequence of phil #4 indicated by underlining, while 2B shows the nucleotide sequence and predicted amino acid sequence of the human IL-1 precursor as deduced from a composite of clones phil #7 and phil #19.

FIG. 3 illustrates the sequence homology of the mouse and human IL-1 precursor proteins.

FIG. 4 is a flow chart showing the construction of an expression vector which directs the synthesis of a modified 154 amino acid carboxy terminal sequence of human IL-1 (phil #1-154*) using pEV-vrf 2 as a vector.

FIG. 5 is a flow chart showing the construction of expression vector (phil #1-154) which directs the synthesis of the 154 amino acid carboxy terminal sequence of human IL-1 without the extraneous amino acids at the amino terminus of phil #1-154*.

BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENT

As indicated above the cloned gene coding for a human IL-1 polypeptide may be obtained by use of a mouse IL-1 cDNA clone as a hybridization probe. In such a procedure, an EcoRl partial human genomic phage library (Fritsch et al. cell 19, 959-972 (1980)) was screened using as the hybridization probe the plasmid pIL-1 1301 (Lomedico et al. supra). Phage plaques were transferred to nitrocellulose filters by standard methods (Maniatis et al. Molecular Cloning A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982) and the filters containing the imobilized DNA were hybridized (10.sup.6 cpm/10 ml/138 mm filter) in 5.times.SSPE (1.times.SSPE=0.18M NaCl, 10 mM Na phosphate pH 7.0, 1 mM Na.sub.3 EDTA), 5.times. Denhardt's (0.1% Ficoll 400, 0.1% bovine serum albumin, 0.1% polyvinylpyrrolidine (w/v)), 0.3% SDS, 20% formamide, 250 microgram/ml of calf thymus DNA, and .sup.32 P-pIL-1 1301 (labelled by nick translation to 1-2.times. 10.sup.8 cpm/microgram) for 48 hours at 37.degree. C. The filters were washed at 30.degree. C. in 0.5.times.SSPE and autoradiographed. After screening 7.9.times.10.sup.5 plaques, 2 positive recombinant phage were identified. These two phage were shown to be identical by restriction endonuclease mapping and named .lambda.-hil 4. Using the hybridization conditions described above, the mouse IL-1 cDNA clone pIL-1 1301 was subsequently shown to specifically hybridize to a 1.4 kb EcoR1-Hind III fragment from the recombinant phage .lambda.-hil 4. This 1.4 kb fragment was subcloned into pBR322 to yield phil #4. The nucleotide sequence proximal to the EcoR1 site of phil #4 was determined (see FIG. 2A) and compared to the sequence of mouse IL-1 mRNA and protein. This analysis showed 75% nucleic acid homology and 66% amino acid sequence homology with the carboxy-terminal 61 amino acids of the mouse IL-1 precursor and its accompanying 3' non-coding region.

The partial human IL-1 gene obtained as above was then employed as a probe to isolate a human IL-1 cDNA clone derived from mRNA obtained from induced normal human peripheral blood leukocytes. The human leukocyte concentrates, collected and prepared from normal donors were obtained from the American Red Cross, Lansing, Mich. The contents from 50-100 leukocyte concentrates were aseptically removed from the collection bags and pooled. The leukocyte pool was mixed with a half volume of a 6% solution of hetastarch (Hespan, American Hospital Supply Corp., Irvine, Calif.) in a separatory funnel and allowed to stand for 3-3.5 hours at room temperature. This procedure differentially sedimented the contaminating red blood cells from the leukocytes. The volume of the leukocyte--Hespan mixture should not exceed two-thirds of the volume of the separatory funnel to insure proper sedimentation. The red blood cells sedimented to the bottom of the separatory funnel and cell separation was complete when a sharp interface band of white cells was apparent just above the red cell layer. The "low density" white blood cells were used for the production of interleukin-1. These cells were removed from the separatory funnel by carefully aspirating only the uppermost layer of cells located above the interface white cells. The low density leukocytes were removed from the Hespan by sedimentation in 250 ml conical centrifuge tubes (Corning Glassworks, Corning, N.Y.) at 500.times.g for 20 minutes. The pellet was resuspended in 9 volumes of a 0.83% solution of ammonium chloride for 5 minutes to lyse the remaining red blood cells. The leukocytes were removed from the ammonium chloride solution by sedimentation at 500.times.g for 10 minutes as described above.

The cell pellet was then resuspended to a concentration not exceeding 3.5.times.10.sup.7 cells/ml in RPMI-1640 medium (GIBCO) prewarmed to 37.degree. C. No fetal calf serum was added to the medium to avoid clumping of the concentrated cells. The freshly isolated leukocytes, free of red blood cell contamination, were diluted in the induction vessel with the appropriate volume of RPMI-1640 medium supplemented with 2% fetal calf serum to make a final concentration of 3.times.10.sup.6 cells/ml. The cells were incubated at 37.degree. C. for one hour and induced for production of IL-1 by the addition of 10 .mu.g/ml E.coli lipopolysaccharide B (LPS, Difco 3880-25). Incubation was continued for 12 hours for mRNA extraction. The cells were then removed from the induction medium by sedimentation at 500.times.g. Cells induced for mRNA extraction may be stored as a frozen pellet at minus 20.degree. C. or lower.

A particularly preferred procedure for isolating poly(A).sup.+ RNA from human peripheral blood leukocytes induced with LPS is the guanidine thiocyanate-CsCl method of Chirgwin et al. (Biochemistry 18, 5294-5299, 1979) and oligo (dT)-cellulose chromatography (Aviv and Leder, Proc. Natl. Acad. Sci. 69, 1408-1412, 1972). Northern blot analysis of this RNA demonstrated that clone phil #4 specifically hybridized to a mRNA approximately 2,100 nucleotides long. Total poly (A) RNA was fractionated by sucrose density gradient centrifugation and mRNA with sizes between 1,000 and 3,000 nucleotides was collected. Using this enriched mRNA pool, a cDNA library of approximately 20,000 clones was constructed in pBR 322 by established procedures (see Gubler and Hoffman, Gene 25, 263-269, 1983). This library was screened using as a hybridization probe the 1.4 kb EcoR1-Hind III insert from plasmid phil #4. Bacterial colonies were transferred to nitrocellulose filters by standard methods (Maniatis et al. supra) and the filters containing the immobilized DNA were hybridized in 5.times.SSPE, 5.times. Denhardts, 0.3% SDS, 50% formamide (w/v), 250 micro g/ml calf thymus DNA and P-phil #4 insert DNA for 16 hours at 37.degree. C. The filters were washed at 50.degree. C. in 0.1.times.SSPE and autoradiographed to identify a single positive clone termed phil #7. This clone contains an insert of 2,200 bp. The nucleotide sequence of this insert (see FIG. 2A) contains the partial sequence of the genomic clone phil #4 and hence corresponds to the cDNA sequence of mRNA encoded by this gene. The nucleotide sequence of this phil #7 insert predicts an open reading frame for a protein of 163 amino acids. This predicted protein is 55% homologous (see FIG. 3) to the carboxy-terminal 160 amino acids of the mouse IL-1 precursor. Hence the sequence provided by phil #7 represents the carboxy-terminal region of the human IL-1 precursor. Primer extension experiments demonstrate that approximately 400 nucleotides from the 5' end of the human IL-1 mRNA are missing from clone phil #7.

Since human IL-1 cDNA clone #7 turned out to be an incomplete copy of human IL-1 mRNA, the following strategy to complete the sequence was adapted: Based on the DNA sequence of clone #7, a synthetic oligonucleotide with the sequence 5'-GGGCGTCATTCAGGATGAATTCGTA-3' was devised and synthesized using solid support phosphoramidite technology. This oligonucleotide is complementary to the sequence coding for amino acids 20-28 as predicted from phil #7. The oligonucleotide spans a region in the cDNA predicted to contain an EcoRI restriction site. Such a site is useful for fusing an extension clone with clone #7 to create a cDNA encompassing the complete protein coding region for the human IL-1 precursor. This oligonucleotide was annealed to size-fractionated poly A.sup.+ RNA from LPS--induced human peripheral blood leukocytes (supra). Annealing conditions were 50 mM NaCl, 10 mM DTT, 0.05 mM EDTA, 550 pmoles oligonucleotide/ml, 250 mcg of poly A.sup.+ RNA/ml for one minute at 90.degree. C., 10' at 43.degree. C., 10' at 20.degree. C. after which the reaction was cooled on ice. cDNA--synthesis and establishment of an extension cDNA--library were performed as described above for clone #7. About 10.sup.5 independent transformants were generated in this way from 5mcg of poly A.sup.+ RNA enriched for human IL-1 mRNA. A total of approximately 2900 were screened with the oligonucleotide described above that had been labelled with polynucleotide kinase and .gamma.-.sup.32 P-ATP according to standard procedures (Maniatis et al., supra). Colony bearing nitrocellulose filters were made according to standard procedures (Maniatis et al., supra). The filters were hybridized with the labelled oligonucleotide under the following conditions: 5.times.SSPE, 10.times. Denhardts, 0.1% SDS, 100 .mu.g/ml yeast soluble RNA; 0.2 pmoles/ml of labelled oligonucleotide (specific activity=.about.1 .mu.Ci/pmole) for 15' at 65.degree. C. and subsequently 2 hours at 37.degree. C. The filters were then washed twice in 2.times.SSPE-0.025% SDS (quick rinses at room temperature) and then in 4.times.SSPE-0.025% SDS at 51.degree. C. for 60'. The filters were then air dried, exposed to x-ray film with the aid of an intensifying screen at -70.degree. C. for 16 hours. Twelve positive colonies were further analyzed by restriction endonuclease cleavage. Clone phil #19 was chosen for further analysis. The sequence of the insert (see FIG. 2B) from phil #19 contains the expected overlap with phil #7 and predicts a single open reading frame coding for 139 amino acids. This region represents the amino-terminal end of the human IL-1 precursor protein, and is highly homologous to the corresponding region of the mouse IL-1 precursor protein (FIG. 3). Hence, combining the sequence information from phil #7 and phil #19. human IL-1 mRNA codes for a protein of 271 amino acids which is significantly related to the 270 amino acid mouse protein (FIG. 3).

Plasmid phil #7 contains the coding information for the carboxy-terminal 163 amino acids of the 271 amino-acid human IL-1 precursor protein. Plasmid phil #19 contains the coding information for the amino-terminal 139 amino acids of this protein. Each plasmid possesses a single EcoR1 restriction endonuclease cleavage site within the sequence (94 nucleotides long) that is common to their inserts. This EcoR1 site can be used to join the information from the two plasmids into a single composite plasmid containing the entire coding region from the human IL-1 precursor protein. Using standard methods, plasmids phil #7 and phil #19 can be individually digested with EcoR1 and Bam HI, and the resultant DNA fragments separated by polyacrylamide gel electrophoresis. The .about.2100 bp EcoR1-Bam HI fragment can be isolated from the phil #7 digest, and the .about.460 bp Bam HI--EcoR1 fragment can be isolated from the phil #19 digest. The two isolated fragments can be ligated together using T4 DNA ligase. The ligase is heat-inactivated and the mixture is treated with Bam HI. This mixture can be ligated to Bam HI-linearized pEV-vrf2 (below) and used to transform E.coli strain MC1061 containing the compatible plasmid pRK248 (cIts) using selection for ampicillin resistance. Bacterial clones can be screened by restriction endonuclease cleavage analysis to identify a plasmid, phil #1-271*, which contains the expected insert in the correct orientation. Plasmid phil #1-271* is modified by site-directed oligonucleotide mutagenesis (see below) to remove the extraneous nucleotides between the initiation ATG codon and the alanine codon (GCC) which is the second amino acid in the 271 amino acid precursor protein, to generate phil #1-271. Bacteria containing phil #1-271, when induced by temperature shift (supra), synthesize the complete 271 amino acid IL-1 precursor protein as set forth in FIG. 2B.

Plasmid pEV-vrf2 is a pBR322 derivative modified using synthetic DNA oligonucleotides to contain a ribosome binding site--initiation codon downstream from a tightly regulated phage .lambda. P.sub.L promoter. Multiple-use restriction endonuclease cleavage sites exist immediately downstream from the initiation codon, allowing for the insertion of coding region sequences to be expressed as fusion proteins with 2-9 extra amino-terminal amino acids. These extraneous amino acids can be removed by site directed mutagenesis, resulting in the expression of the desired protein. The genealogy of pEV-vrf2 is as follows: pBR322.fwdarw.pRC2.fwdarw.pRC23.fwdarw.pEV-vrf2.

pRC2 is a derivative of pBR322 containing a unique BglII site adjacent (on the amp.sup.R side) to the EcoRI site in the plasmid. This plasmid was constructed using known methods in the following manner. 20 .mu.g of pBR322 plasmid DNA was digested with EcoRI and then split into two reactions. In one, the protruding 5' single-stranded termini were removed with S1 nuclease; in the other reaction, the termini were filled-in by incorporating deoxynucleotides with the Klenow fragment of DNA polymerase I. Both reactions were terminated by phenol extraction followed by ethanol precipitation. Approximately 1 .mu.g of DNA from each reaction was mixed with 90 pmoles of phosphorylated BglII linkers (CAGATCTG, purchased from Collaborative Research) and incubated with T4 DNA ligase at 15.degree. for 18 hours. The ligation products were then digested with BglII and PstI and subjected to gel electrophoresis in 1% agarose. The 3600 bp and 760 bp fragments from both reactions were recovered from the gel. For the construction of pRC2, the 3600 bp from the Klenow reaction was ligated to the 760 bp fragment from the S1 reaction. E.coli strain RR1 was transformed with the ligation mixtures, and transformants were selected on LB agar plates containing 50 .mu.g/ml ampicillin. Transformants containing the expected plasmid constructions were identified by restriction analysis of the isolated plasmid DNA. DNA sequence analysis confirmed that the S1 nuclease treatment precisely removed the 5' single-stranded termini.

pRC23 was constructed by inserting into pRC2 a 250 bp BglII-HaeIII fragment containing the .lambda. P.sub.L promoter joined to a pair of complementary synthetic oligonucleotides comprising a model ribosome-binding site (RBS). The HaeIII site is located within the 5' non-coding region of the .lambda.N gene 115 bp downstream of the P.sub.L transcription initiation site (Sanger, et al., 1982). Approximately lg of a 450 bp BglII-HpaI fragment isolated from phage .lambda. DNA was digested with HaeIII. 200 ng of the resulting digestion products were mixed with 60 pmoles each of phosphorylated synthetic oligonucleotides containing the model RBS. These complementary deoxynucleotides (#1=TTAAAAATTAAGGAGG; #2=AATTCCTCCTTAATTTTTAA) were synthesized on solid support using the phosphite methodology (Matteucci, M. D. and Caruthers, M. H. "Synthesis of Deoxyoligonucleotides on a Polymer Support" J. Am. Chem. Soc. (1983) 103:3185-3191). Synthesis was initiated with 1 .mu.mole of the 3'-terminal nucleoside attached to a controlled pore glass support (Pierce CPG/long chain alkylamine resin). This mixture was incubated with T4 DNA ligase at 15.degree. C. for 18 hours, and the ligated molecules were digested with BglII and EcoR1 and separated on a 5% polyacrylamide gel. The 270 bp ligation product was recovered from the gel, mixed with gel purified pRC2 vector that had been digested with BglII and EcoRI, and incubated with T4 DNA ligase at 15.degree. for 15 hours. The ligation mixture was used to transform strain RR1 (pRK248cIts). Transformants selected on ampicillin-containing medium were screened by restriction analysis of the isolated plasmid DNA. The expected plasmid construction, pRC23, was confirmed by further restriction enzyme digestions and by DNA sequence analysis across the EcoRI junction. Plasmid pRC23 contains a unique EcoRI site at the 3' end of the RBS, into which genes containing an ATG at the 5' end can be inserted.

For the construction of pEV-vrf2, pRC23was digested with EcoRI and Hind III and the linearized vector isolated by preparative agarose gel electrophoresis. Two complementary deoxynucleotides (#3=AATTAATATGAATAGAATTCGGATCCATCGATA, #4=AGCTTATCGATGGATCCGAATTCTATTCATATT) were synthesized (supra) combined and heated to 58.degree. C. for 5 minutes in 150 mM NaCl, and cooled slowly to allow annealing. 0.1 pmoles of the synthetic duplexes were added to 0.07 pmoles of the pRC23/EcoRI-HindIII vector and incubated with T4 DNA ligase at 15.degree. C. for 15 hours. Strain RR1 (pRK248cIts) was transformed with the ligation products, and ampicillin-resistant transformants were screened by restriction endonuclease cleavage analysis to identify pEV-vrf2, the expected construction of which was confirmed by DNA sequence analysis. Plasmid pEV-vrf2 contains restriction sites (EcoRI, BamHI, ClaI, and HindIII) located downstream from an appropriately positioned initiation codon--RBS. Hence, appropriately positioned coding region sequences inserted into these restriction sites will be expressed under control of the P.sub.L promoter yielding the corresponding protein with 2-9 extra amino-terminal amino acids. Site directed mutagenesis can be used to remove these extraneous amino acids, as well as to re-orient inappropriately positioned (i.e. the reading frame is not correct) coding sequences.

Synthetic oligonucleotides 5'-GGGCGTTATTCAGGACGAATTCGTA-3' and 5'-ATTGCTCA GG AACATATTAATCC-3' used above were synthesized on an Applied Biosystems model 380A DNA synthesizer. Diisopropyl phosphoramidites were used as the active nucleotidyl component for coupling to the support. See Beacage and Caruthers, Tetrahedron Lett. 22, 1859-1862 (1981). Synthesis on the support was carried out at 0.5 mmol level using a CPG resin with the 3' nucleotide attached to the resin as described by Matteucci and Caruthers, J. Am. Chem. Soc. 103, 3185-3189 (1981). The synthetic cycle was essentially the same as provided by the manufacturer, however, 2% dichloroacetic acid was used in place of 3% trichloroacetic acid as the detritylating reagent. Synthesized oligonucleotides were isolated on 20% polyacrylamide sequencing gels. Isolated oligonucleotides cut out from the gel were eluted in gel elution buffer and desalted on C.sub.18 reverse-phase columns.

With the recombinant DNA thus obtained, living cells may be transformed to amplify the cloned cDNA or to produce IL-1 polypeptide.

Suitable eucaryotic host organisms, which may be employed for production of IL-1, include vertebra, yeast and the like. For instance, monkey cells, e.g. CV-1 cells, transformed by a replication origin defective mutant of SV-40 and expressing the SV-40 large T antigen (COS cells) as discussed by Gluzman (Cell 23, 175-182, 1981), mouse derived cells described by Ohno and Taniguchi (Nucleic Acids Research 10, 967-977 (1982)), and yeast host--vector systems which have been utilized for the expression of interferon genes, discussed by Hitzman et al. (Nature, 293, 717-722 (1981)) may be utilized. In addition, it is possible to use insect cells such as described by Smith et al. (Mol. Cell. Biol. 3, 2156-2165, 1983). Suitable procaryotic host organisms include Escherichia coli, Bacillus subtilis and the like. For amplification of DNA in host organisms, it may be preferred to use E. coli as a host, however other hosts can also be employed.

Suitable vectors used for E. coli include EK type plasmid vectors (stringent type): pSC101, pRK353, pRK646, pRK248, pDF41 etc., EK type plasmid vectors (relaxed type): ColEI, pVH51, pAC105, RSF2124, pCR1, pMB9, pBR313, pBR322, pBR324, pBR325, pBR327, pBR328, pKY2289, pKY2700, pKN80, pKC7, pKB158, pMK2004, pACYCl, pACYC184, dul etc. .lambda. gt type phage vectors: .lambda. gt. .lambda.c, .lambda. gt, .lambda.B, .lambda. WES, .lambda.c, .lambda. gt. .lambda.B, .lambda.WES, .lambda.C, .lambda.WES, .lambda.B, .lambda.ZJvir., .lambda.B', .lambda.ALO, .lambda.B, .lambda.WES. Ts622, .lambda.Dam etc. In general pBR322 has been frequently used as a vector for E. coli.

Transformation of the host cell with the recombinant DNA may be carried out by conventionally used methods as follows:

Where the host is prokaryotic such as E. coli, competent cells which are capable of DNA uptake are prepared from cells harvested during the exponential growth phase and subsequently treated by the CaCl.sub.2 -method by well known procedures. When MgCl.sub.2 or RbCl exists in the transformation reaction medium, the transformation efficiency increases. Transformation can be also performed after forming a protoplast of the host cell.

Where the host used is eucaryotic, transfection methods of DNA as calcium phosphate-precipitates, conventional mechanical procedures such as microinjection, insertion of a plasmid encapsulated in red blood cell hosts or in liposomes, treatment of cells with agents such as lysophosphatidylcholine, or use of virus vectors, or the like may be used.

However, various other microbial strains are useful, including known E. coli strains such as E. coli B. E. coli X 1776 (ATCC No. 31537) and E. coli W 3310 (ATCC No. 27325), and most preferably E. coli RR1' or other microbial strains such as MC 1061, many of which are deposited and available from depository institutions, such as the American Type Culture Collection (ATCC)-cf. the ATCC catalogue listing. See also German Offenlegungsschrift 2644432. These other microorganisms include, for example, Bacilli such as Bacillus subtilis and enterobacteriaceae among which can be mentioned as examples Salmonella typhimurium and Serratia marescens, utilizing plasmids that can replicate and express heterologous gene sequences therein.

As examples, the beta lactamase and lactose promoter systems have been advantageously used to initiate and sustain microbial production of heterologous polypeptides. Details relating to the make-up and construction of these promoter systems have been published by Chang et al., Nature 275, 617 (1978) and Itakura et al., Science 198, 1056 (1977), which are hereby incorporated by reference. More recently, a system based upon tryptophan, the so-called trp promoter system, has been developed. Details relating to the make-up and construction of this system have been published by Goeddel et al., Nucleic Acids Research 8, 4057 (1980). Numerous other microbial promoters have been discovered and utilized and details concerning their nucleotide sequences, enabling a skilled worker to ligate them functionally within plasmid vectors, have been published-see, e.g., Siebenlist et al., Cell 20, 269 (1980).

The expression system hereof may also employ the plasmid YRp7, which is capable of selection and replication in both E. coli and the yeast, Saccharomyces cerevisiae. A useful strain is RH218 deposited at the American Type Culture Collection without restriction (ATCC No. 44076). However, it will be understood that any Saccharomyces cerevisiae strain containing a mutation which makes the cell trp1 should be an effective environment for expression of the plasmid containing the expression system. An example of another strain which could be used is pep4-1. This tryptophan auxotroph strain also has a point mutation in the TRP1 gene.

The experience with expression of mouse IL-1 cDNA in E. coli suggests that the carboxy-terminal portion of the human IL-1 precursor should possess IL-1 biological activity. Clone phil #7 described above, contains the coding information for the carboxy-terminal 163 amino acids of the human IL-1 precursor. There is an Alu I restriction endonuclease cleavage site near the 5' end of the phil #7 insert (see FIG. 2A) within the codon for the ninth amino acid (that is, the 153 amino acid from the carboxy-terminal end of the precursor). The next downstream Alu I site is in the 3' non-coding region (that is, past the termination codon) .about.600 bp away. This .about.600 bp Alu I fragment containing the sequences coding for the carboxy-terminal 154 amino acids of the human IL-1 precursor, was isolated from phil #7 and inserted into the BamHI site of an E. coli expression plasmid (as seen in FIG. 4) in the following manner. Using standard methods, the insert from clone phil #7 was digested with Alu I, and phosphorylated BamHI linkers (CGGATCCG New England Biolabs, Catalog 1021) were ligated to the Alu I cut insert using T4DNA ligase. The ligase was heat-inactivated and the mixture was treated with BamHI to remove excess linkers and to generate cohesive termini. This mixture was electrophoresed on a polyacrylamide gel and the .about.600 bp fragment was isolated. Plasmid pEV-vrf2 was digested with BamHI and the linearized vector was recovered following agarose gel electrophoresis. The .about.600 bp fragment and the BamHI cut vector were ligated together, and used to transform E. coli strain MC 1061 (Casadaban and Cohen J. Mol. Biol. 138, 179-207, 1980) containing the compatible plasmid pRK248cIts (Bernard and Helinski, Meth. Enzym. 68, 482-492, 1979) using selection for ampicillin resistance. Bacterial clones were screened by restriction endonuclease digestion analysis to identify a plasmid phil #1-154* containing the insert in the correct orientation. Plasmid phil #1-154* was partially sequenced to verify that its structure was correct. Bacteria containing phil #1-154* or the parental plasmid pEV-vrf2 were grown in M9 media containing ampicillin at 30.degree. C. until the A.sub.550 reached 0.7, at which time the cultures were shifted to 42.degree. C. for 3 hours. The bacteria from 1 ml of culture were recovered by centrifugation and solubilized in 50 micro liters of 7M guanidine hydrochloride. These crude bacterial extracts were examined for IL-1 activity in the murine thymocyte proliferation assay (Mizel et al. supra). Extracts of bacteria containing only pEV-vrf2 did not stimulate in the assay above background levels. Extracts of bacteria 10 containing phil #1-154* contained 32,000 units of IL-1 activity/ml of guanidine HCl solution. Assuming a specific activity of 6.times.10.sup.6 units per mg, this is equivalent to at least 0.3 mg of IL-1 protein per liter of bacterial culture.

The protein encoded by expression plasmid phil #1-154* contains, in addition to the initiator methionine, 6 extraneous amino acids at its amino terminus. These were removed by site directed mutagenisis as follows:

1-2 .mu.g of plasmid phil#1-154* was subjected to restriction endonuclease digestion in two separate reactions. In one reaction, 1-2 units of both Bgl II and Bam HI created a linearized plasmid with a gap (see FIG. 5), and in the other reaction, 1 unit of Pst I generated a linearized plasmid. Pst I treatment was followed by treatment with 1 unit of Klenow fragment of E.Coli DNA polymerase I. These opened plasmids were purified by electrophoresis through a 0.7% aparose gel and recovered by ethanol precipitation. Each plasmid was resuspended in 5 .mu.l H.sub.2 O. A 1 .mu.l aliquot was taken from each and was combined with 50 ng of phosphorylated synthetic oligonucleotide: 5'-P-ATTGCTCAGGAACATATTATTCC-OH-3' in a 12 .mu.l reaction containing 12 mM Tris.HCl, pH 7.5, 9 mM MgCl.sub.2. 200 mM NaCl and 20 .mu.l beta mercaptoethanol. The reaction was heated to 100.degree. C. for 3' to denature the opened plasmids, and the annealing of the oligonucleotide was permitted by gradually cooling the reaction at 23.degree. C. for 30', followed by 4.degree. C. for 30' and 0.degree. C. for 10'. This resulted in the formation of a heteroduplex form of phil #1-154* with the oligonucleotide annealed to the single-stranded region.

The single-stranded region was made double-stranded and the plasmid was ligated in a 20 .mu.l reaction volume of 75 .mu.M dATP, 75 .mu.M dTTP, 75 .mu.M dCTP, 75 .mu.M dGTP, 500 .mu.M ATP, 2-3 units of the Klenow fragment of E.Coli DNA polymerase I and 1 unit of T.sub.4 DNA ligase. This reaction proceeded at 15.degree. C. for 12-16 hours. Plasmid DNA was recovered by ethanol precipitation and resuspended in 10 .mu.l H.sub.2 O. A 5 .mu.l aliquot was used to transform the MC1061 strain of E.Coli containing the compatible plasmid pRk248cIts to ampicillin resistance. Ampicillin resistant transformants were screened for the new plasmid phil #1-154 in the phil #1-154* background by recovering plasmid DNA from individual transformants and performing Bgl II/Bam HI restriction digestion on this plasmid DNA preparation. Plasmid DNA which contained phil #1-154was used in a second round of MC 1061 pRK248cIts transformation to separate phil #1-154and phil #1-154*. Transformants containing only phil #1-154were recovered and an individual E. Coli colony was used for the production of the phil #1-154protein as described above.

Purification of Recombinant Human Interleukin-1

Like many other recombinant proteins (see Williams, D. C. et al. Science (1982) 215:687-688: Lacal, J. C. et al. Proc. Natl. Acad, Sci. (1984) 81:5305-5309), human interleukin-1 aggregates into insoluble cytoplasmic "inclusion bodies" within E.coli. Hence the purification of recombinant human IL-1 begins with the isolation of these "inclusion bodies" (see Lacal et al., supra).

E. coli cell paste (1 g) was suspended in 5 ml of 1 mM phenylmethylsulfonyl fluoride in buffer A (30 mM Tris-HCl, pH 8, in 5 mM EDTA) and the cells were sonicated six times for a total of 3 minutes using a Sonifier cell disrupter Model 350 (Branson Sonic Power Co.). The cell lysate was centrifuged for 30 minutes at 30,000.times.g to separate the insoluble fraction. The particulate fraction (which contains most of IL-1 activity) was sequentially washed with 5 ml each of 1) buffer A, 2) 1% Triton X-100 in buffer A and 3) 1.75M guanidine HCl. After each wash, the particulate fraction was pelleted by centrifugation at 30,000.times.g for 20 minutes. IL-1 activity was solubilized from the remaining particulate fraction by 3 ml of 5M guanidine HCl, followed by centriguation at 30,000.times.g for 30 minutes. Up to this step, all procedures were carried out at 4.degree. C. The solubilized IL-1 protein was purified to homogeneity by gel filtration chromatography on Sephacryl S-200 or Sephadex G-75 (Pharmacia Fine Chemicals, Piscataway, N.J.) equilibrated and eluted with 5M guanidine HCl. The purified 1-154* IL-1 behaved as a single polypeptide on SDS polyacrylamide gels (Laemmli, U.K. Nature (1970) 227:680-685). When submitted for amino acid compositional analysis and amino-terminal sequence analysis, the expected results were obtained, thus verifying the purity and identity of the protein. In similar fashion E.coli transformed with phil #1-154were used to produce 1-154 protein which was purified as above.

Experiments with the expression products of deletion mutants of the mouse IL-1 gene has provided a basis for determining the minimum sequence which provides a bioactive protein from the carboxy-terminal of the mouse IL-1 precursor. The results are summarized below in Table I.

TABLE I______________________________________Activity of Mouse IL-1Deletion Mutants Protein.sup.1 Activity.sup.2______________________________________ 1-156 6 .times. 10.sup.6 17-156 6 .times. 10.sup.6 30-156 0 1-143, 156 0 17-143, 156 0 30-143, 156 0______________________________________ 1: in this nomenclature, protein 1-156 contains the carboxyterminal 156 amino acids of the mouse IL1 precursor. All the deletion mutants are defined relative to this molecule, hence protein 17-156 is missing the aminoterminal 16 amino acids compared to protein 1-156; protein 1-143,156 is missing amino acids 144-155 compared to protein 1-156. 2: in the thymocyte proliferation assay, units per mg protein.

As seen in Table I the sequence proximal to the carboxy-terminus is needed for activity. A minimum of about 139 amino acids is apparently required to maintain activity as the 17-156 exhibits high activity, whereas deletion of an additional thirteen amino acids at the amino-terminus of this fragment destroys the activity. Thus, by analogy to the mouse molecule data it is believed that the sequence encompassing the carboxy-terminal 139 amino acids of the human IL-1 precursor is the minimum fragment exhibiting IL-1 activity. This would correspond to positions 132-271 of the human IL-1 precursor protein sequence set forth in FIG. 2B. Therefore, one aspect of the present invention relates to peptides exhibiting human IL-1 activity which encompass the aforesaid minimum carboxy-terminal sequence.

The purified recombinant human IL-1 peptides encompassing the sequence needed for biological activity can be employed in a manner known per se to stimulate the immune system of a host subject, such as, for example, by improving host defense response to pathogens, by acting as a vaccine adjuvant and by enhancing host defense against neoplastic diseases. Other clinical uses identified for human IL-1 in the art include promotion of wound healing via stimulation of fibroblast proliferation and improvement of the recovery is of critically ill, protein-malnourished patients.

Purified IL-1 peptides prepared in accordance with this invention may be administered to warm blooded mammals for the clinical uses indicated above. The administration may be by any conventional method such as by parenteral application either intravenously, subcutaneously or intramuscularly. Obviously, the required dosage will vary with the particular condition being treated, the severity of the condition, the duration of the treatment and the method for administration. A suitable dosage form for pharmaceutical use may be obtained from sterile filtered, lyophilized IL-1 peptide reconstituted prior to use in a conventional manner. It is also within the skill of the art to introduce buffers, stabilizers, bacteriostats and other excipients and additives conventionally employed in pharmaceutical parenteral dosage forms.

Claims

1. A human interleukin-1.alpha. a polypeptide free of other human polypeptides and comprising the minimum carboxy-terminus sequence needed for biological activity said minimum sequence corresponding to positions 132-271 of human interleukin-1.alpha. precursor protein.

2. A microbially produced interleukin-1.alpha. polypeptide having interleukin-1 activity and having a total of 155 amino acid residues and the amino acid sequence:

Met Phe Leu Ser Asn Val Lys Tyr Asn Phe MET Arg Ile Ile LysTyr Glu Phe Ile Leu Asn Asp Ala Leu Asn Gln Ser Ile Ile ArgAla Asn Asp Gln Tyr Leu Thr Ala Ala Ala Leu His Asn Leu AspGlu Ala Val Lys Phe Asp MET Glu Ala Tyr Lys Ser Ser Lys AspAsp Ala Lys Ile Thr Val Ile Leu Arg Ile Ser Lys Thr Gln LeuTyr Val Thr Ala Gln Asp Glu Asp Gln Pro Val Leu Leu Lys GluMET Pro Glu Ile Pro Lys Thr Ile Thr Gly Ser Glu Thr Asn LeuLeu Phe Phe Trp Glu Thr His Gly Thr Lys Asn Tur Phe Thr SerVal Ala His Pro Asn Leu Phe Ile Ala Thr Lys Gln Asp Tyr TrpVal Cys leu Ala Gly Gly Pro Pro Ser Ile Thr Asp Phe Gln IleLeu Glu Asn Gln Ala.

3. The polypeptide of claim 2, which is homogeneous.

4. A pharmaceutical preparation suitable for parenteral administration, said preparation comprising a minor, effective amount of an interleukin-1.alpha. polypeptide of claim 2 in homogeneous form and a major amount of a pharmaceutical parenteral carrier material.

5. A microbially produced interleukin-1.alpha. polypeptide having interleukin-1 activity and comprising the minimum sequence needed for interleukin-1 activity, said minimum sequence corresponding to positions 132 to 271 of a human interleukin-1.alpha. precursor polypeptide.

6. The microbially produced polypeptide of claim 5 wherein said minimum sequence corresponding to positions 132 to 271 of the human interleukin-1.alpha. precursor polypeptide is:

132Tyr Glu Phe Ile Leu Asn Asp Ala Leu Asn Gln Ser Ile Ile ArgAla Asn Asp Gln Tyr Leu Thr Ala Ala Ala Leu His Asn Leu AspGlu Ala Val Lys Phe Asp MET Gly Ala Tyr Lys Ser Ser Lys AspAsp Ala Lys Ile Thr Val Ile Leu Arg Ile Ser Lys Thr Gln LeuTyr Val Thr Ala Gln Asp Glu Asp Gln Pro Val Leu Leu Lys GluMET Pro Glu Ile Pro Lys Thr Ile Thr Gly Ser Glu Thr Asn LeuLeu Phe Phe Trp Glu Thr His Gly Thr Lys Asn Tyr Phe Thr SerVal Ala His Pro Asn Leu Phe Ile Ala Thr Lys Gln Asp Tyr TrpVal Cys Leu Ala Gly Gly Pro Pro Ser Ile Thr Asp Phe Gln IleLeu Glu Asn Gln Ala. 271

7. A pharmaceutical preparation suitable for parenteral administration, said preparation comprising a minor, effective amount of an interleukin-1.alpha. polypeptide of claims 5 or 6 and a major amount of a pharmaceutical parenteral carrier material.

8. A microbially produced interleukin-1.alpha. polypeptide having interleukin-1 activity, wherein said polypeptide has the following amino acid sequence:MET Ala Lys Val Pro Asp MET Phe Glu Asp Leu Lys Asn Cys TyrSer Glu Asn Glu Glu Asp Ser Ser Ser Ile Asp His Leu Ser LeuAsn Gln Lys Ser Phe Tyr His Val Ser Tyr Gly Pro Leu His GluGly Cys MET Asp Gln Ser Val Ser Leu Ser Ile Ser Glu Thr SerLys Thr Ser Lys Leu Thr Phe Lys Glu Ser MET Val Val Val AlaThr Asn Gly Lys Val Leu Lys Lys Arg Arg Leu Ser Leu Ser GlnSer Ile Thr Asp Asp Asp Leu Glu Ala Ile Ala Asn Asp Ser GluGlu Glu Ile Ile Lys Pro Arg Ser Ala Pro Phe Ser Phe Leu SerAsn Val Lys Tyr Asn Phe MET Arg Ile Ile Lys Tyr Glu Phe IleLeu Asn Asp Ala Leu Asn Gln Ser Ile Ile Arg Ala Asn Asp GlnTyr Leu Thr Ala Ala Ala Leu His Asn Leu Asp Glu Ala Val LysPhe Asp MET Gly Ala Tyr Lys Ser Ser Lys Asp Asp Ala Lys IleThr Val Ile Leu Arg Ile Ser Lys Thr Gln Leu Tyr Val Thr AlaGln Asp Glu Asp Gln Pro Val Leu Leu Lys Glu MET Pro Glu IlePro Lys Thr Ile Thr Gly Ser Glu Thr Asn Leu Leu Phe Phe TrpGlu Thr His Gly Thr Lys Asn Tyr Phe Thr Ser Val Ala His ProAsn Leu Phe Ile Ala Thr Lys Gln Asp Tyr Trp Val Cys Leu AlaGly Gly Pro Pro Ser Ile Thr Asp Phe Gln Ile Leu Glu Asn GlnAla.271

9. A pharmaceutical preparation suitable for parenteral administration, said preparation comprising a minor, effective amount of an interleukin-1.alpha. polypeptide of claim 8 and a major amount of a pharmaceutical parenteral carrier material.

10. A microbially produced interleukin-1.alpha. polypeptide having interleukin-1 activity, wherein said polypeptide has the following amino acid sequence:

(Met).sub.x Asn Arg Ile Arg Ile Arg Phe Leu Ser Asn ValLys Tyr Asn Phe MET Arg Ile Ile Lys Tyr Glu Phe IleLeu Asn Asp Ala Leu Asn Gln Ser Ile Ile Arg Ala AsnAsp Gln Tyr Leu Thr Ala Ala Ala Leu His Asn Leu AspGlu Ala Val Lys Phe Asp MET Gly Ala Tyr Lys Ser SerLys Asp Asp Ala Lys Ile Thr Val Ile Leu Arg Ile SerLys Thr Gln Leu Tyr Val Thr Ala Gln Asp Glu Asp GlnPro Val Leu Leu Lys Glu MET Pro Glu Ile Pro Lys ThrIle Thr Gly Ser Glu Thr Asn Leu Leu Phe Phe Trp GluThr His Gly Thr Lys Asn Tyr Phe Thr Ser Val Ala HisPro Asn Leu Phe Ile Ala Thr Lys Gln Asp Tyr Trp ValCys Leu Ala Gly Gly Pro Pro Ser Ile Thr Asp Phe GlnIle Leu Glu Asn Gln Ala,

wherein x is 0 or 1.