BACKGROUND OF THE INVENTION

Field of the Invention

Recombinant DNA structures constructed by binding together segments of DNA derived from different biological sources have opened an area for production of a wide variety of only difficulty accessible proteins and providing for bacterial capability of a wide variety of functions, which are not native to the particular bacterium. For example, various eukaryotic enzymes and hormones may be produced by employing a eukaryotic gene specific for such hormone or enzyme, such as somatostatin and insulin. Alternatively, bacteria can be provided with the capability for the production of eukaryotic enzymes for performing chemical functions on a commercial scale.

One of the difficulties encountered in producing proteins foreign to the bacterial host is that in the utilization of signals recognized by the host species to accomplish synthesis of the protein, the protein product may be a covalently linked hybrid protein comprising the protein of the plasmic vector and the protein of the exogeneous gene. Unless there is some simple chemical or enzymatic technique for separating the two proteins, the product is not likely to be physiologically useful or economically efficient. Also, where the naturally occurring protein is a prodrug, means may be required for cleaving the inert portion from the active drug portion.

It is therefore desirable to find techniques which would allow for production of proteins, where the protein is produced in substantially native form and free of covalent bonding to other proteins.

Description of the Prior Art

Chang, et al, Nature, in press (1978) describes phenotypic expression in E. coli. of a DNA sequence coding for mouse dihydrofolate reductase. Shine and Dalgarno, Proc. Natl. Acad. Sci. USA, 71, 1342 (1974) describe 3^1-terminal sequence of E. coli. 16S ribosomal RNA: complementarity to nonsense triplets and ribosome binding sites. Steitz and Steege, J. Mol. Biol. 114, 545 (1977) and Steege, Proc. Nat. Acad. Sci. USA 74 4163 (1977) identify sequences on mRNA in the 5¹ direction from the initiator codon that are complementary to the CCUCC sequence at the 3' end of the 16S ribosomal RNA species proposed to be involved in the binding of mRNA to ribosomes. Komaroff, et al, Proc. Natl. Acad. Sci., (1978) describes a bacterial clone synthesizing proinsulin where the recombinant plasmid has a poly- cytosine tail adjacent an internal codon reciprocal to methionine. Copending U.S. application Serial No. 687,430, filed May 17, 1976 describes the preparation of recombinant plasmids.

SUMMARY OF THE INVENTION

Method and compositions are provided for the production of a foreign protein by a bacterial host. The method employs a vector comprising a replicon having an intact replicator locus and capable of replication in said host, preferably in conjunction with a gene which allows for survival selection, an intact promoter and a DNA sequence that ordinarily is not able to be propagated in said host. The foreign DNA sequence has proximate to the 3^1-terminus of the coding strand a codon which when transcribed defines the initiating methionine codon of the sequence and is modified with a nucleotide oligopolymer which when transcribed simulates a prokaryotic ribosomal binding site to provide initiation of protein synthesis. The termini of the DNA sequence of the vector modified with an oligopolymer complementary to said oligopolymer defines the prokaryotic ribosomal binding site. The modified vector and modified foreign DNA coding sequence are annealed to form a replicating structure which is employed in the transformation or transfection of said bacterial host, whereby the transformed host is capable of expressing said foreign DNA sequence to produce foreign protein.

DESCRIPTION OF THE SPECIFIC EMBODIMENTS

Methods and compositions are provided involving inserting DNA foreign to a host capable of complete protein expression into a vector having free ends where foreign DNA is provided at the 3'-terminus of the coding strand with an oligopolymer which when transcribed defines the prokaryotic 16S ribosomal binding site in proximity to a nucleotide triplet which when transcribed defines the initiator methionine codon to define a protein synthesis starting site. The termini, usually 3', of the vector are extended with an oligopolymer of nucleotides complementary to the above oligopolymer to provide cohesive termini.

The modified foreign gene and modified vector are combined and annealed so as to form a structure capable of replication (propagation) in the host strain. The resulting plasmid need not be ligated, ligation apparently occurring in vivo, but may be introduced directly into the bacterium under transforming conditions. The bacterium may then be cloned, so that the foreign DNA sequence may be expressed, and the structure replicated. Once the structure has been replicated by the bacterium, it need no longer be prepared in vitro.

The term vector intends a DNA sequence having an intact or competent replicon and promoter, so that it is capable of replication and transcription to provide mRNA. The vector may be derived from a plasmid, virus or phage which is compatible with the bacterial host.

The foreign DNA structure will usually be a gene which is derived from a source which does not exchange genetic information with the host. The foreign DNA structure may be derived from a prokaryotic or eukaryotic source, may be synthesized, or combinations thereof. The coding strand intends the strand for transcription or coding for mRNA. The other strand will be referred to as the anticoding or nonsense strand.

Since the plasmid can be used as exemplary of the structure of the recombinant DNA sequence employed for transformation, except that the plasmid and virus are circular, while other DNA sequences derived from phage need not be circular, for the most part the recombinant DNA sequence will be referred to as a plasmid. It should be understood that in referring to a recombinant plasmid, the plasmid is only exemplary and is not limiting. Similarly, as to the foreign DNA sequence, the gene will be used as exemplary of any DNA which provides a useful product and derived from a source which does not exchange information with the bacterial host.

In referring to the DNA sequences, defining shall intend when transcribed to mRNA, the function of the resulting sequence of ribonucleotides.

The foreign DNA sequence will desirably have a phenotypical trait and is modified by the introduction at the 3' termini of an oligopolymer of nucleotides defining a simulated prokaryotic ribosomal binding site, with the terminus being at or adjacent to a nucleotide triplet defining an initiating methionine codon to define a start site. The DNA employed as a vector is also modified at the termini, usually 3', by introducing an oligopolymer having complementary nucleotides to those introduced on the coding strand, so as to serve as cohesive ends for annealing to the modified foreign gene. The modified foreign gene and the modified vector are combined and annealed providing a DNA structure capable of replication in the host species. The ends may be ligated to form covalent bonds or the DNA structure e.g. plasmid, may be used directly without ligation in transformation (includes transfection) of a host bacterium. Alternatively, the oligopolymer sequence may be blunt end ligated to the vector and foreign DNA sequence. The bacteria may then be cloned to provide expression of the foreign gene with the protein product being free of proteins from the replicon. Foreign DNA

The foreign DNA may have one or more DNA sequences, e.g. genes, depending upon the manner in which it was prepared or obtained and the purpose of the transformation. The double stranded DNA may be derived from eukaryotic or prokaryotic cells, viruses and bacteriophage, the source being one which does not normally exchange genetic information with the bacterial host. The fragments employed will generally have molecular weights in the range of about 0.01 to 20x10 d, usually in the range of about 0.1 to 10x10 d.

By introducing one or more exogenous DNA segments into a unicellular organism, the organism will be able to produce polypeptides and proteins ("poly(amino acids)") which the organism could not previously produce. In some instances the poly(amino acids) will have utility in themselves, while in other situations, particularly with enzymes, the enzymatic product(s) will either be useful in itself or useful to produce a desirable product.

One group of poly(amino acids) which are directly useful are hormones. Illustrative hormones include parathyroid hormones, growth.hormone, gonadotropins (FSH, luteinizing hormone, chorionogonadatropin, and glycoproteins), insulin, ACTH, somatostatin, prolactin, placental lactogen, melanocyte stimulating hormone, thyrotropin, parathyroid hormone, calcitonin, enkephalin, and angiotensin.

Other poly(amino acids) of interest include serum proteins, fibrinogin, prothrombin, thromboplastin, globulin e.g. gamma-globulins or antibodies, heparin, antihemophilia protein, oxytocin, albumins, actin, myosin, hemoglobin, ferritin, cytochrome, myoglobin, lactoglobulin, histones, avidin, thyroglobulin, interferon, kinins transcortion, and peptide antigens for use in making vaccines.

Where the genes or genes encode for one or more enzymes, the enzymes may be used for fulfilling a wide variety of functions. Included in these functions are nitrogen fixation, production of amino acids e.g. polyiodothyronine, particularly thyroxine, vitamins, both water and fat soluble vitamins, antimicrobial drugs, chemotherapeutic agents e.g. antitumor drugs, polypeptides and proteins e.g. enzymes from apoenzymes and hormones from prohormones, diagnostic reagents, energy producing combinations e.g. photosynthesis and hydrogen production, prostaglandins, steroids, cardiac glycosides, coenzymes, and the like.

The enzymes may be individually useful as agents separate from the cell for commercial applications, e.g. in detergents, synthetic transformations, diagnostic agents and the like. Enzymes are classified by the I.U.B. under the classification as I. Oxidoreduc- tases; II. Transferases; III. Hydrolases; IV. Lyases; V. Isomerases; and VI. Ligases.

The foreign DNA sequence may be obtained in a wide variety of ways. Conveniently, messenger RNA (mRNA) may be isolated and by employing reverse transcriptase, the DNA can be synthesized. This method has the advantage of providing for a relatively short fragment of DNA, a relatively large source of genetic material, and frequently the presence of the triplet defining the methionine initiator codon. Alternatively, one can synthesize either the mRNA or the DNA, either synthesizing the complete sequence de novo or synthesizing fragments and then allowing the fragments to be covalently joined with the appropriate polymerase or ligase enzyme. Alternatively, with the appropriate genetic material, one could employ endonuclease cleavage, where the restriction enzyme cleaves at a site adjacent to the nucleotide sequence defining the methionine initiator codon.

In the event that the nucleotide sequence defining the initiator methionine codon is cleaved from the gene, the appropriate nucleotide bases may be added to provide such sequence, followed by the addition of the bases or oligopolymer to provide the sequence defining the robosomal binding site or the nucleotides added as a unit. It should be appreciated that while the normal nucleotide triplet on the strand that defines the methionine codon is 5'-dCAT base base triplets other than AUG on the mRNA may in particular situations serve to be translated as methionine e.g. GUG, so that other nucleotide triplets on the sense strand may find use. Desirably, the restriction enzyme employed for isolating the gene from other genetic material will be chosen to provide for the desired nucleotide triplet adjacent the sense strand 3'-terminus.

Vector

In the preparation of the vector or replicon, double stranded plasmid, viral or phage DNA is cleaved with an appropriate restriction enzyme to provide for an intact replicator locus and system, and promoter. If a plasmid, the plasmid chosen will be capable of replicating in a microorganism, particularly a bacterium, which is susceptible to transformation. Various unicellular microorganisms can be transformed, such as bacteria, fungii, plants and algae. That is, those unicellular organisms which are capable of being grown in cultures or by fermentation. Bacteria, which are susceptible to transformation, include members of the Enterobacteriaceae, such as strains of Escherichia coli; Salmonella; Bacillaceae, such as Bacillus subtilis; Pneumococcus; Streptococcus, and Haemophilus influenzae.

After cleavage of the plasmid, depending upon the nature of the restriction enzyme, either square ends or slanted ends (cleavage at different but adjacent sites) may be obtained. Restriction enzymes which provide square ends include Hae III and Alu I, and enzymes providing slant ends include PstI and Hpa II. As indicated previously, a plasmid need not be used, but rather the DNA may be derived from a virus or phage which is compatible with the bacterial host which is to be transformed. The viral DNA will be treated in the same manner as plasmid DNA for providing the vector.

Desirably, the vector is obtained by endonuclease (restriction enzyme) cleavage of the vector source, where the site for cleavage at the 5^1-terminus has as the next successive base complementary base to the 3'-terminal base of the oligopolymer defining the simulated prokaryotic robosomal binding site. In effect, one wishes to recreate the endonuclease cleavage site. In this manner the recircularized plasmid may be cleaved to regenerate the modified gene at its 3^1-terminus employing the same endonuclease. Also, a preferable cleavage site is between the A and G of a CAG sequence.

The two different portions of the DNA used to prepare the recombinant plasmid will then be modified to provide the nucleotides defining the ribosomal binding site or complementary oligopolymer. Where the triplet defining the methionine initiator codon has been removed from the foreign gene, the appropriate triplet, as well as other bases defining appropriate amino acids may be added to the DNA using the appropriate enzyme e.g. ligase. The sequence defining the simulated prokaryotic ribosomal binding site may then be added to the DNA adjacent to, but not necessarily joined to, the triplet defining the initiator methionine codon. The ribosomal binding site has been described in Shine and Dalgarno, supra. Since the binding site on the ribosome has a plurality of cytosines, the nucleotides defining the start site on the coding strand of the DNA will also have a plurality of cytosines, so that the messenger RNA, which is complementary to the DNA will have a plurality of guanosines. The cytosine enriched tail defining the ribosomal binding site may be achieved by forming an oligohomopolymer of cytosines employing a transferase, or alternatively, a fragment which is similar or identical to that defining the 16S ribosomal RNA binding site may be synthetically prepared and then bonded to two of the termini of the DNA employing a ligase.

While conveniently, the deoxyribonucleotides defining the simulated ribosomal binding site may be ligated as a unit or added individually to the 3^1-termini of the foreign DNA sequence, the same final result of providing for a DNA sequence defining a start site may be achieved in many different ways. For example, a ds DNA segment defining the ribosomal binding site, having the triplet defining the methionine codon as appropriate may be blunt or staggered end ligated between the insert (foreign DNA) and the vector.

The need for introduction of the ribosomal binding site to the DNA will be present whenever the genetic material is prepared synthetically, the foreign gene is eukaryotic, or in the preparation or isolation of the gene, the ribosomal binding site has been altered or removed.

In order to provide cohesive ends, the vector will be modified by providing tails complementary to the tails added to the foreign gene. That is, for example, where a poly-dcytosine has been conjugated to the 3¹ ends of the foreign gene, a poly-dguanosine will be added to the 3' ends of the vector.

The number of members on the coding strand defining the ribosomal binding site will vary depending on whether an identical complementary unit is provided or the binding site is only partially complementary, as in the case of a homopolymer. Usually, there will be on the average at least about three base members, and not more than about 50 base members, more usually from about 8 to 40, and generally from about 10 to 25 base members. Furthermore, on the average, the complementary tail attached to the vector need not be of the same length as the length of the tail added to the coding strand, usually being at least about one-half the number of bases, and may be equal to or slightly greater than the number of bases defining the simulated ribosomal binding site, particularly when the number of bases in the tail attached to the foreign gene is in the low portion of the range.

Conveniently, the bases on the coding strand defining the simulated ribosomal binding site is immediately adjacent the codon complementary to the initiator methionine complementary codon. However, interruption is permissible. The interruption should not exceed about thirty bases, preferably not exceed twenty bases, and more preferably not exceed nine bases. The bases defining the simulated ribosomal binding site plus the bases intervening between the simulated ribosomal binding site and the bases defining the initiator methionine codon, which together define a start site, may be in phase or out of phase with the vector preferably out of phase.

Desirably, the vector should have a phenotypical property which allows for selection. Particularly, are those genes which provide for survival selection, for example, by providing for resistance to antibiotics or heavy metals or polypeptides, which have bacteriostatic or bacteriocidal functions. Alternatively, a host can be chosen which lacks an appropriate growth factor, which can be supplied by a gene on the replicon.

After preparation of the two double stranded . DNA sequences, the foreign gene and vector are combined for annealing and/or ligation to provide for a functional recombinant DNA structure. With plasmids, the annealing involves the hydrogen bonding together of the cohesive ends of the vector and the foreign gene to form a circular plasmid which has cleavage sites. The cleavage sites are then normally ligated to form the completely closed and circularized plasmid.

The annealing, and as appropriate, recircularization can be performed in whole or in part in vitro or in vivo. Preferably, the annealing is performed in vitro. The annealing requires an appropriate buffered medium containing the DNA fragments. The temperature employed initially for annealing will be about 40 to 70°C, followed by a period at lower temperature, generally from about 10 to 30°C. The molar ratio of the two segments will generally be in the range of about 1-5:5-1. The particular temperature for annealing will depend upon the binding strength of the cohesive termi. While 0.5hr to 2 or more days may be employed for annealing, it is believed that a period of 0.5 to 6hrs may be sufficient. The time employed for the annealing will vary with the temperature employed, the nature of the salt solution, as well as the nature of the cohesive termini.

The ligation, when in vitro, can be achieved in conventional ways employing DNA ligase. Ligation is conveniently carried out in an aqueous solution (pH6-8) at temperatures in the range of about 5 to 40°C. The concentration of the DNA will generally be from about 10 to 100 g/ml. A sufficient amount of the DNA ligase or other ligating agent e.g. T₄ ligase, is employed to provide a convenient rate of reaction, generally ranging from about 5 to 50U/ml. A small amount of a protein e.g. albumin, may be added at concentrations of about 10 to 200 g/ml. The ligation with DNA ligase is carried out in the presence of magnesium at about 1-lOmM.

At the completion of the annealing or ligation, the solution may be chilled and is ready for use in transformation.

Various techniques exist for transformation or transfection of a bacterial cell with plasmid DNA. See particularly co-pending application Serial No. 687,430, filed May 17, 1976, pages 15 to 16, which disclosure is incorporated herein by reference, as well as the references cited previously.

The transformed bacteria may then be cloned in accordance with conventional methods, particularly em_- ploying the phenotypical property for selection. The resulting clones are then selected for their capability to produce the functional protein and used for expression of the desired protein in a form free of protein derived from the vector and replication of the plasmid.

EXPERIMENTAL

(All temperatures not otherwise indicated are in centigrade. The following abbreviations are employed with their indicated meanings: 2ME-2-mercaptoethanol; SSC-saline sodium citrate; DHFR-dihydrofolate reducatase; cDNA-complementary DNA; TBE-tris-borate EDTA; PABA-Penassay broth agar.) Example 1. Preparation of pBR332 plasmid with DHFR

cDNA insert.

DNA complementary to DHFR mRNA was synthesized essentially as described by Buell et al, Biol. Chem. 253, 2471 (1978), using avian myeloblastosis virus (AMV) reverse transcriptase and polysomal RNA obtained by indirect immunoprecipitation of DHFR-synthesizing polysomes from methotrexate-resistant AT-3000 S-180 mouse cells (Alt et al, Biol. Chem., 253, 1357 (1978)). The RNA had been estimated to contain DHFR mRNA as 20% of its mRNA. The reaction was carried out in 100pl (50mM Tris, pH8.2 at 42°, 140mM KC1, lOmM MgCl₂, 30mM 2ME, 100µl/ml oligodT (12-18) (Collaborative Research)), 500pM each deoxynucleotide triphosphate (dNTP) (dCTP was adjusted to -4Ci/mm with 32P-dCTP)), 340pg polisomal RNA (estimated to contain 5µg polyA-RNA), and 45 units AMV reverse transcriptase. The reaction was incubated at 42° for 30min and stopped by the addition of 0.25M EDTA pH8.0 to 10mM.

Approximately 1.4pg cDNA was synthesized. To the reaction mixture was added E. coli tRNA (30pg), followed by extraction with an equal volume of phenol (saturated with TEN) (2X), followed by ether, before being passed over a Sephadex G-50 fine column (0.7x20cm) in 10mM Tris pH7.4, 2mM EDTA, 10mM NaCl (TEN). The void volume was collected and precipitated with 2.5 volumes of ethanol. After centrifugation, the cDNA was dissolved in 200pl TEN containing 110mm NaOH and heated at 70° for 25min to hydrolyze the RNA, followed by cooling, neutralization with equimolar HC1 (buffered with Tris, pH7.4 to 40mM) made to 0.4M NaCl and precipitated with ethanol.

After centrifugation, the cDNA was dissolved in 50pl 5mM Tris pH7.4, O.lmM EDTA and then used as template for the synthesis of the second strand by E. coli DNA polymerase I essentially as described in Seeburg et al, Nature, 270, 486 (1977). (The reaction was carried out at 42° for 10 minutes in 100pl(50mM Tris pH8.2, 20mM KC1, 7mM MgCL₂, 10mM 2ME, O.lmM EDTA) using 1,1µg cDNA, 10 units E. coli DNA polymerase I, and 200pM of each dNTP with dCTP adjusted to 30 ci/mM as above. Approximately 0.85pg of the second strand was synthesized. The reaction was stopped and extracted as above before being passed over a Sephadex G-50 fine column (0.5x7cm) in TEN containing only O.lmM EDTA. Column fractions containing the ds cDNA were adjusted to 40mM NaOAc pH4.5, 300mM NaCl, 3mM 2n(OAc)₂ and then treated with Aspergillus oryzae Sl nuclease(5U/ml as described by Wickens et al, J. Biol. Chem. 253, 2483 (1978)). Digestion was carried out at 37° for 60min and stopped by the addition of EDTA to 10mM.

After extraction with phenol and ether (as previously described) and adjusted to 0.3M NaCl, followed by precipitation with ethanol, approximately 1.0µg ds cDNA was obtained. Aliquots of a) first strand product, b) first strand product after base treatment, c) second strand product and d) second strand product after S1 nuclease treatment were examined on a 1.5% agarose gel under alkaline conditions as described by McDonnel et al, J. Mol. Biol., 110, 119 (1977).

Terminal addition of dCTP to the ds cDNA by terminal deoxynucleotidyl transferase (TdT, Chang and Bollum, J. Biol. Chem. 246, 909 (1971)) was carried out by a modification of the Co procedure (Roychoudhury, et al, Nuc. Acids Res., 3, 101 (1976)). The reaction was performed in 500µ1 containing 140mM cacodylic acid, 30mM Tris base, 100mM KOH (final pH7.6), 0.1mM dithiothreitol, 150pm dCTP (adjusted to 8 Ci/mm with ³H-dCTP (Amersham)), 1mM CoCl₂ (added to prewarmed reaction mix prior to enzyme addition), approximately l.Opg ds cDNA (assuming a number average MW of approximately 600 base pairs, this provides 10 pM 3' termini/ml) and 0.5µl TdT (2.3 x 10⁵ units/ml). The reaction was allowed to proceed at 37° for 10 minutes before being cooled and sampled to determine incorporation. Approximately 30dC residues were added per 3' terminus. The reaction was stopped (EDTA to lOmM), extracted, desalted and precipitated with ethanol as above.

Aliquots of e) second strand product, f) second strand product after S1 nuclease treatment and g) dC-tailed ds cDNA were analyzed on a 1.7% agarose gel in Tris-acetate-NaCl. The dC-tailed ds cDNA was then preparatively electrophoresed on a similar gel and the '1500 base pair' region cut out of the gel and electrophoretically eluted into a dialysis bag as described (see McDonnel et al, supra). The eluted material was extracted as above, concentrated by lyophilization and precipitated with ethanol. After centrifugation, the '1500 base pair' dC-tailed ds cDNA (approximately 80ng) was redissolved in 10mM Tris HC1 pH7.4, 0.25mM EDTA, 100mM NaCl (annealing buffer).

pBR322 plasmid DNA, isolated as described (Kuperstock and Helinski, Biochem. Biophys. Res. Commun., 54, 1451 (1973) was digested with a 1.5 fold excess of PstI endonuclease under conditions suggested by the vendor (New England Biolabs) and the linear plasmid DNA was cut out and eluted as described above from a 0.7% agarose gel in TBE (Sharp and Sambrook, Biochemistry, 12, 3055 (1974). The plasmid DNA was 'tailed' with dG residues following the procedures described above. Approximately 15-20 dG residues were added per 3' terminus. Following extraction with phenol and ethanol as previously described, the dG-tailed vector was passed over a Sephadex G-50 file column (0.5x7cm) in annealing buffer and the void volume was collected. Equimolar amounts of dC-tailed ds cDNA and dG-tailed vector DNA were allowed to anneal essentially as described in Sanger and Coulson, FEBS Lett., 87, 107 (1978), except that the vector concentration was kept at 75ng/ml in the annealing reaction. Circularization was monitored by electron microscopy and was typically about 20-40%. This annealed DNA was used directly for transformation into _X1776 or _X2282. Example 2. Transformation and cloning with DHFR

containing plasmid

pBR322 plasmid DNA that had been annealed in vitro with dc tailed DHFR cDNA (designated 1°) was introduced into 1776 or _x2282, using a modification of a previously described transfection procedure (Enea et al, J. Mol. Biol., 96, 495 (1975)). One ml of an overnight bacterial culture was innoculated into 100ml of L broth supplemented with diaminopimelic acid (DAP, 50pg/ml) and (for 1776 only) thymidine (4µg/ml). Bacterial cultures were grown until exponential phase at 35° and then harvested by centrifugation at 4°. Cells were washed in 0.3 volume 10mM NaCl, resuspended in 30ml freshly prepared MCN buffer (70mM MnCl₂; 40mM sodium acetate, pH5.6 and 30mM calcium chloride) and chilled on ice for 20 minutes. Cells were collected, resuspended in 1ml MCN and added in 200µl aliquots to 50pl DNA in TEN (lOmM Tris-HCl, pH7.5; O.lmM EDTA, 50mM NaCl) or MCN buffer. After chilling at 0° for 30 minutes, reactions were incubated at 27° for 5 min, chilled again for 30 min, and 50µl samples were plated onto Penassaybroth agar supplemented with DAP, thymidine (for _x1776), and antibiotics as indicated.

When _x2282 was used, the selective medium was M9 minimal agar supplemented with 0.5% casamino acids, biotin (2pg/ml), DAP (50pg/ml) and trimethoprim (Tp) (2.5-10µg/ml) plus tetracycline (Tc) or kanamycin (Km) as indicated in the table. Plates containing transformants were incubated at 32° and colonies were scored 2 to 3 days after plating. pBR322 plasmid DNA lacking the cDNA insert was used as a control. cDNA preparations labeled as 2° consisted of plasmid DNA isolated from a non-fractionated population of clones that had previously been transformed with chimeric molecules carrying a cDNA insert.

Example 3. Detection of colonies containing DHFR cDNA inserts by in situ hybridization

Colonies were screened for DHFR sequences using a modification of an in situ hydridization procedure (Grunstein and Hagness, Natn. Acad. Sci. U.S.A. 72, 3961 (1975)). Tc-resistant colonies were transferred to nitro-cellulose filters (Millipore, HAWG) that had been placed on PABA plates containing Tc(10pg/ml). (Filters had been washed twice by boiling in H₂0 and autoclaved prior to placing on plates). After 2-3 days of bacterial growth at 32°, the filter was removed from the plate and placed on a Whatman #3 pad saturated with 0.5N NaOH. After 7', the filter was sequentially transferred to a series of similar pads saturated with 1M Tris pH7.4 (twice, 7' each); 1.5M NaCl, 0.5M Tris, pH7.5 (once, 7'); and 0.30M NaCl, 0.03M Nacitrate (2 x SSC), (once, 7'). After the excess liquid was removed by suction the filter was placed on a pad containing 90% ethanol, dried by suction and backed in vacuo at 80° for 2hrs.

Prior to hybridization, filters were pretreated for 3-6 hrs at 65° in hybridization buffer that contained 5 x SSC pH6.1, 0.2% sodium dodecyl sulfate (SDS), 0.02% Ficol 400 (Pharmacia) and 8pg/ml E coli tRNA. Hybridizations were performed with individual filters in 1.5ml hybridization buffer containing 2 x 10⁴cpm 32 P-labelled purified DHFR cDNA (Alt et al, Biol. Chem., 253, 1357 (1978) in a sealed plastic bag at 65° for 24 hours. The filters were then washed in hybridization buffer (once, 60' at 65°); in 5 x SSC, pH6.1 (three times, 60' each at 65°); and in 2 x SS, pH7.4 (twice, 10' at room temperature); air dried, and prepared for autoradiography. A collection of 1776 colonies were shown to contain a DHFR cDNA insert.

Example 4. Inhibitor analysis of DHFR from bacterial cells

Stationary phase cultures of _X2282 expressing trimethoprim resistance were grown in the presence of Tp (lpg/ml) in minimal medium, washed with isotonic saline, and suspended in 50mM potassium phosphate buffer pH7.0 containing 10mM benzamidine and 10mM phenyl methyl sulfonyl fluoride (3 volumes buffer to 1 volume cells). The suspension was sonicated and centrifuged at 10,000rpm for 15min. The supernatant was centrifuged for lhr at 100,000xg before being studied. An R₂ methotrexate resistant mouse cell extract was prepared. Enzyme activity was measured by the radioactive folic acid assay previously described (Alt et al, J. Biol. Chem. 251, 3063 (1976)). Protein was determined by the method of Lowry. Approximately 3 units of activity from the _χ2282 extract or 5 units from the methotrexate resistant mouse cel_' extract were incubated with inhibitor for 10min at 24° before assaying for folate reductase activity; background values were determined by measuring enzyme activity in the presence of 10mM methotrexate. The results for enzyme activity were closely comparable in the presence of varying concentrations of trimethoprim for the mouse abstracts and _χ2282 as evidenced by the amount of enzyme needed to reduce 1nM of folate in 15min at 37°.

The subject method provides for the preparation of a functional enzyme free of protein or polypeptide chain from the plasmid vector which might interfere with or significantly change the character of the enzyme from the eukaryotic source. Thus, the subject invention provides a method for introducing a DNA sequence defining a ribosomal binding site adjacent a triplet defining an initiator methionine codon to provide a unit which allows for ribosomal initiation of protein production at the initial site of a DNA sequence foreign to the host. In this manner, the difficult problem of separating the covalently bound vector protein from the foreign protein is avoided.

Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be obvious that certain changes and modifications may be practiced within the scope of the appended claims.