GB2104901A - Expression vectors - Google Patents

Abstract

An expression vector for introducing a gene into a procaryotic organism, comprising a closed loop of DNA having the capability of replication in a procaryotic cell and including: (a) promoter, operator and translation initiation sequences, which are recognizable by the procaryotic host cell; (b) a Shine-Dalgarno sequence and an initiator codon within the translation initiation sequence the initiator codon being proximate to the end of the translation initiation sequence distal from the promoter and operator sequences; and (c) a recognition sequence for a restriction endonuclease, which is proximate to the initiator codon, and which, upon cleavage, leaves said initiator codon, or its complement, within the translation initiation sequence. r

Description

SPECIFICATION Expression vectors Background of the invention In the field of genetic engineering, a technique commonly used to achieve expression of a foreign protein by a procaryotic organism is to transform that organism with a plasmid that has been modified to contain genetic information specifying the desired foreign protein. Various eucaryotic genes have been successfully inserted into plasmids and, upon introduction into procaryotic organisms, cause those organisms to express a foreign protein. Thus the production of human interferon, human growth hormone, human insulin, chick ovalbumin, somatostatin and the like by genetically modified strains of Escherichi coli has been reported.Taniguichi, T., et al., Proc. Natal. Acad. Sci. USA, 77, 5230 4233 (1980); Villa-Komaroff, L., et al., Proc. Natal. Acad. Sci. USA, 75, 3727-3731(1978); and Goeddel, D.

V., et al. Nature, 281, 544 (1979).

Constructing an organism capable of expressing foreign protein involves a number of steps. First of all, the gene which carries instructions for the biosynthesis of the desired protein is identified and isolated. Procedures for the identification and isolation of genes vary considerably, but any procedure by which the desired gene can be obtained in substantially pure form can be employed.

A method commonly used for obtaining eucaryotic genes begins with the isolation of messenger RNA (mRNA). Protein synthesis is initiated in cells by the process of transcription, which involves the enzymatic synthesis of an RNA chain corresponding to a DNA template. The RNA thus formed interacts with the protein-synthesizing apparatus of the cell to produce protein. The RNA specified by a particular gene (and thus specifying a particular protein) is termed messenger FINA. When the cell is producing protein, mRNA is present in much larger concentrations than the gene which carries the protein-synthesizing information. Additionally, the mRNA is generally in a form more available for isolation and purification than is chromosomal DNA.Conventional purification techniques such as density gradient sedimentation, affinity chromatography, and electrophoresis can be used to isolate mRNA from cell lysates or homogenates. These techniques have been employed, for instance, to recover insulin mRNA from pancreas cells (Villa-Komaroff, L., et al., supra) and interferon mRNA from human Leukocyte (Goeddel, D. V., et al., Nature, 287, 41 1 (1980)) and fibroblast cell lines (Taniguchi, T., et al., supra).

The mRNA so obtained may be then used to prepare a single-stranded DNA molecule (known as complementary DNA (cDNA)) (Ullrich, et al., Science, 196, 131 3 (1977)). The resulting single-stranded DNA molecule has a 3'-terminal hairpin structure which provides a primer for subsequent synthesis of a second DNA strand using the enzyme, DNA polymerase. The loop connecting the two DNA strands is then broken (e.g., using the enzyme S1 nuclease) resulting in a cDNA segment identical to the original gene. The resulting cDNA is generally isolated and purified, e.g. by gel electrophoresis, and its structure may be verified by sequencing analysis or restriction mapping. The ends of DNA segments obtained in this manner may be modified, as hereinafter described, to facilitate insertion into a transfer vector.

Genes may be introduced into procaryotic cells via transfer vectors, which are generally plasmids or bacteriophages into which the gene of interest has been inserted. When the purpose of introducing the gene into an organism is to amplify the gene (i.e., obtain usable quantities of the gene by cloning), the transfer vector is sometimes termed a "cloning vector". When the purpose of introducing the gene into an organism is to obtain protein expression, the transfer vector is sometimes referred to as an "expression vector". The present invention concerns novel transfer vectors, and particularly, plasmid expression vectors. Plasmids are relatively small closed loops of DNA.An increased understanding of the action of restriction endonuclease enzymes and the discovery and isolation of a number of such enzymes have enabled molecular biologists to characterize plasmids and to modify them for improved utility as transfer vectors. For instance, unwanted DNA can be deleted, thus reducing the size of the plasmid; restriction sites can be tailored for insertion of a gene; and selectable genes, such as antibiotic resistance, can be inserted into a plasmid.

Mere splicing of a gene into a plasmid does not assure protein expression. Protein synthesis requires the presence of various control signals in proximity to the gene. Guarente, L., et al., Cell, 20, 543-553 (1980). These signals initiate, regulate, and terminate both mRNA transcription and mRNA translation (i.e., protein synthesis). It has been found that control signals associated with synthesis of a particular protein by a bacterium may be employed for controlling the synthesis of a foreign protein by the bacterium. This finding has been used advantageously to construct versatile plasmid expression vectors. Roberts, T. M., et al., Proc. Natl. Acad. Sci. USA, 76, 760-764 (1979); Guarante, et al. supra.

The control signals associated with protein synthesis are illustrated graphically in Figure 1 of the drawings, which represents a segment of double-stranded DNA associated with control of the enzymes involved in the biosynthesis of tryptophan (trp). "P" represents a base-pair sequence known as the promoter region. During protein synthesis, the promoter region signals the initiation of mRNA transcription. "0" represents a base pair sequence known as the operator region. This region, in conjunction with a repressor, controls the amount of mRNA transcription. For instance, in tryptophan biosynthesis excess tryptophan interacts with a repressor and decreases the level of trp mRNA transcription.Although the promoter and operator regions are graphically illustrated as separate, some operator regions may be subsequences within the promoter sequence. "S-D" represents a base pair sequence known as the Shine-Delgarno sequence. This sequence is the signal by which the proteinsynthesizing apparatus (the ribosome) recognizes the mRNA. Thus the Shine-Dalgarno sequence is not required for mRNA synthesis, but is required for protein synthesis. "ATG" represents the methionine, or initiator, codon. This codon is the signal for the start of protein synthesis in bacteria.

These signals are, in large part, idiosyncratic, in that different organisms utilize different codes.

For this reason, when a eucaryotic gene containing the eucaryotic regulatory regions is inserted into a plasmid and then inserted into a procaryotic cell, that cell might not express protein, or even mRNA because of failure to recognize the required control sequences. On the other hand, a promoter-operator region associated with synthesis of a particular protein in a procaryotic organism may be joined to a eucaryotic gene, and a plasmid clone containing such combination can utilize the protein synthesizing apparatus of such procaryotic organism to express the foreign protein.

Molecular biologists have taken advantage of this discovery to synthesize so-called "portable promoters" Roberts, T. M., et al. supra. Such portable promoters, which are illustrated in Figure 2, contain promoter, operator and Shine-Dalgarno regions which can be recognized by a particular host organism. They are bounded on each end by specific restriction enzyme recognition sequences, which allow them to be excised and inserted elsewhere after appropriate enzymatic treatment. Preferably, the opposite ends of such a portable promoter have base pair sequences which correspond to different restriction enzymes. In Figure 2, these ends are labelled X and Y, corresponding to different restriction enzymes, arbitrarily designated X and Y.Such a structure insures the proper orientation of the portable promoter relative to the gene to be expressed; it also provides for a single insertion site in the resulting plasmid (as hereinafter discussed). A plasmid, as illustrated in Figure 3, having restriction sites X and Y corresponding to each end of the portable promoter may be opened by digestion with appropriate restriction enzymes and the portable promoter inserted therein, e.g., using the enzyme T4 ligase. The resulting plasmid (Figure 4) can be used as an expression vector.

To utilize such an expression vector, the desired gene is isolated, as described above, and its ends are chemically modified to have sequences corresponding to the Y restriction enzyme. If the gene does not have initiation and termination codons, each end must be modified to include such codons. After opening the plasmid with restriction enzyme Y, the modified gene may be inserted to produce the desired plasmid (Figure 5).

Transformation of an appropriate procaryotic organism with such a plasmid can result in expression of the foreign protein under control of the tryptophan control signals. Virtually any bacterial promoter-operator-Shine-Dalgarno region can be employed for this purpose.

As explained above, genes which are inserted into heretofore known transfer vectors should carry initiation and termination signals. In the case of termination signals, this requirement is not particularly bothersome, because, when isolating the gene, a restriction enzyme can be selected such that the termination codon is included, and there is little concern if additional DNA (after the termination codon) is also obtained.

On the other hand, the spacing between the Shine-Dalgarno sequence and the initiation codon can be very important (Guarente, L., et ai., supra). To find a restriction enzyme which excises a gene from a longer DNA strand in such a way that the initiation codon is obtained without also obtaining objectionably long "leader" portions of DNA is fortuitous. More often, the gene is cut in such a manner that the initiation codon and a portion of the gene is lost. The lost DNA, the initiation codon and suitable leader usually must be added back enzymatically or chemically.

Summary of the invention In accordance with the present invention, a novel expression vector for introducing a gene into a procaryotic organism is provided. The expression vector comprises a closed loop of DNA, having signals which enable it to be replicated in a procaryotic host cell, and further includes (a) promoter, operator and translation initiation sequences, which are recognizable by the procaryotic host cell; (b) a Shine-Dalgarno sequence and an initiator codon within the translation initiation sequence the initiator codon being proximate to the end of the translation initiation sequence distal from the promoter and operator sequences; and (c) a recognition sequence for a restriction endonuclease, which is proximate to the initiator codon, and which, upon cleavage, leaves said initiator codon, or its complement, within the translation initiation sequence.

Detailed description of the invention The novel expression vector of this invention is illustrated in Figure 6 of the drawings. Figure 6 shows a circular plasmid having promoter, operator, Shine-Dalgarno regions and the methionine (ATG) initiator codon. A site for a restriction endonuclease (arbitrarily designated Z) is proximate to the initiator codon. Such a transfer vector can be very versatile, because, upon opening the plasmid with restriction endonuclease Z and insertion of a gene, the plasmid will have all of the signals needed for mRNA transcription and for translation of the gene which encodes the desired foreign protein.

In the plasmid expression vector of this invention, the composition and length of the segment between the Shine-Dalgarno sequence and the initiator codon can be accurately and reproducibly controlled. Thus, this segment may be tailored to the particular promoter-operator system employed, to optimize gene expression. Moreover, the expression vector has an insertion site located such that cleavage of the plasmid with the appropriate restriction endonuclease leaves the initiator codon in the translation initiation sequence. Thus, the manipulation of the expression vector for insertion of a gene does not disturb the translation initiation sequence, and this sequence remains intact, regardless of the gene being inserted.As used herein, "translation initiation sequence" includes the Shine-Dalgarno sequence (ribosome binding site), the initiator codon (usually ATG), and the DNA segment between those two sequences.

The expression vectors of this invention may be derived from wild type plasmids or modifications thereof. Advantageously, the plasmid selected as starting material will have replication functions which are unaffected by subsequent manipulations. These replication functions insure than the final transfer vector will be subject to the desired mode of replication control, i.e., will be present in multiple copies or a single copy per cell or in a controllable number of copies per cell. Such plasmid starting material is also preferably relatively small in size. By being small, the plasmid is capable of accepting large gene insertions, transformation of cells is facilitated, and the plasmid does not divert unnecessarily large amounts of cellular energy and nutrients to the production of unwanted macromolecules.

Advantageously, the plasmid will be smaller than about ten kilobases, and preferably range from about two kilobases to about five kilobases.

To prepare an expression vector of the present invention, control signals associated with RNA transcription and translation are inserted into the starting plasmid. Although such control signals may be in any form recognizable by the intended host organism, certain systems are preferred. The promoter-operator regions associated with procaryotic protein synthesis vary in efficiency. Generally, a highly efficient system is preferred, because such a system results in a high level of protein expression.

Known promoter-operator systems which are particularly useful in synthesizing the novel transfer vectors of the present invention are the Escherichia systems for the control of lactose (lac) and tryptophan (trp) metabolism and the promoter-operator region (P,) of bacteriophage A. The lac and trp systems are generally preferred for the synthesis of the present expression vectors.

These control signals may be obtained by excision from any DNA containing them. The method of excision will depend upon the restriction enzymes which are available and the presence of useful restriction sites bounding the system. The promoter, operator and Shine-Dalgarno regions are preferably recovered as one intact segment of DNA. The excised DNA segment containing the desired control signals is then inserted into the plasmid, using conventional procedures for gene insertion.

Alternatively, a plasmid which already contains the desired control signals may be employed as the starting material for synthesizing the expression vectors of this invention. A particularly preferred plasmid starting material is designated pGL101 and is described by Guarente, L., et al., supra, herein incorporated by reference. That plasmid, which is illustrated graphically in Figure 7, contains restriction sites for each of the enzymes EcoRl and Ecvull, and carries genes for ampicillin resistance. This plasmid also contains the promoter-operator-Shine-Dalgarno sequences of an E. colliac operon. The system was derived from a lac UV5 mutant by digestion with Alul restriction endonuclease. The UV5 mutation renders the lac promoter insensitive to catabolic repression by catabolic gene activator protein (CAP).

The Alul ends of the promoter-operator-Shine-Dalgarno segment were modified to provide EcoRl and BamHI cohesive ends, and this segment was inserted into plasmid pBR322 (Bolivar, F., et al., Gene, 2, 74 (1977)) after digestion with EcoRI and BamHI, thereby producing plasmid pGL101.

A plasmid, such as pGL101, or a promoter-operator-Shine-Dalgarno segment such as that used to construct plasmid pGL101 may advantageously be modified to contain the initiator codonrestriction site configuration of the expression vectors of the present invention. To accomplish this modification, a restriction enzyme site which contains the initiator codon (e.g., ATG) as part of its recognition sequence can be added to the segment containing the promoter-operator-Shine-Dalgarno sequences. Alternatively, a recognition sequence for an enzyme of the type for which the recognition sequence is separated from the cutting location by a certain number of non-specific nucleotide base pairs may be utilized. The non-specific portion of the sequence would contain the initiator codon in this case.For example, the restriction enzyme Hgal cuts at 5 and 10 base pairs from the sequence GACGC; therefore, a restriction site for this enzyme could include the initiator codon in the non-specific 5 to 10 base pair segment. The restriction site is advantageously within three base pairs of the initiator codon, and preferably, the restriction site is positioned such that upon cleavage with the restriction endonuclease, the translation initiation sequence ends with the initiator codon or its complement.

The sequence containing the initiator codon may conveniently be inserted into a plasmid or added to a linear DNA segment as a blunt-ended synthetic linker, i.e., a segment of DNA synthesized from individual nucleotides. For instance, in the case of plasmid pGLl Ol the plasmid may be opened at the recognition site adjacent to the Shine-Dalgarno sequence with Pull, and the synthetic linker inserted by blunt end ligation. (Note: Pvull makes flush end cuts).

A preferred restriction enzyme recognition sequence is the one for the enzyme Sphl. This recognition sequence is a hexanucleotide which contains the ATG initiator codon. The recognition sequence for this enzyme is shown in Figure 8, and the cutting locations are depicted by the dotted line.

This novel restriction site-initiator codon configuration provides several distinct advantages. A unique insertion site is introduced into the plasmid, the presence of the initiator codon is assured, and the distance between the Shine-Dalgarno sequence and the initiator codon can be accurately controlled.

Although the significance of the distance between the Shine-Dalgarno sequence and the initiator codon is not well understood, it is considered to be very important for efficient protein expression.

Preferably, the distance approximates the natural distance for the promoter-operator system employed, which generally ranges from about 0 to about 22 base pairs, preferably from about 3 to about 7 base pairs. This distance can conveniently be controlled by adjusting the size of the synthetic linker utilized for the addition of the initiator codon. The sequence length can then be optimized by selection based upon the performance of organisms containing plasmids having sequences of varying lengths.

Because a gene rarely has a particular restriction site at the beginning of its coding sequence, some modification of the gene is generally required before it can be inserted into the expression vector.

If the desired restriction sites occur near the beginning of the gene and at or after the end of the gene, restriction enzyme digestion will result in the recovery of a somewhat longer or shorter DNA segment than the actual gene. The excised gene can conveniently be inserted into the expression vector, but the protein expressed by the organism into which the vector is inserted will be somewhat different from the natural protein; typically, it will have a shorter or longer amino acid sequence than the naturally occurring protein. In many cases, these differences will not be significant, because the resulting protein will have substantially equivalent activity.On the other hand, when the structure of the protein is critical, the excised gene may advantageously be modified to restore its original structure, or alternatively, the resulting protein may be modified to produce the natural structure.

More often, the foreign gene will have no convenient restriction sites corresponding to the unique restriction insertion site of the expression vector. In this situation, the gene is then generally excised using one or more restriction enzymes which do cut at convenient locations near the beginning and end of the gene. The ends of the gene are then modified by "chewing back" with an exonuclease to remove excess DNA and by adding linkers containing base pair sequences corresponding to the restriction insertion site of the expression vector. If, in isolating the gene, segments of the gene itself are removed, the genetic information contained in those segments (if known) may also be added back with synthetic linkers, if desired.

In order to add linkers containing the desired restriction enzyme site to each end of the gene, each end is advantageously made blunt, i.e. single-stranded extensions are removed. If the end modifications described above do not result in blunt ends, the ends can be made blunt by known procedures (Ullrich, et al., supra). Once having obtained the blunt-ended gene, previously synthesized linkers can be added by blunt end ligation using, for example, the enzyme T4 DNA ligase. Then, digestion of the modified gene with the restriction enzyme corresponding to the site contained in the linker removes any excess linkers and results in a gene having cohesive ends. The modified gene can then conveniently be inserted into the transfer vector, which has been opened with the restriction enzyme.Insertion of the gene into the expression vector of the present invention reforms the initiator codon.

It is apparent that the expression vectors described herein can be very versatile. They can be used for introducing a wide variety of eucaryotic genes into procaryotic organisms. The expression of foreign protein by such organisms can be placed under the control of efficient promoter-operator systems. Moreover, once a plasmid containing a promoter-operator-Shine-Dalgarno-initiator codon sequence is obtained, the entire sequence may conveniently be cloned and excised for insertion into other vectors, utilizing known genetic engineering techniques. Thus, specialized transfer vectors can be tailored to specific situations.

Although the present invention has been described in connection with certain specific plasmids, genes, and control sequences, it is not intended to be so limited, but intended to be interpreted broadly in accordance with the appended claims. The invention is further illustrated by the following examples, which are not intended to be limiting.

Example I The objective of the experiment described in this example was to place the gal K (galactokinase) gene of E. colí under the control of the lac operater/promoter using an expression vector constructed in accordance with the present invention. In brief, the protocol was divided into three operations: (1) The vector pGLl 01 (Taniguichi et al., Proc. Natl. Acad. Sci.USA 77: 5230, 1 980) was modified, such that a synthetic DNA sequence encoding the endonuclease Sphl recognition site was ligated at a unique Pvull site operator-distal to the lac ribosome binding site, thereby creating a hybrid translation initiation site including the ATG codon, according to the following scheme (Note: the new vector is designated pGX951):

Pvull cleavage site pGL101 lacolp AGGAACAGGATC ~~~~~~~~~ TCCTTGTCCTAG j PvuII endonuclease iac /P AGGAACAG TCCTTGTC +Sphl LINKER GCATGC CGTACG LIGATION lac o/p AGGAACAGGCATGC TCCTTGTCCGTACG Sphl RESTRICTION Translation Initiation Site 1) PGX95l lac o/p AGGAACAGGCATG TCCTTGTCC (2) A restriction fragment containing the galK (galactokinase) gene was modified, such that its Nterminal sequence included a ligated Sphl linker which generated a complement to a new ATG initiation codon and a new second codon, retaining all subsequent original sequence in correction reading frame. The following scheme illustrates the galK gene modification::

original N-terminal codon Hindlll n Accl site AGAAATGAGTCTGA gal K gene site TCTTTACTCAGACT I I ! , Hinfl AGTCTGA galK Accl GACT fill-in of DNA POLYMERASE (KLENOW) restriction DNA POLYMERASE (KLENOW) ends AGTCTGA TCAGACT +linker AAGCATGCTT TTCGTACGAA I rigation Sphl site AAGCATG CTTAGTCTGA AAGCATGCTT TTCCTACGAATCAGACT TTCCTACGAA Sphl new codon P' CTTAGTCTGA gal K AACGATG CTACGAATCAGACT TTG (3) Ligation of (2) and (1) according to the following scheme, reconstitutes a complete translation initiation site permitting in-phase synthesis of galactokinase:

Vector: lac o/p AGGAACAGGCATG pGX951 TCCTTGTCC insert: + CTTAGTCTGA gal K CTACGAATCAGACT LIGATION gal K lac o/p AGGAACAGGCATGCTTAGTC ~~~~~~~ TCCTTGTCCGTACGAATCAG TRANSLATION INITIATION SITE Experimental procedures 1. Preparation of the vector pGX951 Plasmid pGL101 was prepared as described by Taniguichi, et al., Proc. Natl. Acad. Sci. USA, 77: 5230, (1980). Purified plasmid DNA was prepared from cleared cell lysates (Hershfield et al., Proc.

Natl. Acad. Sci. USA, 71: 3455, (1974)) by two consecutive bandings in CsCI gradients. A reaction mixture was prepared. having a volume of 50 Hl and containing 15 y9 of plasmid pGL101 DNA, 25 units endonuclease Pvull (New England Biolabs) and buffer salts, final concentration of 50 mM NaCI, 6mM Tris-HCI pH 7.4, 6 mM MgCI2, 10 mM mercaptoethanol and 100 y9 bovine serum albumin per ml. This buffer formulation is referred to herein as "50 mM salts buffer".Following incubation for one hour at 37 CC, 0.8 units of calf intestine alkaline phosphatase (Boehringer-Mannheim) were added and incubation continued for 30 min at 65 . The mix was resolved by electrophoresis in 1% agarose (Sea Plaque, Marine Colloids) by standard methods (Selker et al., J. Bacteriol., 129: 388, (1977). Following electrophoresis, the linear plasmid was visualized by ethidium bromide staining, cut from the gel and extracted by the method of Langridge et al., Anal. Biochem., 103: 264, (1980). The synthetic oligonucleotide (GceCATGccC) ("linker") was prepared by the triester method (Itakura, et al., J. Biol. Chem., 250: 4592, (1975)).The synthetic linker was kinased in a 20 yI reaction volume containing 5 ssg linker, 1 mM disodium ATP(P-L Biochemicals), 10 units T4 polynucleotide kinase (P-L biochemicals), 50 mMtris-HCI pH 7.6,10 mMMgCL2, 5 mM dithiothreitol, 0.1 mM spermidine (Sigma) and 0.1 mM disodium EDTA (ethylenediamine tetraacetic acid). The reaction mix was incubated at 37CC one hour, then at 650C for 5 min.A 4 yI portion of this mix was added to a solution containing 2 g of plasmid pGL101 cut with Pvull endonuclease (above), 1 mM disodium ATP, 10 mM dithiothreitol, 100 g bovine serum albumin per ml, 300 units of T4 DNA ligase (New England Biolabs), 50 mMtris-HCl pH 7.6 and 5 mM MgCI2 in a volume of 20 iul. The reaction mix was incubated 14 hours at 1 20C. This mixture was then used to transform E. coli strain N1 00. The cells were made competent for transformation by the method of Cohen, et al., Proc. Natal. Acad. Sci., USA, 69: 211 0, (1972).A 10 yl portion of the ligation mix was added to 0.2 ml of ice-cold cell suspension and incubated on ice for one hour. The mix was heat-shocked 2 min at 420C and left at 240C for 10 min. Three ml of LB broth (10 gm tryptone, 5 gm yeast extract, 10 gm NaCI per liter) were added, and the cells incubated for one hour at 370C. Volumes of 0.1 ml were plated on plates containing 1.5% agar, LB broth, and 100 yg/mi of ampicillin. After 16 hours of incubation, approximately 200 colonies were scored, compared to ten in a control mix lacking the linker. Ten of these colonies were grown overnight in 10 ml volumes of LB broth. Lysates were prepared as described.The isolated DNA was shown to be cut once by Sphl endonuclease, and uncut by Pvull, compared to the starting plasmid pGL101 which was uncut by Sphl and cut once by Pvull. The new plasmid was designated pGX951. A purified DNA preparation of 50 fly was cut by Sphl and treated with calf intestine alkaline phosphatase as described above.

2. Ligation of synthetic linker to a prepared fragment containing gal K.

The source of gal K was plasmid pKO-I DNA, purified as above. The plasmid and restriction data were obtained from Dr. Keith McKenney (McKenney, K., et al.,............... al ). The gal K gene was obtained from a restriction fragment bounded by Hindlil and Acc/ sites (Fig. 9). The reaction mix contained "50 mM salts" buffer (above), 50 y9 pKO-I plasmid DNA and 70 units Hindlil endonuclease (Bethesda Research Labs) in a volume of 125 l. After 2 hr. at 370C the DNA was precipitated with 2 volumes ethanol, collected by centrifugation, dried, and redissolved in 55 l H20.A volume of 20,ul of buffer salts (5x concentrated "50 mM salts buffer") was added plus 25 l of endonuclease Accl (10 units, New England Biolab). Incubation was for 2.5 hr. at 37 C, then at 650C five minutes. The DNA was again precipitated with ethanol as above, and dried. A partial digest was then carried out with endonuclease Hinfl; this was done because in addition to the particular Hinfl site at the N-terminus of gal K, there is an internal Hinfl site which had to be left intact. The reaction mix of 240 jul contained 5.7 units Hinfl (Bethesda Research Lab), 9.6 ssg of the H!ndlll-Acclfragment DNA and the "50 mM salt" buffer. Incubation was for 2.5 min. at 370C followed by 1 5 min. at 650C.The DNA was extracted with 50 yI phenol, the residual phenol removed with ether and the DNA precipitated with 2 volumes of ethanol. The pellet was washed with 4 volumes 70% ethanol and dried in vacuo. The "partial" mixture was then "filled in" by DNA polymerase to convert the sticky restriction ends to blunt ends for linker ligation. A reaction mix of 40 yI contained 3.2 ,ug of the Hinf--l "partial" DNA, above, 4 jul polymerase buffer (Bethesda Research Labs nick translation kit), 1 0 mM MgCL2, 4 units DNA polymerase "Klenow" (Boehinger-Mannheim) and 0.2 mM dATP, dCTP, dGTP and TTP (Bethesda Research Labs). This mix was incubated 20 min at 230C, then 1 5 min. at 650C.Ligation of a synthetic oligonucleotide TTCGTACGAA) was carried out following kinasing of the linker as above. A reaction mix of 40 ,ul contained "50 mM salts" buffer as above (1), 3 ug of the DNA partial digest, 2 yg kinased linker and 1 500 units T4 DNA ligase (New England Biolabs). This was incubated 20 hrs. at 200C, then 1 5 min. at 650C. To the cooled mix was added 20 units Sphl endonuclease and incubation contained at 370C for 1 hr. The mix was extracted with 20 flu phenol, residual phenol removed with ether, and 2 volumes ethanol used to precipitate the DNA.Following a 70% ethanol wash and drying in vacuo the DNA was redissolved in 20 jbl ligase buffer (above) containing 2.5 ,ug plasmid pGX951 (cut with Sphl and phosphatased as above), 1000 units T4 DNA ligase (New England Biolabs), and 0.5 mM ATP. The mix was incubated 16 hrs. at 200C, then 1 5 min. at 650C. Upon cooling 10 units of endonuclease Smal (Bethesda Research Labs) was added to linearize any molecules containing sequence between the Hindu and Hinfl sites (Figure 9). Incubation was for 1 hr. at 370C. After 15 min.

at 650C the chilled mix was used to transform 200 jut of E. collstrain N1 00 galK made competent and transformed as above. The cells were plated on McConkey agar containing 1% galactose and 100 ,ug/ml ampicillin. Following 16 hr. incubation colonies were observed. Red colonies which were now able to express the gal K gene were purified and DNA preparations made, as above. The DNA was subjected to restriction analysis to verify the expected pattern. The restriction map of plasmid pKO-l in the gal K region is shown in Figure 9 of the drawings. The numbers in Figure 9 refer to distance in base pairs. Symbois: E=endonuclease EcoRI, H=Hindlil, HF=Hinfl, A=Accl, S=Smal. The line labelled "gal K"shows the position of the gal K (galactokinase) gene. The galactokinase-producing strain prepared by this procedure has been designated pGx951, and has been deposited with the American Type Culture Collection, Rockville, MD as ATCC No. 31917.

Claims (25)

Claims

1. An expression vector for introducing a gene into a procaryotic organism, which comprises a closed loop of DNA, having signals which enable it to be replicated in a procaryotic host cell and further including: (a) promoter, operator, and translation initiation sequences which are recognizable by the procaryotic host cell; (b) a Shine-Dalgarno sequence and an initiator codon within the translation initiation sequence the initiator codon being proximate to the end of the translation initiation sequence distal from the promoter and operator sequences; and (c) a recognition sequence for a restriction endonuclease which cleaves proximate to the initiator codon leaving said initiator codon, or its complement, within the translation initiation sequence.

2. The expression vector of claim 1 , wherein said restriction endonuclease cleaves within three base pairs of the initiator codon or its complement.

3. The expression vector of claim 1, wherein the restriction endonuclease cleaves in such a manner that the translation initiation sequence ends with the initiator codon or its complement.

4. The expression vector of claim 1, wherein the initiator codon is contained within the recognition sequence for said restriction endonuclease.

5. The expression vector of claim 1, wherein said initiator codon is contained within the recognition sequence for the restriction endonuclease, Sphl.

6. The expression vector of any of claims 1 to 3, wherein the recognition sequence for said restriction endonuclease is separated from the restriction endonuclease cleaving site by a plurality of non-specific nucleotide base pairs.

7. The expression vector of claim 6, wherein the restriction endonuclease is Hgal.

8. The expression vector of any of claims 1 to 5, wherein the initiator codon is separated from the Shine-Dalgarno sequence by from C to about 22 base pairs.

9. The expression vector of claim 8, wherein the initiator codon is separated from the Shine Dalgarno sequence by from about 3 to about 7 base pairs.

1 0. The expression vector of any of claims 1 to 9 wherein said gene is a procaryotic gene, a eucaryotic gene, or a synthetic DNA sequence that codes for a specific polypeptide.

11. The expression vector of any claims 1 to 5 wherein the procaryotic host cell is Escherichia coli, and the promoter, operator and Shine-Dalgarno sequence are derived from the /ac or trp operons of E. coli.

1 2. The expression vector of any of claims 1 to 5, wherein the procaryotic host cell is Escherichia coli, and the promoter, operator and Shine-Dalgarno sequences are derived from a bacteriophage.

13. A bacterium of the genus, Eschericia, containing the expression vector of claim 11.

14. A bacterium of the genus, Escherichia, containing the expression vector of claim 12.

1 5. A substantially pure DNA segment consisting essentially of: (a) promoter, operator and translation initiation sequences which are recognizable by a procaryotic host organism; (b) a Shine-Dalgarno sequence and an initiator codon within the translation initiation sequence the initiator codon being proximate to the end of the translation initiation sequence distal from the promoter and operator sequences; and (c) a recognition sequence for a registration endonuclease which cleaves proximate to the initiator codon, or its complement, leaving said initiator within the translation initiation sequence.

1 6. The DNA segment of claim 15, wherein said restriction endonuclease cleaves within three base pairs of the initiator codon or its complement.

1 7. The DNA segment of claim 15, wherein the restriction endonuclease cleaves in such a manner that the translation initiation sequence ends with the initiator codon or its complement.

1 8. The DNA segment of claim 15, wherein the initiator codon is contained within the recognition sequence for said restriction endonuclease.

1 9. The DNA segment of claim 15, wherein said initiator codon is contained in the recognition sequence for the restriction endonuclease Sphl.

20. The DNA segments of any of claims 1 5 to 17, wherein the recognition sequence for said restriction endonuclease is separated from the restriction endonuclease cleaving site by a plurality of non-specific nucleotide base pairs.

21. The DNA segment of claim 20, wherein the restriction endonuclease is Hgal.

22. The DNA segment of any of claims 1 5 to 19, wherein the initiator codon is separated from the Shine-Dalgarno sequence by from 0 to about 22 base pairs.

23. The DNA segment of claim 22 wherein the initiator codon is separated from the Shine Dalgarno sequence by from about 3 to about 7 base pairs.

24. The DNA segment of any of claims 1 5 to 19, wherein the promoter, operator and Shine Dalgarno sequences are derived from the lac or trp operons of Eco!.

25. The DNA segment of any of claims 1 5 to 19, wherein the promoter, operator and Shine Dalgarno sequences are derived from a bacteriophage.