AU620873B2 - Expression vectors for the preparation of infused proteins in microorganisms - Google Patents

Description

01 S.1 of Company and Signa~tuirsof ,is Officars its PreOcribed Sy its A41oles of Asociation.

by D. B. Mischlewski 620873 Form COMMONWEALTH OF AUSTRALIA PATENTS ACT 1952-69 COMPLETE SPECIFICATION

(ORIGINAL)

Class Application Number: Lodged: I t. C~ass Cppplete Specification Lodged: 0 Accepted: Published: 13riptity: g~eated Art: 0 oNamne of Applicant: *A-dress of Applicant: SActual Inventor: Address for Service BEL-RINGWERKE AKTIENGESELLSCHAFT D-3550 Marburg, Federal Republic of Germany EGON AMANN and KAL-JOSEF ABEL EYJVA0k)M K P ,Watermark Patent Trademark Attorneys 50 QUEEN STREET, MIELBOURNE, AUSTRALIA, 3000.

Complete Specification for the invention entitled: EXPRESSION VECTORS FOR THE PREPARATION OF INFUSED PROTEINS IN

MICROORGANISMS

The following statement is a full description of this invention, including the best method of performing it known to us I Ij,..

.4 BEHRINGWERKE AKTIENGESELLSCHAFT Hoe 88/B 015 Ma 678 Dr. Lp/EBk Expression vectors for the preparation of unfused proteins in microorganisms The signals for the initiation of transcription and translation differ in prokaryotic and eukaryotic genes.

This is why, as a rule, little or no expression is observed when eukaryotic genes are transferred into prokaryotic cells. In order to achieve efficient expression of eukaryotic genes in prokaryotes, the latter are often ligated to 5' portions of prokaryotic genes, which presupposing that the two translation reading frames coincide results in the synthesis of fusion proteins.

It is then possible, by BrCN cleavage, acid cleavage or °'45 else by insertion of specific protease cleavage sites between the two protein portions of the fusion protein and subsequent protease treatment, to extract the desired eukaryotic protein portion from the fusion protein. This is as a rule possible only for smaller proteins, because ,0 BrCN cleavage and acid cleavage are unsuitable for larger proteins by reason of the large number of resulting 4444 o protein fragments. A common problem with protease cleavage is the low yields of the desired protein and the high costs of the protease which, moreover, can be employed tl t,5 only when very pure. The present invention therefore has the object of obtaining foreign proteins in high yield in the unfused state by direct exPression in prokaryotes by means of particularly suitable nxpression vectors.

The expression of a cloned gene requires efficient transcription thereof and efficient translation of the mRNA thereof. Necessary for the latter is the presence of a particular sequence in the mRNA, which is called the "ribosome-binding site" (RBS). This site is composed of a region which is 3-9 base-pairs long and is called the "Shine-Dalgarno sequence" (SD) (Shine and Dalgarno (1975) Nature 254, 34-38) and which is followed at a distance of 3-11 base-pairs by an ATG start codon which encodes the i .in 14 i 2 amino-terminal methionine. The SD sequence is complementary to the 3' end of the 16S rRNA, and there is probably duplex formation between these two RNA sequences, which in turn is regarded as the prerequisite for initiation of translation (Steitz (1979) in: Biological Regulation and Control, R. Golderger ed., New York, Plenum Press). This is why, as a rule, complicated manipulations must be carried out immediately upstream of the eukaryotic ATG start codon in order to express a heterologous unfused protein. Thus, for example, upstream insertion of the missing prokaryotic RBS is needed to ensure efficient binding of the 5' end of the mRNA with the 3' end of the 16S ribosomal RNA. The combination of the fused prokaryo- ,tic SD sequence with the eukaryotic ATG start codon and a few bases of the following coding sequence is called the "hybrid ribosome-binding site". Both the distance from the prokaryotic SD sequence to the ATG start codon and the base composition upstream and downstream of the SD sequence represent crucial parameters for maximizing the expression. Since there are as yet no reliable computer programs for predicting the ideal sequence of the hybrid ribosome-binding site for high-level expression, the latter must be determined experimentally. This can take place, on the one hand, by use of genetic methods using p-galactosidase (Guarente et al. (1980) Cell 543-533) or, on the other hand, by employing .ynthetic oligodeoxynucleotides. The latter fre-uently requires empirical testing of many different sequences in order to achieve maximum expression of the desired gene product.

One of the genes with the highest level expression in E.

coli is the product of the lacZ gene, p-galactosidase.

The amount of p-galactosidase may account for up to 5 of the total protein in haploid strains and up to 20 in partially diploid strains (Zabin and Fowler (1978) in: The Operon, Miller, Reznikoff, ed., Cold Spring Harbor Laboratory). The present invention is now based on the assumption that a protein with high-level expression, such as p-galactosidase, also has an mRNA optimized for -1-4t I IA 1 nrotains in E. coli. containini 3 high-level expression and possesses there in particular an optimized RBS. The invention makes use of the RBS, which has been selected in a "natural" way for high-level expression, of the lacZ mRNA in conjunction with a synthetic DNA sequence inserted upstream, in order to achieve a high yield of expressed foreign protein in unfused form. The following representation compares the relevant 5' region of the natural lacZ sequence (Gilbert and Maxam (1973) Proc. Natl. Acad. Sci. USA 70, 3581- 3584; Maizels (1973) Proc. Natl. Acad. Sci. USA 70, 3585- 3589; Dickson et al. (1975) Science 187, 27-35) with the corresponding sequence of the new vectors: Sequence of the lacZ DNA: 5' CAATTTCACACAGGAAACAGCT ATG A 3' Seqe]nce ofthe] veto DNA: 5 1 Sequence of the vector DNA:5' CAATTTCACACAGGAAACAGACC ATG G 3' The synthetic DNA which has been inserted as doublestranded oligodeoxynucleotide downstream of the lacZ RBS has the sequence: 5' CCATGG 3' S3' GGTACC and thus the recognition region for the restriction endonuclease NcoI. The synthetic DIA is positioned in such a way that the 'ATG' start codon which is contained in the NcoI recognition site is located at a listance Sfrom the prokaryotic RBS which is favorable for highlevel expression. The NcoI site occlrs in the plasmid only once.

The NcoI site can now be made use of for high-level expression of foreicrn enes in a varie'y of ways. One aspect of the invention comprises the plasmid vectors being particularly suitable for the direct expression of eukaryotic genes. This is based on the fact that many eukaryotic genes likewise have an NcoI site at their ATG start codon. The concensus sequence of the 5' region F 1 ^t 1

I

4 immediately upstream of the ATG start codon of eukaryotic mRNA is stated by Kozak (Marilyn Kozak (1987) Nucleic Acids Research 15, 8125-8148):

G

(GCC)GCCACCATGG

Of 699 sequences of eukaryotic mRNA identified by Kozak, 124 17.7%) carry an NcoI site at their ATG start codon. These can be expressed directly in the new vectors. In 194 of the remaining 575 cases 33.7%) the second codon starts with a (possible amino acids at this position: Gly, Glu, Asp, Ala, Val). The DNA sequen- 1 ces of these 33.7% can be converted, while retaining the amino acid sequence, by mutagenesis upstream of the ATG into an NcoI site. Thus, it is possible in 51.4% of all *the 699 eukaryotic genes investigated by Kozak to obtain, either directly or after mutagenesis DNA .d0 fragments which can be cloned in the new prokaryotic I vectors and whose expression can be brought about. The expression products are unfused and unchanged in their 4 4'o 4amino acid sequence. The remaining 48.6% of the eukaryotic genes can, of course, after introduction of an NcoI site by mutagenesis also be expressed in the new vectors, S. but in these cases the second amino acid is changed from x into Gly, Glu, Asp or Ala.

Another possibility of cloning in the new vectors is provided by cutting the sequence with NcoI and subsequently filling in the overlapping end enzymatically using Klenow polymerase. This results in an exposed 'ATG': t CCATG 3' 3' GGTAC It is now possible to ligate onto this exposed 'ATG' any foreign DNA which likewise has a blunt end, in which case there is no longer a need for its own ATG to be present.

SPst Hind III t

P

i l *-t 5 If this DNA carries the complete information or parts of -the information of a structural gene, these can be expressed as long as they are in the correct reading frame with respect to the manipulated 'ATG' vector codon.

The specific nature of this cloning permits, as already described above, the expression of the desired protein without additional or modified amino acids. Another possible application is provided by the use of commercially available NcoI linkers, which can be obtained in various lengths (8mer, 10mer, 12mer) (Analects Vol. 14, No. 1; Pharmacia) and can be used to combine the "open reading frame" of any desired structural gene with the ATG start codon of the new vectors. The use of NcoI linkers also permits the employment of the new vectors for the expression of cDNA banks for immunoscreening Susing known methods (for example Broome and Gilbert S(1978) Proc. Natl. Acad. Sci. USA 75, 2746-2749; Erlich et al. (1978) Cell 13, 681-689; Kemp and Cowman (1981) Proc. Natl. Acad. Sci. USA 78, 4520-4524).

In order to make available a large number of restriction cleavage sites downstream of the NcoI site, the EcoRI- HindIII polylinker region of the commercially available vector pUC18 was inserted into the plasmid vectors downstream of the NcoI site (see Fig.). It is consequently possible to make use of a wide variety of cleavage sites in the 3' region of the gene which is to be 9 expressed, in order to guarantee directed cloning.

In a few specific examples of the applications of the new vectors, it is also possible to use the EcoRI site to express an unfused foreign protein. For this purpose, the vector DNA can be cut with EcoRI and the protruding single-stranded end be digested off with nucleases specific for single strands, for example with mung bean nuclease. This results in the following sequences: 5' AACAGACCATGG 3' in pTrc99A L I*I BBHBBIB BBBHIBB 6 AACAGACCATGGG 3' in pTrc99B AACAGACCATGGGG 3' in pTrc99C It is now possible to ligate onto these ends a foreign DNA with blunt ends in such a way that the reading frame conforms with the ATG start codon predetermined by the vector DNA. This manipulation may result, in the individual case, in there being no foreign amino acids at the N terminus of the expressed heterologous protein. In other individual cases there may be a foreign amino acid fused N-terminally to the heterologous protein downstream of the initiator methionine.

i Another possible application of the new vectors is ,'"015 provided by the fact that the abovementioned polylinker is available in all three reading frames with respect to S, the ATG start codon of the NcoI site. The consequence of this is that DNA fragments of structural genes which have been obtained by hydrolysis of restriction enzymes I .:20 occurring in the polylinker region can be cloned and 0i 0 expressed in the correct reading frame with respect to the vector ATG. However, this possible application S, applies only to the case where the presence of a few foreign vector-encoded amino acids at the N terminus of the desired protein creates no disturbance. For this type of "short fusion expression" the new vectors at least offer the advantage of an expression rate which is about 5-10 times higher than with similar lac promoter vectors (De Boer et al. (1983) Proc. Natl. Acad. Sci. USA 80, 21- 25; Amann et al. (1983) Gene 25, 167-178).

Brief description of the vectors The vectors all possess the trc promoter, a derivative of the tac promoter with comparably high activity (Mulligan et al. (1985) J. Biol. Chem. 260, 3529-3538; Brosius et al. (1985) J. Biol. Chem. 260, 3539-3541). The tac and

I'

ii 7 trc promoters differ in a single base: whereas the distance between the -35 and -10 regions is 16 base-pairs in the tac promoter, these two regions in the trc promoter are 17 base-pairs apart. The tac promoter and its activity by comparison with lac, trp and tet promoters have already been described (De Boer et al. (1983) Proc.

Natl. Acad. Sci. USA 80, 21-25; Russel and Bennett (1982) Gene 20, 231-243; Amann et al. (1983) Gene 25, 167-178).

The vectors use, downstream of the trc promoter, the lacZ ribosome-binding site, followed by an NcoI site which contains a translation start codon. Also following are the EcoRI-HindIII polylinker from pUC18 as well as the transcription terminators of the rrnB operon (Brosius et al. (1981) J. Mol. Biol. 148, 107-127). The "plasmid backbone" is composed of a pBR322 portion which embraces Sa ColEl-type origin of replication as well as the J ampicillin-resistance gene (bla gene). Since there has been deletion of the pBR322 DNA up to position 2066 (PvuII site), and thus of parts of the rop region, in the '0 vectors, there is an increase in the plasmid copy number (Balban et al. (1986) Gene 50, 3-40; Twigg and Sheratt (1980) Nature 283, 216-218) which has a beneficial effect on the level of expression ("gene dose effect"). In some lacI q host strains there is observed to be partial derepression of the trc promoter, which may be ascribed to the increased copy number of the new vectors by comparison with the known tac promoter vectors ptacll and ptacl2 (Amann et al. (1983) Gene 25, 167-168). This is the reason for the construction of the vectors according to the invention of the pTrc99 series, which carry as additional information the lacI q allel. These vectors show complete repression of the trc promoter in the absence of inducer even in E. coli host strains which carry no lacI' allel. This is why these vectors are especially suitable for expressing gene products toxic for E. coli, because there is no basal expression, with the consequence of negative selection, during the cloning and cultivation phase. The table shows the complete nucleotide sequence for pTrc99A: 8- Table 1 51 101 151 201 251 301 351 401 4 451 501 551 4 651 4 4 75 44 4 601 4 4 14 1 4 651 7001 10751 41101 1141 12801 124 51 9301

GTTTGACAGC

GGCAGCCATC

TAATTCGTGT

C CGACAT CAT

TAATCATCCG

ACACAGGAAA

GAGTCGACCT

ATTTTCAGCC

ACAGAATTTG

CGAACTCAGA

CATGCGAGAG

CGAAAGACTG

CTGAGTAGGA

GCCCGGAGGG

TrTAAG CAGAA

CTTTTTGTTT

CAATAACCCT

TATTCAACAT

TTC CTGTTTT

GATCAGTTGG

TAAGATCCTT

CTTTTAAAGT

CAAGAG CAAC

GTACTCACCA

AATTATGCAG

CTTCTGACAA

CATGGGGGAT

TTAT CAT CGA

GGAAGCTGTG

CGCTCAAGGC

AACGGTTCTG

GCTCGTATAA

CAGAC CATGG

GCAGGCATGC

TGA-ACAGAT

C CT G GCGG CA AG'TGAAkACGC

TAGGGAACTG

GGCCTTTCGT

CXAATCCGCC

TGGCGGGCAG

GGCCATCCTG

ATTTTTCTAA

GATAA.ATCCT

TTCCGTGTCG

TGCTCAC CCA

GTGCACGAGT

GAGAGTTTTC

TCTGCTATGT

TCGGTCGCCG

GTCACAGAAA

TGCTGC CATA

CGATCGGAGG

CATGTAACTC

CTGCACGGTG

GTATGG'CTGT

G CAC TC CC GT

GCAA-ATATTC

TGTGTGGAAT

AAT TC GAG CT

PLAGCTTGGC"T

TAAAT CAGAA

GTAGCGCGGT

C GTAG C GCC G C CAGG CAT CA

TTTATCTGTT

GGGAGCGGAT

GACGCCCGCC

ACGGATGGCC

ATACATTCAA

T CAATAA TAT CC CTTATTC C GAA-AC GCTGG

GGGTTACATC

GCCCCGAAGA

GGC GC GGTAT CATACAC TAT

AGCATCTTAC

ACCATGAGTG

AC CGAAGGAG

GCCTTGATCG

CAC CAAT GCT

GCAGGTCGTA,

TCTGGATALAT

TGAAATGAGC

TG TGAG C GGA C GG TACC CG G

GTTTTGGCGG

CGCAGAAGCG

GGTCCCACCT

ATGGTAGTGT

AATAAAAC GA

GTTTGTCGGT

TTGAACGTTG

ATAAACTGCC

TTTTTGCGTT

ATATGTATCC

TGAAAAAGGA

CTTTTTTGCG

TGAAAGTAAA

GAACTGGATC

ACGT7TTCCA

TATCCCGTGT

T CT CAGAl.ATG

GGATGGCATG

ATAACACTGC

CTA.ACCGCTT

TTGGGA.ACCG

TCTGGCGTCA

AATCACTGCA

GTTTTTTGCG

TGTTGACAAT

TXACAkATTTC GGATC CT C-TA

ATGAGAGAAG

GTCTGATAAA

GACCCCATGC

GG GGT C TC CC

AAGGCTCAGT

GAACGCTCTC

CGAAGCAACG

AGGCATCAAA

TCTACAAACT

GCTCATGAGA

AGAGTATGAG

GCATTTTGC C

AGATGCTGAA

TCAACAGCGG

ATGATGAGCA

TGACGCCGGG

ACTTGGTTGA

ACAGTAAGAG

GGCCAACTTA

TTTTGCACAA

GAG CTGAATG 1351 AAGCCATACC AAACGACGAG CGTGACACCA CGATGCCTAC AGCAATGGCA 9- I I I I

II~~

I'

I I

II~I

II

I I

I~II

I I I

II

I II I I

'I

1401 1451 1501 1551 1601 1651 1701 1751 1801 1851 1901 1951 2001 2051 2101 2151 2201 2251 2301 2351 2401 2451 2501 2551 '2601 2651 2701 2751

ACAACCTTC

GCAACA-ATTA

TGCC CT CGCC CCT GAG CGCTG

CCCCTCCCGT

ATGAAC GAAA

TGGTAACTGT

ACTTCATTTT

TCATGAC CAA CCC GTACAAA

AATCTGC-TGC

TGCCGGATCAk

AGACCAGA

CCACTTCA.AG

TGTTACCACT

GACTCAAGAC

GGGTTCGTC

GATACCTACA

AAGGCGGACA

GAGGGAGCTT

TTCGCCACCT

CGGAGCCTAT

CTTTTGCTGG

CTGTGGATAA

AGCCGAACGA

C CTGATGCGG

TATGGTGCAC

TATACACTCC

G CAAAC TAT T

ATAGACTGGA

C CTTC CCGCT b G TC T CGCCGG ATCGCTAG TTA

TAGACACATC

CACAC CAAGT

TAATTTAAAA

AATCCCT"TAA

AGATC.AAAGG

TTGCAAACAAN

AGACCTACCA

TAC CAAATAC

AACTCTCTAG

GCCTCCTGCC

GATAGTTACC

ACACACCCCA

GCGTGACCTA.

GGTAT C CCT

CCAGGGGGAA

CTGACTTGAG

GGAAAAACGC

CCTTTTGCTC

CCGTATTACC

C CGAGCGCAG

TATTTTCTCC

TCTCAGTACA

GCTATCGCTA

AACTGGCGAA

TC GAGCGCGA CGC TCGTT TA TAT CATTG CA

TCTACACCAC

CCTCACATAG

TTACTCATAT

CCATCTACCT

CCTCACTTTT

AT CTT CTT CA A.AAAAC CAC C

ACTCTTTTTC

TCT C CTT CTA CAC CCCCTA C

AGTGCCATA

CGATAAGCG

GCTTCCACC

TGACAAAC

AACCCAGG

ACGCCTGGCTA

C CT C ATTTT

CAGCA.ACC

ACATGTTCTT

GCCTTTGAGT

CGAGTCAGTG

TTACGCATCT

ATCTGCTCTG

CGTGACTGGG

CTACTTACTC

TAAACTTGCA

TTGCTGAI'AA

C CAC TGGG C CCCCACT CAC

CTCCCTCACT

ATACTTTACA

GAACATCCTT

CCTTCCACTC

CATC CTTTTT

CCTACCACCG

CCA-ACCTAAC

CTGTACCCT

ATAC CT CGCCT

ACTCCTCTCT

CACCGTCCC

AAC GAG CTAC C CACGCCTTC C

GTCCCAACAC

TCTTTATACT

TGTGATGCTC

GCCTTTTTAC

TC CTG CGTTA

GAGCTGATAC

AGCGAGGAAG

GTGCGGTATT

ATGCCGCATA

TCATGGCTGC

TAG CTTC CC C

CCACCACTTC

AT C TC ACCC

CACATGGTAA

GCA.ACTATG

GATTAACCAT

TTCATTTAAA

T T TGATAkAT C

ACCTCACAC

TTCTCCCT

CTCCTTTCTT

TGCCTTCAC

AGTTACCCCA

CTC TAAT CC TAGC CGCTT C CCTGAAC CCC

ACCGAACTGA

CGAACCGAGA

GAGAGCGCAC

C CTr-TCGCC T

GTCAGGGGG

GGTTCCTGGC

TCCCCTGATT

CG CT CGCCGC C GGAAGAC

TCACACCCA

GTTAAGC CAG

GCCCCGACAC

i Ii

I.

V

~Af1 r Cf

II

1a S S S

SI

4155 S C C S SC CS S S S 45 4 5 44 54 4 55 C s 2801 2851 2901 2951 3001 3051 3101 3151 3201 3251 3301 3351 3401 3451 3501 3551 3 60 1 3651 3701 3 7731 3801 3851 3901 3951 4001 4051 4101

CCGCCAACAC

CC CT TACAGA

TTTCACCGTC

AAGC CA.AG C

TTTCGCCGTA

GAA TCT GAAA

CTTATCACAC

AAAkAC GC GGG

CAACCGCGTG

TTG CCA CC'T C ATTAAAT CT C

AGAACGAAGC

CC A C GC CT

GCCATTGCTG

TGTCTCTGAC

GTACGCGACT

GCGCTGTTAG

TGCGC TGGCAT

GGGAA.GGCGA

CTGAATGAGG

GGCGCTGGGC

CGGATATCTC

ATCCCGCCGT

CAGCGTGGAC

ATCAGCTGTT

AATAC GCAAA

GGCACGACAG

CCGCTGACC

CACCTCTCA

ATCACCCAAA

&,GCATGCATT

TGG CAT GA TA C CAGTAACGT C GTTTC C CGC

AAA.AAGTGGA

GCACAACAAC

CAGTCTGGCC

GC-GCCGATCA

GGCCTCGAAG

CAGTGGGCTG

TGGAAGCTGC

CAGACAC CCA

GGGCGTGGAC

C GGC CCATT

AAATATCTCA

CTGGAGTGCC

GCATCGTTCC

GCAATGCCC G GGTkGTGGGA, CAAC CAC CAT

CGCTTGCTGC

GCCC CGTCTCA

CCGCCTCTCC

GTTT CC CGAC

GCCCTGACGG

C C iT CT C C G

CGCCCAGGC

TACCTTGACAk G CCC C GAA

TATACCATGT

GTGGTGAAC C

AGCGGCGATG

TGG C GG GCAA C TGCACGC CGC

ACTGGGTGCC

C CTG TAAA GC

ATCATTAACT

C TG CAC TAAkT

TCAACAGTAT

CATCTGGTCG

AAGTTCTGTC

CTCGC CAT CA

ATGTCCGGTT

CACTGCGATG

CCATTACCGA

TACGACGATA

CAAACAGGAT

AACT CT CTCA

CTGGTGAAAA

CCGCGCGTTG

TGGAAAGCGG

GCT-TGTCTGC

GAGCTGCATG

AGCA--GATCAA

CCATCCALATC

GAGAGTCAAT

CGCAGAGTAT

AGGCCAGCCAk GC GAGC TGA

ACAGTCGTTG

CGTCGCAAAT

A CGTGGTGG

GGCGGTGCAC

ATCCGCTGGA

CTTCCGGCGT

TATT TTC TC C

CATTGGGTCA

TCGGCGCGTC

AATTCAGCCG

TTCAACAAAC

CTGGTTGCCA

GTCCGGGCTG

C CGAAGACAG

TTTCGCCTGC

GGGCCAGGCG

GAAAAAC CAC

GCCGATTCAT

GCAGTGAGC C

TCCCGGCATC

TGTCACGAGGT

TTCC CC C

GTGCAAAACC

TCAkGCGTGGT

GCCGGTGTCT

CGTTTCTGCG

ATTACATTC C CT CATTGC C

TGTCGCGGCG

TGTCGATGGT

AAT CTT CTC G

TGACCAGGAT

TATTTCTTGA

CATGAAGACG

CCAGCAAATC

TGCGTCTGGC

ATAG CGGAAC

CATCCAAATG

ACGATCAGAT

CGCGTTGGTG

CTCATGTTAT

TGGGGCAAAC

GTGAAGGGCA

CCTGGCGCCC

TAATGCAGCT

CAACGCAATT

4151 AATGTGAGTT AGCGCGAATT GATCTG 1 11 Example 1: Construction of the expression vectors pTrc99A, pTrc99B and pTrc99C pTrc89-1 (European Patent Application EP 0,236,978) was linearized with BglII, and the overlapping ends were filled in using Klenow polymerase. The DNA was subsequently treated with phosphatase. The plasmid placIl (Wang et al. (1983) Cold Spring Harbor Symp. Quan. Biol. 47, 85-91) was digested with EcoRI, the overlapping ends were likewise filled in using Klenow polymerase, and the fragment which is about 1100 base-pairs in size and.

carries the complete lacI allel (Calos (1978) Nature 274, 762-765) was ligated and transformed with the linearized pTrc89-1 DNA. The resulting plasmid (with the direction of transcription of the lacI gene identical to the direction of transcription of the trc promoter) was pTrc90-3. The junctions of the vector DNA with the lacI q fragment were sequenced; this information is also included in the sequence in the table. The sequence of I'O0 pTrc90-3 embraces 4134 base-pairs. pTrc90-3 was digested with NcoI and HindIII, and the large fragment was ligated with the synthetic oligonucleotides A, B or C in three separate mixtures and subsequently transformed into competent cells of the E. coli K12 strain W31101ac': Oligo A 111 1111 CTTAAGGCTTCGA Oligo B 5' CATGGGAATTCGA 111111111 CCTTAAGCTTCGA Oligo C 5' CATGGGGAATTCGA CCCTTAAGCTTCGA 'i 12 The resulting plasmids pTrc98A (oligo pTrc98B (oligo B) and pTrc98C (oligo C) were digested with EcoRI and HindIII, and the large fragment was isolated in each case and ligated with the EcoRI-HindIII polylinker fragment, which is 55 base-pairs in size, of the commercially available vector pUC18 (Yanisch-Perron et al. (1985) Gene 33, 103-119). The resulting plasmids are pTrc99A (4176 base-pairs), pTrc99B (4177 base-pairs) and pTrc99C (4178 base-pairs). The consequence of the ligation of the synthetic oligonucleotides A, B and C is that the vectors differ respectively by one base between NcoI and the EcoRI recognition site, which in turn results in a shift in the translation reading frame. The presence of the lacI gene on the expression vector (pTrc99 series) is of interest when the E. coli strains to be used for expression possess no endogenous lac repressor, and in those cases where the expressed protein exerts a toxic effect on the E. coli host cell and, for this reason, complete repression is necessary in the phase of cloning and of ,210 cultivation. The figure shows the scheme of construction of the vectors.

Example 2: Expression of the calcium-binding protein calmodulin Calmodulin is a highly conserved calcium-binding protein which occurs ubiquitously in eukaryotes (Baba et al.

(1984) Mol. Biol. Evol. 1, 442-455; Klee et al. (1980) Annu. Rev. Biochem. 49, 488-518; Means et al. (1982, SPhysiol. Rev. 62, 1-39). It is involved in many calciumcontrolled intracellular bioregulatory processes, including the regulation of cyclic nucleotides (Cheung (1980) Science 207, 19-27; Klee et al. (1980) Annu. Rev.

Biochem. 49, 448-518). The DNA sequence of calmodulin cDNA isolated from a rat brain cDNA gene bank is known (Sherbany et al. (1987) DNA 6, 267-272). It codes for a protein with 149 amino acids. Although the calmodulin cDNA sequence of the rat differs by 66 nucleotides from .i t.

I

13 the human cDNA, there is 100 homology of the derived amino acid sequences. This shows that there have been "silent exchanges" and that the protein must indeed be a protein which has been highly conserved during evolution.

The calmodulin cDNA of rats has an NcoI site at its start codon: Met Ala Asp Gin GTGCCTCGCC ATG GCT GAC CAG 3' To express the calmodulin cDNA in E. coli, the clone prCM79 was digested with NcoI and SmaI, and the f-..gment 457 base-pairs in size was isolated and ligated into the correspondingly digested vector pTrc99A. The resulting oi plasmid pTrc99A-Cal expresses after induction in E. coli Sthe calmodulin protein which is about 17 kD in size.

a0t I nD

I

1 B t 0 1 4 B s a44

Claims (6)

1. An expression vector for the expression of foreign proteins in E, coli, containing: a) the trc promoter, b) the lacZ ribisome-binding site, c) an Ncol cleavage site which contains a translation start codon located downstream of the laz Z ribosome-binding site, d) a restriction enzyme cleavage site polylinker in all three reading frames, e) plus transcription terminators, f) with the abovementioned elements being integrated in a plasmid backbone with origin of replication, antibiotic-resistance gene and the laclq allel. i t S4 4 t I C t t I t t "t11 i^ i proteins in E. coli, containing i^ -1-4- IIPTE JLAI- ID DIINNf Th 1hE f3ARE G FOLLOW3. proteins in E. coli, containing a) the trc promoter, b) the lacZ ribosome-binding site, c) an NcoI cleavage site whic contains a trans- lation start codon, d) a restriction en ae cleavage site polylinker in all three r ing frames, e) plus t scription terminators, f) w the abovementioned elements being integrated Sin a plasmid backbone with origin of replication, antibiotic-rsiatance gene and the lacdQ allel. t 2. An expression vector as claimed in claim 1, with pBR322 DNA as plasmid backbone.

3. An expression vector as claimed in claim 2, with deletion of the pBR322 DNA up to position 2066.

4. An expression vector as claimed in claim 1, with the EcoRI-HindIII polylinker from pUC18. An expression vector as claimed in claim 1, with transcription terminators of the rrnB operon.

6. P.asmids pTrc99A, pTrc99B and pTrc99C.

7. A process for the preparation of expression vectors as claimed in claim 1 to 6, which comprises ligation of the said elements to

8. The use of expression vectors as claimed in claim 1 to 6 for the expression of foreign proteins in E. coli. DATED this 6th day of June 1989. BEHRINGWERKE AKTIENGESELLSCHAFT WATERMARK PATENT TRADEMARK ATTORNEYS i 50 QUEEN STREET i E CMELBOURNE. VIC. 3000. CE K