CA1341124C

CA1341124C - Genetic engineering process for the preparation of hirudins, and means for carrying out this process

Info

Publication number: CA1341124C
Application number: CA000488468A
Authority: CA
Inventors: Dieter Brauer; Paul Habermann; Eugen Uhlmann; Friedrich Wengenmayer; Joachim Engels
Original assignee: Hoechst AG
Current assignee: Bayer Pharma AG
Priority date: 1984-08-10
Filing date: 1985-08-09
Publication date: 2000-10-24
Anticipated expiration: 2017-10-24
Also published as: KR870002257A; FI853056A0; NZ213049A; NO853155L; IE58485B1; NO177434B; EP0171024A1; ES545994A0; AU4597785A; DE3429430A1; IE851973L; AU584628B2; HUT38968A; GR851962B; DK175337B1; IL76059A; ATE67245T1; ES8605040A1; KR930009337B1; DK364585D0

Abstract

It is possible to obtain hirudin in a genetic engineering process using a specific DNA sequence. The gene is advantageously synthesized in the form of several fragments which are ligated enzymatically to give two larger part sequences which are incorporated in hybrid plasmids and amplified in host organisms. After re-isolation of the part sequences, they are combined to give the total gene which is incorporated into a hybrid plasmid, and expression of the latter is induced in a host organism.

Description

a ' A GENETIC ENGINEERING PROCESS FOR THE PREPARATION OF
HIRUDINS, AND MEANS FOR CARRYING OUT THIS PROCESS
Hirudin is a polypeptide which is obtained from Hirudo medicinalis, has a specific antithrombin activity and is used as an anticoagulant.
It has now been found that polypeptides of the general foreula I
(X)~ A-B-C-Tyr~-D-Asp-Cys-E-Glu-Ser-Gly-Gln-Asn-Leu-Cys-Leu--Cys-Glu-Gly-Se~r-Asn--Val-Cys-Gly-Gln-Gly-Asn-Lys-Cys-Ile--Leu-Gly-Ser-Asp-F-G-Lys-Asn-Gln-Cys-Val-Thr-Gly-Glu--Gly-Thr-Pro-Lys-Pro-Gln-Ser-His-Asn-Asp-Gly-Asp-Phe-Glu--H-Ile-Pro-Glu-Glu-Tyr-Leu-Gln-(Z)n OH
in which (I) '10 0 - 50, m =

n = 0 - 100, rE:ferably 0 and p X represe nts dentical different residues of gene-i or tically coda ble amino cids, a A denotes Ile or a directbond, '15 denotes Ile, Thr or a irect bond, B d C denotes 'Thr,Val, Ile, Leu or Phe, D denotes 'Thr or a directbond, E denotes 'Thr or Ile, F denotes Gly or a directbond, ~?0 denotes Glu or a directbond, G

H denotes Glu or Pro, and Z represe nts dentical different residues of i or genetic ally codable no acids, and ami in whichthe 6 Cys residues may be -e'.5 linked pairwise ~~ia disulfide bridges, can aso be prepared by genetic engineering when a gene which codes for a polypeptide of tl~e formula I or part sequences thereof is incorporated in ~~n expression plasmid. This polypeptide is also called "hirudin" in the following text.
:S 0 - lA -The invent_Lon relates co a' process for the preparation of a polypeptide of the Formula I, as defined above, which compr~~ses incorporation into an expression plasmid of a gene which codes for a polypeptide of the Formula I, it being possible for other amino acids to be replaced by one or more genetically codable amino acids.
The invention also relates to the DNA sequence I
(see Page 18) which codes for the entire amino acid sequence of the pol.ypept:ide of the Formula I, the DNA
sequence IIA (see F~age 20) and IIB (see Page 21) which are used for the synthesis of sequence I, and the DNA
sequences IA, IB ar,.d IC (see Page 19).
The invention will now be described in further detail by reference to the appended drawings:
Figure 1 shows a hybrid plasmid containing the gene fragment HIR-I.
Figure 2 shows a hybrid plasmid containing the gene fragment HIR-II.
Figure 3 shows a hybrid plasmid containing the DNA sequence I.
Figure 4 shows tree expression plasmid pCK-5196.
Figure 5 shows a hybrid plasmid containing DNA
sequence IA.
Figure 6 shows a hybrid plasmid containing DNA
sequence IB.
Figure 7 shows the DNA-sequence I, encoding the complete amino aci<i sequence of hirudin.
Figure 8 shows the DNA-sequences IA, IB and IC, encoding the amino acid sequences from position 1 to 64 of hirudine and the: amino acids Met, Arg and Met in position 0, respectively. DNA-fragment with the DNA-sequence IC is used to produce a synthetic gene for the expression of hiru<iin according to Figure 7. The DNA-fragments with the DNA-sequences IA and IB are used to produce synthetic genes for the expression of hirudine-fusion proteins. ~'he non-hirudine part of such fusion proteins can be removed by treatment with BrCN (DNA-sequence IA) or by proteolysis of the fusion protein with :~...

trypsin (DNA-sequence IB).
Figure 9 shows the DNA-sequence IIA of a DNA
fragment used for the synthesis of a DNA-fragment with the DNA sequence I.
Figure 10 :3hows 'the DNA-sequence IIB of a DNA-fragment used for the synthesis of a DNA-fragment with the DNA-sequence I.
The requis:Lte gene can be chemically synthesized by known methods, isolated from the genome and processed, X341 X24 _ or the mRNA can be isolated from induced cells by known methods and the cDNA can be obtained obtainE~d from this.
The route via cDNA is preferred, and especially chemical synthesis, in particular by the phosphate method. It is furthermore preferred to synthesize a gene which codca for the polypeptide of the formula I, in which m is 1 and n is zero, X represents Met or Arg, B represents Thr or a direct bond, C represents Thr, and G represents Glu. Another preferred embodiment of the invention relates to polypeptides of the formula I, in which (X)m-A-B-C- does not represent Val-Val- when n is zero, D and E denote Thr, F
1o denotes Gly, and G and H denote Glu.
It is particularly preferred to synthesize polypeptides of the formula I, in which m is 1, X denotes Met or Arg, A and B denote direct bonds, C, D
and E denote Thr, F denotes Gly, G and H denote Glu, and n denotes zero.
The DNA sequence I which is preferred for this purpose is shown at Page 18 and is used in the following text to illustrate the invention.
It is known that the genetic code is "degenerate", i.e. only two amino acids are coded for by a single nucleotide sequence, whereas 2 to 6 triplets are assigned to the remaining 18 genetically codable amino acids.
Moreover, the host cells of different species do not always make the same use of the possibilities of variation resulting from this. Thus, there is an immense variety of codon possibilities for the synthesis of the genes.
It has now been found that the DNA sequence I (see Page 18), which codes for the entire amino acid sequence, and the DNA part sequences II A and I I B, (sc:e Pages 20 and 21 ) which are used for the synthesis of sequence I, are particularly advantageous for the genetic engineering synthesis of the particularly preferred form of hirudin. A
"protruding" DNA sequence corresponding to the restriction endonuclease Xbal is located at the ',i' end of the coding strand of DNA sequence I, whereas the single-stranded ~~rotrudir~g sequence corresponding to the restriction 3 o enzyme Sal I is located at the 3' end of the coding strand. These two ?341 124 different recognition sequences ensure the insertion of the DNA in plasmids in 'the desired orientation. (However, it is also possible to aelect identical recognition sequences and to undertake appropriate selection after 5~ characterization of the plasmid DNA by appropriate restric-tion or DNA sequence analysis or by expression.) Between these recognition sequences and the codons for the amino acid sequence is located at the 5' end of the coding strand the codon for the amino acid methionine 1Ci (Which is numbered 0 in DNA sequence I). This can be replaced by a pre:~equence (also called a signal or leader sequence) of a bacterial or other host-intrinsic protein (review article: Pe n man and Halvorson; J. Mol. Biol.
167 (1983), 391), which brings about the secretion of the 15 desired polypeptide from the cytoplasm, and in this excre-tion process is sE~lit off by a signal peptidase naturally occurring in the host cell. Then, according to the inven-tion, at the end of the coding strand the triplet coding for glutamine is lEOllowed by a stop codon or, preferably, 2CI as shown in DNA serquence I, two stop triplets. In addi-tion, a nucleotide' sequence corresponding to the restric-tion enzyme SstI its incorporated in DNA sequence I between these stop codons and the protruding SaII end.
A unique iinternal cleavage site for the restric-25~ tion enzyme Bam H:I (in codon 30/31) permits the subcloning of two gene fragments HIR-I and HIR-II (see DNA sequence II), which can be incorporated into cloning vectors which have been thoroughly investigated, such as, for example, pBR 322, pUC 8 or pUC 12. In addition, a number of 3C1 other unique recot~nitio~n sequences for restriction enzymes are incorporated within the gene, and these provide, on the one hand, acccess to part sequences of hirudin and permit, on the other hand, variations to be carried out (Table 1):

_ 4 _ 1341 124 Table 1:
Restriction Cleavage after nucleotide No.
enzyme (coding strand) Rsa Ia) 8 Acc I 12 Hinf I 27 Xho IIa) 57 Fnu 4 HI 71 Hae III 73 1C1 est NI 75 Hph I 117 Rsa Ib) 137 Kpn I 139 Taq I 172 Xho IIb) 178 Dde I 184 Mbo II 186 Sst I 210 a) unique wii:h respect to part sequence HIR-I
2~~ b) " " " " " " HIR-II
DNA sequence I can be synthesized from 14 oligo-nucleotides with Lengths of from 25 to 35 nucleotides (see DNA sequence II), by first chemically synthesizing them and then linking them enzymatically via "sticky ends" of 4 to 6 nucleotides..
In addition, in DNA sequence I account has been taken of the fact that 'for those amino acids to which several codons arer assigned the latter are not equivalent but, on the contrary, they show different preferences in the particular host cell, such as E. coli. Furthermore, palindromic sequences have been reduced to a minimum.
Thus, the gene structure of DNA sequence I is readily accessible from relatively small units, it makes possible the subcloning of two gene fragments into well known vectors, anci it permits their combination to give the total gene and, where appropriate, modifications thereof.
Depending on the incorporation of the synthetic gene into the cloning vector, the desired peptide having the amino acid sequence of hirudin is either expressed directly (DNA
sequence IC, see Paye 19) or is expressed in the form of a fusion protein with a bacterial protE~in such as f3 - galactosidase or partial sequences thereof. These fusion proteins can then be chemically or enzymatically cleaved in a known m;~nner. Ivf, for example, the amino acid methionine (DNA
sequence IA, see page 19) is located in position 0 of sequence I, chemical cleavage with cyanogen bromide can be carried out, and if the amino acid arginine (DNA sequence IB, see page 19) is located in this position, then enzymatic cleavage with trypsin can be carried out.
1o Preferred variations of the amino acid sequences are shown in Table II, m and n being zero:
Table 2:
A B C D E F G H
a) Ile Thr Thr Thr Thr Gly Glu Glu b) Ile - Thr Thr Thr Gly Glu Glu c) Ile - Thr Thr Ile Gly Glu Glu d) Ile - Thr Thr Thr Gly Glu Pro e) Ile - Thr Thr Ile Gly Glu Pro f) - - Thr - Thr Gly - Glu 2o g) - - Thr Thr Thr Gly - Glu Depending on the composition of the synthetic gene, these modifications can readily be brought about at the DNA level by replacement of the appropriate gene fragments by de-novo-synthesized DNA sequences, utilizing appropriate restriction enzyme cleavage sites.
2 ~ An example of a variation of the amino acid sequence which may be mentioned is the (known) polypeptide of the formula I, in which m is 1, X
and C represent Val, ~~ and B represent direct bonds, D and E denote Thr, F
denotes Gly, G and H denote Glu, and n denotes zero. This modification can readily be brought about at the DNA level via the cleavage site corresponding 3 o to the restriction enzyme Acc I.
The incorporation of the synthetic genes or the gene fragments into cloning vectors, for' example into the _ 6 _ ~34t f24 commercially available plasmids pUC 8, pUC 12 and peR 322, or other generally accessible plasmids, such as ptac 11 and pKE; 177.3, is carried out in a manner known per se. It is also possible beforehand to provide the chemically synthesized genes with suitable chemically synthesized control regions which make expression of the proteins possible.. In this context, reference may be made to the textbook by~ Maniatis (Molecular Cloning, Maniatis et al., Cold Spring Harbor, 1982). The transformation into suitable host: organisms, advantageously E. coli, of the hybrid plasmids thus obtained is likewise known per se and is describerd in detail in the abovementioned text-book. The isolation of the expressed protein and its purification can be carried out by processes known per se.
The gene fragments HIR-I and HIR-II obtained accor-ding to the invention, the hybrid plasmids obtained there-with, and the transformed host organisms are new and the invention relates to them. The same applies to the new DNA sequences which are modifications of DNA sequence I.
Further embodiments of l:he invention are set out in the patent claims.
Some other embodiments of the invention are illus-trated in detail in the examples which follow, from which the multiplicity a~f the possible modifications and combina-tions are evident to those skilled in the art. In these examples, percentage data relate to weight unless other-wise specified.
Examples 1. Chemical synthesis of a single-stranded oligo-nucleotide The synthesis of the structural units of the gene is illustrated using the example of structural unit Ia of the gene, which comprises nucleotides 1-32 of the coding strand. The nucleoside at the 3' end, in the present case thymidine (nucleotide No. 32), is covalently bonded via the 3'-hydroxy group by known methods (M. J. Gait et al., Nucleic Acids Rers. 8 (1980) 1081-1096)) to silica gel ('('~JFRACTOS~L~f supplied by Merck). For this pur-pose, initially the silica gel is reacted with ~ ha cPe Mn~~k 3-(triethoxysilyl)propylamine with elimination ~f~e~h~n~l~ ' an Si-0-Si bond being produced. The thymidine is reac-ted as the 3'-0-succinoyl-5'-dimethoxytrityl ether with the modified carrier in the presence of paranitrophenol !i and N,N'-dicyclohexylcarbodiimide, the free carboxy group of the succinoyl group acylating the amino radical of the propylamino group.
In the following synthesis steps, the base com-ponent is used as the monomethyl ester dialkylamide or chloride of the 5'-0-dimethoxytritylnucleoside-3'-phos-phorous acid, the adenine being in the form of the N6-benzoyl compound, the cytosine in the form of the N4-benzoyl compound, the guanine in the form of the N2-isobutyryl compound, and the thymine, which contains no 1!i amino group, being without a protective group.
50 mg of the polymeric carrier which contains 2 ~rmol of bound t~~hymidine are treated successively with the following agents:
a) nitrometh;ane, b) saturated zinc bromide solution in nitromethane containing 1X water, c) methanol, d) tetrahydr~~furan, e) acetonitrile, 2!i f) 40 rmol of the appropriate nucleoside phosphate and 200 ~rmol of tetrazole in 0.5 ml of anhydrous acetonitrile (5 minutes>, g) 20X acet is anhydride in tetrahydrofuran containing 40X lutidine and 10X dimethylaminopyridine (2 minute:;), h) tetrahydrofuran, i) tetrahydrofuran containing 20X water and 40%
lutidine, j) 3X iodine in collidine/water/tetrahydrofuran in 3!i the ratia by volume 5:4:1, k) tetrahydrofuran, and U methanol.
The term "'phosphate" in this context is to be understood to be 'the monomethyl ester of the deoxyribose--g- 1341 124 3'-monophosphorous acid, the third valency being saturated by chlorine or a tertiary amino group, for example a morpholino radical. The yields of the individual synthe-sis steps can be determined in each case after the detrity-lation reaction b) by spectrophotometry by measurement of the absorption of the dimethoxytrityl cation at s wave-length of 49b nm.
When the synthesis of the oligonucleotide is complete, the methyl phosphate protective groups of the oligomer are eliminated using p-thiocresol and triethyl-amine.
The oligonucleotide is then separated from the solid carrier by treatment with ammonia for 3 hours.
Treatment of the ~oligomers with concentrated ammonia for 1!i 2 to 3 days quantitatively eliminates the amino protective groups on the bases. The crude product thus obtained is purified by high-pressure liquid chromatography (HPLC) or by polyacrylamide gel electrophoresis.
The other structural units Ib-IIh of the gene are .'0 also synthesized entirely correspondingly, their nucleo tide sequences being evident from DNA sequence II.
2. Enzymatic linkage of the single-stranded oligo-nucleotides to give the gene fragments HIR-I and HIR-II
?5 For the phosphorylation of the oligonucleotides at the 5' terminal end, 1 nmol of each of oligonucleo-tides Ib-Ie was treated with 5 nmol of adenosine triphos-phate, with four units of T4 polynucleotide kinase in 20 girl of 50 mM tris.HCl buffer (pH 7.6), 10 mM magnesium :50 chloride and 10 mM dithiothreitol (DTT), at 37°C for 30 minutes. The enzyme is inactivated by heating at 95°C
for 5 minutes. The oligonucleotides Ia, If, IIa and IIh, which form the "protruding" sequence in DNA sequences IIA
and IIB, are not phosphorylated. This prevents, in the 35 subsequent ligation, the formation of larger subfragments than correspond to DNA sequence IIA or IIB.
The oligonucleotides Ia-If are ligated to form the subfragment HIR-I as follows: 1 ~nmol of each of the oligonucleotides Ia and If and of the 5'-phosphates of Ib, Ic, Id and Ie are dissolved together in 45 Nl of buffer containing 50 mM ~tris.HCl (pH 7.6>, 20 mM magnesium chloride, 25 mM potassium chloride and 10 mM DTT. For the annealing of ithe oligonucleotides corresponding to DNA
sequence IIA, the solution of the oligonucleotides is heated at 95°C fon 2 minutes and then slowly (2-3 hours) cooled to 20°C. 'then, for the enzymatic linkage, 2 ~l of 0.1 M DTT, 8 Nl of 2.5 mM adenosine triphosphate (pH 7) and 5 rl of T4 DNA ligase (2,000 units) are added, and the mixture i:~ incubated at 22oC for 16 hours.
The oligonucleotides IIb to IIg are phosphorylated and then ligated together with the oligonucleotides IIa and IIh to give the subfragment HIR-II analogously.
The gene 'Fragments HIR-I and HIR-II are purified by gel electrophoi~esis on a 10X polyacrylamide gel (with-out addition of u:~ea, 20 x 40 cm, 1 mm thick), the marker substance used being ~X 174 DNA (supplied by BRL) cut with Hinf I, or pBR 32.? cut with Hae III.
3. Preparation of hybrid plasmids which contain the 2(I gene fragments HIR-I and HIR-II
a) Incorporation of the gene fragment HIR-I in pUC 12 The comme:~cially available plasmid pUC 12 is opened in known manner using the restriction endonucle~jses Xba I and Bam HI in accordance with 2'i the manuf~3cturer's instructions. The digestion mixture is fractionated on a 5X polyacrylamide gel by eUectrophoresis in known manner, and the fragments are visualized by staining with ethidium bromide or by radioactive labeling ("nick trans-3(1 lation",cf. Maniatis, loc. cit.). The plasmid band is then cut out of the acrylamide gel and separated from the polyacrylamide by electrophoresis. The fractionation of the digestion mixture can also be carried out on 2X low-melting agarose gels (as 35 described in Example 5a» .
1 Ng of plasmid is then ligated at l6oC overnight with 10 n~~ of gene fragment HIR-I which has pre-viously been phosphorylated as described in Example I:I. The hybrid plasmid as shown in _ ' Figure I is obtained.

b) Incorporation of the gene fragment HIR-II in pUC 8 In analogy to a), the commercially available plas-mid pUC 8 is cut open with Bam HI and Sal I and '.5 ligated with the gene fragment HIR-II which has previously been phosphorylated as described in Example 2. The hybrid plasmid as shown in Figure 2 is obtained.

4. Synthesis of the complete gene and incorporation '10 in a plasmid a) Transforimation and amplification The hybrid plasmids thus obtained are transformed into E. ~coli. For this purpose, the strain E.

coli K 1.? is made competent by treatment with a 115 70 mM calcium chloride solution, and the suspen-sion of the hybrid plasmid in 10 mM tris.HCl buffer (pH 7.5), which is 70 mM in calcium chloride, is added.. The transformed strains are selected as usual utilizing the resistance and sensitivity i'.0 to antibiotics conferred by the plasmid, and the hybrid vectors are amplified. After the cells have been killed, the hybrid plasmids are isolated, cut open with the originally used restriction enzymes, and tihe gene fragments HIR-I and HIR-II

f5 are isolated b;y gel electrophoresis.

b) Linkage of the gene fragments The subfragmen~ts HIR-I and HIR-II obtained by amplification are enzymatically linked as des-cribed in Example 2 (annealing at 60oC), and the 30 synthetic. gene with the DNA sequence I Which has been thu:~ obtained is introduced into the cloning vector pIIC 12. A hybrid plasmid as shown in Figure 3 is obtained.

c) Preparation of a gene for CVall Va127-hirudin , 35 The AccI-SaII fragment (nucleotides 13-212) is obtained as described in Example 5a)from the plas-mid as shown in Figure 3, and this fragment can then be ligated, using the following adaptor 11 ~3~' X24 S' CT AGA ATG GTT GTA T 3' 3' T TAC CAA CAT ATA 5' Xbal AccI
into the E~lasmid pUC 12 which has previously been opencad with XbaI and SaII.
d) Preparation of a gene for variant a) in Table 2 f. The following DNA sequence is synthesized for this purpose:
Glu phe Het Ile Thr Thr 5' GG G.AA TTC ATG ATC ACA ACG T 3' 1 C~ CC C'~T AAG TAC TAG TGT TGC ATA
Smal AccI
This sequcance can be incorporated as a Smal-AccI
adaptor into the DNA from the plasmid as shown in Figure 3, which has been opened with the restric-15~ tion enzymes SmaI and AccI.
e) Preparation of a gene for variant b) in Table 2 The follo~~ing DNA sequence is first synthesized:
Glu Fhe !!et Ile Thr 5' GG GAA TTC ATG ATC ACG T 3' CC CTT AAG TAC TAG TGC ATA
Smal AccI
This, D~IA sequence which has been shortened by the 20 ACA triplet is incorporated, as described under d) into the pla.smid.
f) Preparation of .a gene for variant c) in Table 2 The startling DNA sequence is that previously des-cribed under e) (variant b) in Table 2). After 25~ cleavage with AccI and SaII, two fragments are obtained <~nd isolated. The AccI-SaII-thirudin) fragment 'is cut with the enzyme HinfI, and the larger of the fragments obtained is isolated.
This fragrnent, the residual plasmid DNA opened 3C1 with AccI~~SaII, and the synthetic DNA sequence Thr Asp CYs Iie 5' AT ACT GAC TGC ATC G 3' TGA CTG ACG TAG CTT A
Accl ' Hinf I
are tigatnd in the same vessel. The resulting hybrid pl~~smid codes for variant c) in Table 2.
g) Preparation of a gene for variant d) in Table 2 The starting DNA sequence is again that previously described under e) and thus codes for variant b) in Table ~?, this sequence being cut with AccI and SaII. Th~~ AccI-SaII-(hirudin) fragment is, how-ever, rea~:ted with TaqI and OdeI, and the fragment 1t1 which has been cut out is replaced by the synthetic DNA sequence Glu Pro Ile Vii' C GAA CCG ATC CC 3 ~
TT GGC TAG GGA CT
Ta~I Dde I
The AccI-;3alI-(hirudin) fragment which has thus been modified is now ligated with the residual 1.'~ plasmid DIVA. The resulting plasmid codes for variant d;l in Table 2.
h> Preparation of a gene for variant g) in Table 2 The BamHI-SaII-(hirudin) fragment is isolated from the hybrid plasmid as shown in Figure 2, and 2t1 is then cut with KpnI. The larger of the two resulting fragments is isolated, and is ligated with the synthetic DNA sequence ;5er Asp Gly Lys Asn Gln Cys Val 5~ GA 'rCC GAC GAA AAG AAC CAG TGC GTT
G CTG CTT TTC TTG GTC ACG CAA
BamHI
Thr Gly Glu Gly 5~ ACT GGC GAA GGT AC 3' TGA CCG CTT C Kpn I

141 124._ The ligation product is now linked with the XbaI-BamHI-(hirudin) fragment isolated from the plas-mid DNA shown in Figure 1, and is introduced into pUC 12 as described under 4b). The hybrid plas-mid which is obtained codes for variant g) in Table 2.
i) Preparation of a gene for variant f) in Table 2 The XbaI-SaII-(hirudin) fragment is isolated from the DNA which codes for variant g) in Table '10 2, and is reacted with HinfI. The larger of the two resulting fragments is then ligated with the synthetic DNA sequence A.rg !!et Thr Tyr Aep Cys Thr 5' CT A.GA ATG ACG TAT GAC TGC ACT G 3' T TAC TGC ATA CTG ACG TGA CTT A
XbaI Hinf I
and the ligation product is inserted into the 1!i plasmid pUC 12 which has been opened with XbaI
and SaII. The hybrid plasmid codes for variant f) in Table 2.
j) Preparation of other variants The following may be mentioned as examples of 20 other possible variants:
The DNA sequences which code for variants c) and d) in Table 2 contain the BamHI recognition site corresponding to position 97 by of DNA sequence I.
Now, utilizing this cleavage site, it is possible 2!i to link together as desired the AccI-BamHI and BamHI-SaII part sequences of the two variants which have been described above.
Transformation of all these variants into E. coli and their expression is possible in analogy to the descrip-30 tion for DNA sequence I.
5. Construction of hybrid plasmids for the expression of DNA sequences IA, IB and IC
a) Incorporation of DNA sequence IC in pKK 177.3 (direct expression) '~'' The expression plasmid pKK 177.3 (plasmid ptac 11, Amman et al., Gene 25 (1983) 167, into the Eco RI
recognition site of which a sequence which con-tains a :5a1 I cleavage site has been incorporated synthetically) is opened with the restriction !i enzymes IEco RI and Sal I. The DNA sequence IC is obtained from the plasmid as shown in Figure 3 as follows. The plasmid is cut first with the restric-tion enz;~me Sal I and then with the enzyme Acc I, and then the small Acc I-Sal I fragment is sepa-'10 rated from the plasmid band by gel electrophoresis on a 2X low-melting agarose gel, the DNA being recovere~~ by dissolution of the gel at elevated temperature (as stated by the manufacturer). This ONA fragment can be converted into DNA sequence IC
using the following adaptor 5' AATTC ATG ACG T 3' 3' (3 TAC TGC ATA 5 ~
Eca RI Acc I
A hybrid plasmid in Which there is an expression or regulation region upstream of the insert is produced by ligation of the plasmid pKK 177.3, 20 which has been cut open, with the gene IC. After addition of a suitable inducer, such as isopropyl -thiogalactopyranoside (IPTG), a mRNA which leads to expression of the methionyl-polypeptide corres-ponding to DNA sequence IC is formed.
25 b) Incorporation of the hirudin genes IA and IB into the expression plasmid pCK-5196 (fusion construction) The expression plasmid pCK-5196 (Fig. 4> is a derivative of the plasmid pAT 153 (Twigg and :30 Sherrat) which itself is a high copy-number deriva-tive of the known plasmid pBR 322. It contains the known Lac UV5 promoter with the known 5ecuence of ~ -galactosidase up to nucleotide No. 1554 (corresponding to amino acid No. 518:Trp) inserted :35 between the tetR gene (Hind III from pBR 322) and the terminator of the ampRgene (.~ha III in Pos. 3231 in pBR 322). Another feature of this vector is the polylinker sequence vhich originates from the known plasmid pUC 13 and is inserted between the (3 -galactosidase sequence and the terminator sequence. Transcription from the lac UV5 promoter takes place counter to the tetR
gene.

This plasmid contains at amino acid No. 518 of -galactosidase the polylinker part Xba I-1D Bam HI-Sma I-Ss~t I from the plasmid pUC 13, which serves as the cloning site for eucaryotic genes.

The plasmid pCk:-5196 is opened using the restric-tion enzymes Xba I and Sst I and is ligated with DNA

sequence IA which can be isolated from the plasmid shown in Fig. 3~ by Xba I/SSt I digestion. A

hybrid plasmid as shown in Fig. 5 is obtained, and this contains downstream of the lac UV5 control region the codons for the first 518 amino acids of ~3 -galactosidase followed by the codons for 2~D Ser-Arg-Met-(hirudin 1-64).

A hybrid plasmid as shown in Fig. 6 can be obtained in the same manner, this containing in the plasmid pCK.-5196 an insert of DNA sequence IB in which, however, the codon for amino acid No. 1 (threonine) is now preceded by the codon for the amino acid arginine. The DNA sequence IB is obtained from the plasmid shown in Fig. 3 by cutting the Latter with the restriction enzymes Sst I and Acc I and ligating with the following 3~D adaptor:

5' CT AGA CGT ACG T 3' 3' T GCA TGC ATA 5' Zba I Acc I

6. Transformation of the hybrid plasmids.
Competent E. coli cells are transformed with 0.1 to 1 erg of the hybrid plasmids which contain sequence I, or IA, IB or IC, and in the case of tac plasmids are plated out onto agar plates containing ampicillin, and in the case of the pCK-5196 hybrid plasmids are plated out onto agar plates containing tetracycline. Clones which contain the correctly integrated hirudin sequences in the particular plasmids can then be identified by DNA rapid work-up (Maniatis,loc. cit.).

7. Expression of the polypeptides having hirudin activity Following transformation of the specified tac hybrid pl~asmids into E. coli, the polypeptide expressed carries, in addition to the hirudin amino aci~~ sequence, an additional methionyl group at the amino terminal end, but this group can be eliminated by cleavage with cyanogen bromide. On 1!i the other hand, following transformation of E.

coli with the hybrid plasmid shown in Fig. S or Fig. 6, tike fusion proteins obtained comprise 518 amino acids of p -galactosidase and hirudin which are linked together via the sequence Ser-Arg-Met 2(1 or Ser-Ar~~-Arg. These fusion proteins can be cleaved to give hirudin and p -galactosidase frag-ments using cyanogen bromide or trypsin.

8. 4lorking up and purification The bacterial strains which have been cultivated 2: to the desired .optical density are incubated for a suffici~ant time, for example 2 hours, with a suitable ~inducer, for example IPTG. The cells are then billed with 0.1X cresol and 0.1 mM

benzylsul~fonyl fluoride.

30 After cenl:rifugation or filtration, the mass of cells is l:aken up in a buffer solution (50 mM

tris, SO a~M EDTA, pH 7.5) and is disrupted mechanically, for example using a French press or ~DYNO mi~ l l~(supplied by Wi lly Bachofer, Basle), 35 whereupon the insoluble constituents are removed by centrii'ugation. The protein containing the hirudin is purified from the supernatant by custo-~nary procE~sses. Ion exchange, adsorption, or gel filtration columns or affinity chromatography on ' ~ T'w~f~ ~H a~~k .

thrombin or antibody columns are suitable. The enrichment and purity of the product are checked by analysis using sodium dodecyl sulfate/acryl-amide gels or HPLC. Fusion proteins containing a ~3 -galactosidase fraction can be detected even in the crude extract of the lysed bacteria on the basis of the difference in their migration beha-vior from that of ~3 -galactosidase (1-518). The hirudin produced by genetic engineering is charac-terized b;y a comparison with the authentic sub-stance isolated from leeches or by means of tests based on the anticoagulant properties of hirudin.

_,8_ 134112~~' DNA sequence I

Triplet 0 1 2 3 4 5 No.

Am o a ~t ~. rte. ~. ~p ~s i c n i d NucleotideNo. 1 10 20 Cod.strand 5 CT AGA ATG ACLU TAT ACT GAC 'IGC

Non-cod. rand 3 T TAC TGC ATA 'IGA CTG ACG
st 6 7 8 9 to 11 12 13 14 15 ~r Glu o r Gly Gln Asn Leu Cps Lru Cys ACT GAA TCT GGT CAG AAC CIA 'IGC CTG TGC

TGA CTT AGA CCA G2'CTIG GAC ACG GAC ACG

Glu G7y Ser Asn Val Cys Gly Gln Gly Asn 60 7o 80 QAA GGA TCT AAC GTR ZGC GGC CAG GGT AAC

26 27 28 29 3o 31 32 33 34 35 I,ysCya Ile Leu Gly Ser Asp Gly Glu Iys AAA ZOC ATC CTT GGA TCC GAC GGT GAA AAG

TTT ACG TAG GAA CCT AGG CTG CCA CTT TT'C

Aan aln Cya Yal ~r Gly (~Iu Gly Thr Pro AAC CAG 'IGC QTT ACT GGC. GAA GGT ACC CCG

I~yaPro Gln Ser His Aan Aap Gly Asp Phe 150 160 17o AAA CCG CAG TCT CAT AAC GAC QGC GAC TIC

TTT GGC Q'IC AGA QTA TTG CTG CCG CTG AAG

~6 57 58 59 60 61 62 63 64 flu Glw Ile Pro Glu Glu Tyr Leu Gln Stp GAA QAG ATC CCT GAG GAA TAC CTT CAG TAA

CZT CTC TAG GGA CTC CTT ATG GAA GTC ATT

Stp TAG JI~3C T~ 3 ATC TCG A~3C A(3C T 5 ONA sequence IA:

Met (Amino acids 1-64) 5' CT AGA ATG (Nucleotides 9-200) TAA TAG AGC T 3' 3' T TAC (Complement . nuc l . ) p'~' TAC 5' XbaI
Sstl DNA sequence IB:

Arg (Amino acids 1-64) 5' CT AGA CGT (Nucleotides 9-200> TAA TAG AGC T 3' T TAC (Complement. nu c l . ) ATT ATC 5,' ~aI SstI
DNA sequence IC:

Met (Amino acids 1-64) 5' AA TTC ATG (Nucleotides 9-200) TAA TAG AGC TCG 3' G TAC (Complement. nucl.) p~ ATC TCG AGC AGC T 5' EcoRI Sall DNA-Sequenz II A (HIF
E-- I a Nucleotid-Nr. 10 20 Cod. Strang 5' CT AGA ATG ACG TAT ACT GAC TGC
nicht cod. Strang 3' T TAC TGC ATA TGA CTG ACG
Xbal - Ib Ic-ACT GAA TCT L GGT CAG AAC CTG TGC CTG TGC
TGA CTT ACA CCA GTC TTG GAC ACG GAC ACG
I b ---~ I I d Ic ~ ~- le GAA GGA TCT AAC GTT TGC GGC CAG GGT AAC
CTT CCT AGA TTG CAA ACG CCG GTC CG~A TTG
I d - ~ If le AAA TGC ATC CTT G Bam H I 3' TTT ACG TAG GAA CCT AG 5' I f - ---X

DNA-Sequenz II B WHIR-II~
II
a E-Nucleotid-Nr. 100 110 cod. Strang 5' GA TCC GAC GGT GAA AAG

nicht cod. Strang3' Baml-il G CTG CCA CTT TTC

E- I I
b -I I I
I
c a ~ ~-AAC CAG TGC GTT ACT GGC GAA GGT ACC CCG

TTG GTC ACG CAA TGA CTT CCA TGG GGC
CCG

II b II
d ~ F

l le c ~ ~--I

AAA C C G CAG TCT CAT AAC GAC G G GAC TTC
C

TTT GGC GTC AGA GTi~ TTG CTG CCG CTG AAG

I I d I
I
f - -- ~ E-Il Il e- ~ E- g GAA GAG ATC CCT GAG GAA TAC CTT CAG TAA

CTT CTC TAG GGA CTC ATG GAA GTC ATT
CTT

I I f -- ~ I
I
h II g TAG AGC TCG Sall 3' ATC TCG AGC AGC T 5' IIh

Claims

1. A process for the preparation of a polypeptide having hirudin activity, of a formula (X)m-A-B-C-Tyr-D-Asp-Cys-E-Glu-Ser-Gly-Gln-Asn-Leu-Cys-Leu--Cys-Glu-Gly-Ser-Asn-Val-Cys-Gly-Gln-Gly-Asn-Lys-Cys-Ile--Leu-Gly-Ser-Asp-F-G-Lys-Asn-Gln-Cys-Val-Thr-Gly-Glu--Gly-Thr-Pro-Lys-Pro-Gln-Ser-His-Asn-Asp-Gly-Asp-Phe-Glu--H-Ile-Pro-Glu-Glu-Tyr-Leu-Gln-OH
in which m is 0, 1 or 3, and X represents Met, Arg, Ser-Arg-Met or Ser-Arg-Arg A denotes Ile or a direct bond, B denotes Thr or a direct bond, C denotes Thr, D denotes Thr or a direct bond, E Denotes Thr or Ile, F denotes Gly, G denotes Glu or a direct bond, H denotes Glu or Pro, which comprises incorporation into an expression plasmid of a gene which codes for the polypeptide of the formula and isolation of the expressed polypeptide.

2. The process as claimed in claim 1, wherein the non-coding strand of the incorporated gene hybridizes with the coding strand of the gene coding for the polypeptide.

3. The process as claimed in claim 1, wherein the gene has been synthesized chemically or enzymatically via an mRNA, encoding a gene as defined in claim 1.

4. The process as claimed in claim 3, wherein the chemical synthesis is the phosphate process.

5. The process as claimed in claim 1, 2 or 3, wherein the gene contains two stop codons downstream of the codon for the carboxy terminal amino acid.

6. The process as claimed in claim 4, wherein the gene contains two stop codons downstream of the codon for the carboxy terminal amino acid.

7. The process as claimed in claim 1, 2 or 3, wherein, in the polypeptide of the formula as defined in claim 1, D anal E denote Thr, F denotes Gly, and G
and H denote Glu.

8. The process as claimed in claim 4, wherein, in the polypeptide of the formula as defined in claim 1, D
and E denote Thr, F denotes Gly, and G and H denote Glu.

9. The process as claimed in claim 1, 2 or 3, wherein the gene corresponds to the DNA sequence I as shown in Figure 7.

10. The process as claimed in claim 4, wherein the gene corresponds to the DNA sequence I as shown in Figure 7.

11. DNA sequence IA as shown in Figure 8.

12. DNA sequence IB as shown in Figure 8.

13. DNA sequence IC as shown in Figure 8.

14. DNA sequence I as shown in Figure 7.

15. DNA sequence IIA as shown in Figure 9.

16. DNA sequence IIB as shown in Figure 10.

17. A fusion peptide which contains all or part of the polypeptide of the formula as defined in claim 1, said polypeptide having the properties of a functional hirudin-like polypeptide.

18. A fusion peptide as claimed in claim 17, which contains a bacterial fraction comprising all or part of the amino acid sequence of p-galactosidase.

19. A hybrid plasmid which contains a DNA sequence which codes for all or part of the polypeptide of the formula, as defined in claim 1, said polypeptide having the properties of a functional hirudin-like polypeptide.

20. A host microorganism which contains a hybrid plasmid as claimed in claim 19.

21. A host microorganism as claimed in claim 20 which is of the species E. coli.