CH659826A5 - PROCESS FOR OBTAINING A MICROORGANISM TRANSFORMED BY A PLASMID COMPRISING A GENE OF 27-DESAMIDOSECRETINE STRUCTURE. - Google Patents

Description

The present invention relates to a process for obtaining a strain of a microorganism transformed by a plasmid comprising a structural gene for a deamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is valine.

The invention also relates to a strain of transformed Escherichia coli obtained by this method, as well as a process for producing 27-desamidosecretin comprising culturing this strain to transformed Escherichia coli and recovering the produced 27-desamidosecretin.

Furthermore, the invention relates to 27-deamidosecretin obtained by the process which has just been mentioned, as well as a chimeric plasmid comprising a structural gene, expressing 27-deamidosecretin, and a double-stranded polydeoxyribonucleotide, as intermediates of the process for obtaining the transformed microorganism strain.

It is therefore a process for the production of 27-desamidosecretin (“27” being sometimes omitted in the designation of this compound in the description below) by a technique of genetic engineering using a gene for desamidosecretin. obtained by chemical synthesis.

Secretin is a compound known as one of the gastrointestinal hormones. Hitherto, physiological activities of this compound are known, such as its action promoting the secretion of water and bicarbonate by the pancreas and it has been used in a practical way for the evaluation of the pancreatic function or as a therapeutic agent for the treatment of duodenal ulcer.

Secretin is a polypeptide having the following formula:

5 10

His - Ser - Asp - Gly - Thr - Phe - Thr - Ser - Glu - Leu - Ser-

15 20

Arg - Leu - Arg - Asp - Ser - Ala - Arg - Leu - Gin - Arg - Leu -

25 27

Leu - Gin - Gly - Leu - Val - NH2

In order to obtain this secretin, methods such as the extraction1 of a secretin existing in the natural state (usually porcine secretin, have been implemented, but it has also been determined that the amino acid sequence bovine secretin is the same as that of porcine secretin) and chemical synthesis2. However, each of these methods involves difficulties regarding cost, speed, production efficiency, etc. In addition, according to the method of the prior art in which the secretin from the small intestine of pigs is extracted and purified, there is a difficulty resulting from the fact that the large number of gastrointestinal hormones (motilin, VIP , CCK-PZ, GIP, GLI, etc.)

5

io

15

20

25

30

35

40

45

50

55

60

65

659,826

4

obstructs precise determination of secretin by the biological or radioimmunological determination method.

In the meantime, recent advances in so-called genetic engineering techniques, in which a synthetic gene is used, are so important that we are now beginning to produce commercially, by means of these techniques, different polypeptides having physiological activity. Consequently, the possibility of producing secretin by a genetic engineering technique would make it possible to discover clues which could lead to the solution of the problems mentioned above which are encountered in the production methods in accordance with the prior art.

However, although it has generally been established that the production of a polypeptide by a genetic engineering technique, using a synthetic gene, is made possible by the operations of performing the chemical synthesis of a gene structural, to insert this gene as a vector into an appropriate plasmid, to transform an appropriate host and to produce and recover the desired polypeptide by culturing the transformation product, it is not necessarily possible to predict with certainty whether the a given polypeptide having a desired physiological activity can be introduced by this method.

In the case of secretin, since the C end of the secretin polypeptide is a valine amide, as mentioned above, the end of the DNA structural gene representing the polypeptide must necessarily include the codon corresponding to the valine and, therefore, the polypeptide produced is considered to have a C-terminus consisting of valine. With such reasoning, it is not certain that the polypeptide obtained will exhibit the physiological activity of secretin.

The present invention results from the confirmation of the fact that a structural gene expressing a secretin derivative the C end of which is constituted by valine which is not in the form of amide, namely desamidosecretin, can be obtained by synthesis chemical, that such a gene can be inserted as a vector into a suitable plasmid to form a chimeric plasmid, that the transformation of an appropriate host cell by the chimeric plasmid as well as the production and recovery of deamidosecretin by culture of the transformation product are possible, that the desamidosecretin thus produced has an activity similar to that of secretin and, moreover, that the expression system of the lactose operon can be used in the production of desamidosecretin and that the 'The yield of desamidosecretin can be increased to a large extent through the use of this expression system.

Thus, the 27-desamidosecretin obtained by the process according to the invention comprises a polypeptide represented by the amino acid sequence:

His - Ser - Asp - Gly - Thr - Phe - Thr - Ser - Glu - Leu - Ser -Arg - Leu - Arg - Asp - Ser - Ala - Arg - Leu - Gin - Arg - Leu -Leu - Gin - Gly - Leu - Val - OH

wherein Val-OH shows that the C-terminus of the polypeptide comprising valine is a carboxylic acid.

The present invention also relates to a process for obtaining a strain of a microorganism transformed by a plasmid comprising a structural gene for a deamidosecretin corresponding to a polypeptide in which the amino acid located at the end C of secretin is valine, this process being characterized by the fact that it comprises the following operations:

(1) chemical synthesis of a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is valine,

(2) insertion of this gene into a vector plasmid capable of proliferating in a predetermined host cell, so as to produce a chimeric plasmid capable of proliferating in this cell, and

(3) transformation of the host cell with this chimeric plasmid.

The process for producing 27-deamidosecretin comprises culturing the transformed cell thus obtained and recovering the deamidosecretin produced.

A preferred variant of the process for obtaining the strain of transformed microorganism comprises the following operations:

(1) chemical synthesis of a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid

5 located at the C end of the secretin is valine,

(2) obtaining a vector plasmid capable of using the lactose operon and capable, moreover, of proliferating in a predetermined host cell,

(3) insertion of this structural gene into the vector plasmid so as to produce a chimeric plasmid capable of proliferating in this cell, and

(4) transformation of the host cell by this chimeric plasmid.

Thus, the process for producing 27-desamidosecretin includes

generally takes the following steps:

(1) obtaining a chimeric plasmid comprising a fragment consisting of a structural gene for a deamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is valine, this chimeric plasmid being capable of

20 proliferate in a predetermined host cell and express the structural gene for a desamidosecretin in the host cell,

(2) transformation of the host cell with this chimeric plasmid, and

(3) culture of the transformed cells thus obtained and recovery of the deamidosecretin produced.

A typical example of the host cells which can be used in the transformation process defined above is constituted by Escherichia coli which belongs to the genus Escherichia, the lactose operon originating from Escherichia coli.

Thus, the preferred variant of the process for the production of 27-deamidosecretin can be defined as a process comprising the culture of a microorganism producing deamidosecretin, this microorganism consisting of Escherichia coli in which incorporated a plasmid containing the structural gene for deamidosecretin and the recovery of deamidosecretin produced when using the lactose operon in Escherichia coli.

The invention also relates to an Escherichia coli culture intended to be used for the implementation of the method, this Escherichia coli having been transformed so as to present a phenotype such that the culture of this microorganism produces 27-deamidosis- 40 moron. Thus, the invention relates to an Escherichia coli characterized in that it has been transformed by means of a plasmid in which a structural gene has been incorporated or inserted, obtained by chemical synthesis, of deamidosecretin corresponding to the polypeptide in which the the amino acid at the C end consists of valine, this plasmid 45 being capable of using the lactose operon and also being capable of proliferating in a predetermined host cell.

The invention also relates to a double-stranded polydeoxyribonucleotide comprising the structural gene expressing 27-deamidosecretin.

The invention also relates to a plasmid comprising the structural gene expressing 27-deamidosecretin.

In accordance with the description of the invention given above, it is possible to eliminate the extraction and purification difficulties encountered in the case of the extraction of the natural secretin which has been used, as the only commercially exploitable process, through the use, as raw material, of the gene extract obtained by recombination.

During the implementation of the method according to the invention, it is possible to vary one or more amino acid species constituting a peptide, which makes it possible to study the correlations between the structures and the activities of peptide hormones. In other words, any desired derivative substituted by an amino acid of a peptide exhibiting physiological activity can be easily obtained. It can be considered that the formation of a polypeptide derivative by genetic manipulation is the best method, taking into account that the formation by chemical synthesis of a polypeptide derivative of high molecular weight is very difficult to carry out according to the methods of which we currently have.

5

659,826

Furthermore, the use of the host-vector system used in the preferred variant of the process for the production of deamidosecretin, as mentioned above, has made it possible to produce deamidosecretin having a secretin activity corresponding to approximately 3 x 10 4 molecules per host cell, which can be considered a yield suitable for practical use. In addition, as regards the fused protein which will be described below, a yield of 2.85 x 105 molecules / cell is obtained, which constitutes a value which suggests that the production of deamidosecretin with a high yield is made. possible by the expression of character using the chimeric plasmid according to the invention in a bacterium, without destructive effect (proteolytic) by the proteases.

There are also polypeptides other than secretin which have a C-terminus in the form of an amide, and the present invention provides a suggestion regarding the production of amide derivatives of such polypeptides as well as the expression of their physiological activity.

The 27-desamidosecretin obtained by the process according to the invention is a polypeptide corresponding to the amino acid sequence represented by the formula (I) indicated below:

His - Ser - Asp - Gly - Thr - Phe - Thr - Ser - Glu - Leu - Ser -Arg - Leu - Arg - Asp - Ser - Ala - Arg - Leu - Gin - Arg - Leu -Leu - Gin - Gly - Leu - Val - OH

In the formula (I), the symbol His and the other symbols of the same kind are the usual symbols indicating amino acids such as histidine, etc.

This polypeptide differs from the secretin polypeptide in that the C-terminated valine is not in the form of an amide (Val-NH2) but in the form of a carboxyl group (Val-OH).

Deamidosecretin, which has a physiological activity similar to that of secretin, in particular the effect of promoting the secretion of pancreatic juice, can itself be used as a polypeptide having a physiological activity analogous to that of secretin. On the other hand, it is also possible to use desamidosecretin after amidation of the carboxyl group at the C end of the secretin. The amidation can be carried out by a purely chemical method or by an enzymatic method or by a biological method in vivo.

Although desamidosecretin can be obtained by modification of the C-terminus of secretin, this compound is preferably produced by a genetic engineering technique, in accordance with the present invention, by performing the chemical synthesis of a structural gene for this polypeptide, the preparation of a plasmid making it possible to express the polypeptide, the transformation of a host cell by means of this plasmid, the production of the desired polypeptide by culturing the transformed cells and the recovery of deamidosecretin.

The DNA base sequence that makes up the structural gene for secretin is unknown.

Consequently, among the codons which designate the amino acid constituting the peptide, those which satisfy the following conditions are chosen for the synthesis of DNA:

(i) it must be set so that the region enriched in base pairs A - T does not immediately follow the region enriched in base pairs G - C, and

(ii) it must also be adjusted so that each synthetic fragment, as described below, does not exhibit an undesirable complementary base sequence intramolecularly or intermolecularly.

It is also desirable, in order to facilitate the choice of the transformed strains, for the structural gene to be arranged so as to contain one or more restriction enzyme recognition base sequences. In the case of desamidosecretin, it is preferable that the gene has recognition base sequences of the Hinf I and Hae II type.

From this point of view, typical examples of codons designating amino acids in the structural gene for desamidosecretin are the following:

Amino acid Codon

His CAC

Ser TCT, TCA

Asp GAT

Gly GGT

Thr ACT, ACC

Phe TTC

Glu GAA

Leu TTG, CTC, CTA, CTG, TT A, CTT

Arg CGT, CGC

Ala GCA

Gin CAA, CAG

Val GTT

Consequently, according to a preferred form of the structural gene for desamidosecretin according to the invention, the base sequence is that which is indicated below in the experimental examples as well as in the diagram representing the structure of the gene (in this diagram , the base sequence is of course the group going from the CAC triplet, corresponding to His, to the GTT triplet, corresponding to Val).

More specifically, the invention relates, in particular, as an intermediate product, to a double-stranded polydeoxynucleotide comprising a structural gene expressing 27-desamidosecretin, as mentioned above, a particularly advantageous embodiment of the gene for structure with the base sequence represented

felt below:

His

Ser

Asp

Gly

Thr

Phe

Thr

CAC

-TCA

-GAT

-GGT

-ACT -

-TTC

- ACC-

GTG

-AGT

-CTA

-CCA

-TGA-

-AAG

- TGG -

Ser

Glue

Leu

Ser

Arg

Leu

Arg

TCA

-GAA

-CTA

-TCT

-CGT

-CTA

-CGT -

AGT

-CTT

-GAT

-AGA

-GCA

-GAT

-GCA -

Asp

Ser

To the

Arg

Leu

Gin

Arg

GAT

-TCA

-GCA

-CGC

- CTC

-CAG

- CGC -

CTA

-AGT

-CGT

-GCG

-GAG

-GTC

- GCG -

Leu

Leu

Gin

Gly

Leu

Val

TTG

-CTG

-CAA

-GGT

-CTC

-GTT

AAC

-GAC

-GTT

-CCA

-GAG

-CAA

-

The manner of expressing such a structural gene is described in detail, for example in the published Japanese patent application No. 92696 / 1979. In the case where the plasmid pBR 322 is used as the plasmid in which the gene must be incorporated, the desamidosecretin having to be expressed as a protein fused with its lactamase operon, the site in which the gene must be incorporated is advantageously the site of recognition of the restriction enzyme Pst I in the operon. More specifically, the ATG codon of methionine which constitutes the site to be attacked by cyanogen bromide is located on the side of the 5 ′ end of the structural gene, while one or more stop codons are located at the 'end 3'. Subsequently, so as to be synchronized3 with the frame starting at the start codon of the lactamase gene, the gene on the 5 ′ side of the ATG is given bases chosen at random in equal numbers 2 + 3n (n = 0, 1 , 2, ...) as well as a sequence of Pst I recognition bases to form cohesive ends at the two ends respectively. In general, structural genes are designed and synthesized4 with their two unprotected cohesive ends but, if desired, these two ends can be blocked, which makes it necessary to provide two or more bases chosen at random to the parts even more distant. outside the recognition base sequences in order to increase the hydrolysis capacity of the restriction enzyme.

5

10

15

20

25

30

35

40

45

50

55

60

65

659,826

6

The optional base pairs to be placed on the 5 'side of the ATG codon are 3m pairs, (3m + 1) or (3m + 2), m being an integer equal to 0.1 or more.

In view of the foregoing considerations, a preferred embodiment of the desamidosecretin gene which is suitable for carrying out the invention is as described below in the experimental examples.

More specifically, a particularly advantageous form of the gene for desamidosecretin has the structure shown below:

Met

ACCTGCAGCC - ATG -TGGACGTCGG-TAC-

His

Ser

Asp

Gly

Thr

Phe

Thr

CAC

-TCA

-GAT

-GGT

-ACT-

■ TTC -

ACC-

GTG

-AGT

-CTA

-CCA

-TGA-

- AAG-

TGG -

Ser

Glue

Leu

Ser

Arg

Leu

Arg

TCA

-GAA

-CTA

-TCT

-CGT

-CTA -

CGT -

AGT

-CTT

-GAT

-AGA

-GCA

-GAT-

GCA -

Asp

Ser

To the

Arg

Leu

Gln

Arg

GAT

-TCA

-GCA

-CGC

-CTC

-CAG-

-CGC-

CTA

-AGT

-CGT

-GCG

-GAG

- GTC -

GCG-

Leu

Leu

Gin

Gly

Leu

Val

TTG

-CTG

-CAA

-GGT

-CTC

- GTT -

AAC

-GAC

-GTT

-CCA

-GAG

-CAA-

END

END

TGA

-TAG

- GGCTGCAGGT

ACT

-ATC-

• CCGACGTCCA

For the synthesis of the gene designated as described above, each of the two strands + and - can be divided into several fragments. These fragments can be chemically synthesized and then the respective fragments can be linked together. Preferably, each strand is divided into about 16 fragments, each of which comprises 9 to 16 bases, so that there will be an overlap of 6 to 7 bases.

As a method of synthesizing each of the fragments, mention may be made of the diester method 5, the triester method 6, the solid phase method 4, the liquid phase method or the method in which an enzyme is used 7. From the point of view of synthesis time, yield and purification, the most suitable method is the solid phase method according to the triester method.

With regard to the details of the synthesis, reference should be made to the references indicated above, experimental examples being described below.

When synthesizing an oligonucleotide, separation and purification of the final product generally becomes more difficult as the length of the strand increases. In particular, according to the solid phase synthesis process, suitably protected oligonucleotide blocks are condensed in successive stages and, therefore, the purification cannot be easily carried out by the usual methods such as gel filtration, gel electrophoresis, the use of an ion exchange column, high speed liquid chromatography, etc.

In a reverse phase column, the retention time varies to a large extent depending on whether the oligonucleotide has or does not have a lipophilic protecting group. Consequently, when an oligonucleotide block having a stable protective group under the conditions allowing the elimination of the other protective groups is used, in the final condensation step, and then these other groups are removed protectants, a mixture of oligonucleotides containing only the stable protective groups can be obtained in the desired end product. By taking advantage of the lipophilic nature of the protective group, the desired end product can be separated from the mixture of unreacted species , by passage through a reverse phase column, an operation which is followed by removal of the protective group, so as to produce the desired oligonucleotide.

By proceeding in this way, the oligonucleotide thus synthesized can be separated and purified with good efficiency from the mixture of unreacted species.

Mutual ligation of the synthetic fragments thus prepared can be carried out successively using DNA ligase. For the preparation of the substrates of the synthetic fragments for the enzyme, it is necessary to phosphorylate the 5 'hydroxyl groups in the fragments. For this purpose, a polynucleotide kinase is generally used, but it is also possible to carry out chemical phosphoryla-tion5. While the ligation of the fragments is generally carried out using a DNA ligase, it is also possible to resort to the method according to which the phosphoric acid groups at the 5 'ends are inactivated by an appropriate method (for example, imidazolylation) and one performs chemically ligated with the strand on the opposite side serving as the base9.

When implementing the preferred variant of the process for the production of 27-deamidosecretin according to the invention, it is possible to use different plasmids containing all or part of the lactose operon of the DNA of the chromosome of E. coli, these plasmids also being able to proliferate in YE. coli. The preparation of these plasmids can be carried out by the usual methods well known in the field of molecular biology. DNA containing the lactose operon can be obtained directly from the chromosome of E. coli but also transducer phages containing all or part of the lactose operon are also obtained (for example, Pldl, F'-lac, 080 dplac, Mi80 dlac, Xplac, etc.) and can therefore be taken from these phages the necessary part of the lactose operon. In addition, for the preparation of the plasmid capable of proliferating in E. coli, it is necessary to ligate the required part of the lactose operon indicated above with another plasmid from VE. coli (eg pBR 322, pSC 101, A.dVl) so as to form a single vector plasmid.

In an exemplary implementation of the invention, the transducer phage Xplac523 is used as the lactose operon containing DNA. The DNA of XplacS can be obtained, for example, from YE. coli PK 1512, which is a lysogenic bacterium, using a known method20. This phage> .plac5 corresponds to the region going from the middle region of the gene i to the middle region of the gene y of the lactose operon and it advantageously does not have any other £ genes. coli than the lactose operon20. Consequently, this phage is particularly suitable for implementing the invention. As a plasmid from E. coli, pBR 322 is used which is chosen because it is one of the easiest plasmids to obtain, its base sequence is well known and has labeling genes resistant to ampicillin and tetracycline resistant. In addition, when these genes are ligated, each of the genes (Xplac5 and pBR322) is treated with the restriction enzymes EcoRI and HindIII and the 3.8 Md fragment (megadaltons15) is taken respectively. for Xplac5 and the larger fragment16 for pBR322, which are ligated together for the preparation of the desired vector plasmid. In this description, this vector is designated by the letters pRE.

We chose the lactose operon from the chromosome of E. coli for expression of the desired desamidosecretin due to the fact that a foreign gene can be inserted into the z gene of the lactose operon, at the recognition site of the restriction enzyme of EcoRI for expression in the form of a protein fused with ß-galactosidase43'17, that a protein can be produced in large quantities, that the induced production can be carried out using an appropriate host microorganism and that the product thus obtained can be easily and stably recovered in the form of a fusion protein in the practically pure state.

Consequently, it is desirable for the vector prepared by the method according to the invention to have a single recognition site for the restriction enzyme EcoRI and, for this purpose, use is made of DNA fragments obtained by cutting Xplac5 and pBR322 using

5

10

15

20

25

30

35

40

45

50

55

60

65

7

659,826

Eco RI and HindIII, respectively, as described above, these fragments in turn being ligated together.

The chimeric plasmid is prepared in the following manner:

The structural gene for desamidosecretin, indicated above, is incorporated in an appropriate position in the vector designed for the expression of a foreign gene, as described above. The incorporation operation itself can be carried out according to a process well known in the field of molecular biology. With regard to the details about the process used, reference should be made to the experimental examples given below.

In accordance with an embodiment of the invention, pBR322 is used as the vector plasmid and the gene is incorporated at the PstI recognition site of this plasmid so as to produce a chimeric plasmid. In the present description, the chimeric plasmid is designated by pMG. The main reasons for choosing PBR322 as the vector plasmid are that this plasmid is one of the most easily obtained, that its base sequence is well known, and that it has genes ampicillin resistant and tetracycline resistant labeling. Plasmid pBR322 has been deposited with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland, 20852, U.S.A., under reference number ATCC37017. The main reasons which motivated the choice of the PstI site as the place of incorporation of the gene consists in the fact that it allows the use as such of the operon lactamase, that it allows an easy search for the product of transformation of the fact that resistance to ampicillin (Apr) is changed to sensitivity to ampicillin (Aps) and that there is only one PstI site in the plasmid pBR322.

In the preferred embodiment of the process for producing 27-deamidosecretin, described above, pRE1, which will be described in more detail below, is used as vector plasmid, and a gene containing the structural gene for desamidosecretin, at its EcoRl recognition site, so as to produce a chimeric plasmid. In the present description, this chimeric plasmid is designated by pLS.

Still in this preferred embodiment of the method, the gene containing the structural gene indicated above is preferably used, that is to say, the gene having the base sequence indicated in the diagram. In order to incorporate this gene, which has recognition sites for the restriction enzyme PstI, at its two ends, into the plasmid pRE1, it is necessary to convert the recognition sites for the restriction enzyme PstI into sites for recognition of the EcoRI enzyme.

Thus, in accordance with this preferred mode of implementation of the process for producing desamidosecretin, a double-stranded DNA is necessary to constitute the interlayer sequence (spacer)

between the gene containing the structural gene and the plasmid pRE1. More specifically, double-stranded DNA is necessary in order to have two recognition sites for the restriction enzymes PstI and EcoRI, the base pairs between the two recognition sites for the restriction enzymes being 3n + 1 (n = 1, 2, 3, etc.) so as to be synchronized with the reading frame for translation starting with the start codon of the β-galactosidase gene (the nature of the base pairs being able to be chosen at random, as long as no inconsistent codon appears).

However, it is only necessary that the part corresponding to the spacer finally has the function indicated above and it is possible to obtain the spacer by the method of synthesis of the structural gene indicated above. In accordance with an embodiment of the invention, constituting an example of a synthesis process of this kind, the spacer was obtained by proceeding in the manner described below.

First of all, the structure of a single-stranded DNA is worked out, in accordance with the scheme indicated below, this DNA having sites for recognition of the restriction enzymes PstI and £ coRI.

5 '1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 3' CTGAATTCA G C TC TG CAGAG

Eco RI

PstI

The bases in position 1, 2, 9-12, 19, 20 can be chosen at random, as long as they satisfy the following conditions:

Condition (1): 1 and 10, 2 and 9, 11 and 20, and 12 and 19, are each base pairs complementary to the others; io Condition (2): the sequence of 9, 10, 11 is not an inconsistent codon.

The single-stranded DNA thus conceived (designated, in the present description, by the term “pre-binding fragment”) allows, thanks to its specific complementarity, to obtain a long double-stranded DNA structure. chain with a certain number of bends (part cut without empty gap formed on one of the two arms of DNA). Consequently, one can form from this structure a double-stranded DNA without empty space by making act on it a DNA ligase. The double-stranded DNA thus prepared is a double-stranded polyDNA alternately having the recognition sites of the restriction enzymes Pst \ and iscoRI, that is to say the sequence of bases 1, 20 indicated above. .

It is therefore possible to obtain, by treatment of the double-stranded DNA with the restriction enzyme PstI, a binding fragment of PstI - £ t oRI -25 Pst I having cohesive ends.

In general, a binding fragment presenting recognition sites of two restriction enzymes is capable of various uses which have been carried out, in accordance with the prior art, using two different species of oligomer. The indications given in the present description allow the same result to be obtained using a single kind of oligomer, in accordance with the method described above, the single-stranded DNA being capable of expression as shown below. :

35 X'n ... X'2X \ X ^ .. XnYm restriction enzyme A recognition site

... Y2 Y,

restriction enzyme B recognition site

40 Y ^ YV.Y'm

X and X ', Y and Y' each representing any of the desired bases (A, G, C, T) mutually complementary (n and m = 0, 1, 2, etc.).

45 In this case, A represents the enzyme co coRI and B represents the enzyme PstI, and n + m = 3p + 1 (p = 1.2, etc.), p being preferably equal to 1 or 2.

The direction of the 27-desamidosecretin gene incorporated into the plasmid can be determined by cleavage at the specific site of the structural gene 50 by means of an enzyme (Hae II) having the property of recognizing this specific site, cleavage of another site. at a given position outside the structural gene and analysis of the dimensions of the fragment obtained.

As a characteristic example of the host cell which can be transformed using a chimeric plasmid in which the structural gene of desamidosecretin is incorporated, as described above, such as the pMG described above, mention may be made of: the strain of E. coli Kl2C600 (FERM BP - 115, filed with the Institute of Fermentation Research, Agency of Industriai Science and Techno-60 logy, Japan). The strain of E. coli K12C600 is described in the literature24 and its bacteriological properties are also described in this reference.

Another example of a chimeric plasmid in which the structural gene of desamidosecretin is incorporated, as described above, consists of a pLS such as pLS58 which can be used in a preferred embodiment of the invention for the production of desamidosecretin. A typical example of the host cell that can be transformed using pLS58 is

659,826

8

constituted by the strain of E. coli XA35, belonging to the genus Escherichia coli, this strain having been obtained from the known strain, K12 of Escherichia coli22 and having the properties indicated below, its other properties not being different from those of strain K. 12:

[Smr, Lac "" (i3 ~, z ~)]

The transformation by means of a chimeric plasmid into which the structural gene of desamidosecretin, according to the invention, is incorporated, is possible in the case of all strains of i ?, coli. However, in order to obtain desamidosecretin in the form of a protein fused with ß-galactosidase, a strain is preferably used in which the gene for ß-galactosidase is lacking, in order to avoid the presence of ß- normal galactosidase mixed with protein. Generally, the production of the protein is regulated by the repressor gene (gene i) of the lactose operon. Consequently, when using a kind of E. wild coli, the induced production of the fused protein is possible by means of an inducer, for example IPTG (isopropyl thiogalactoside). When a strain with a gene for sensitivity to high temperatures is used, the production of the fused protein can be induced by raising the temperature. When you use a strain in which the i gene is lacking, you can still produce the fused protein18.

In accordance with the preferred embodiment of the process for producing desamidosecretin according to the invention, the strain XA35 of E. is used. coli which is a strain in which the ß-galactosidase genes and the i gene are lacking.

It should also be noted that the transformation with the plasmid into which the structural gene of desamidosecretin is incorporated is not limited to the case where the host is constituted by E. coli but that one can also choose an appropriate host in a wider range of bacterial species, by choosing an appropriate vector, as is well known in the field of molecular biology. A typical example of such a host cell is described in published patent application No. 92696/1979, mentioned above.

The transformation operation itself can be carried out according to the usual techniques well known in the field of molecular biology. With regard to the details about the process used, reference should be made to the experimental examples described below.

A typical example of transformed cells consists of the cells obtained by transformation of the strain K12C600 with the chimeric plasmid pMG. In the present description, these transformed cells are designated by the name K12C600 (pMG103).

The genetic properties of the transformed K12C600 (pMG103) cells are as follows:

[m-, r ~, F-, lacY, Leu, Thr, tonA, supE, recBC]

Another example of transformed cells consists of those obtained by transforming the XA35 strain of E. coli, these transformed cells can be used in the preferred mode of implementation of the process for the manufacture of desamidosecretin according to the invention, using the chimeric plasmid pLS58. These transformed cells are designated, in the present description, by the term E. coli XA 35 (pLS 58).

The strain XA35 (pLS58) of E. transformed coli differs from strain XA35 of E. coli with regard to the following properties, as has been demonstrated in the experimental examples described below:

[Apr, Lac +]

Deamidosecretin can be produced by culturing, according to a usual method, the transformed bacteria obtained as described above. With regard to the details about this process, reference can be made to the experimental examples described below.

Experimental examples Example 1:

Development of the structure of the desamidosecretin gene

The structure of a gene having the base sequence constituted by the combination of blocks A, B and C is developed, as shown in the attached diagram. These blocks respectively include the fragments indicated below:

Block

A B C

Fragment

S-l, S-4, S-6 S-5, S-7, - S-10, S-12 S-l 1, S-13 - S-16

15 For the development of the gene structure, the procedure is as described below:

(1) Selection of codons

The codons are chosen as shown in the diagram.

(2) The ATG codon for methionine is added to the N terminus so that the polypeptide obtained by synthesis will be cut at this site by chemical treatment (+ CNBr).

(3) At the C end, two translation termination codons (TAG or TGA) are added so that there is no production of

25 unnecessary peptides.

(4) In order to allow synchronization with the frame starting with the start codon of the lactamase gene, two suitably chosen base pairs are added upstream of the methionine [in general, 2 + 3n (n = 0, 1 , 2, 3, ...)].

(5) PstI sites are added at both ends.

It is also possible to add, immediately before the PstI site, any desired number of base pairs if appropriate when synthesizing each of the fragments.

(6) Finally, two pairs of

35 bases chosen at random to improve the hydrolysis efficiency of the restriction enzyme PstI.

The gene designated as indicated above contains the Hinfl site and the Hae II site in the structural gene for 27-deamidose-cretin.

40

Chemical synthesis of the fragment

1) Summary

The fragments are synthesized by the solid phase method described in the literature63. However, the fragments obtained by synthesis are isolated and purified by proceeding according to the improved method described below.

The yields of the synthesis of the respective fragments are indicated in the table below:

50

Length

of

Render

Fragment

Sequence of chain bases (%)

S-l

ACCTGCAGCCATGCAC

16

33

S-2

GCTGCAGGT

9

52

S-3

TCAGATGGTACTT

13

56

S-4

ATCTGAGTGCATG

13

42

S-5

TCACCTCAGAACTAT

15

43

S-6

GAGGTGAAAGTACC

14

47

S-7

CTCGTCTACGTGATT

15

38

S-8

AGACGAGATAGTTCT

15

27

S-9

CAGCACGCCTCCAGC

15

32

S-10

CGTGCTGAATCACGT

15

43

S-ll

GCTTGCTGCAAGGT

14

22

S-12

AGCAAGCGCTGGAGG

15

44

S-13

CTCGTTTGATAGG

13

47

S-14

AAACGAGACCTTGC

14

42

S-l 5

see S-2

S-16

ACCTGCAGCCCTATC

15

32

9

659,826

2) Purification

To 50 mg of resin having undergone the synthesis in solid phase, 0.5 M of tetramethylguanidine of aldoxime α-picolinic and 100 to 200 μl of a mixture (1: 1) of dioxane and water are added, and let the mixture sit at room temperature for a few hours. Then, 2 ml of a concentrated aqueous ammonia solution are added to the mixture and the whole is left to stand at 55 ° C., overnight, in a hermetically sealed container with a stopper. The resulting mixture is filtered so as to separate the resin and the filtrate is concentrated, which is then subjected to filtration through a gel. The elution is carried out using 50 mM of TEAB buffer (pH 7.5) and the fractions eluted are collected in the empty volume. These fractions are concentrated and introduced into “HPLC” from a column in reverse phase C-18 (Waters: “Radial Pack A”, diameter 8 cm × 10 cm) so as to elute in a diacetate buffer. 0.01 M ethylene amine (pH 7.8) with a concentration gradient of acetonitrile from 10 to 32% with a flow rate of 2 ml / min for 16 min. The eluted fractions are collected for 11 to 12 min. During this operation, oligo-nucleotides without a trityl group are eluted in the form of an injection peak. The eluted fraction is concentrated, and after adding 1 ml of 80% acetic acid thereto, it is left to stand at room temperature for 15 min. Tritanol is removed by extraction, the aqueous layer is concentrated and it is applied again to the column in reverse phase. The elution is carried out, under the same conditions as above, with a concentration gradient of acetonitrile from 0 to 20% and the fractions eluted are collected for 12 to 13 min.

Phosphorylation

Each fragment (30 g) is dissolved in a 30 mM buffer of Tris-HCl (pH 7.5) and added to the solution 60 Ci (19.8 pmol)

of ATP [y32P] and 2 (il (9 units) of polynucleotide kinase T4, so as to bring the total amount to 50 μl, and this operation is followed by an incubation at 37 ° C. for 20 min. 10 equivalents are then added to the mixture, relative to the fragments,

ATP and T4 polynucleotide kinase (9 units) and this operation is followed by incubation at 37 ° C for 20 min. The reaction is stopped by heating at 100 ° C for 2 min and the product is purified by gel filtration. Each fragment is confirmed by 20% gel electrophoresis.

Ligation of fragments

Each of the 0.05A26O fragments S1, S-2, S-3, S-4 and S-6 is dissolved in a buffer (20 mM Tris-HCl, pH 7.5 MgCl2 10 mM, 10 mM DTT [dithiothreitol, ATP (adenosine) triphosphate) 0.2mM]) so as to constitute a solution in a total quantity of 30 dj. 1. To the solution 1 (xl (150 units) of T4 DNA ligase is added and the mixture is left to stand at 10 ° C. overnight. The reaction is confirmed by electrophoresis on 8% polyacrylamide gel. synthesis of blocks B and C.

The reaction mixtures of blocks A and B are mixed together and, after addition of 3 µl of 0.2 mM ATP and 1 (xl (150 units)

of T4 DNA ligase, the mixture is allowed to stand at 10 ° C overnight. After confirmation of the reaction by electrophoresis on 8% polyacrylamide gel, the reaction mixture of block C is added and the reaction is allowed to proceed in a similar manner overnight. The progress of the reaction is confirmed by electrophoresis on 5% polyacrylamide gel.

The reaction mixture is heated to 68 ° C for 15 min, then cooled to room temperature. 15 (4.1 of 0.5 M NaCl and 80 units of the restriction enzyme Pst I are then added, then the mixture is left to stand at 37 ° C. overnight. The reaction is stopped by heating at 90 ° C. for 1 min. and the reaction mixture is subjected to separation by electrophoresis on 5% polyacrylamide gel. The strip having the longest chain length is cut and transferred to electrophoresis by a 1% low melting agarose gel in order to perform the extraction of the strip obtained.

Cloning

Plasmid pBR322 (4 g) is added to a mixture (total amount: 50 (ll) of 100 mM Tris-HCl buffer, 5 mM MgCl2 and 50 mM NaCl and the reaction is carried out using 6 units of the restriction enzyme Pst I, at 37 ° C. for 2 hours, after which the reaction is stopped by heating at 68 ° C., for 15 min. The synthetic gene indicated above is added to the reaction mixture and the ligation reaction is carried out by carrying out the procedure similar using T4 DNA ligase.

The E. strain is transformed. coli K12C600, proceeding according to the Kuschner10 method, in a P-3 "type enclosure device using 50 μl of the reaction mixture obtained which constitutes the equivalent of 0.68 ng of pBR322 DNA.

The strain transformed is chosen on a type L plate (1% Bactotripton, 0.5% Bactolevure extract, 0.5% NaCl, 1.5% Bac-toagar) containing 10 (ig / ml of tetracycline (Te) and 50 strains are studied, among the transformed strains obtained, in order to determine their resistance to ampicillin (Ap), and 45 transformed strains are isolated having the Tc '/ Ap * properties. the latter strains by the name C600 (pMG 101) -C600 (pMG 145).

Among the transformed strains having the Tcr / Aps properties, 8 strains are randomly chosen [C600 (pMG 101) - C600 (pMG 108)] and the plasmid DNA is isolated by equilibrium sedimentation through a gradient density of cesium chloride containing ethidium bromide. By way of reference, the ADB of the plasmid thus isolated is designated by the name pMG 101 - pMG 108.

Management confirmation

The plasmid (5 g) is dissolved in a mixture (30 (il) of 10 mM Tris-HCl buffers, 66 mM MgCl2, 6 mM P-mercaptoethanol and 60 mM NaCl and 30 units of the restriction enzyme Hae are added to the solution. III, this operation being followed by heating at 37 ° C. for 1 hour. Fragments of 375 bp are extracted by electrophoresis on agarose gel with low melting point at 1.5%. similar, the fragments thus extracted, using 18 units of the restriction enzyme Hae II and the hydrolyzed product is identified at 191 bp and 176 bp, as determined by electrophoresis on 5% polyacrylamide gel, as plasmid having the correct direction.

Identification of the fused protein using minicells

We transform, by means of the plasmids pMG 103 and pBR 322, according to the method of Kuschner10, an E. coli producing minicells (F ~, Thr +, ara +, Leu +, azis, tonA \ minA, minB, gal +, X ', sfr '. malA, xyl, mtl, thi, sup ~) so as to obtain a transformation product having the Tcr / Aps properties.

A preculture of the transformation product is carried out in 20 ml of minimum Davis culture medium containing 0.5% of casamino acid, thymine (20 ng / ml), thiamine (2 ng / ml) and tetracycline (10 ng / ml) at 37 "C for 16 hours and, after inoculation of 10 ml of the preculture in 50 ml of the same medium (but not containing tetracycline), the culture is carried out until ODS! 0 = 0.6-0.8 The culture broth is subjected to centrifugation, using a cooled HITACHI brand centrifuge (model RPR-9, at 3000 revolutions per min) for 12 min, in order to separate the host cells and, then, to a centrifugation at 8500 rpm for 25 min, so as to obtain granules of minicells. The granules of minicells are washed by suspending them in 25 ml of buffer To (IM Tris-HCl pH 7.3, 150 mg MgSO 4, 7H, 0.1 ml 1% gelatin, CaCl2 IM 0.25 ml / l-H2O) The suspension is subjected to centrifugation at 10,000 rpm (in an RPR-20 type rotor) for 15 min, and l 'we suspend d, in 1 ml of buffer T1, the precipitate of minicells resulting. The suspension thus obtained is subjected to centrifugation in a density gradient in steps of 10% - 15% - 20% (at 4000 rpm, in an RPRS-4 type rotor, for 25 min) so as to recover the minicells. The suspension of minicells is then suspended in 20 ml of T1 buffers and the similar operations are repeated.

5

10

15

20

25

30

35

40

45

50

55

60

65

659,826

10

The minicells are suspended in 1 ml of Davis minimum culture medium, each containing the following 18 amino acids (alanine, valine, isoleucine, proline, phenylalanine, tryptophan, glycine, serine, threonine, cystine, tyrosine, asparagine, asparagine tick, glutamine, glutamic acid, lysine, arginine, histidine) in final concentration of 50 ng / ml for each of these acids, 20 pg / ml of thymine and 2 ng / ml of thiamine and heated with stirring to 37 ° C for 12 min. The mixture is further heated for 10 min after addition to the suspension of 20 Ci (S35) -methionine (NEN (New England Nuclear): 1260.1 Ci / mmol) and 5 nCi (H3) -leucine (NEN: 115.2 Ci / mmol). Immediately after, the same buffer is added to the mixture containing 50mM NaN3, 100 ng / ml of methionine and 100 ng / ml of leucine. A centrifugation is then carried out at 3000 revolutions / min, for 20 min, so as to recover the mini-cells which are in turn suspended in 20 μl of a buffer sample [H20 6 ml, 1.25M Tris-HCl ( pH 6.8) 1 ml, 0.4 g SDS (sodium do-decylsulfate), 2 ml glycerol, 1 ml 2-mercaptoethanol, 4 mg BPB (bromophenol blue)] and heated to 100 ° C for 3 min. 18 μl of this suspension are then subjected to separation by electrophoresis on SDS-poly-acrylamide.

In the cells of E. coli K12C600 (pBR322) (FERM-P 6017), ß-lactamase was synthesized and a band corresponding to a molecular weight of approximately 29 kilodaltons appears on the elec-trophoresis gel, but the presence of d is not detected such a protein of about 29 kilodaltons in extracts of E. cells. coli K12C600 (pMG 103). Instead of this protein, a polypeptide of about 24 kilodaltons appears in the form of a new band. It is clearly the gene produced by coding by pMG 103 in relation to the disappearance of the β-lactamase band. PMG 103 has a structure in which part of the pBR322 ß-lactamase gene (encoding 182 amino acids from the amino terminus of the signal protein of the ß-lacta-mase structural protein) is ligated downstream of the latter (for transcription) to the 27-deamidosecretin gene (encoding 27 amino acids), and its resulting gene can be detected as a fused protein of the ß-lactamase fragment (182 amino acids) - 27-deamidosecretin ( 27 amino acids). The value calculated for the molecular weight of the fused product is 23.4 kilodaltons and it can therefore be considered that the new band which appears in the extracts of it, coli K12C600 (pMG 103) corresponds to this fused protein.

Protein purification

The E. coli K12C600 (pMG 103) is subjected to a culture with stirring in 3 liters of Luria culture broth at 37 ° C. and centrifuged when the concentration of microorganisms reaches approximately 1 × 10 9 cells / ml. The resulting agglomerates are washed with 10 mM Tris-HCl buffer (pH 8.0) / phenylmethylsulfonyl fluoride ImM and centrifugation is carried out again. The agglomerates are resuspended in the same buffer. The suspension is treated with EDTA in a final concentration of 10 mM, at 0 ° C for 5 min, then it is treated with lysozyme of egg white, with a final concentration of 0.1 mg / ml, at 0 ° C for 30 min, so as to perform the lysis formation of the spheroplast. The resulting product is treated by means of an ultrasound generator device, of the Kubota brand, “Insonator 200 M” model, at maximum power (200 W) for 10 min, so as to completely destroy the bacterial cells. This operation is followed by separation by centrifugation (using a cooled HITACHI brand centrifuge, model 20 PR52, with an RPR 20-2 rotor, at 18,000 rpm for 30 min, at 0 ° C) so as to eliminate cell debris. To the supernatant obtained (110 ml) is added magnesium chloride in a final concentration of 50 mM, then the mixture is subjected to centrifugation (by means of the rotor indicated above, at 8000 rpm, for 10 min, at 0 ° C). To the supernatant, 400 ng of DNase I and 5 mg of RNase I are added so as to carry out a treatment at 4 ° C. for 1 hour, after which the proteins and the peptides are separated by salification using a saturated solution. 80% ammonium sulfate. The precipitate obtained by centrifugation (using an RPR 20-2 rotor, at 8000 rpm, for 10 min, at 0 ° C.) is dissolved in 10 mM Tris-HCl buffer (pH 8.0). desalt using the same buffer and centrifuged under the conditions indicated above. Then added to the supernatant acetone at a final concentration of 75% and the precipitate formed is treated overnight with 1.0 g of cyanogen bromide in 20 ml of 80% formic acid. After evaporation, the residue is extracted with 40 ml of isopropanol, then with an equal volume of methanol. A further evaporation is then carried out and the residue is dissolved in 0.1N acetic acid. After elimination by centrifugation of the insoluble parts (3000-4000 G, for 5 min), gel filtration is carried out through a column of “Sephadex G-25” (produced by the company Pharmacia CO.) And the fractions having molecular weights from 1000 to 10,000, and then lyophilized.

Biological assay

The lyophilized samples thus obtained are assayed by a biological assay method12 in which the lyopilised sample 20 is dissolved in a total volume of 2 ml of an isotonic sodium chloride solution, and is injected intravenously into the groin of the parts. respective 0.25 ml aliquots to 5 rats, the quantities of pancreatic juice secreted are measured through an artificial pancreatic tube, at intervals of 10 min, and the increase in quantity of secretion is determined by as the biological activity of secretin. As a reference secretin, Secrepan (Eisài, Japan) is injected intravenously, in the same rat, in quantities of 0.25 U / kg and 0.50 U / kg and the quantities of pancreatic juice are measured. secreted, to determine the correspondence with the 30 Eisai units (which are practically equal to the unit of Crick Harper Rasp). It can be seen that the samples obtained from E. coli K12C600 (pMG 103) cause an increase in the secretion of pancreatic juice of 202% ± 76 (mean + standard deviation), 10 min after injection, and 180% + 87.6 20 min after injection. On the other hand, Secrepan causes an increase of 162% + 28 and 206% + 58, respectively after 10 min and 20 min in the case of an injection of 0.25 U / kg and 159.8% ± 83 , 7 and 250.8% + 216.5, respectively 10 min and 20 min after injection of a dose of 0.5 U / kg. These values, which are the result of a t-type test, reveal a significant difference for p <0.05. On the other hand, when the sample obtained from extracts of the bacterial cells obtained from the clone containing only pBR322 and not containing the deamidosecretin gene was subjected to the test by the same method, there is n there is no increase in the amount of secreted pancreatic juice. This indicates that a substance with a biological activity similar to that of secretin is produced by E. cells. coli K12C600 (pMG 103) when such a substance is not produced in the cells of E. coli K12C600 (pBR 322). By postponing the increase in the amount of 50 pancreatic juice secreted by the samples obtained from E. coli K12C600 (pMG 103) on a line obtained by plotting the points on a semi-logarithmic scale graph, the increase in the quantity of secreted pancreatic juice being plotted on the ordinate and the Secrepan units on the x-axis (logarithmic scale), 55 on can determine the secretin unit by the extrapolation method (provided a linear relationship is obtained), the value thus obtained being from 0.17 to 0.24 Eisai unit (corresponding to 0.17 to 0.24 Crick unit Harper Raper) / 0.25 ml / kg of rats. When the calculation is made taking into account the body weights of the rats used, this value is calibrated from 0.051 to 0.072 U / 0.25 ml / rat, the total activity amounting to 0.408 to 0.576 U for the quantity 2 ml total. Since it is known13 that the purified secretin having the highest degree of purity at a specific activity of 16,000 CHR / mg units, the values of 0.408 to 0.576 U correspond to an amount of 25.5 to 36 ng of 65 secretin. This corresponds to 7.15 to 10.1 pmol molecules of secretin, which indicates that the biological synthesis of a desamidosecretin having an activity corresponding to 4.3 to 6.1 × 1012 molecules of secretin has been carried out. The fact that desamidosecretin is prepared

11

659,826

using about 3 x 1012 i ™ cells, coli K12C600 (pMG 103) as the starting material, at least 1.4 to 2.0 molecular equivalents of secretin are recovered per bacteria cell.

Radioimmunoassay

The sample is diluted in a mixture of 50 mM Tris-HCl buffer (pH 8.0) and 0.1% bovine serum albumin. The radio-immunoassay of the diluted sample is carried out using a “Secretin kit Daiichi” assay set produced by the Japanese company Daiichi Radioisotope Research Institute, following the procedure indicated, so as to confirm the activity of the sample. .

Example 2:

Ligation operations are carried out between the fragments, proceeding in a similar manner to that described in example 1.

Preparation of XplacS DNA and pBR322 DNA

The DNA of A, plac5 is obtained from lysogenic bacteria E. Coli PK 1512 (IFO 14149), the preparation method followed corresponding to the method of Oshima20.

PBR322 DNA is obtained from i ?. Coli K-12 C600 (pBR322) (FERM-P 6017) using the method of M. Kahn et al. 21.

Preparation of the pREl

10 Xg of DNA from A, plac5 are added to a mixture (in total amount equal to 20 dj.1) of 100mM Tris-HCl buffer, pH 7.5, 7mM MgCl2 and 50mM NaCl, and the reaction is carried out using 10 units of the restriction enzyme .EcoRI and 10 units of the restriction enzyme HindIII at 37 ° C, for 2 hours. Then purified by electrophoresis on agarose gel. at 1%, DNA fragments of 3.8 mega-gadaltons.

Add 1%. of pBR322 DNA to a mixture (in total amount equal to 10 n) similar to that indicated above and the reaction is carried out using a unit of the restriction enzyme £ coRI and a unit of l restriction enzyme HindIII, at 37 ° C, for 2 hours. After which, 2.6 megadalton DNA fragments are purified by electrophoresis on a 1% agarose gel.

The fragments thus obtained are added to a mixture (total quantity: 10 (11) of 50 mM Tris-HCl buffer (pH 7.8), 10 mM MgCl 2, 20 mM DTT and ImM ATP, and the reaction is carried out using 30 units. T4 DNA ligase at 14 ° C for 24 hours.

Transformation of strain XA 35 of E. is carried out using the reaction mixture. coli, in accordance with the method of Kuschner10, in an enclosure device of the p-111 type.

The strain transformed is chosen on a type L plate (Bac-totrypton 1%, extract of bacteria yeast 0.5%, NaCl 0.5%, Bactoagar 1.5%) containing Ap (20 ng / ml) and the transformed strains obtained are replicated on an EMB-lac plate (Bacto-EMB 2.25% culture broth, Bactoagar 1.5%). The strains transformed into Lac + (red colony) are also chosen. Among these strains, 5 are chosen at random to prepare a plasmid which is analyzed with different kinds of restriction enzymes, which shows that the four strains consist of 3.8 megadalton fragments of DNA from \ plac5 ligated with EcoRI - Hind III DNA fragments from pBR 322. One of these strains is designated by the name E. coli XA 35 (pRE1). This means that the transformed strain E. coli XA 35 (pRE1) is different from the strain XA 35 of E. coli, indicated above with regard to the following properties:

Apr, Lac +

Link sequence

The synthesis is carried out of a fragment having the basic sequence CTGAATTCAGCTCTGCAGAG. The synthesis and the purification are carried out according to the method of synthesis and purification of the respective structural genes as described above (yield: 40%). The single-stranded DNA thus obtained is designated by the term "prelink sequence".

The pre-binding sequence (15 ng) is incubated in the presence of 20 units of T4 polynucleotide kinase at 37 C, for 40 min, in a mixture (total quantity: 30 nO of 50 mM Tris-HCl buffer (pH 7.5) , MgCl2 10mM, spermidine 0, lmM, EDTA 0, lmM, DTT 10MM and 0.2 mM ATP The reaction was stopped by heating at 60 ° C. for 2 min. After allowing the mixture to stand at room temperature for 1 hour, 300 units, 1 nU of DNA ligase are added and the reaction is carried out at 14: C. overnight. The reaction is stopped by heating at 60 ° C. for 2 min. The mixture is left to stand at room temperature for 1 hour and 20 units of restriction enzyme Pst I are added so as to carry out the reaction for 5 hours The reaction is stopped by heating at 60 ° C. for 2 min.

15 ng of the Pst I-coRI-Pstl binding sequence are thus obtained.

Preparation of the EcoRi-EcoRI structural gene

The Pst I-EcoRI-PSTI binding sequence is prepared as described above (4 ng) and the Pstl-PstI structural gene mentioned in Example 1 (0.6 ng) is added to a mixture (total amount: 35 m of 100 mM Tris-HCl buffer, pH 7.5 7 mM MgCl 2 and 50 mM NaCl, and the reaction was carried out using 300 units of T4 DNA ligase, at 14 ° C, for 24 hours.

The reaction is then stopped by heating at 68 ° C. for 10 min and the reaction mixture is gradually cooled.

The restriction enzyme i? CoRI (100 units) is added to the reaction mixture thus obtained so as to carry out the reaction at 37 ° C. for 5 hours, then the DNA fragment having 120 base pairs is purified by gel electrophoresis. of 5% polyacrylamide.

PLS preparation and cloning

0.25 ng of pRE1 is added to a mixture (total amount: 4 n ′) of 100 mM Tris-HCl buffer, pH 7.5, 7 mM MgCl 2 and 50 mM NaCl and the reaction is carried out using one unit of the EcoRI restriction enzyme at 37 ° C, for 1 hour.

Plasmid pRE 1 (cut with EcoRI) thus prepared (0.25 ng) and 0.02 ng of the EcoRl-EcoRl structural gene are added to a mixture (total amount: 10 nO of 50 mM Tris-HCl buffer, pH 7 , 8, MgCl2 10 mM, DTT 20 mM, ATP ImM and the reaction is carried out using 30 units of T4 DNA ligase, at 14 ° C., for 24 hours.

Using the reaction mixture thus obtained, strain XA 35 of E. coli by the method of Kuschner10 in a P-3 enclosure device.

The strain transformed is chosen on a type L plate containing Ap (20 ng / ml) as described above. The transformed strains obtained are replicated on the EMB-lac plate, using the procedure described above, with a view to carrying out the subsequent selection of the Lac + strain. Among these strains, 200 are chosen to prepare the plasmid DNA, the structure of which is analyzed using the restriction enzyme Pst I or Hae II, which confirms that 11 strains have the structural gene indicated above. .

Determination of direction

The following experiments are carried out on each of the plasmids of the 11 strains having the structural gene.

The plasmid (20 ng) is added to a mixture (total amount: 20 μl) of 10 mM Tris-HCl buffer, pH 7.5, 66 mM MgCl 2 and 60 mM NaCl and 20 units of enzyme are added to the solution. Hae restriction 11. The incubation is carried out at 37 ° C. for 2 hours. From the mixture thus obtained, fragments of about 1 megadalton and fragments of about 0.7 megadalton are extracted by electrophoresis on a 1% agarose gel. The respective fragments are cut using the restriction enzyme i? CoRI, proceeding as described above, and subjected to electrophoresis on 15% polyacrylamide gel. The plasmid is identified in which, by electrophoresis, DNA fragments having 40 base pairs are obtained from the fragment of approximately 1 megadalton and from the DNA fragment having 80 base pairs, obtained from the fragment

5

10

15

20

25

30

35

40

45

50

55

60

65

659,826

12

about 0.7 megadalton, as being the plasmid in which the structural gene is inserted in the correct direction.

These experiments make it possible to deduce that 5 of the plasmid strains among the 11 strains presenting the structural gene have the correct direction. One of these strains is chosen at random and is designated by the name E. coli XA 35 (pLS 58).

Identification and purification of the fused protein

We submit \. coliXA 35 (pLS 58) to a culture with stirring in 1 liter of Luria broth containing 20 ng / ml of ampicillin, at 37 ° C, until the number of bacteria reaches 5 x 108 cells / ml. The bacteria cells are then removed by centrifugation and suspended in a mixture (total quantity: 100 ml) of 10 mM Tris-HCl buffer (pH 7.5) and 10 mM MgCl 2 and the suspension is treated using a generator Kubota brand ultrasound, “Insonator 200 M” model, for 30 minutes, so as to destroy the bacteria cells.

The solution thus obtained (2 μl) is subjected to electrophoresis on 7.5% SDS polyacrylamide gel, which makes it possible to observe a protein band corresponding to 120,000 daltons. No such band is observed in the extract similarly obtained from bacteria having no plasmid (strain XA 35 of E. coli). It is therefore considered that this protein is derived from the plasmid pLS 58.

It is also confirmed that the protein is substantially the same size as that of β-galactosidase19 obtained from E. coli, which is not in contradiction with the dimensions (117 173 daltons) of the fused protein desamidosecretin and ß-galactosidase.

We therefore conclude that this protein of approximately 120,000 daltons constitutes the fused protein and we proceed to the purification of this protein.

The extract obtained by ultrasonic destruction is subjected to centrifugation and the precipitate is recovered. The fused protein is substantially recovered from the precipitate. The precipitate is suspended in a mixture (total amount: 100 ml) of Tris-HCl buffer (10 mM pH 7.5) and MgCl2 ImM and subjected to centrifugation followed by recovery of the precipitate. This operation is repeated twice to remove the water-soluble parts.

The precipitate obtained is dissolved in a mixture (total quantity: 100 ml) of 20mM Tris-HCl buffer (pH 7.5), MgCl2 ImM, 10MM NaCl and 7M urea, and it is passed through a column (diameter: 5 cm x 3 cm) of DEAE cellulose balanced by the same buffer.

After washing with 20 ml of the same buffer, the column is further washed with 100 ml of a similar buffer in which the NaCl concentration is modified so as to take the value 50 mM. The fused protein is then eluted using 150 ml of a similar buffer in which the NaCl concentration is modified so as to take the value 150 mM.

Ss-mercaptoetha-nol is added to the elution product thus obtained in a final concentration of 1% and the mixture is dialyzed 3 times against 5 liters of water. The dialysis product is centrifuged so as to recover the fused protein as a precipitate.

The thus-fused protein thus produced a practically unique band by SDS polyacrylamide gel electrophoresis, in an amount of about 28 mg per liter of culture broth, calculated from the absorption at 280 nm. It is calculated that this value corresponds to approximately 280,000 molecules / cell.

The purified protein is dissolved in 2 ml of 70% formic acid and treated with 100 mg of cyanogen bromide for 18 hours. After evaporation, the residue is extracted with 5 ml of isopropanol, then with 5 ml of methanol, and this extraction is followed by lyophilization.

Biological assay

Using the biological assay method12, the lyophilized sample is assayed by dissolving it in a total volume of 25 ml of isotonic sodium chloride solution, injecting 5 rats intravenously into the groin , aliquots of 50 μl each, the quantities of secreted pancreatic juice flowing through an artificial pancreatic tube are measured at 5 min intervals and the increase in flow is noted as biological activity of secretin. As the reference secretin, Secrepan (Eisai, Japan) is injected intravenously beforehand, in the same rats, in an amount equal to 1 unit, 2 units and 4 units / rat and the quantities of secreted pancreatic juice are measured in order to to determine the correspondence between the Eisai unit (practically equal to the Crick Harper Rasp unit).

It is found that the samples obtained from E. coli XA 35 (pLS 58) lead to an increase in the secretion of pancreatic juice. On the other hand, when we dose, using the same method, the sample obtained from extracts of bacteria cells obtained from the clone containing only pRE1 and not containing the deamidosecretin gene, we do not observe no increase in the amount of secreted pancreatic juice. This indicates that a substance with biological activity similar to that of secretin is produced in the cells of E. coli XA 35 (pLS58), whereas no such substance occurs in the cells of E. coli XA 35 (pREl).

By reporting the increased amounts of pancreatic juice secreted by the samples obtained from E. coli XA 35 (pLS 58) on a straight line obtained on a semi-logarithmic graph, the increased quantities of secreted pancreatic juice being plotted on the ordinate and the Secrepan units on the abscissa (logarithmic scale), it can be determined, by interpolation, that l he secretin unit is 3 Eisai units / 50 nVrat> which corresponds to approximately 1500 Eisai units in the form of 25 ml of isotonic sodium chloride solution containing the lyophilized sample. Since it is known that the maximum purified secretin has a specific activity of 16,000 CHR / mg units, 1,500 units correspond to approximately 26 nmol of secretin molecules, which indicates that biological synthesis has been carried out. '' a deamidosecretin having an activity corresponding to 1.6 x 1016 secretin molecules. Because desamidosecretin is prepared using approximately 5 x 1011 E. cells. coli XA 35 (pLS 58), at least 3 x 10 4 molecular equivalents of secretin are recovered per bacteria cell.

It can be concluded that the gene for synthetic 27-deamidosecretin is expressed by the action of the promoter of the lactose operon of E. coli in E. cells coli XA 35 (pLS 58) and that its translation product, that is to say 27-deamidosecretin, has an activity similar to that of secretin.

Radioimmunoassay

The sample is diluted in a mixture of 50 mM Tris-HCl buffer (pH 8.0) and 0.1% bovine serum albumin. The radio-immunoassay of the diluted sample is carried out using a “Secretin Kit Daiichi” type test set produced by the Japanese company Daiichi Radioisotope Research Institute, in the prescribed manner, so as to confirm its activity.

Gene structure diagram

The following diagram represents the base sequence of the gene comprising the structural gene for 27-deamidosecretin in the form of separate blocks A, B and C, a being a base pair chosen at will, b the Pst I site, it is a base pair chosen at will, d is a start codon, e is the Hinf I site, and f is the Hae II site.

\

5

10

15

20

25

30

35

40

45

50

55

60

13

659,826

r

GENE STRUCTURE DIAGRAM

Block A__

Mot Illa Sor Aap Gly Thr Plio S - / 1 | S - 3 1

A C T G

C T G C A G G A C G T C

C C G G

A T G TAC

L

b

-S - J-

vs

_J L

-C A C-T C A-G A T-G G T-A C T-T

-G T G-A G T-C T A-C C A-T G A-A A G-T G G-A G

S - V-

Blo

Phe Thr Sor Glu Leu Ser Arc Lou Arg Asp Ser Ala -S - S H ~ 7

S - 7-

I b - J 11 o - '[T

T C-A C C-T C A-G A A-C T A-T C T-C G T-C T A-C G T-G A T-T C T-C T T-G A T-A G A-G C A-G A T-G C A-lc T A-A G

Arg Lou Gin Arg -S -?

-S - t-

o

-S - / £> ■

A-G C A-C G C-C T C-C T-C G T-G C G-G A G-G

I I

~!

A G-C

T C-G C G-A

A C-G A

- / 2 ■

Block

Arg Lou Lou Gin Gly Lou Val FIN FIN i S - // n S - A? -

• Have-

G C-T T G-C T G-C

AT

A-G

G

T-C

T

C-G

T

T-T

G

AT

AT

G-G G

VS

T

G

VS

AT

G

G

T

C-G

T

T-C

VS

A-G

AT

G-C

AT

A-A

VS

YOUR

T

C-C C

G

AT

VS

G

T

VS

VS

AT

■ S - / V--

_1L

-Yes

Deposit of microorganisms

The E. coli K12C600, E. coli XA 35 and its E. coli XA 35 transformation product (pRE 1) by the plasmid pRE 1 are deposited in accordance with the provisions of the Budapest Treaty on the international recognition of micro- organizations for patent protection, at the Fermentation Research Institute, Agency of Industriai Science and Technology (FERM), 1-3, Higashi 1-chome, Yatabemachi, Tsukuba-gun, Ibaraki-ken,

Japan, under the deposit number FERM BP-115, dated June 9 to 1981, FERM BP-116, dated January 7, 1982 and FERM BP 117,

dated January 7, 1982, respectively. The plasmid pBR 322 is deposited with the FERM in the form of the transformation product of E. coli K12C600 by pBR 322, i.e. E. coli K12C600 (pBR 322), under the number FERM P-6017, dated June 9, 1981. 45

THE. coli K12C600 (pMG 103), which is the transformation product by the chimeric plasmid pMG, is deposited with the American Type Culture Collection (ATCC), 12301, Parklawn Drive, Rock-ville, Maryland 20852, USA, under the number ATCC 39042, dated January 26, 1982. 50

THE. coli PK 1512, which is the lysogenic bacterium A.plac5, is deposited with the Institute for Fermentation, Juso-Hommachi, 2-17-85, Yodogawa-Ku, Osaka-Shi, Japan, under the number IFO 14149, dated February 1, 1982.

THE. coli XA 35 (pLS 58), which is the transformation product 55 by the chimeric plasmid, is deposited in accordance with the provisions of the Budapest Treaty with ATCC, under the number ATCC 39040, dated January 11, 1982.

60

Bibliography

1) Acta Chemica Scandinavica, 15, 1790 (1961);

2) a) Chemical industry, 1966,1757,

b) Journal of the American Chemical Society, 89, 6753 (1967);

3) Proceedings of the National Academy of Sciences of the United 65 States of the America, 75, 3737 (1978);

4) a) Science, 198.1056 (1977),

b) Proceedings of the National Academy of Sciences of the United States of America, 75, 5765 (1978)

c) Nature, 281, 18 (1979),

d) Biochemistry, 1980, 6096 (1980);

5) Science, 203, 614 (1979);

6) a) Nucleic Acids Research, 8, 5491 (1980),

b) Nucleic Acids Research, 8, 5193 (1980),

c) Tetrahedron Letters, 21, 4159 (1980),

d) Nucleic Acids Research, 8, 2331 (1980);

7) a) The Journal of Biological Chemistry (J. Biol. Chem.), 241, 2014 (1966),

b) Nucleic Acids Research, 1, 1665 (1974);

8) Nucleic Acids Research, 8, 5753 (1980);

9) Chemical and Pharmaceutical Bulletin. 26, 2396 (1978);

10) Genetic Engineering, 1978, 17 (1978);

11) Guidlines for Recombinant DNA Experiment, published in August 1979;

12) The Japanese Journal of Pharmacology, 21, 325 (1971);

13) a) Chemische Berichte, 105, 2508 (1972),

b) Chemische Berichte, 105.2515 (1972);

14) Tetrahedron Letters, 1978, 2727 (1978);

15) Proceedings of the National Academy of Sciences of the United States of America, 73, 3900 (1977);

16) Gene, 2, 95 (1977);

17) Proceedings of the National Academy of Sciences of the United States of America, 76, 106 (1979);

18) The operon, 31 (1980) (by J. H. Miller, W.S. Resnikofî; published by the Cold Spring Harbor Laboratory);

19) Proceedings of the National Academy of Sciences of the United States of America, 74, 1507 (1977);

20) Additional volumes of "Protein, nucleic acid & enzyme", last vol., P. 19 (1973);

21) Methods in Enzymology, 68, 265 (1979);

22) Microbiological Reviews, 44, 1-50 (1980);

23) Nature, 224, 768 (1969);

24) Nature, 217, 1110 (1968).

R

Claims (36)

659,826 2 1. Method for obtaining a strain of a microorganism transformed by a plasmid comprising a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is the valine, characterized in that it comprises the following operations: (1) chemical synthesis of a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is valine, (2) insertion of this gene into a vector plasmid capable of proliferating in a predetermined host cell, so as to produce a chimeric plasmid capable of proliferating in this cell, and (3) transformation of the host cell with this chimeric plasmid. 2. Method according to claim 1, characterized in that the host cell is an E. coli belonging to the genus Escherichia. 3. Method according to claim 2, characterized in that VE. coli is VE. coli K 12C600, FERM BP-115. 4. Method according to claim 2, characterized in that the vector plasmid is the plasmid pBR322. 5. Method according to claim 4, characterized in that the chimeric plasmid is pMG in which said structural gene is incorporated into the Pst I recognition site of the plasmid pBR322. 6. Method for obtaining a strain of a microorganism transformed by a plasmid comprising a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is the valine, characterized in that it comprises the following operations: (1) chemical synthesis of a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid located at the C end of the secretin is valine, (2) obtaining a vector plasmid capable of using the lactose operon and capable, moreover, of proliferating in a predetermined host cell, (3) insertion of this structural gene into the vector plasmid so as to produce a chimeric plasmid capable of proliferating in this cell, and (4) transformation of the host cell by this chimeric plasmid. 7. Method according to claim 6, characterized in that the host cell is an E. coli belonging to the genus Escherichia. 8. Method according to claim 7, characterized in that the host cell is VE. coli XA35, FERM BP-116 which is a strain in which the β-galactosidase gene is lacking and in which the i gene is also lacking. 9. Method according to claim 7, characterized in that the lactose operon comes from VE. coli. 10. The method of claim 9, characterized in that the vector plasmid contains all or part of the lactose operon of the DNA of the VE chromosome. coli and that it is able to proliferate in VE. coli. 11. The method of claim 10, characterized in that the vector plasmid is prepared from a transducer phage, containing all or part of the lactose operon, and a plasmid from VE. coli. 12. Method according to claim 11, characterized in that the transducer phage is chosen from the following phages: pldl, F'-lac, 080 dplac, Â.h80dlac and Âplac. 13. Method according to claim 11, characterized in that the plasmid originating from VE. coli is chosen from the following plasmids: pBR 322, pSC 101 and Mvl. 14. Method according to claim 12, characterized in that the phage is Xplac 5. 15. The method of claim 13, characterized in that the plasmid is pBR 322. 16. The method of claim 14, or claim 15, characterized in that the vector plasmid is pRE which is constituted by the 3.8 Md fragment of Xplac 5 linked to the largest of the fragments obtained by digestion of pBR 322 with Eco RI and Hind III. 17. The method of claim 16, characterized in that the chimeric plasmid is pLS which is an insertion product of said structural gene in the plasmid pRE at its Eco RI recognition site. 18. The method of claim 1, characterized in that the structural gene is a gene expressing 27-deamidosecretin, part of a double-stranded polydeoxyribonucleotide. 19. Method according to claim 18, characterized in that the structural gene has the following sequence of bases: His Ser Asp Gly Thr Phe Thr CAC -TCA -GAT - GGT -ACT -TTC - ACC- GTG -AGT -CTA - CCA - TGA - AAG- TGG - Ser Glue Leu Ser Arg Leu Arg TCA - GAA -CTA - TCT -CGT -CTA - CGT - AGT - CTT -GAT -AGA -GCA -GAT- GCA - Asp Ser To the Arg Leu Gln Arg GAT -TCA - GCA -CGC - CTC -CAG- CGC - CTA -AGT - CGT - GCG -GAG - GTC - GCG- Leu Leu Gin Gly Leu Val TTG -CTG - CAA -GGT -CTC - GTT - AAC -GAC -GTT -CCA -GAG - CAA - 20. The method of claim 18 or claim 19, characterized in that the structural gene contains a codon designating methionine, at its 5 'end, and one or more codons to designate a translation stop, at its end 3 '. 21. The method of claim 20, characterized in that the 5 'end of the methionine codon and the 3' end of the translational stop codon of the structural gene are 3m (3m +1) or (3m + 2 ) base pairs (m being equal to 0 or an integer at least equal to 1), these bases being chosen at random. 22. The method of claim 21, characterized in that the double-stranded polydeoxyribonucleotide has a base sequence for recognition of any desired restriction enzyme at its two ends and has, at the outside of the resulting ends, at least two pairs bases so as to produce blunt ends chosen at random. 23. The method as claimed in claim 18, characterized in that the double-stranded polydeoxyribonucleotide has the following structure: Met ACCTGCAGCC - ATG-TGGACGTCGG-TAC- His Ser Asp Gly Thr Phe Thr CAC -TCA -GAT -GGT -ACT- -TTC - ACC- GTG -AGT -CTA -CCA -TGA - AAG- TGG - Ser Glue Leu Ser Arg Leu Arg TCA -GAA -CTA -TCT -CGT -CTA - CGT - AGT -CTT -GAT -AGA -GCA -GAT- GCA - Asp Ser To the Arg Leu Gln Arg GAT -TCA -GCA -CGC -CTC -CAG- ■ CGC - CTA -AGT -CGT -GCG -GAG - GTC - GCG- Leu Leu Gin Gly Leu Val TTG -CTG -CAA -GGT -CTC - GTT - AAC -GAC -GTT -CCA -GAG - CAA - END END TGA -TAG - GGCTGCAGGT ACT - ATC - CCGACGTCCA. 24. Escherichia coli transformed by a plasmid into which is incorporated a structural gene, desamidosecretin, corresponding to the polypeptide whose amino acid located at the C end is valine, this plasmid being capable of proliferating in a host cell predetermined, obtained by the method according to claim 1. 5 10 15 20 25 30 35 40 45 50 55 60 65 3 659,826 25. Escherichia coli transformed by a plasmid into which is incorporated a structural gene, desamidosecretin, corresponding to the polypeptide whose amino acid located at the C end is valine, this plasmid being capable of using the lactose operon and also being able to proliferate in a predetermined host cell, obtained by the method according to claim 6. 26. E. coli according to claim 25, characterized in that it is E. coli XA35 (pLS58), ATCC 39040. 27. A process for the production of 27-desamidosecretin, characterized by the use of the transformed Escherichia coli, according to one of claims 24 to 26, comprising the culture of VE. coli transformed and recovery of the desamidosecretin produced. 28. 27-desamidosecretin consisting of a polypeptide represented by the following amino acid sequence: His - Ser - Asp - Gly - Thr - Phe - Thr - Ser - Glu - Leu -Ser - Arg - Leu - Arg - Asp - Ser - Ala - Arg - Leu - Gin -Arg - Leu - Leu - Gin - Gly - Leu - Val - OH wherein Val-OH indicates that the C-terminus of the valine polypeptide is a carboxylic acid, obtained by the method of claim 27. 29. A chimeric plasmid, comprising a structural gene expressing 27-desamidosecretin, as an intermediate product in the preparation process according to claim 1. 30. A chimeric plasmid, comprising a structural gene expressing 27-DAS as an intermediate product in the preparation process according to claim 6. 31. Double-stranded polydeoxyribonucleotide, comprising a structural gene expressing 27-desamidosecretin, as an intermediate product in the preparation process according to claim 1. 32. Double-stranded polydeoxyribonucleotide according to claim 31, characterized in that the structural gene has the following sequence of bases: His Ser Asp Gly Thr Phe Thr CAC -TCA -GAT -GGT -ACT- -TTC - ACC- GTG -AGT -CTA -CCA -TGA AAG - TGG- Ser Glue Leu Ser Arg Leu Arg TCA -GAA -CTA -TCT -CGT -CTA - CGT - AGT -CTT -GAT -AGA -GCA -GAT- GCA- Asp Ser To the Arg Leu Gln Arg GAT -TCA -GCA -CGC -CTC -CAG- • CGC - CTA -AGT -CGT -GCG -GAG - GTC - GCG- Leu Leu Gin Gly Leu Val TTG -CTG -CAA -GGT -CTC - GTT - AAC -GAC -GTT -CCA -GAG - CAA- 33. Double-stranded polydeoxyribonucleotide according to claim 31 or claim 32, characterized in that the structural gene contains a codon designating methionine, at its 5 'end, and one or more codons to designate a translation stop, its 3 'end. 34. Double-stranded polydeoxyribonucleotide according to claim 33, characterized in that the 5 'end of the methionine codon and the 3' end of the translation stop codon have 3m, (3m +1) or (3m + 2 ) base pairs (m being equal to 0 or an integer at least equal to 1), these bases being chosen at random. 35. Double-stranded polydeoxyribonucleotide according to claim 34 having a base sequence of recognition of any desired restriction enzyme at its two ends and having, outside the resulting ends, at least two pairs of hases. in a way. to produce blunt ends chosen has hassrrd "" "" 36. Double-stranded polydeoxyribonucleotide according to claim 31, having the following structure: Met ACCTGCAGCC-ATG-TGGACGTCGG-TAC- His Ser Asp Gly Thr Phe Thr CAC -TCA -GAT -GGT -ACT -TTC - ACC- GTG -AGT -CTA -CCA -TGA -AAG - TGG - Ser Glue Leu Ser Arg Leu Arg TCA -GAA -CTA -TCT -CGT -CTA -CGT - AGT -CTT -GAT -AGA -GCA -GAT - GCA - Asp Ser To the Arg Leu Gin Arg GAT -TCA -GCA -CGC -CTC -CAG - CGC - CTA -AGT -CGT -GCG -GAG -GTC - GCG - Leu Leu Gin Gly Leu Val TTG -CTG -CAA -GGT -CTC -GTT AAC -GAC -GTT -CCA -GAG - CAA - END END TGA -TAG - GGCTGCAGGT ACT-ATC-CCGACGTCCA.