GB2103220A - 27-Desamidosecretin and production of same through recombinant DNA technique - Google Patents

Abstract

27-desamidosecretin defined by a specific amino acid sequence is disclosed. Production of the protein through the recombinant DNA technique is also disclosed with the transformed E. coli strain, double- stranded polydeoxyribonucleotides and plasmids utilized in the production.

Description

SPECIFICATION 27-Desamidosecretin and production of same through recombinant DNA technique Technical field This invention relates to 27-desamidosecretin ("27-" may sometimes be omitted hereinafter) and production thereof. More particularly, the present invention relates to a method of producing desamidosecretin through genetic engineering technique with the use of a structural gene for desamidosecretin which has been chemically synthesized.

Background art Secretin is a compound known as one of gastrointestinal hormones. It has heretofore been known to have physiological activities such as an action to promote secretion of water and bicarbonates from the pancreas and has been practically used to pancreatic function test or as a therapeutical agent for duodenal ulcer.

Secretin is a polypeptide having the following formula: 5 10 15 His-Ser-Asp-Gly-Thr-P he-Thr-Ser-GI u-Leu-Ser-Arg-Leu-Arg-Asp-Ser- 20 25 27 Ala-Arg-Leu-Gln-Arg-Leu-Leu-Gln-GlyLLeu-ValLNH2 In order to obtain this secretin, such methods as extraction) of a naturally occurring secretin (conventionally porcine secretin, but the amino acid sequence of bovine secretin has also been identified to be the same as that of porcine secretin) and chemical synthesis2) have been practiced.

Each of these methods, however, involves problems with respect to cost, rapidity, production yield, etc.

Further, according to the method of the prior art in which the secretin from the samall intestine of hogs is extracted and purified, a difficulty has been encountered in that the multitude of gastrointestinal hormones (motilin, VIP, CCK-PZ, GIP, GLI, etc.) obstructs accurate bioassay or radioimmunoassay of the secretin.

Meanwhile, recent developments of the so-called genetic engineering technique with the use of a synthetic gene are so marked that various physiologically active polypeptides are now beginning to be commercially producted by utilization of this technique. Accordingly, if secretin can be produced by genetic engineering, it would be possible to discover clues to the solution of the problems as mentioned above in the production methods of the prior art.

However, while it has been generally clarified that the production of a polypeptide through genetic engineering technique with the use of a synthetic gene is possible by the steps of chemical synthesis of a structural gene, insertion of the gene into an appropriate plasmid as a vector, transformation of an appropriate host and production and recovery of a desired polypeptide by culturing of the transformant, it is not necessarily possible to know assuredly beforehand whether or not a particular polypeptide desired can be produced to have a desired physiological activity according to this method.

Regarding this problem in the case of secretin, since the C-end of the secretin polypeptide is an amide of valine as mentioned above, the end of the structural gene DNA expressing the polypeptide must necessarily comprise the codon corresponding to valine, and therefore the polypeptide produced is considered to have a C-end which is valine. With such reasoning, it is doubtful whether the polypeptide obtained will exhibit the physiological activity possessed by secretin.

Summary of the invention The present invention has been accomplished by confirming that a structural gene expressing a secretin derivative with its C-end being valine not in the form of an amide, namely, desamidosecretin, can be chemically synthesized; that said gene can be inserted into an appropriate plasmid as a vector to form a chimera plasmid; that transformation of an appropriate host cell with the chimera plasmid as well as production and recovery of desamidosecretin by culturing of the transformant are possible; that the desamidosecretin produced has an activity similar to that of secretin; and further that the expression system of lactose operon can be utilized in the production of the desamidosecretin and the yield of the desamidosecretin can be increased to a great extent by utilization of this expression system.

Thus, the 27-desamidosecretin according to the present invention comprises a polypeptide represented by the amino acid sequence H is-Ser-Asp-Gly-Thr-P he-Thr-Ser-Gl u-Leu-Ser-Arg-Leu-Arg-Asp-Ser-Ala-Arg Leu-Gln-Arg-Leu-Leu-Gln-Gly-Leu-Val-OH wherein Val-OH shows that the C-end of the polypeptide comprising valine is a carboxylic acid.

The present invention also provides a method of producing the 27-desamidosecretin, which comprises the steps of: (1) synthesizing chemically a structural gene for the desamidosecretein corresponding to the polypeptide in which the amino acid at the C-end of secretin is valine, (2) inserting this gene into a vector plasmid capable of proliferation in a predetermined host cell to make a chimera plasmid capable of proliferation in the cell, (3) transforming the host cell with the chimera plasmid, and (4) culturing the resultant transformant and recovering the desamidosecretin produced.

Another and preferable method of producing the 27-desamidosecretin according to the present invention comprises the steps of: (1) synthesizing chemically a structural gene for the desamidosecretin corresponding to the polypeptide in which the amino acid at the C-end of secretin is valine, (2) providing a vector plasmid capable of utilizing the lactose operon and also capable of proliferation in a predetermined host cell, (3) inserting the structural gene into the vector plasmid to make a chimera plasmid capable of proliferation in the cell, (4) transforming the host cell with the chimera plasmid, and (5) culturing the resultant transformant and recovering the desamidosecretin produced.

The method of producing the 27-desamidosecretin provided by the present invention comprises, in its broadest aspect, the steps of: (1) providing a chimera plasmid which comprises a fragment of a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid at the C-end of secretin is valine, the chimera plasmid being capable of proliferation in a predetermined host cell and being capable of expressing the structural gene for a desamidosecretin in the host cell; (2) transforming the host cell with said chimera plasmid; and (3) culturing the resulting transformant and recovering the desamidosecretin produced.

A typical example of the host cells to be used in the transformation method as defined above is Escherlchia coli belonging to the genus Escherichia, and the lactose operon is from Escherichia coli.

Thus, the preferable method of producing the 27-desamidosecretin according to the present invention may be defined alternatively as a method which comprises culturing a desamidosecretinproducing microorganism belonging to Escherichia coli in which has been incorporated a plasmid containing the desamidosecretin structural gene and recovering the desamidosecretin produced in utilization of the lactose operon in the Escherichia coli.

In another aspect, the present invention also provides a culture of Escherichia coli to be used in practice of the present invention, which Escherichia coli has been transformed so as to have such a phenotype that the strain produces upon culturing thereof the 27-desamidosecretin. The present invention thus provides an Escherichia coli, characterized by having been transformed with a plasmid which has incorporated or inserted therein a chemically synthesized structural gene of the desamidosecretin corresponding to the polypeptide of which the amino acid at the C-end is valine, which plasmid is capable of utilizing the lactose operon and also capable of proliferation in a predetermined host cell.

In further aspect of the present invention, there is provided a double-stranded polydeoxyribonucleotide which comprises the structural gene expressing the 27-desamidosecretin.

In still further aspect of the present invention, there is provided a plasmid which comprises the structural gene expressing the 27-desamidosecretin.

In accordance with the present invention as summarized above, the problems in extraction and purification encountered in the case of extraction of natural secretin which has been relied upon as a sole commercially feasible means can thus be avoided when the extract of the recombinant gene according to the present invention is used as the starting material.

According to the method of the present invention, it is possible to change one kind or several kinds of amino acids constituting a peptide, whereby it becomes possible to study the correlations between the structures and activities of peptide hormones. This means, in other words, that there is provided a method of obtaining easily any desired amino acid-substituted derivative of a physiologically active peptide. Formation of a polypeptide derivative by manipulation of genes may be said to be the best method in view of the fact that formation of a high molecular polypeptide derivative by chemical synthesis is very difficult according to the presently available methods.

Further, by use of the host-vector system employed in the preferable method for production of the desamidosecretin as mentioned above, it has been made possible to produce desamidosecretin having a secretin activity corresponding to about 3 x 104 molecules per one host cell, which may be said to be a practically applicable yield. Also, in terms of the fused protein as described hereinafter, there is a yield of 2.85x 1 05 molecules/cell, which value suggesting that a high yield production of the desamidosecretin is made possible by producing trait expression with the use of the chimera plasmid according to the present invention in bacteria without destructive (proteolytic) action by proteases.

There also exist polypeptides other than secretin having a C-end in the form of an amide, and the present invention provides a suggestion with respect to the production of desamide derivatives of such other polypeptides as well as expression of physiological activities thereof.

Detailed description of the invention 1. Desamidosecretin The 27-desamidosecretin according to the present invention is a polypeptide represented by the amino acid sequence formula (I) below: H is-Ser-Asp-G Iy-Thr-Phe-Thr-Ser-Glu-Leu-Ser-Arg-Leu-Arg-Asp-Ser-Ala-Arg- Leu-Gl n-Arg-Leu-Leu-G ln-G ly-Leu-Val-OH (I) In the formula (I), His and others are symbols recognized in the art as indicating amino acids such as hystidine, etc.

The polypeptide is different from the secretin polypeptide in that the valine at the C-end is not in the form of an amide (Val-NH2) but in the form of carboxyl (Val-OH).

The desamidesecretin, which has a physiological activity similar to secretin such as an action to promote secretion of pancreatic juice, can itself be used as a secretin-like physiologically active polypeptide. On the other hand, the desamidosecretin can also be utilized after it has been amidated at the C-end carboxyl moiety secretin. Amidation may be performed by a pure chemical method, an enzymatic method or a biological method in vivo.

2. Production of the desamidosecretin The desamidosecretin, while it can be obtained by modification of the C-end of secretin, is preferably produced through genetic engineering technique according to the present invention by chemically synthesising a structural gene for this polypeptide, preparing a plasmid so that the polypeptide may be expressed, transforming a host cell with the plasmid, producing the desired polypeptide by culturing the transformant, and recovering the desamidosecretin.

11 Structural gene (1) Design of gene The base sequence of DNA of which the structural gene of secretin is constituted is unknown.

Therefore, from among some codons designating the amino acids constituting the peptide, those satisfying the following conditions are selected for synthesis of the DNA: (i) it should be so controlled that the region enriched in A-T base pairs will not subsequently follow the region enriched in G-C base pairs; and (ii) it should also be so controlled that each synthetic fragment as hereinafter described will not have an undesirable complementary base sequence intramolecularly or intermolecularly.

It is also desirable for convenience of easy selection of transformed strain to design the structural gene so that one or more restriction enzyme recognition base sequences will be contained therein. In the case of the desamidoscretin, it is preferable that the gene have recognition base sequences of Hinf i and Hae II.

From such a point of view, typical examples of codons designating amino acids in the desamidosecretin structural gene are as follows.

Amino acid Codon His CAC Ser TCT, TCA Asp GAT Gly GGT Thr ACT, ACC Phe TTC Glu GAA Leu TTG, CTC, CTA, CTG, TTA, CTT, Arg CGT, CGC Ala GCA Gin CAA, CAG Val GTT Accordingly, a preferred embodiment of the structural gene of the desamidosecretin of the invention has a base sequence as shown hereinafter in the experimental examples and the drawing (in the drawing, the base sequence is of course the moiety from CAC corresponding to His to GTT corresponding to Val).

That is, the present invention relates, in one aspect, to a double-stranded polydeoxynucleotide comprising a structural gene expressing the 27-desamidosecretin as mentioned above and a preferred embodiment of the structural gene has a base sequence as shown below.

His Ser Asp Gly Thr Phe Thr Ser Glu CAC-TCA-GAT-GGT-ACT-TTC -ACC-TCA-GAA- GTG-AGT -CTA-CCA-TGA-AAG-TGG-AGT-CTT- Leu Ser Arg Leu Arg Asp Ser Ala Arg CTA-TCT-CGT-CTA-CGT-GAT-TCA-GCA-CGC- GAT-AGA-GCA-GAT-GCA-CTA-AGT- CGT-GCG- Leu Gln Arg Leu Leu Gln Gly Leu Val CTC-CAG-CGC-TTG -CTG-CAA-GGT- CTC- GTT GAG GTC-GCG-AAC-GAC-GTT-CCA-GAG-CAA The method of expressing such a structural gene is described in detail in, for example, Japanese Patent Laid-Open No. 92696/1979.In the case where pBR 322 is used as the plasmid in which the gene is to be incorporated, and the desamidosecretin is to be expressed as a fused protein with the lactamase operon thereof, the site at which the gene is to be incorporated is suitably the restriction enzyme Pst I recognition site in the operon. That is, the codon ATG for methionine which is the site to be attacked by cyanogen bromide is provided on the 5'-end side of the structural gene, while one or more stop codons are provided on the 3'-end. Subsequently, so as to be synchronized3) with the frame beginning with the start codon of the lactamase gene, bases randomly selected in number of 2+3n (n=O, 1, 2,... ) are imparted to the gene on the 5'-side of ATG, and further Pst I recognition base sequence for forming cohesive ends at both ends, respectively.In general, structural genes are designed and synthesized4) with both cohesive ends exposed, but, if desired, both ends may be made blunt ends, whereby it is necessary to provide two or more bases randomly selected at further outside parts of the recognition base sequence for enhancement of hydrolysis efficiency of the restriction enzyme.

The optional pairs of bases to be provided on the 5'-side of ATG are 3m pairs, (3m+1) pairs or (3m+2) pairs where m is an integer of O or 1 or more.

Having thus weighed these considerations, a preferred embodiment of the gene for the desamidosecretin suitable for use in the present invention is that as shown hereinafter in the experimental examples.

That is, a preferred embodiment of the gene for the desamidosecretin has the structure shown below.

Met ACCTGCAGCC - ATC TGGACGTCGG - TAC His Ser Asp Gly Thr Phe Thr Ser Glu CAC-TCA-GAT-GGT-ACT-TTC -ACC-TCA-GAA- GTG-AGT-CTA-CCA-TGA-AAG-TGG-AGT- CTT- Leu Ser Arg Leu Arg Asp Ser Ala Arg CTA-TCT-CGT-CTA-CGT-GAT-TCA-GCA-CGC- GAT-AGA-GCA-GAT-GCA-CTA-AGT-CGT-GCG- Leu Gln Arg Leu Leu Gln Gly Leu Val CTC-CAG-CGC-TTG-CTG-CAA-GGT-CTC-GTT - GAG-GTC-GCG-AAC-GAC-GTT-CCA-GAG-CAA- END END TGA - TAG - GGCTGCAGGT ACT-ATC- CCGACGTCCA (2) Synthesis For the synthesis of the gene which has been designed as described above, each of both + and strands may be divided into some fragments. These fragments can be chemically synthesized and then respective fragments are linked together. It is preferred to divide each strand into about 1 6 fragments, each comprising 9 to 1 6 bases so that 6 to 7 bases will overlap each other.

As the method for synthesis of each fragment, Invention may be made of the diester method5), the triester method6, the solid phase method4), the liquid phase method or the method in which an enzyme is employed7). In view of synthesis time, yield, and purification, the solid phase method according to the triester method is the most suitable.

As to details about synthesis, reference should be made to the literatures referred to above and the experimental examples as described hereinafter.

(3) Purification When an oligonucleotide is synthesized, separation and purification of the final product becomes generally more difficult with elongation of the strand length. Particularly, in the solid phase synthetic method, oligonucleotide blocks suitably protected are condensed stepwise, and therefore purification cannot be easily practiced by conventional techniques such as gel filtration, gel electrophoresis, ionexchange column, high speed liquid chromatography, etc.

In the reverse phase column, the retention time differs greatly depending on whether the oligonucleotide has a lipophilic protective group or not. Accordingly, with the use, in the final condensation step, of an oligonucleotide block having a protective group stable under the conditions for removing other protective groups followed by an appropriate removal of the other protective groups, a mixture of oligonucleotides having the stable protective groups only in the desired final product can be obtained. By utilization of the lipophilic nature of the protective group, the desired final product can be separated from the mixture of the unreacted species through the reverse phase column, which step is followed by removal of the protective group to produce the desired oligonucleotide.

According to this method, the oligonucleotide thus synthesized can be separated and purified with good efficiency from the mixture of the unreacted species.

(4) Phosphorylation and ligation The thus prepared synthetic fragments are successively ligated with each other by the use of a DNA ligase. In order to make the synthetic fragments substrates for the enzyme, it is necessary to phosphorylate 5'-hydroxyl groups in the fragments.

For this purpose, a polynucleotide kinase is generally employed, but chemical phosphorylation may also be possible8). While ligation of fragments is generally practiced by the use of a DNA ligase, it is also possible to use the method in which the phosphoric acid groups at the 5'-ends are activated by a suitable method (for example, imidazolylation) and chemically ligated with the strand on the opposite side as template) 2) Preparation of a vector having the lactose operon In the preferable method for production of the 27-desamidosecretin according to the present invention, it is possible to use various plasmids containing all or part of the lactose operon from E. coli chromosome DNA and being capable of proliferation in E. call Preparation of those plasmids can be performed according to conventional methods well known in the field of molecular biology. The DNA containing the lactose operon may be obtained directly from the E. call chromosome, but various transducing phages containing all or a part of the lactose operon (e.g., Pldl, F'-lac, j80dplac, Sh80dlac, Aplac, etc.) are obtained, and therefore a necessary portion of the lactose operon may conveniently be taken out from these phages. Further, in order to make the plasmid capable of proliferation in E. coli, it is necessary to ligate the necessary portion of the above lactose operon with another plasmid from E.

coli (e.g., pBr322, pSC1O1, )LdV1) to form a single plasmid vector.

In one example of the present invention, the transducing phage Aplac523 > is used as the DNAcontaining lactose operon. Aplac5 DNA can be obtained from, for example, E. coli PK1 512 which is a lysogenic bacterium according to conventional method201. This Aplac5 has the region from the midway in the i-gene to the midway of the y-gene of the lactose operon, advantageously having no other E. coli gene than the lactose operon20), and therefore is preferable in the present invention. As the plasmid from E. coli, use is made of pBR322, which has been selected for such reasons as that it is one of the most readily available plasmids, that all of the base sequence is determined, and that it has ampicillin resistant and tetracycline resistant marker genes.Further, in ligation of these genes, each gene (Aplac5 and pBR322) is treated with restriction enzymes EcoRI and Hindlll, and the fragment of 3.8 Md (mega dalton15 for Aplac5 and the larger fragment15) for pBR322 are taken out, respectively, and linked together to make the desired plasmid vector. This vector is named pRE in the present invention.

The lactose operon from E. coli chromosome has been selected for expression of the desired desamidosecretin for such reasons as that a foreign gene can be inserted into the z-gene in the lactose operon at the recognition site of the restriction enzyme of EcoRI to be expressed in the form of a protein fused with p-galactosidase4al 17). that a protein can be produced in a large amount, that induced production may be effected by when an appropriate host microorgansim is used, and that the product can be easily and stably recovered as a fused protein and substantially in pure form.

Thus, it is desirable that the vector prepared in the present invention have only one recognition site of the restriction enzyme EcoRI, and for this purpose use is made of DNA fragments obtained by cleaving Aplac5 and pBR322 with EcoRI and Hindlli, respectively, as described above, which are in turn linked together.

3) Chimera plasmid (1) Preparation At an appropriate position in the vector designed for a foreign gene to be expressed as described above, the aforesaid desamidosecretin structural gene is incorporated. The incorporating operation per se can be conducted according to a conventional method well known in the field of molecular biology.

As to details about the method employed, reference should be made to the experimental examples as hereinafter described.

According to one embodiment of the present invention, pBR322 is used as the vector plasmid, and the gene is incorporated at the Pstl recognition site thereof to produce a chimera plasmid. In the present invention, this chimera plasmid is named pMG.

The principal reasons why pBR322 was selected as the vector plasmid are that it is one of the most readily available plasmids, that all of the base sequence is determined, and that it has ampicillin resistant and tetracycline resistance marker genes. pBR322 plasmid is under deposition at the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852, U.S.A. under the accession number ATCC3701 7. The principal reasons why the Pstl site was selected as the place for incorporation of the gene are that the lactamase operon can be utilized as such, that searching for the transform ant can easily be done because ampicillin resistance (Apr) is changed to ampicillin susceptibility (ApS), and that there is only one Pstl site in pBR322.

In the preferable method for production of the 27-desamidosecretin as described above according to another embodiment of the present invention, pRE1 which will be described later in more detail is used as the vector plasmid, and a gene containing the desamidosecretin structural gene is incorporated at the EcoRI recognition site thereof to produce a chimera plasmid. In the present invention, this chimera plasmid is named pLS.

Also in this preferred embodiment of the present invention, the gene containing the above structural gene is the preferable one, that is, having the base sequence as shown in the drawing. In order to incorporate this gene, which has recognition sites of restriction enzyme Pst I at both ends thereof, into pRE 1, it is necessary to convert the restriction enzyme recognition sites Pst I to EcoRI.

(2) Linker Thus, in the preferred method for production of the desamidosecretin of the present invention, a double-stranded DNA for the linker between the gene containing the structural gene and pRE 1 is needed. That is, the double-stranded DNA is necessary to have two recognition sites of restriction enzymes of Pst I and EcoRI, with the base pairs between both of the restriction enzyme recognition sites being 3n+ 1 (n=1,2, 3... ) so as to be synchronized with the reading frame for translation beginning with the start codon of p-galactosidase gene (the kinds of base pairs may be randomly selected, as long as no nonsense codon appears).It is only necessary, however, that the portion for the linker will ultimately have the above function, and it is possible to obtain the linker by the method by which the above structural gene is synthesized. In one embodiment of the present invention, as one example of such a synthetic method, the linker was obtained by applying a procedure as described below.

First, a single-stranded DNA as shown below having recognition sites of restriction enzymes Pst I and EcoRI is designed.

wherein 1, 2, 9-12, 1 9, 20 may be randomly selected as long as they satisfy the following conditions.

Condition(1): 1 and 10,2and9, 11 and 20, and 12 and 19areeachpairofbases complementary to each other; Condition (2): the sequence of 9, 10, 11 is not a so-called nonsense condon.

The single-stranded DNA thus designed (this is called a prelinker in the present invention), due to complementarity of its own, assumes a long chain double-stranded DNA structure having a number of nicks (cleaved points without gap formed on one of the two strands of DNA). Accordingly, a doublestranded DNA without gap can be formed therefrom by an action thereon of a DNA ligase. The thus prepared double-stranded DNA is a double-stranded polyDNA having alternately the recognition sites of the restriction enzymes Pst I and Eco RI, namely recurring base sequences of the above 1-20.

Therefore, by the treatment of the double-stranded DNA with the restriction enzyme Pst I, a linker of Pstl-EcoRl-Pst I having a cohesive ends can be obtained.

Generally speaking, a linker having recognition sites of two restriction enzymes has various uses and their objects have been achieved in the prior art with the use of two kinds of oligomers. The idea herein mentioned enables achievement of its object with the use of one kind of oligomer according to the method as described above, and the single-stranded DNA is capable of expression as shown below.

X'n... X'2 X'1 Restriction enzyme A recognition site X1 X2... Xn Ym Y2Y1 Restriction enzyme B recognition site Y'1 Y'2 ... Y'm wherein X and X', Y and Y' are any desired bases (A, G, C, T) having complementarity to each other (n and m=0,1, 2...).

In case of the present invention, A is EcoRI and B Pst I, and n+m=3p+1 (p=1, 2... ), wherein p is desired to be 1 or 2.

(3) Determination of direction Determination of direction of the 27-desamidosecretin gene incorporated in the plasmid may be conducted by cleaving the specific site contained in the structural gene with an enzyme (Hae II) recognizing said specific site, cleaving another site at a certain position outside of the structural gene, and analyzing the size of the fragment obtained.

4) Transformation (1) Host cell A typical example of the host cell to be transformed with the use of a chimera plasmid having incorporated therein the desamidosecretin structural gene as described above, such as the above pMG, is an E. coli strain K12C600 (FERM BP-1 1 5, deposited at the Institute of Fermentation Research, Agency of Industrial Science and Technology, Japan). The E. coli strain K12C600 is described in the literature241 and its bacteriological properties are also disclosed therein.

Another example of the chimera plasmid having incorporated therein the desamidosecretin structural gene as described above is pLS such as pLS 58 to be used in the preferred method for production of the desamidosecretin according to the present invention. A typical example of the host cell to be transformed with the use of pLS 58 is the E. coli strain XA35 belonging to Escherichia coli, which was derived from the known strain, E. coli K12 strain22), having the property as shown below, the remaining properties not being different from those of the K 12 strain.

[Smr, Lac-(i3-, z-)] Transformation with the chimera plasmid having incorporated therein the desamidosecretin structural gene according to the present invention is possible in all E. coli strains. However, in order to recover the desamidosecretin as a fused protein with ,B-galactosidase, it is preferable to use a strain deficient in the gene of p-galactosidase for prevention of the presence of normal ss-galactosidase mixed in the protein. Generally speaking, production of protein is controlled by the repressor gene (i-gene) of the lactose operon. Therefore, when using a wild type E. coli, induced production of the fused protein is possible with an inducer, e.g.IPTG (isopropyl thiogalactoside); when using a strain with a gene of high temperature susceptibility, the fused protein can be induced by elevating the temperature; or when using a strain deficient in the i-gene, the fused protein can always be produced18).

In the preferred method for production of the desamidosecretin according to the present invention, the E. coli strain XA35 which was a strain deficient in the genes of ,B-galactosidase and i-gene was employed.

It should also be noted that the transformation with the plasmid having incorporated therein the desamidosecretin structural gene is not limited to E. coli as host, but an appropriate host can also be selected from a wider spectrum of bacteria species when an appropriate vector is selected, as is well known in the field of molecular biology. A typical example of such a host cell can be found in the disclosure in the aforesaid Japanese Patent Laid-Open No. 92696/1 979.

(2) Transformation The transformation operation itself can be conducted according to conventional techniques well known in the field of molecular biology. As to details about the method employed, reference should be made to the experimental examples hereinafter described.

(3) Transformant A typical example of the transformants is a transformant obtained by transforming K12C600 with pMG. In the present invention, this is named K12C600 (pMG103).

The transformant K12C600 (pMG103) has genetic properties as shown below: [m-, r-, F-, lacY, Leu, Thr, tonA, supE, recBC,] Another example of the transformants is a transformant obtained by transforming the E. coli strain XA35, to be used in the preferred method for production of the desamidosecretin of the present invention as described above, with pLS 58. This is named E. coli (pLS58) in the present invention.

The transformant E. coli strain XA35 (pLS 58) is different from the E. coli strain XA35 in the following properties as clarified in the experimental examples hereinafter described.

[Apr, Lac+] 5) Production of the desamidosecretin The desamidosecretin can be produced by culturing the thus transformed bacteria according to a conventional method. As to details about the method, reference should be made to the experimental examples described hereinbelow.

3. Experimental examples Example 1 Design of the desamidosecretin gene 1) A gene with the base sequence consisting of a combination of blocks A, B and C as shown in the accompanying drawing was designed. These blocks respectively comprise framgents as shown below.

Block Fragment A S-1-S-4, S-6 B S-S, 5-7-5-10, S-12 C S-i 1,S-13-S-16 2) The procedure of designing was as described below.

(1) Selection of codons Codons were selected as shown in the drawing.

(2) The codon ATG for methionine was added to the N-end so that the polypeptide synthesized will be cleaved at this site by the chemical treatment (+CNBr).

(3) To the C-end, there were added two translation termination codons (TAG or TGA) so that no superfluous peptide will be produced.

(4) For the purpose of synchronizing with the frame beginning with the start codon of the lactamase gene, two pairs of bases selected (generally 2+3n (n=0, 1, 2, 3,... ) were added upstream of the methionina (5) Pst I sites were added to both ends.

Any desired number of base pairs may also be added immediately before the Pst I site on the downstream side if it is convenient in synthesis of each fragment.

(6) Finally, two pairs of optionally selected bases were added to both ends in order to enhance hydrolysis efficiency of the restriction enzyme Pst I.

The gene designed as described above contains the Hinf I site and the Hae II site in the 27desamidosecretin structural gene.

Chemical synthesis of fragment 1) Synthesis Synthesis of fragments was conducted according to the solid phase method disclosed in the literature6al. However, isolation and purification of the synthetic fragments were performed according to the improved method as described below.

The synthesis yields of respective fragments-are shown in the following Tabie: Fragment Base sequence Chain length Yield ( /0) 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 Fragment Base sequence Chain length Yield { /OJ S-11 GCTTGCTGCAAGGT 14 22 S-12 AGCAAGCGCTGGAGG 15 44 S-13 CTCGTTTGATAGG 13 47 S-14 AAACGAGACCTTGC 14 42 S-15 Same as S-2 S-16 ACCTGCAGCCCTATC 15 32 2) Purification To 50 mg of the resin which underwent the solid phase synthesis, 0.5 M of a-picolinic aldoxime tetramethylguanidine and 100 to 200 ul of a mixture of dioxane: water (1:1)14) were added, and the mixture was left to stand at room temperature for several hours. Then, 2 ml of conc. aqueous ammonia was added to the mixture, which was in turn left to stand at 550C overnight under tight sealing with a stopper. The resulting mixture was filtered for separation of the resin, and the filtrate was concentrated and subjected to gel filtration. Elution was carried out with 50 mM TEAB buffer (pH 7.5), and the fractions eluted into the void volume were collected.These were concentrated and applied to HPLC of a reversed phase column C-18 (Waters: "Radial Pack A", diameter 8 cmx cm) to accomplish elution in 0.01 M ethylenediamine diacetate buffer (pH 7.8) with a concentration gradient of acetonitrile from 10% to 32% at a flow rate of 2 ml/min. over 16 minutes. the fractions eluted within 11 to 12 minutes were collected. During this procedure, oligonucleotides having no trityl group were eluted as injection peak. The eluate was concentrated and, after addition thereto of 1 ml of 80% acetic acid, left to stand at room temperature for 1 5 minutes. Tritanol was removed by extraction, and the aqueous layer was concentrated and applied again to the reverse phase column.Under the same conditions as previously described, elution was performed with a concentration gradient of acetonitrile of from 0% to 20%, and the fractions eluted within 12 to 13 minutes were collected.

Phosphorylation Each fragment (30 y9) was dissolved in 30 mM Tris-HCI buffer (pH 7.5), and 60 uCi (19.8 pmol) of [y32p] ATP and 2 ,rl (9 units) of T4 polynucleotide kinase were added to the solution to make up a total quantity of 50 u1 which step was followed by incubation at 370C for 20 minutes. Subsequently, 10 equivalents relative to the fragment of ATP and T4 polynucleotide kinase (9 units) were added to the mixture, which step was followed by incubation at 370C for 20 minutes. The reaction was stopped by heating at 1 000C for 2 minutes, and the product was purified by gel filtration. Each fragment was confirmed by 20% gel electrophoresis.

Ligation of fragments Each of 0.05A26o of S-1, S-2, S-3, 5-4 and S-6 was dissolved in a buffer (20 mM Tris-HCI pH 7.5, 10 mM MgCl2, 10 mM DTT (dithiothreitol, 0.2 mM ATP (adenosine triphosphate)) to make up a solution with a total quantity of 30 ul. 1 yl (1 50 units) of T4-DNA ligase was added to the solution, and the mixture was left to stand at 1 00C overnight. The reaction was confirmed by 8% polyacrylamide gel electrophoresis. Similarly, blocks B and C were synthesized.

The reaction mixtures of block A and of block B were mixed together, and, after addition of 3 u1 of 0.2 mM ATP and 1 41 (150 units) of T4 DNA ligase, the mixture was left to stand at 1 OOC overnight.

After the reaction was confirmed by 8% polyacrylamide gel electrophoresis, the reaction mixture of block C was added, and the reaction was allowed to proceed similarly overnight. The progress of the reaction was confirmed by 5% polyacrylamide gel electrophoresis.

The reaction mixture was heated at 680C for 15 minutes and then cooled to room temperature.

Then, 1 5 u1 of 0.5M NaCI and 80 units of the restriction enzyme Pst I were added, and the mixture was left to stand at 370C overnight. The reaction was stopped by heating at 900C for one minute, and the reaction mixture was subjected to separation by 5% polyacrylamide gel electrophoresis. The band with the longest chain length was cut out and transferred to 1% low melting point agarose gel electrophoresis for extraction of the band obtained.

Cloning The plasmid pBR 322 (4 y9) was added to a mixture (total quantity: 50 yI) of 100 mM Tris-HCI buffer, 5 mM MgCl2 and 50 mM NaCI, and the reaction was carried out with the use of 6 units of the restriction enzyme Pst I at 370C for 2 hours. Therefore the reaction was stopped by heating at 680C for 15 minutes. The above synthetic gene was added to the reaction mixture and the ligating reaction was carried out similarly with the use of T4 DNA ligase.

By using 5 ul of the resultant reaction mixture, which was equivalent to 0.68 yg of pBR322 DNA, transformation of the E. coli strain K12C600 was performed according to the method of Kuschner' ) in a P-3 physical containment facility11).

The transformed strain was selected on L-plate (1% Bactotripton, 0.5% Bactoyeast extract, 0.5% NaCl, 1,5% Bactoagar) containing 1 0 g/ml of tetracyline (Tc), and 50 strains among the transformed strains obtained were examined for resistance to ampicillin (Ap), and 45 strains of TCr/Aps transformed strains were isolated. For the purpose of reference, these were named C600 (pMG 101 )-C600 (pMG 145).

From among the transformed strains of Tcr/Aps, 8 strains (C600 (pMG 101)-C600 (pMG 108)) were randomly selected, and the plasmid DNA was isolated by equilibrium sedimentation through cesium chloride density gradient containing ethidium bromide. The plasmid DNA thus isolated was named pMG 101-pMG 108 for the purpose of reference.

Confirmation of direction The plasmid (5 jug) was dissolved in a mixture (30 ,xl) of 10 mM Tris-HCI buffer, 66 mM MgCl2, 6 mM ss-mercaptoethanol and 60 mM NaCI, and the 30 units of the restriction enzyme Hae Ill were added to the solution, which step was then followed by heating at 370C for one hour. Fragments of 375 bp were extracted by 1.5% low melting point agarose gel electrophoresis. The fragments thus extracted were similarly hydrolyzed with the use of 18 units of the restriction enzyme Hae II, and the product hydrolyzed to 191 bp and 176 bp determined upon 5% polyacrylamide gel electrophoresis was evaluated as plasmid having the correct direction.

Identification of fused protein with minicells A minicell-producing E. coli (F-, thr+, ara+, leu+, azis, tonAs, minA, minB, gal+, ,t-, sfrr, malA, xyl, mtl, thi, sup-) was transformed with pMG 103 and pBR 322 according to the method of Kuschner10 to obtain a transformant of Tcr/Aps.

The transformant was pre-cultured in 20 ml of Davis's minimum culture medium containing 0.5% casamino acid, thymine (20 g/ml), thiamine 2 yg/ml) and tetracycline (10 g/ml) at 370C for 16 hours, and, with inoculation of 10 ml of the pre-culture into 500 ml of the same medium (but containing no tetracycline), cultivation was further conducted to 0D620=0.6-0.8. The culture broth was subjected to centrifuging by means of a Hitachi Refrigerated Centrifuge (Model RPR-9, at 3,000 rpm) for 12 minutes to romove host cells and further to centrifuging at 8,500 rpm for 25 minutes to obtain minicell pellets.

The minicell pellets were washed by suspending in 25 ml of To-buffer (1 M Tris-HCI pH 7.3, MgSO4. 7H20 150 mg, 1% gelatin 1 ml, 1 M CaCI2 0.25 ml/l liter-H20). The suspension was subjected to centrifuging at 10,000 rpm (in an RPR-20 rotor) for 15 minutes, and the resulting minicell precipitates were suspended in 1 ml of T1 buffer. The resulting suspension was subjected to the 1 0%-1 5%-20% stepwise density gradient centrifugation (at 4,000 rpm, in an RPRS-4 rotor for 24 minutes) to recover minicells. Further, the minicell suspension was suspended in 20 ml of TI-buffer, and similar procedures were repeated.

The minicells were suspended in 1 ml of Davis's minimum culture medium containing each 50 ,ug/ml as final concentration of 18 amino acids (alanine, valine, isoleucine, proline, phenylalanine, tryptophan, glycine, serine, threonine, cystine, tyrosine, asparagine, aspartic acid, glutamine, giutamic acid, lysine, arginine, histidine), 20 ,ug/ml of thymine and 2 ,ug/ml of thiamine, and heated with shaking at 370C for 12 minutes. After 20 uCi (35S)-methionine (NEN) New England Nuclear): 1260.1 Ci/mmol) and 5 uCi (3H)-leucine (NEN: 11 5.2 Ci/mmol) were added to the suspension, the mixture was further heated for 10 minutes.Immediately thereafter, the same buffer containing 50 mM NaN3, 100 ,ug/ml of methionine and 100 yg/ml of leucine was added to the mixture. Then centrifugation was carried out at 3,000 rpm for 20 minutes to recover minicells, which were in turn suspended in 20 ,u1 of a sample buffer [H20 6 mi, 1.25 M Tris-HCI (pH 6.8) 1 ml, SDS (sodium dodecylsulfate) 0.4 g, glycerol 2 ml, 2 mercaptoethanol 1 ml, BPB (brom phenol blue) 4 mg] and heated at 1 000C for 3 minutes. Then 18 l of this suspension was subjected to separation by SDS-poly-acrylamide electrophoresis.

In the cells of E. coli K1 2C600 (pBR 322) (FERM-P 6017), p-lactamase had been synthesized and a band corresponding to a molecular weight of about 29 kilodaltons appeared on the electrophoresis gel, but no such protein of about 29 kilodaltons was detected in the extracts from the cells of E. coli K1 2C600 (pMG 103). Instead, a polypeptide of about 24 kilodaltons appeared as a new band. This was clearly the gene product produced by being coded by pMG 103 and was related to the disappearance of the p-lactamase band.The pMG 103 has a structure in which a part of the p- lactamase gene of pBR 322 (coding 182 amino acids from the amino end of the signal peptide of the structural protein of p-lactamase) is linked downstream thereof (for transcription) to the 27 desamidosecretin gene (coding 27 amino acids), and its gene product must be detected as the fused protein of p-lactamase fragment (182 amino acids)-27-desamidosecretin (27 amino acids). The molecular weight of the fused product is calculated as 23.4 kilodaltons, and therefore the newly appeared band in the E. coli K12C600 (pMG 103) extracts may be considered to be this fused protein.

Purification of protein The E. coli K12 C600 (pMG 103) was subjected to shaking cultivation in 3 liters of Luria broth at 370C and centrifuged when the microorganism concentration reached about 1 x109 cells/ml. The resultant pellets were washed with 10 mM Tris-HCI buffer (pH 8.0)/1 mM phenylmethylsulfonyl fluoride and centrifuged again. The pellets were suspended again in the same buffer. THe suspension was treated with EDTA of a final concentration of 10 mM at 0 C for 5 minutes and subsequently treated with egg white lysozyme of a final concentration of 0.1 mg/ml at OOC for 30 minutes to effect spheroplast formation-lysis.The rsultant lysate was treated by means of Kubota ultrasonic wave generating device "Insonator 200 M" at the maximum output (200 W) for 10 minutes to destroy completely the bacterial cells. This step was followed by centrifugal separation (by the use of a Hitachi Refrigerated Centrifugal Machine, Model 20 PR52, with RPR20-2 rotor, at 18,000 rpm, for 30 minutes, at OOC) to remove the cell debris.To the supernatant obtained (110 ml), magnesium chloride of a final concentration of 50 mM was added, and then the mixture was subjected to centrifugation (by means of the above rotor at 8,000 rpm, for 10 minutes, at OOC). To the supernatant, 400 y9 of DNase I and 5 mg of RNase I were added to carry out treatment at 40C for one hour, and thereafter proteins and peptides were salted out with 80% saturated ammonium sulfate. The precipitates obtained by centrifugation (by means of RPR 20-2 rotor, at 8,000 rpm, for 10 minutes, at OOC) were dissolved in 10 mM Tris-HCI (pH 8.0), desalted with the same buffer, and centrifuged (under the above conditions).

Then, acetone of a final concentration of 75% was added to the supernatant, and the precipitates formed were treated overnight with 1.0 g of cyanogen bromide in 20 ml of 80% formic acid. After evaporation, the residue was extracted with 40 ml of isopropanol, and then with an equal volume of methanol. Subsequently, evaporation was carried out again, and the residue was dissolved in 0.1 N acetic acid. After removal of insolubles by centrifugation (3,000 4,000 G, for 5 minutes), gel filtration was conducted through "Sephadex G-25" column (produced by Pharmacia Co.), and the fractions with molecular weights of 1 ,000 to 10,000 were collected and lyophilized.

Bioassay The lyophilized sample thus obtained was assayed by a bioassay method12 in which the lyophiiized sample obtained was dissolved in 2 ml as a total volume of isotonic sodium chloride solution; aliquots each of 0.25 ml were injected intravenously at the groin into five rats; the quantities of pancreatic juice secreted to flow out from an artificial pancreatic tube were measured at intervals of 1 0 minutes; and the increased quantity was determined as the biological activity of the secretin. As the standard secretin, Secrepan (Eisai, Japan) in quantities of 0.25 U/Kg and 0.50 U/Kg were previously injected intravenously into the same rats, and the quantities of the pancreatic juice secreted were measured to determine correspondence to the Eisai unit (substantially equal to the Crick Harper Raper unit).As a result, the samples obtained from E. coli K 12 C600 (pMG 103) brought about an increase of secreted pancreatic juice of 202%+76 (average + standard deviation) 10 minutes after injection and of 1 80%+87.6, 20 minutes after injection. On the other hand, Secrepan brought about an increase of 1 62%*28 and 206%+58, respectively, after 10 minutes and 20 minutes in case of 0.25 U/Kg injection; and 159.8%+83.7 and 250.8%+216.5, respectively, 10 minutes and 20 minutes after injection of 0.5 U/Kg. These values, as a result of a t-test, were found to give a significant difference by p < 0.05.On the other hand, when the sample obtained by starting from the extracts of the bacterial cells obtained from the clone containing only pBR 322 and free of the desamidosecretin gene was tested according to the same method, no increase of the quantity of pancreatic juice secreted was observed. This indicates that a substance having a secretin-like biological activity is produced in the cells of E. coli K1 2 C600 (pMG 103), while no production of such a substance is observed in the cells of E. coli K12 C600 (pBR 322).By plotting the increased quantities of pancreatic juices secreted by the samples obtained by the E. coli K12 C600 (pMG 103) on a straight line obtained by plotting on a semilogarithmic graph paper with the increased quantities of pancreatic juice secreted as ordinate and the Secrepan units as abscissa (logarithmic scale), the secretin unit can be determined by the extrapolation method (provided that the linear relation has been realized) to be 0.17 to 0.25 Eisai units (6.0.17 to 0.25 Crick Harper Raper units)/0.25 ml/Kg rat. When calculated by taking into consideration the body weights of the rats employed, this value is calibrated as 0.051 to 0.072 U/0.25 ml/rat, the total activity amounting to 0.408 to 0.576 U for the total quantity of 2 ml.Since the secretin purified to the highest purity is known13 to have a specific activity of 1 6,000 CHR units/mg, the values of 0.408 to 0.576 U correspond to 25.5 to 36 ng of the secretin quantity. This corresponds to 7.15 to 10.1 pmol of secretin molecules, indicating that a desamidosecretin having an activity corresponding to 4.3 to 6.1 xl 012 secretin molecules was biosynthesized. As the desamidosecretin was prepared with the use of about 3x 1012 cells of E. coli K 12 C600 (pMg 103) as the starting material, at least 1.4 to 2.0 molecules equivalent to secretin/one bacteria cell was recovered.

Radioimmunoassay The sample was diluted with 50 mM Tris-HCI (pH 8.0)/0.1 % bovine serum albumin.

Radioimmunoassay of the diluted sample was conducted by the use of "Secretin kit Daiichi" produced by Daiichi Radioisotope Research Institute, Japan according to the method prescribed to conform its activity.

Example 2 The steps to the ligating reaction between fragments were carried out similarly as in Example 1.

Preparation of Aplac5 DNA and pBR322 DNA Aplac5 DNA was obtained from lysogenic bacteria E. coli PK 1512 (IFO 14149), and the procedures for preparation were conducted according to the method of Oshima20).

pBR322 DNA was obtained from E. coli K-i 2 C600 (pBR 322) (FERM-P 6017) and the procedures for preparation were conducted according to the method of M. Kahn et al21).

Preparation of pRE1 Aplac5 DNA (10 ,ug) was added to a mixture (total quantity: 20 jul) of 100 mM Tris-HCI buffer pH 7.5, 7 mM MgCl2 and 50 mM NaCI, and the reaction was carried out with the use of 10 units of the restriction enzyme EcoRI and 10 units of the restriction enzyme Hind Ill at 370C for 2 hours.

Subsequently, by 1% agarose gel electrophoresis, 3.8 megadalton of DNA fragments were purified.

pBR322 DNA (1 ug) was added to a mixture (total quantity: 10 u1) similar to the above mixture, and the reaction was carried out with the use of 1 unit of the restriction enzyme EcoRI and 1 unit of the restriction enzyme Hind Ill at 370C for 2 hours. Subsequently, by 1% agarose gel electrophoresis, 2.6 megadalton of DNA fragments were purified.

The resultant fragments were added to a mixture (total quantity: 1 0 u1) of 50 mM Tris-HCI buffer pH 7.8, 10 mM MgCl2, 20 mM DTT and 1 mM ATP, and the reaction was carried out with the use of 30 units of T4DNA ligase at 1 40C for 24 hours.

By using the reaction mixture, transformation of the E. coli strain XA 35 was performed according to the method of Kuschner' ) in a p-1 physical containment facility11).

The transformed strain was selected on L-plate (1% Bactotrypton, 0.5% Bactoyeast extract, 0.5% NaCI, 1.5% Bactoagar) containing Ap (20 dtg/ml), and the transformed strains obtained were replicated on a plate of EMB-lac (2.25% Bacto-EMB broth, 1.5% Bactoagar). Further, the strains converted to Lac+ (red colony) were selected. From among these strains, 5 strains were randomly selected to prepare a plasmid, which was analyzed with several kinds of restriction enzymes, whereupon it was found that the four strains consisted of 3.8 megadalton fragment of Aplac5DNA ligated to Eco Rl-Hind Ill fragment of pBR 322 DNA. Among these strains, one strain was named E. coli XA 35 (pRE 1).That is, the transformant E. coliXA 35 (pRE 1) is different from the aforesaid E. coli strain XA 35 in the following properties.

[Apr, Lac+] Linker Synthesis of a fragment with the base sequence of CTGAATTCAGCTCTGCAGAG was carried out.

Synthesis and purification were conducted according to the methods for synthesis and purification of the respective structural genes as described above (Yield: 40%). In the present invention, this singlestranded DNA was named a prelinker.

The prelinker (15 y9) was incubated together with 20 units of T4-polynucleotide kinase at 370C for 40 minutes in a mixture (total quantity: 30 jul) of 50 mM Tris-HCI buffer pH 7.5, 10 mM MgCI2, 0.1 mM spermidine, 0.1 mM EDTA, 10 mM DTT and 0.2 mM ATP. The reaction was stopped by heating at 600C for 2 minutes. After the mixture was left to stand at room temperature for one hour, 300 units, 1 yI, of T4 DNA ligase was added and the reaction was carried out at 140C overnight. By heating at 600C for 2 minutes, the reaction was stopped. The mixture was left to stand at room temperature for one hour, and 20 units of the restriction enzyme Pst I were added to carry out the reaction for 5 hours.

By heating at 600C for 2 minutes, the reaction was stopped.

Thus, 1 5 jug of the Pst l-Eco Rl-Pst I linker was obtained.

Preparation of the EcoRI-EcoRI structural gene The Pst l-EcoRI-Pstl linker as prepared above (4 ,ug) and the Pstl-Pstl structural gene mentioned above in Example 1 (0.6 ,ug) were added to a mixture (total quantity: 35 41) of 100 mM Tris-HCI buffer pH 7.5, 7 mM MgCI2 and 50 mM NaCI, and the reaction was carried out with the use of 300 units of T4 DNA ligase at 1 40C for 24 hours.

Then, the reaction was stopped by heating at 680C for 10 minutes, followed by gradual cooling of the reaction mixture.

To the resultant reaction mixture, the restriction enzyme Eco RI (100 units) was added to carry out the reaction at 370C for 5 hours, and thereafter the DNA fragment with 120 base pairs was purified by 5% polyacrylamide gel electrophoresis.

Preparation of pLS and cloning 0.25 g of pREI was added to a mixture (total quantity: 4 ul) of 100 mM Tris-HCI buffer pH 7.5, 7 mM MgCI2 and 50 mM NaCI, and the reaction was carried out with the use of one unit of the restriction enzyme Eco RI at 370C for 1 hour.

The thus prepared plasmid pRE 1 (cleaved with EcoRI) (0.25 g) and 0.02 g of the EcoRI-EcoRI structural gene were added to a mixture (total quantity: 10 ul) of 50 mM Tris-HCI buffer pH 7.8, 10 mM MgCl2, 20 mM DTT and 1 mM ATP, and the reaction was carried out with the use of 30 units of T4 DNA ligase at 1 40C for 24 hours.

By using the resultant reaction mixture, transformation of the E. coli strain XA35 was performed according to the method of Kuschner' > in a P-3 physical containment facility1').

The transformed strain was selected on L-plate containing Ap (20 yg/ml) (as described above).

The resultant transformed strains were replicated on the EMB-lac plate (as described above) for further selection of the Lac+ strain. From among these strains, 200 strains were selected to prepare plasmid DNA's, of which structures were analyzed by the restriction enzyme Pst I or Hae II to confirm that 11 strains had the above structural gene.

Determination of direction The following experiment was carried out for each plasmid of the 1 1 strains having the structural gene.

The plasmid (20 ,ug) was added to a mixture (total quantity: 20 ,ul) of 10 mM Tris-HCI buffer pH 7.5, 66 mM MgCI2 and 60 mM NaCI, and further 20 units of the restriction enzyme Hae 11 were added to the solution. Then incubation was carried out at 370C for 2 hours. From the resultant mixture, there were extracted by 1% agarose gel electrophoresis fragments of about 1 megadalton and fragments of about 0.7 megadalton. The respective fragments were cleaved with the restriction enzyme EcoRI according to the procedure described above, and were subjected to 15% polyacrylamide gel electrophoresis.The plasmid in which, upon the electrophoresis, DNA fragments with 40 base pairs were obtained from the fragment of about 1 megadalton and the DNA fragment with 80 base pairs obtained from the fragment of about 0.7 mega dalton was evaluated as the piasmid in which the structural gene had been inserted in the correct direction.

From these experiments, it was found that five strains of plasmid among the 1 1 strains of plasmid having the structural gene had the correct direction. One of these strains was randomly selected and named E. coliXA35 (pLS58).

Identification and purification of fused protein E. collXA35 (pLS58) was subjected to shaking cultivation in 1 liter of Luria-broth containing 20 ,ug/ml of ampicillin at 370C until the number of bacteria cells reached 5x108 cells/ml. Then, the bacteria cells were collected by centrifugation and suspended in a mixture (total quantity: 100 ml) of 10 mM Tris-HCI buffer (pH 7.5)/10 mM MgCI2, and the suspension was treated by means of a Kubota ultrasonic wave generating device "Insonator 200 M" for 30 minutes to destroy the bacterial cells.

The thus obtained solution (2 yI) was subjected to 7.5% SDS polyacrylamide gel electrophoresis, whereby a band of a protein with a size of 120,000 dalton was recognized. No such band was recognized in the extract similarly obtained from the bacteria having no plasmid (E. call strain XA35).

Therefore, this protein was considered to have been derived from the plasmid pLS58.

The protein was also confirmed to have asize substantially the same as that of p-galactosidase19 from E. coli, thus being not in contradiction to the size (11 7,873 dalton) of the fused protein of the desamidosecretin and p-gelactosidase.

Accordingly, it was concluded that this protein with about 120,000 dalton was the fused protein, and purification of this protein was carried out.

The extract obtained by ultrasonic destruction was subjected to centrifugation, and the precipitates were recovered. The fused protein was substantially recovered in the precipitates. The precipitates were suspended in a mixture (total quantity: 100 ml) of 10 mM Tris-HCI buffer pH 7.5 and 1 mM MgCl2 and subjected to centrifugation, which was followed by recovery of the precipitates. This operation was further repeated twice to remove water-solubles.

The resultant precipitates were dissolved in a mixture (total quantity: 100 ml) of 20 mM Tris-HCI buffer pH 7.5, 1 mM MgCl2, 10 mM NaCI and 7 M urea, and applied to a column (5 cmXx3 cm) of DEAE celiulose equilibrated with the same buffer.

After washing with 20 ml of the same buffer, the column was further washed with 100 of a similar buffer in which the NaCI concentration was changed to 50 mM. Then, the fused protein was eluted with 1 50 ml of a similar buffer in which the NaCI concentration was changed to 1 50 mM.

To the resultant eluate, p-mercaptoehanol was added to the final concentration of 1%, and the mixture was dialyzed three times against 5 liters of water. The dialysate was subjected to centrifugation to recover the fused protein as precipitates.

The resultant fused protein produced a substantially single band by SDS polyacrylamide gel electrophoresis in a quantity of about 28 mg per one liter of the culture broth, as calculated from the absorption at 280 nm. This value is calculated as corresponding to about 280,000 molecules/cell.

The purified fused protein was dissolved in 2 ml of 70% formic acid and treated with 100 mg of cyanogen bromide for 1 8 hours. After evaporation, the residue was extracted with 5 ml of isopropanol and then with 5 ml of methanol, which extraction was followed by lyophilization.

Bioassay The lyophilized sample thus obtained was assayed by a bioassay method12) in which the lyophilized sample was dissolved in 25 ml as a total volume of isotonic sodium chloride solution; aliquots each of 50 ,ul were injected intravenously at the groin into five rats; the quantities of pancreatic juice secreted and flowing out from the artificial pancreatic tube were measured at intervals of 5 minutes; and the increment was determined as biological activity of secretin. As the standard secretin, Secrepan (Eisai, Japan) in quantities of 1 unit, 2 units and 4 units/rat were previously injected intravenously into the same rats, and the quantities of pancreatic juice secreted were measured to determine correspondence to the Eisai unit (substantially equal to the Crick Harper Raper unit).

As a result, the samples obtained from E. coliXA35 (pLS58) brought about an increase of secreted pancreatic juice. On the other hand, when the sample obtained by starting from the extracts of the bacterial cells obtained from the clone containing only pREI and free of the desamidosecretin gene was tested according to the same method, there was observed no increase of the amount of pancreatic juice secreted. This indicates that a substance having secretin-like biological activity is produced in the cells of E. coli XA35 (pLS58), while no production of such a substance occurs in the cells of E. coli XA35 (pREI).

By plotting the increased quantities of pancreatic juices secreted by the samples obtained by the E. coliXA35 (pLS58) on a straight line obtained by plotting on a semilogarithmic graph paper with the increased quantities of pancreatic juice secreted as ordinate and the Secrepan units as abscissa (logarithmic scale), the secretin unit can be determined by the interpolation method to be 3 Eisai units/50 yi/rat, which amounts to about 1,500 Eisai units as 25 ml isotonic sodium chloride solution containing the lyophilized sample.Since the secretin purified to the highest purity is known13 to have a specific activity of 16,000 CHR units/mg, 1 ,500 units correspond to about 26 nmole of the secretin molecules, indicating that a desamidosecretin having an activity corresponding to 1 .6x 1015 secretin molecules was biosynthesized. As the desamidosecretin was prepared with the use of about 5x 1011 cells of E. coliXA35 (pLS58) as the starting material, at least 3x 1 04 molecules equivalent to secretin/one bacteria cell was recovered.

From this result, it can be concluded that the synthetic 27-desamidosecretin gene is expressed by the action of the E. coli lactose operon promoter in the cells of E. coli XA35 (pLS58), and the translated product thereof, viz. 27-desamidosecretin, exhibits activity similar to secretin.

Radioimmunoassay The sample was diluted with 50 mM Tris-HCI (pH 8.0)/0.1% bovine serum albumin.

Radioimmunoassay of the diluted sample was conducted by the use of "Secretinkit Daiichi" produced by Daiichi Radioisotope Research Institute, Japan according to the method prescribed to confirm its activity.

4. Design drawing of gene In the following diagram, there is shown the base sequence of the gene comprising the 27desamidosecretin structural gene separately in blocks A, B and C, wherein a is an optional base pair, b the Pst I site, can optional base pair, d a start codon, e the Hinf I site, and fthe Hae II site.

Design drawing of the gene Block A

Block B

Block C

5. Deposition of microorganisms E. coli K12C600; E. coliXA35 and its transformant E. collXA35 (pRE 1) with the plasmid pRE 1 are deposited under the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure at the Fermentation Research Institute, Agency of Industrial Science and Technology (FERM), 1-3, Higashi 1-chome, Yatabemachi, Tsukuba-gun, Ibaraki-ken, Japan under FERM BP-115 dated June 9, 1981; FERM BP-1 16 dated January 7, 1 982;; and FERM BP-1 17 dated January 7, 1982, respectively. The plasmid pBR322 is deposited at FERM in the form of a transformant of E. coli K1 2C600 with pBR322, namely as E. coli K12C600 (pBR322) under FERM P-6017 dated June 9, 1981.

E. coli K12C600 (pMG 103) which is the transformant with the chimera plasmid pMG is deposited at the American Type Culture Collection (ATCC), 12301, Parklawn Driven, Rockville, Maryland 20852, U.S.A. under ATCC 39042 dated January 26, 1982.

E. coli PK 1512 which is the Aplac5 lysogenic bacteria is deposited at the Institute for Fermentation, Juso-Hommachi 2-17-85, Yodogawa-ku, Osaka-shi, Japan under IFO No. 14149 dated February 1, 1982.

E. coliXA 35 (pLS58) which is the transformant with the chimera plasmid is deposited under the Budapest Treaty at the ATCC under ATCC 39040 dated January 11, 1982.

6. References 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 Sates 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 Rescarch, B, 5753, (1980); 9) Chemical and Pharmaceutical Bulletin, 26,2396, (1978); 10) Genetic Engineering, 1978, 17, (1978); 11) Guidelines for Recombinant DNA Experiment, issued on August, 1979; 12) The Japanese Journal of Pharmacology, 21, 325, (1971); 13) a) ChemischeBerichte, 105, 2508, (1972), b) Chemische Berichte, 105, 2515, (1972); 14) Tetrahedron Letters, 1978,2727, (1978); 1 5) Proceedings of the National Academy of Sciences of the United States of America, 73, 3900, (1977); 16) Gene, 2,95, (1977); ; 1 7) Proceedings of the National Academy of Sciences of the United States of America, 76, 1 06, (1979); 18) The operon, (1980) (written by J. H. Miller, W. S. Resnikoff; published by Cold Spring Harbor Laboratory); 1 9) Proceedings of the National Academy of Sciences of the United States of America, 74, 1 507 (1977); 20) Extra Volumes of "Protein, nucleic acid S enzyme", Last vol. p. 19, (1973); 21) Methods in Enzymology, 68, 285, (1979); 22) Microbiological Reviews, 44, 1-50, (1980); 23) Nature, 224, 768, (1969); 24)Nature, 217, 1110, (1968).

Claims (39)

Claims

1. 27-Desamidosecretin, comprising a polypeptide represented by the amino acid sequence formula: His-Ser-Asp-Gly-Thr-Phe-Thr-Ser-Glu-Leu-Ser-Arg-Leu-Arg-Asp-Ser Ala-Arg-Leu-Gln-Arg-Leu-Leu-Gln-Gln-Gly-Leu-Val-OH wherein Val-OH indicates that the C-end of the polypeptide comprising a valine is a carboxylic acid.

2. A method of producing 27-desamidosecretin, which comprises the steps of: (1) synthesizing chemically a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid at the C-end of secretin is valine, (2) inserting said gene into a vector plasmid capable of proliferation in a predetermined host cell to make a chimera plasmid capable of proliferation in said cell, (3) transforming the host cell with said chimera plasmid, and (4) culturing the resulting transformant and recovering the desamidosecretin produced.

3. A method according to claim 2 wherein the host cell is an E. coli belonging to the genus Escherichia.

4. A method according to claim 3 wherein the E. coli is E. coli K1 2C600, FERM BP-1 1 5.

5. A method according to claim 3 wherein the vector plasmid is pBR322.

6. A method according to claim 5 wherein the chimera plasmid is pMG which has incorporated therein said structural gene at the Pst I recognition site of pBR322.

7. A method of producing a 27-desamidosecretin comprising the steps of: (1) synthesizing chemically a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid at the C-end of secretin is valine, (2) providing a vector plasmid capable of untilizing the lactose operon and also capable of proliferation in a predetermined host cell, (3) inserting said structural gene into the vector plasmid to make a chimera plasmid capable of proliferation in said cell, (4) transforming the host cell with said chimera plasmid, and (5) culturing the resultant transformant and recovering the desamidosecretin produced.

8. A method according to claim 7 wherein the host cell is an E. coli belonging to the genus Escherichia.

9. A method according to claim 8 wherein the E. coli is E. coli XA35, FERM BP-1 1 6, which is a strain deficient in the gene of ,B-galactosidase and also deficient in the i-gene.

10. A method according to claim 8 wherein the lactose operon is from E. coli.

11. A method according to claim 10 wherein the vector plasmid contains all or a part of the lactose operon from the E. coli chromosome DNA and is capable of proliferation in the E. coli.

12. A method according to claim 11 wherein the vector plasmid is prepared from a transducing phage containing all or a part of the lactose operon and a plasmid from the E. coli.

1 3. A method according to claim 12 wherein the transducing phage is selected from the group consisting of pill, F'-lac, 480dplac, Ah80dlac and AWplac.

14. A method according to claim 12 wherein the plasmid from E. coli is selected from the group consisting of pBR 322, pSC 101 -and Advl.

1 5. A method according to claim 13 wherein the phage is Aplac5.

1 6. A method according to claim 14 wherein the plasmid is pBR 322.

1 7. A method according to claim 1 5 or 16 wherein the vector plasmid is pRE which comprises the 3.8 Md of the fragment of Aplac5 linked with the larger fragment of the fragments obtained through digestion of pBR322 with EcoRI and Hind Ill.

18 A method according to claim 17 wherein the chimera plasmid is pLS which is a product of insertion of said structural gene into said pRE at its EcoRI recognition site.

19. An Escherichia colitransformed with a plasmid: said plasmid having incorporated therein a chemically synthesized structural gene of the desamidosecretin corresponding to the polypeptide of which the amino acid at the C-end is valine, and being capable of utilizing the lactose operon and also capable of proliferation in a predetermined host cell.

20. An E. coli according to claim 19 which is E. coli XA35 (pLS58), ATCC 39040.

21. A double-stranded polydeoxyribonucleotide comprising a structural gene expressing 27desamidosecretin.

22. A double-stranded polydeoxyribonucleotide according to claim 21 wherein the structural gene has the base sequence: His Ser Asp Gly Thr Phe Thr Ser Glu CAC-TCA - GAT - GGT - ACT - TTC - ACC - TCA - GAA - GTG - AGT - CTA - CCA - TGA - AAG - TGG - AGT - CTT- Leu Ser Arg Leu Arg Asp Ser Ala Arg CTA-TCT-CGT-CTA-CGT-GAT-TCA-GCA-CGC- GAT-AGA-GCA-GAT-GCA-CTA-AGT-CGT-GCG- Leu Gln Arg Leu Leu Gln Gly Leu Val CTC-CAG-CGC-TTG-CTG-CAA-GGT-CTC-GTT GAG-GTC-GCG-AAC-GAC-GTT-CCA-GAG-CAA.

23. A double-stranded polydeoxyribonucleotide according to claim 21 or 22 wherein the structural gene contains a codon designating methionine at the 5'-end and one or more codons for designating translation stop at the 3'-end.

24. A double-stranded polydeoxyribonucleotide according to claim 23 wherein the 5'-end of the methionine codon and the 3'-end of the translation stop codon has 3m base pairs, (3m+1) base pairs or (3m+2) base pairs (where m is O or an integer at least 1) of randomly selected bases.

25. A double-stranded polydeoxyribonucleotide according to claim 24, having a recognition base sequence of any desired restriction enzyme at both ends and having outside of the resultant ends at least two base pairs to produce blunt ends randomly selected.

26. A double-stranded polydeoxyribonucleotide according to claim 21, having the following structure: Met ACCTGCAGCC - ATC TGGACGTCGG - TAC His Ser Asp Gly Thr Phe Thr Ser Glu CAC-TCA-GAT--GGT-ACT-TTC TTC -- ACC -- TCA -- GAA - GTG -AGT- CTA-CCA-TGA-AAG-TGG-AGT- CTT - Leu Ser Arg Leu Arg Asp Ser Ala Arg CTA-TCT-CGT-CTA-CGT-GAT-TCA-GCA-CGC- GAT-AGA-GCA-GAT-GCA- CTA-AGT- CGT-GCG- Leu Gln Arg Leu Leu Gln Gly Leu Val CTC-CAG-CGC-TTG-CTG-CAA-GGT-CTC-GTT- GAG-GTC-GCG-AAC-GAC-GTT-CCA-GAG-CAA- END END TGA-TAG- GGCTGCAGGT ACT-ATC-CCGACGTCCA.

27. A plasmid comprising a structural gene expressing the 27-desamidosecretin.

28. A plasmid according to claim 27 wherein the plasmid is from pBR322.

29. A method of producing 27-desamidosecretin which comprises: (1) providing a chimera plasmid which comprises a fragment of a structural gene for a desamidosecretin corresponding to a polypeptide in which the amino acid at the C-end of secretin is valine, the chimera plasmid being capable of proliferation in a predetermined host cell and being capable of expressing the structural gene for a desamidosecretin in the host cell, (2) transforming the host cell with said chimera plasmid, and (3) culturing the resulting transformant and recovering the desamidosecretin produced.

30. A method according to claim 29 wherein said chimera plasmid is capable of utilizing the lactose operon.

31. A method according to claim 30 wherein said chimera plasmid is produced by inserting said structural gene into a vector plasmid capable of utilizing the lactose operon and also capable of proliferation in said host cell.

32. A method according to claim 31 wherein the vector plasmid contains all or a part of the lactose operon from the E. coli chromosome DNA and is capable of proliferation in the E. coli.

33. A method according to claim 32 wherein the vector plasmid is prepared from a transducing phage containing all or a part of the lactose operon and a plasmid from the E. coli.

34. A method according to claim 33 wherein the transducing phage is selected from the group consisting of pill, F'-lac, 80dplac, Sh80dplac and Aplac.

35. A method according to claim 33 wherein the plasmid from E. coli is selected from the group consisting of pBR322, pSC101 and Advl.

36. A method according to claim 34 wherein the phage is Aplac5.

37. A method according to claim 35 wherein the plasmid is pBR322.

38. A method according to claim 36 or 37 wherein the vector plasmid is pRE which comprises the 3.8 Md of the fragment of Aplac5 linked with the larger fragment of the fragments obtained through digestion of pBR322 with EcoRI and Hindlll.

39. A method according to claim 38 wherein the chimera plasmid is pLS which is a product of insertion of said structural gene into said pRE at its EcoRI recognition site.