HRP940770A2 - Process for the production of stranger protein in streptomycetes - Google Patents

Abstract

Gene structures which code for the signal sequence and approximately the first ten amino acids of tendamistat as well as a required protein are expressed in high yield in streptomyces host cells, and the fusion proteins are secreted into the medium.

Description

From the European patent application, published under the number (EP-A) 0 289 936, it is known that fusion proteins can be prepared in such a way that the structural genes are joined to the desired protein at the end of the 3 coding chain, in the given case of the modified tendamistat gene, where the gene construct is expressed in a streptomycete host cell and the secreted fusion protein is isolated from the supernatant. In a preferred embodiment, the tendamistat gene is truncated at the 3' end. For truncation, cleavage sites for the restriction enzymes BstEII in the triplet region 31 and 32, StuI in the triplet region 43 and 44, as well as Sau3A in the triplet region 52 and 53 are used.

In the further development of this inventive idea, the preparation of a fusion protein in which the tendamistat part is followed by a shortened proinsulin, whose C chain consists of only one or two lysine residues ("mini-proinsulin") has already been proposed. In further development, it was proposed to also shorten the tendamistat part in fusion proteins of this type (EP-A-0 367 163, published on May 9, 1990).

We have now surprisingly established that fusion proteins with a very short tendamistat portion are persistent in the streptomycete cell and are secreted into the medium. Due to the very short tendamistat chain, the thus obtained fusion proteins behave like "mature" proteins and are generally found in the medium in a tertiary structure.

EP-A-0 177 827 discloses a synthetic signal sequence for the transport of proteins into expression systems, which is characterized by the fact that it corresponds to DNA in the main natural signal sequence, although however it has one or more cleavage sites for endonucleases, which natural DNA does not owns. If the gene for the protein to be transported is joined to such a DNA sequence, the fusion gene is inserted into a vector and a host cell is transformed with it that transports the expressed protein from the cytoplasm, eukaryotic, prokaryotic or viral proteins can be produced in prokaryotic and eukaryotic cells. . Using the example of the alkaline phosphatase periplasmic protein, it was shown that when expressing in E. coli, it is very useful to place codons for approximately the first 40 amino acids of the alkaline phosphatase in front of the structural gene of the desired protein, also for the presequence. However, in many examples, even fewer additional amino acids are sufficient, eg, approximately 10, preferably approximately 5. The corresponding monkey proinsulin fusion protein is approximately 90% transported to the periplasmic space.

It has also already been proposed (WO 91/03 550, published 21.3.1991), that the fusion protein is prepared by constructing a mixed oligonucleotide, which codes for the ballast part of the fusion protein, that oligonucleotide is introduced into the vector so that it is functionally attached to regulatory region and to the structural gene of the desired protein, suitable host cells are transformed with this thus obtained plasmid population and those clones that show a high gain of the encoded fusion protein are selected.

The oligonucleotide consists primarily of 4 to 12, especially 4 to 8 triplets.

Attempts have already been made to prepare fusion proteins with a short ballast part. For example, a gene fusion was prepared that encodes a fusion protein from the first 10 amino acids of β-galactosidase and somatostatin. It was shown, however, that this short β-galactosidase fragment was not sufficient to protect the fusion protein from degradation by the host's own proteases (US-A 4,366,246, column 15, paragraph 2). Accordingly, EP-A 0 290 005 and 0 929 763 describe fusion proteins whose ballast part consists of β-galactosidase with more than 250 amino acids.

We have now surprisingly found that fusion proteins from approximately the first 10 amino-terminal amino acids of tendamistat and the desired protein, eg proinsulin, are stable in streptomycete host cells and are secreted into the medium, from which they can be obtained in high yields. Surprisingly, this also applies to relatively small proteins such as "mini-proinsulins".

"Approximately 10 amino acids" means that a smaller number of amino acids is also taken into account, for example the first 7 N-terminal amino acids of tendamistat, although preferably not more than 10. Preference is given to fusion proteins in which in the tendamistat part at position 7 and/or 9 stands for proline (as in the natural sequence).

In accordance with the already known or proposed forms of performance, it is of course possible to choose a larger tendamistat ballast part, whereby, of course, the advantage of a smaller "ballast" is increasingly lost.

It is possible and very useful to change the natural sequence of amino acids of the tendamistat part, i.e. to replace or omit amino acids or to incorporate amino acids that appear in the natural sequence of amino acids. Furthermore, the sequence of amino acids in the signal peptide can be varied.

In terms of the invention, particular useful fusion constructs can be readily determined with simple preliminary experiments.

Furthermore, the idea of the invention can also be realized in other cells of gram-positive bacteria, for example in cells of bacilli and staphylococci, with the use of signal sequences, which these host cells recognize.

The fusion proteins obtained according to the invention are in the medium in dissolved form, which offers many advantages in further processing and purification. Thus, further enzymatic processing with separation of the ballast part is easily performed directly with the secretion product, where it is not necessary to carry out the processing phases, which are necessary for insoluble fusion proteins. However, before further processing, concentration or purification can also be carried out, eg by affinity chromatography, then also by ultrafiltration, precipitation, ion exchange chromatography, absorption chromatography, gel filtration or high pressure liquid chromatography.

Examples

The invention is further explained in the following examples.

The starting material for the plasmid constructions is the plasmid pKK500, which is proposed in EP-A 0367 163. This plasmid differs from the plasmid pKK400, known from EP-A 0 289 936, by the replacement of the proinsulin gene with an analogous gene, which codes at the C chain site. only the amino acid lysine, as well as by incorporating a terminal sequence behind that "mini-proinsulin" genome. Tables 1 and 2 of EP-A 0 367 163, showing the "mini-proinsulin" gene, i.e. the terminal sequence, are attached to this description.

Plasmids pKK400 and pKK500 in the signal sequence of the α-amylase inhibitor gene contain an XmaIII cleavage site (in the triplet region -5 to -7).

Example 1

Plasmid pKK500 was treated with restriction enzymes EcoRI and XmaIII and the large fragment was separated on a 0.8% agar gel by gel electrophoresis and isolated by electroelution. This fragment binds to DNA fragment (1) (SEQ ID NO:1)

[image]

synthesized by the phosphoramide method, and after binding, the mixture is transformed into E. coli. Plasmid pKK510 is obtained. It encodes a preproinsulin in which the signal sequence of tendamistat is followed by the first 7 amino acids of tendamistat and then the chain of mini-proinsulin.

Example 2

Analogous to the procedure for transforming plasmid pKK400 into expression plasmid pGF1, described in EP-A 0 289 936, plasmid pKK510 is translated into expression plasmid pKF1.

The isolated DNA of plasmid pKK510 is cut with restriction enzymes SphI and SstI and a small fragment with the fusion gene is isolated. A common commercial expression plasmid pIJ 702 (available from the John Innes Foundation, Norwich, England) is cut with the same enzyme and a large fragment is isolated. This isolated fragment is blinded, after blinding the mixture is transformed into S. lividans TK24 and a plasmid is isolated from the white (i.e., unable to produce melanin) transformants, resistant to thiostrepton. Clones carrying the inserted insert were tested in triple culture for their production of fusion proteins.

The expression of the encoded fusion protein is carried out in a manner known per se. The transformed strain is incubated for 4 days at 25°C in a shake flask and the mycelium is separated from the culture solution by centrifugation, after electrophoresis of 20 µm culture filtrate in a 15% polyacrylamide gel, the resulting fusion protein can be prepared by precipitation with COOMASSIE®-blue, visible as additional protein lysis, which does not appear in the comparative experiment in which the strain was transformed with pIJ 702 alone.

The culture filtrate is treated with lysylendoproteinase, Des-(B30)-Thr-insulin can be demonstrated by gel electrophoresis, which is verified with authenticity control.

Furthermore, a fusion protein with antibodies to insulin can be demonstrated in the culture filtrate either by "immunocopying" or by radioimmunoassay for insulin.

Example 3

Proceed as in examples 1 and 2, except that synthetic fragment (2) with SEQ ID NO:2 is used

[image]

[image]

This plasmid encodes a fusion protein that differs from the protein in examples 1 and 2 in that the first 7 amino acids of tendamistat are followed by asparagine (instead of the natural amino acid alanine) and then the ninth amino acid in tendamistat, proline. Replacing alanine with asparagine also introduces an additional positive charge into the ballast portion of the fusion protein. Surprisingly, approx. 20 to 30% higher utilization than in example 2.

Example 4

Proceed as in examples 1 and 2, except that synthetic fragment (3) with SEQ ID NO:3 is used

[image]

[image]

and the plasmid pKK330, i.e. pKF3, is obtained. This plasmid differs from the plasmid from examples 1 and 2 in that it encodes the first 9 natural amino acids of tendamistat. Compared to example 2, yields are higher by approx. 10%.

Example 5

The fusion protein, encoded by pKK500, between the tendamistatog part and the proinsulin B chain contains a linker sequence, which codes for the amino acids Asn-Ser-Asn-Gly-Lys. Replacement of that terminal Lys and Lys, which represents chain C, with Arg is done as described below. The only StyI cutting site in the region of codons B30 to A1 in the proinsulin sequence is used.

The isolated pKK500 plasmid DNA was cut with StyI, digested with S1 nuclease, denatured over the molecular ends, and excess nuclease extracted with phenol-chloroform. Then the linearized plasmid is cut with EcoRI, the large fragment is separated by electrophoresis and isolated by electroelution. This fragment binds to the synthetic fragment (4) (SEQ ID NO:4)

[image]

and the mixture is transformed into E. coli after gluing. The desired clones are checked by restriction analysis of the resulting plasmid, with the newly created site used for SstII cutting. The sequences in the entire SphI-SstI fragment are then separated.

For the expression of the coded fusion protein, the fragment checked by sequence analysis is blinded with the vector pIJ 702, cut with the same enzymes, resulting in the expression vector pGF4.

Evidence of the secreted fusion protein, encoded by pGF4, can be performed by assaying with an α-amylase inhibitor on plates (EP A 0 161 629, example 3) or from the culture supernatant in analogy to example 2.

Example 6

If, analogously to example 5, fragment (4) is inserted into vectors pKK510, 520 and 530, and vectors pKK610, 620 and 630 are obtained. Insertion of respective SphI-SstI fragments with the coding sequence for fusion proteins into vector pIJ 702 gives expression vectors Pkf11, 12 and 13. The expression of secreted fusion proteins is checked as in example 2.

Example 7

To increase the expression of the derivative of plasmid pIJ 702, the melanin promoter was removed from it by digestion with PstI and SphI and replaced with a synthetic fragment (5) (SEQ ID NO:5)

[image]

This results in a tandem construct from a synthetic and tendamistat promoter. The plasmid was designated pGR110.

If, after cutting with SphI and SStI, synthetic fragments (1), (2) and (3) are inserted into pGR110, expression vectors pGR200, 210 and 220 are created. Analogous to fragment (4), expression vectors pGR250, 260 and 270 are obtained.

Example 8

To prepare human insulin from an insulin precursor with a combination of trypsin or an enzyme of equal action and carboxypeptidase B, it is useful to quickly cleave off the amino-terminal ballast part during the grafting reaction, to support the grafting reaction in the direction of k B31 (Arg)-insulin. For this, a modification of amino acids in front of amino acid B1 (Phe) is offered:

Proceed as in example 1 and plasmid pKK500 is opened with restriction enzymes EcoRI and DraIII. The original fragment is then replaced with fragment (6) (SEQ ID NO:6)

[image]

synthesized by the phosphoramide method. Cloning in E. coli and expression in Streptomyces lividans is carried out as in example 1 and 2. Plasmid pKK640 and expression plasmid pKF14 are created.

The plasmid obtained in example 5 (after incorporation of fragrane (4)) can be treated analogously. Thus, plasmids pKK650 and pKF15 are obtained.

Table 1

[image]

Table 2

[image]

Claims (8)

1. Method for the preparation of fusion proteins, indicated by the fact that the structural gene is attached to the desired protein at the codons for the signal sequence and approximately the first 10 amino-terminal amino acids of tendamistat, this gene structure is expressed in the streptomycete host cell and the secreted fusion protein is isolated from the supernatant , whereby tendamistat gene sequences can be modified. 2. A gene structure, characterized in that it contains a signal sequence and approximately the first ten codons for tendamistat and a structural gene for another protein. 3. Gene structure according to claim 2, characterized in that codons for amino acids have been omitted, replaced or added in the tendamistat part. 4. Gene structure according to claim 2 or 3, characterized in that there is a linker sequence between the tendamistat gene and the structural gene of the desired protein. 5. Vector, characterized in that it contains the gene structure according to claim 2, 3 or 4. 6. A streptomycete cell, characterized in that it contains the vector according to claim 5. 7. Fusion protein, characterized by the fact that it contains a part of approximately the first ten amino acids of tendamistat, which are located at N, attached - in the given case via a bridge sequence - to the desired protein. 8. The use of a fusion protein according to claim 7, that is, a fusion protein obtainable according to claim 1, characterized in that it is used for the preparation of the desired protein.