CA2040810C

CA2040810C - Process for producing foreign proteins in streptomyces

Info

Publication number: CA2040810C
Application number: CA002040810A
Authority: CA
Inventors: Klaus-Peter Koller; Gunther Riess
Original assignee: Hoechst AG
Current assignee: Hoechst AG
Priority date: 1990-04-21
Filing date: 1991-04-19
Publication date: 2001-07-24
Anticipated expiration: 2011-04-19
Also published as: KR910018551A; TW213487B; PL169178B1; LV10494A; LTIP1523A; YU48435B; YU69691A; HU210358B; CN1055952A; JPH04228086A; ZA912937B; JP3319605B2; BR9101587A; EP0453969B1; FI911882A; NO911557L; IL97903A0; ATE142263T1; HRP940770A2; NZ237882A

Abstract

Process for producing foreign proteins in streptomyces Genetic structures which code for the signal sequence and about the first ten amino acids of tendamistat as well as a desired protein are expressed in stregtomyces host cells with a high yield, and the fusion proteins are secreted into the medium.

Description

2~D4~~1(~
HOECHST AKTIENGESELLSCHAFT HOE 90/F 122 Dr.RL/gm Description Process for producing foreign proteins in streptomyces The European Patent Application with the publication number (EP-A) 0,289,936 discloses the production of fusion proteins by coupling the structural gene for the desired protein to the 3'-end of the coding strand of the optionally modified tendamistat gene, expressing this genetic structure in a streptomyces host cell and isolat-ing the secreted fusion protein frofi the supernatant. In a preferred embodiment the tendamistat gene is truncated at the 3'-end. For the truncation, the cleavage sites for the reet.riction enzyme BstEII in the region of triplets 31 and 32, StuI in the region of triplets 43 and 44, and Sau3A in the region of triplets 52 and 53 are used.
In a further development of this inventive concept it has already been proposed to produce a fusion protein in which a truncated proinsulin whose C-chain is composed of only one or two lysine residues ("mini-proinsulin") follows the tendamistat portion. Another further develop-ment which has been proposed is to truncate the tendamistat portion, too, in fusion proteins of this type (EP-A 0,367,163 published on May 9,1990).
Surprisingly, it has now been faund that fusion proteins with a very short tendamistat portian are stable in streptomyces cells and are secreted into the medium. The fusion proteins obtained in this way behave like "mature"
proteins because of the very short tendamistat chain and generally are present in the medium in the correct tertiary structure.
EP-A 0,177,827 discloses a synthetic signal sequence for transporting proteins in expression systems, wherein the ' DNA is virtually identical to a natural signal sequence but has one or more cleavage sites for endonucleases, ~o~o~zo -which are not contained in the natural DNA. When the gene for the protein to be transported is coupled to such a DNA sequence, this fusion gene is incorporated into a vector, and a host cell which transports the expressed protein out of the cytoplasm is transformed therewith, it is possible to produce eukaryotic, prokaryotic or viral proteins in prokaryotic and eukaryotic cells. Using the periplasmic protein alkaline phosphatase as an example, it is shown that it is advantageous in the expression in E. coli to place the codons for about the first 40 amino acids of alkaline phosphatase immediately downstream of the pre-sequence and upstream of the structural gene for the desired protein. However, in many cases even fewer additional amino acids are sufficient, for example about 10, preferably about 5. About 90~ of a corresponding fusion protein with simian proinsulin was transported into the periplasmic space.
It has also been proposed (WO 91/03550, published March 21, 1991) to produce fusion proteins by constructing a mixed oligonucleotide which codes for the ballast portion of the fusion protein, introducing this oligonucleotide into a vector in such a way that it is functionally coupled to a regulatory xegion and the structural gene for the desired protein, transforming suitable host cells with this plasmid population obtained in this way and select-ing those clones which show a high yield of coded fusion protein. The oligonucleotide preferably consists of 4 to 12, in particular 4 to 8, triplets in this case.
It has already been attempted to produce fusion proteins with a short ballast portion. Thus, a gene fusion which codes for a fusion protein from the first 10 amino acids of p-galactosidase and somatostatin has, for example, been produced. However, it became apparent that this short p-galactosidase fragment was not sufficient to ' protect the fusion protein from digestion by endogenous host proteases (US-A 4,366,246, column 15, paragraph 2).

Accordingly, fusion proteins whose ballast portion consists of a ~-galactosidase fragment having more than 250 amino acids are described in EP-A 0,290,005 and 0,292,763.
However, fusion proteins of about the first 10 amino-terminal amino acids of tendamistat and a desired protein, for example a proinsulin, are, in fact, unexpec-tedly stable in streptomyces host cells and are secreted into the medium, from which they can be isolated in high yields. Surprisingly this is also true for relatively small proteins such as "mini-proineulins".
"About 10 amino acids" is intended to mean in this case that even fewer amino acids are suitable, for example the first 7 N-terminal amino acids of tendamistat, but preferably not more than 10. Fusion proteins in whose tendamistat portion proline is present in position 7 and/or 9 (as in the natural sequence) are preferred.
However, it is, of course, possible to choose a larger tendamistat ballast portion in accordance with the embodiments already known or proposed, the advantage of low "ballast" being lost more and more, of course.
It is possible and even advantageous to vary the natural amino acid sequence of the tendamistat portion, i.e. to exchange or delete amino acids, or to insert amino acids which do nat occur in the natural amino acid sequence.
Furthermore, it is possible to vary the amino ac,td sequence in the signal peptide.
Particularly advantageous fused constructions can be readily determined by simple preliminary experiments if the concept of the invention is known.
It is furthermore possible to realise the concept of the invention also in other Gram-positive bacterial cells, for example in bacillus or staphylococcus cells using signal sequences which are "recognized" by these hosts.

2~4~8~.0 The fusion proteins obtained according to the invention are present in the medium in a dissolved form, which has many advantages in processing and purification. Thus, enzymatic processing with cleaving of the ballast portion can, for example, take place directly on the secretion product, and working-up stags, such as the ones necessary for insoluble fusion proteins, do not have to be carried out. It is also possible to carry out concentration or purification processes, for example affinity chroma-tography but also ultrafiltration, precipitation, ion exchange chromatography, adsorption chromatography, gel filtration or high-pressure liquid chromatography, first, before further processing.
In the examples which follow the invention is illustrated in more detail.
The starting material for the plasmid constructions is plasmid pKK500 which was proposed in EP-A 0, 367,163 . This plasmid differs from plasmid pKR400 known from EP-A 0,289,936 in that the proinsulin gene is replaced by an analogous gene which, instead of the C chain, merely encodes the amino acid lysine, and in that a terminator sequence is inserted immediately downstream of this "mini-proinsulin" gene. Tables 1 and 2 from EP-A 0,367,163, in which the "mini-proinsulin" gene and the terminatar sequence, respectively, are Shawn, are enclosed as an annex to the description.
The plasmids pRK400 and pKK500 contain a XmaIII cleavage site in the signal sequence of the a-amylase inhibitor gene (in the region of triplets -5 to -7).
Example 1 Plasmid pKK500 is opened up with the restriction enzymes EcoRI and XmaIII, and the large fragment is separated by ' gel electrophoresis on a 0.8% agarose gel and isolated by electroelution. This fragment is ligated with the DNA

fragment (1) (SEQ ID NOal) Ala Gly Pro Ala Ser Ala 5° G GCC GGG CCG GCC TCC GCC
3' CCC GGC CGG AGG CGG
(XmaIII) Asp Thr Thr Val Ser Glu Pro GAC ACG ACC GTC TCC GAG CCG 3' CTG TGC TGG CAG AGG CTC GGC TTA A 5' (EcoRI) 20408.0 which has been synthesized by the phosphoramidite method, and the lfgation mixture is transformed into E. cola.
Plasmid pKK510 is obtained. This plasmid encodes a preproinsulin in which the signal sequence of tendamistat is followed by the first 7 amino acids of tendamistat which are followed by the mini-proinaulin chain.
Example 2 In analogy with the process described in EP-A 0,289,936 for transferring plasmid pKK400 into expression plasmid pGFi, plasmid pKK510 is transferred into expression plasmid pKFI:
The isolated plasmid DNA of pIQC510 is cut with the restriction enzymes SphI and Sstl, and the small fragment with tha fusion gene is isolated. The commercial expres-sion plasmid pIJ 702 (obtainable from John Inner Founda-tion, Norwich, England) is cut with the same enzymes and the large fragment is isolated. These two isolated frag-ments are ligated, the ligation mixture transformed into S, lividans TK24 and the plasmid is isolated from the thiostregton-resistant white (i.e. not capable of forming mQlanin) transformants. Clones which carry the introduced ' 25 insert are tested for the formation of fusion proteins in a shake culture.

- 2~4~F~1(~
The encoded fusion protein is expressed in a manner known per se: if the transformed strain is incubated in a shaken flask at above 25°C for 4 days and the mycelium is separated from the culture solution by centrifugation, it is possible, after electrophoresis of 20 ~1 of culture filtrate in a 15~ polyacrylamide gel, to visualize by dyeing with ~COOMASSIE Rlue the fusion protein formed in the culture supernatant as an additional protein band which does not occur in a control experiment in which the strain was transformed only with pIJ 702.
If the culture filtrate is treated with lysyl endo-proteinase, it is possible to detect de-(B30)-Thr-insulin, which is verified by an authentic control on gel electrophoresis.
Furthermore, it is possible to detect the fusion protein in the culture filtrate with insulin antibodies either in an immunoblot or with an insulin RIA.
Example 3 The procedure according to Examples 1 and 2 is carried out, but the synthetic fragment (2) with the SEQ ID N0:2 _5 _1 Als Gly Pro Ale Sex Ale 5' G GCC CCG GCC TCC
GGG GCC

3' , CCC GGC CGG AGG
CGC

(XmaIII) Asp Thr Thr Val Ser Glu Pro Asp Pro GAC ACG ACC GTC TCC GAG CCC GAC CCG ~ 3' CTG TGC TGG CAG AGG CTC GGG CTG GGC TTA A 5' (EcoRI) is used and the plasmids pRR320 and pKF2, respectively, are obtained in this way.

20401,0 These plasmids encode a fusion protein which differs from the one according to Examples I and 2 in that the first 7 amino acids of tendamistat are followed by asparagine (instead of the natural amino acid alanine) and that this is followed by the ninth amino acid in tendamistat, proline. By exchanging alanine for asparagine, an addi tional positive charge is therefore introduced into the ballast portion of the fusion protein. Surprisingly, yields about 20 to 30$ higher than in Example 2 are obtained.
Example 4 If the procedure according to Examples 1 and 2 is carried out, but the synthetic fragment (3) with the SEQ ID N0:3 Ala Gly Pro Ala Ser Ala 5' G GCC GGG CCG GCC TCC GCC
3' CCC GGC CGG AGG CGG
(XmaIII) ~l5 Asp Thr Thr Val Ser Glu Pro Ala Pro GAC ACG ACC GTC TCC GAG CCC GCA CCG 3' GTG TGC TGG CAG AGG CTC GGG CGT GGC TTA A 5' is used, the plasmids pKK330 and pKF3, respectively, are obtained. These plasmids differ from those according to Examples 7, and 2 in that they encode the first 9 natural amino acids of tendamistat. In comparison with Example 2, yields about 10~ higher are obtained.
Example 5 The fusion protein encoded by pKR500 contains between the tendamistat portion and the B chain of proinsulin a linker sequence which codes for the amino acids -Asn-Ser-Asn-Gly-Lys. This terminal Lys and the Lys representing the C chain are replaced by Arg as described below. In this procedure, the single StyI cleavage site in the region of colons B30 to Al in the proinsulin sequence is used.
Isolated plasmid DNA from pKK500 is cut using Styl, digested with S1 nuclease to remove protruding ends and the excess nuclease is extracted using phenol-chloroform.
The linearized plasmid is then subsequently cut with EcoRI, and the large fragment is electrophoretically separated off and isolated by electroelution. This fragment is ligated with the synthetic fragment (4) (SEQ ID N0:4) Asn Ser Asn Gly Arg Phe Val Asn Gln His Leu Cys Gly Ser His AAT TCG AAC GGC CGC TTC GTC AAC CAG CAC CTG TGC GGC TCG CAC
GC TTG CCG GCG AAG CAG TTG GTC GTG GAC ACG CCG ACC GTG
(EcoRI) 15 Leu Val Glu Ala Leu Tyr Leu Val Cya Gly Glu Arg Gly Phe Phe CTC GTG GAG GCC CTC TAC CTG GTG TCC GGG GAG CGC GGC TTC TTC
GAG CAC GTC GGG GAG ATG GAC CAC ACC CCC CTC GGG CCG AAG AAG
S30 C(831) Tyr Thr Pro Lys Thr Arg TAC ACC CCC AAG ACC CGC
ATG TGG GGG TTC TGG CCG
and the ligation mixture is transformed into E. cola. The desired clones are tested by restriction analysis of the plasmid contained, using the newly developed SstII
20 cleavage site. Furthermore, the entire SphI-SstI fragment is sequenced.
In order to express the encoded fusion protein, the - 9 - 204~81.0 fragment, which has been checked by sequence analysis, is ligated into the vector pIJ 702, which has been cut with the same enzymes, resulting in the expression vector pGF4.
The secreted fusion protein encoded by pGF4 can be detected, on the one hand, by the a-amylase inhibitor plate test (EP ~ 0,161,629, Example 3} and, on the other hand, from the supernatant of the shake culture in analogy with Example 2.
Example 6 If fragment (4) is, in analogy with Example 5, inserted into the vectors pKK510, 520 and 530, the vectors pKR610, 620 and 630 are obtained. The incorporation of the respective SphI-SstI fragments with the coding sequence for the fusion proteins into the vector pIJ 702 results in the expression vectors pKFll, 12 and 13. The expres-sion of the secreted fusion proteins is tested in analogy with Example 2.
Example 7 In order to increase the expression of derivatives of the plasmid pIJ 702, the m~lanin promoter is deleted there-from by digestion with Pstl and SphI and is replaced by the synthetic fragment (5) (SEQ ID N0:5) ratr sc~a CTGCAGTGATCAGGGGGACCCTTGTGCGAATTTCCGTTACGGGTTTGGGTGGTAGGG
GACGTCACTAGTCCCCCTGGGAACACGCTTAAAGGCAATGCCCAAACCCACCATCCC
SphI

a5 ACGCACCCGAAGAGGAGGCCCCAGCATGC
TGCGTGGGCTTCTCCTCCGGGGTCGTACG

~o~o~~.o _ 10 _ A tandem construction of the synthetic and the tenda-mistat promoter is thereby obtained. The plasmid is called pGR110.
If the synthetic fragments (1), (2) and (3) are, after cutting with SphI and SstI, inserted into pGR110, the expression vectors pGR200, 210 and 220 result. In an analogous way, the expression vectors pGR250, 260 and 270 are obtained with fragment (4).
Example 8 If it is intended to produce human insulin from the insulin precursors by combining trypsin, or an enzyme with an identical effect, and carboxypeptidase B, it is advantageous to cleave off rapidly the ballast portion in the course of the cleavage reaction in order to favor the cleavage reaction leading to the B31 (Arg)-insulin. For this purpose, a modification of the amino acids upstream of amino acid B1 (Phe) is suitable:
The procedure is analogous to Example 1 and the plasmid pKK 500 is opened using the restriction enzymes EcoRI and DraIII. The original fragment is then replaced by DNA
fragment (6) (SEQ ID NOs6) E1 ' Asn Ser Ala Arg Phe Vnl Azsn Cln His Leu Cyss Gly Ser Hia Leu 5' AAT TCG GCC CGC TTC GTC AAC CAG CAC CTG TGC GGC TCG CAC CTC 3' 3' GC CGG GCG AAG CAG TTG GTC GTG GAC ACG CCG AGC GTG 5' (EcoRI) (DraIII) which has been synthesized by the phosphoramidite method.
Cloning into E. coli and expression in Streptomyces lividans are carried out in accordance with Example 1 and Example 2, respectively. Plasmid pRR640 and expression plasmid pKFl4 result.

2~4~~1(~

The plasmid which results according to Example 5 (after incorporation of fragment (4)) can be treated in an analogous way. The plasmids pKK650 and pKFlS are obtained in this way.

~:~~~~'.~

Annex (from EP-A 0,367,163):
Table 1 g1 10 ASNSER ASN GLYLYS PHEVAL ASNGLN HIS LEUCYS GLYSER HIS

AATTCG AAC GGCAAG TTCGTC AACCAG CAC CTGTGC GGCTCG CAC

GC TTG CCGTTC AAGCAG TTGGTC GTG GACACG CCGAGC GTG

(EcoRI) LEUVAL GLU ALALEU TYRLEU VALCYS.GLY GLUARG GLYPHE PHE

CTCGTG GAG GCCCTC TACCTC GTGTGC GGG GAGCGC GGCTTC TTC

GAGCAC CTC CGGGAG ATGGAC CACACG CCC CTCGCG CCGAAG AAG

C A~ 40 TYfi THR PRO LYS THR LYS GLY ILE VAL GLU GLN CYS CYS THR SER
TAC ACC CCC AAG ACC AAG GGC A?C GTG GAG CAG TGC TGT ACG TGC
ATG TGG GGG TTC TGG TTC CCG TAG CAC CTC GTC ACG ACA TGC AGG
ILE C'YS SEH LEU TYR GLN LEU GLU ASN TYR CYS ASN STP STP
ATC TGC TCC CTC TAC CAG CTC GAG AAC TAC TCC AAC TAG TAA
TAG ACG AGC GAG ATG GTC GAG CTC TTG ATG ACG TTG ATC ATT
GTC GAC CTG CAG CCA ' CAG CTG GAC GTC GGT TCG A
SalI (HindIII) Table 2 5' -CGATAAA COGATACAATTAAAGGGTCCTTT TCGAG CCITIZTTT'TTGGAGATTTTCAACGTGGATC
GCTpTTTGGCTATGTTAATITCCGAGGAAAACCTCGGAAAAAAAA.ACCTCTAAAAGTTGCACCTAG-5'

Claims

1. A fusion protein which comprises a tendamistat portion, other than the signal peptide of tendamistat, and a desired protein, wherein the first seven to ten amino acids of the N-terminal portion of tendamistat are coupled to the desired protein.

2. A fusion protein as claimed in claim 1, wherein the first seven to ten amino acids of the N-terminal portion of tendamistat are coupled via a bridge sequence to the desired protein.

3. A process for making a desired protein, which comprises cleaving a fusion protein as defined in claim 2.

4. A process for preparing a fusion protein of claim 1, which comprises ligating a structural gene for a desired protein onto the 3' end of the coding strand of a DNA
encoding the first seven to ten amino acids of the N-terminal portion of tendamistat, expressing the ligated gene and DNA in a Streptomycetes host cell, and isolating secreted fusion protean from the supernatant.

5. A process as claimed in claim 4, wherein the DNA
encoding the first seven to ten amino acids of the N-terminal portion of tendamistat is ligated to the gene for the desired protein via a DNA encoding a bridge sequence.

6. A process as claimed in claim 5, wherein the desired protein is a proinsulin derivative in which the B chain is connected to the A chain via a bridge sequence comprising Lys or Lys-Lys.

7. A process as claimed in claim 4, wherein the desired protein is a proinsulin derivative in which the B chain is connected to the A chain via a bridge sequence comprising Lys or Lys-Lys.

8. A substantially purified DNA which encodes the fusion protein as claimed in claim 1.

9. A vector containing a substantially purified DNA as claimed in claim 8.

10. A Streptomycetes cell containing a vector as claimed in claim 9.

11. A fusion protein as claimed in claim 1, wherein the desired protein is a proinsulin derivative in which the B
chain of the proinsulin derivative is connected to the A
chain of the proinsulin derivative via a bridge sequence comprising Lys or Lys-Lys.

12. A fusion protein as claimed in claim 11, wherein the first seven to ten amino acids of the N-terminal portion of tendamistat are ligated to the proinsulin derivative via a bridge sequence.

13. A substantially purified DNA which encodes the fusion protein as claimed in claim 11.

14. A vector containing a substantially purified DNA as claimed in claim 13.

15. A Streptomycetes cell containing a vector as claimed in claim 14.

16. A process for preparation of a desired protein, which comprises cleaving the fusion protein as claimed in claim 1 to separate the tendamistat portion from the desired portion of said fusion protein.

17. A process for the production of a fusion protein, which comprises ligating the structural gene for a desired protein to a DNA coding for the signal sequence and the first seven to ten amino-terminal amino acids of tendamistat, expressing this DNA sequence in a Streptomycetes host cell and isolating the secreted fusion protein from the supernatant.

18. A substantially purified DNA encoding a signal sequence, the first seven to ten amino acids of the N-terminal portion of tendamistat, and another protein.

19. A substantially purified DNA encoding a signal sequence, the first eight to ten amino acids of the N-terminal portion of tendamistat, and another protein, wherein the eighth amino acid of the N-terminal portion of tendamistat has been replaced by aspartic acid.

20. The fusion protein of claim 1, wherein said first seven to ten amino acids of the N-terminal portion of tendamistat consists of the first seven amino acids of the N-terminal portion of tendamistat.

21. The fusion protein of claim 1, wherein said first seven to ten amino acids of the N-terminal portion of tendamistat consists of the first nine amino acids of the N-terminal portion of tendamistat, and wherein the eighth amino acid of the N-terminal portion of tendamistat has been replaced by aspartic acid.

22. The fusion protein of claim 1, wherein said first seven to ten amino acids of the N-terminal portion of tendamistat consists of the first nine amino acids of the N-terminal portion of tendamistat.

23. The process of claim 17, wherein said first seven to ten amino-terminal amino acids of tendamistat consists of the first seven amino-terminal amino acids of tendamistat.

24. The process of claim 17, wherein said first seven to ten amino-terminal amino acids of tendamistat consists of the first nine amino-terminal amino acids of tendamistat, and wherein the eighth amino-terminal amino acid of tendamistat has been replaced by aspartic acid.

25. The process of claim 17, wherein said first seven to ten amino-terminal amino acids of tendamistat consists of the first nine amino-terminal amino acids of tendamistat.

26. The substantially purified DNA of claim 18, wherein said first seven to ten amino acids of the N-terminal portion of tendamistat consists of the first seven amino acids of the N-terminal portion of tendamistat.

27. The substantially purified DNA of claim 18, wherein said first seven to ten amino acids of the N-terminal portion of tendamistat consists of the first nine amino acids of the N-terminal portion of tendamistat.