NO881963L - PROCEDURE FOR THE PREPARATION OF GENETICALLY CODABLE POLYPEPTIDES. - Google Patents

Description

The invention relates to the genetic engineering production of polypeptides partly with an insoluble fusion protein with a part of ε-galactosidase, where this fusion protein is not in suspension as usual, but in solution where cleavage to release the desired proteins is provided.

Method for the production of a genetically coded polypeptide by linking the structural genes for this polypeptide in the correct reading frame for a gene for p<->galactosidase at a regulatory region, placing this gene structure in a bacterium, expressing an insoluble fusion protein, isolating the fusion proteins after cell termination and recovery of the desired polypeptides by chemical or enzymatic cleavage is largely known. According to the invention, the insoluble fusion protein is brought back into solution under specific conditions, which facilitates processing and provides new possibilities for processing.

K. Nagai et al., Proe. Nati. Acad. Pollock. USA 82 (1985) 7252-7255; Nature 309 (1984) 810-812, have already described dissolution of a fusion protein. On the other hand, it is a fusion protein of X cll protein and p-globin, which are relatively easily soluble and which only temporarily become insoluble at a high salt concentration.

It is surprising that dissolution is also possible in the case of the highly insoluble fusion proteins which, according to the invention, are inserted, whereby these proteins are kept in solution. One can thus on the one hand use the advantages associated with the high stability of the insoluble fusion protein in the host cell, or on the other hand use processing routes, which require soluble protein into the medium or which are greatly favored here.

The method according to the invention is characterized by reacting the insoluble fusion protein, suitably at a temperature from 0°C to room temperature, preferably to 10°C, with urea until it is dissolved, and separating the dissolved from the remainder, and removing the urea by dialysis from the solution and exposes the resulting fusion protein solution to cleavage.

In the following, several advantageous preparations according to the invention will be explained in more detail.

The lower limit of the urea concentration depends on the solubility of the insoluble fusion proteins, which is usually about 6 M. The upper limit depends on the solubility of the urea in the system, which is about about 8 M.

The amount of inserted fusion proteins per ml urea solution depends on the solubility of the fusion proteins and can consist of up to 10 mg/ml. In most cases, the fusion protein concentration of about 1 to about 2.5 mg/ml urea solution is favorable.

Dialysis is performed at a temperature from 0 to ICC, preferably 4°C, against a normal buffer solution. The pH value is approximately between 8 to 9, preferably 8.3 to 8.6. As with the addition of salt such as table salt is appropriately provided.

By this method, the fusion proteins are in solution and are very highly enriched, so that they can be recovered with an average purity of over 75%.

A particular advantage according to the present invention includes that the protein solution provided in this way can immediately be subjected to chemical or preferably enzymatic cleavage. It is advantageous, for example, to carry out the cleavage with factor Xa, which is for example described by Nagai et al., a.a.o. or in the European Patent with the number (hereinafter EP-A) 0 161 973. The same applies to cleavage with other enzymes such as trypsin or IgA protease (from Nelsserla gonorrhoea), so that the protein solution provided according to the invention maintains the conditions, such that no damage to the desired proteins occurs.

The method according to the invention also provides advantages in terms of the chemical cleavage of the fusion proteins, which usually takes place faster in solution than in suspension. In this way, it becomes e.g. easier to carry out the cleavage of a fusion protein-pre-β-lactamase inserted according to the invention with hydroxylamine.

It has surprisingly turned out that the method according to the invention is not subject to any limitations with regard to the inserted fusion protein. The nature of the desired proteins in the inserted fusion protein thus plays no decisive role.

To provide a sufficiently insoluble fusion protein, the "β-galactosidase gene" can be the unaltered wild-type gene, but usually a truncated gene is sufficient. Such abbreviated ε-galactosidase genes are known, for example, from EP-A 0 001 930 and from Br5ker, Gene Anal. Technology 3 (1986) 53-57. Very advantageous are the gene constructs for a 3-galactosidase fragment of over 250 amino acids, which is significantly smaller than the entire β-galactosidase sequence, preferably. gene structures encoding a fragment of up to about 800, especially up to about 650 amino acids. Particularly favorable gene constructs are those which consist of ε-galactosidase fragment from an amino-terminal and/or a carboxy-terminal part of the sequence.

The gene constructs for a shortened γ-galactosidase are thus advantageous because the "Ballast" proportion of cellular proteins becomes smaller, and the yield of desired protein also becomes higher. Figure 1 shows advantageous abbreviated p<->galactosidase genes, whereby they are linked with sequence designated A immediately, and with sequence designated B above a linker on the gene for the desired polypeptide. Figure 2 shows the construction of the vectors pLZPWB (containing the listed gene construct in Figure A 2). Figure 3 and its continuation, Figure 3A, show the construction of the vector pLZPWB/Xa/VL and pLZHP/Xa/VL/A. Figures 2 and 3 are, to make it clearer, drawing 1 is not to the correct scale.

An advantageous embodiment of the method according to the invention includes constructing the gene for a fusion protein, where after the DNA sequence for the 3-galactosidase fragment of more than 250 amino acids follows a DNA section, which codes for an amino acid or a group of amino acids, which enables chemical or enzymatic cleavage of the desired proteins, which causes the structural gene for the desired protein to emerge. Using known methods, this gene for the fusion protein is inserted into an E. coli expression vector, and transformed into E. coli, which thus expresses the encoded fusion protein. When the cells have reached the desired concentration of e.g. O.D.559=1.0, incubated with isopropyl-3-D-thiogalactopyranoside (IPTG) at a final concentration of 1 mM for sufficient time and finally centrifuged. The following is then resuspended in approximately 1/10 of the original volume with sodium phosphate buffer, after which the cell suspension is homogenized in a known manner. The homogenization can e.g. followed by twice passing through a "French Press". A stirring ball mill is also suitable, e.g. from Typ(<R>)DYN0 (Willi Bachofen, Basel).

A solution consisting of 400 g of sucrose per liters of said sodium phosphate buffer in a fourfold volume in relation to the lysates are then covered with the lysate and centrifuged (Sorvall-Rotor GSA, 4000-13000 rpm, 4 °C). The precipitate consisting of highly enriched fusion protein is suspended in as small a volume of tris buffer (10 mM, pH 8.3) as possible, and a relatively large volume of tris buffer, preferably a volume of the same size as the original lysates, where the original substance contains (10 mM tris-HCl, pH 8.3; 8 M urea), added dropwise with good mixing and preferably with stirring. This causes the fusion protein to go into solution. This urea solution is centrifuged again (Sorvall centrifuge, Rotor SS 34, 20,000 rpm, 1 hour, 4"C). The supernatant is dialyzed against tris buffer (10 mM tris-HCl, pH 8.3) at 4°C This results in the fusion proteins being in solution.

The fusion protein which was enriched in this way can, if desired, be further purified using known methods, but which are usually not necessary. Usually, the protein solution thus provided is immediately subjected to chemical or enzymatic cleavage.

The invention is explained in more detail in the following examples. The percentage designations in the following stand for the weight, unless otherwise stated.

Example 1

DNA constructs

20 jag plasmid pUC9 (cf. Vieira et al., Gene 19 (1982) 259-268; is subjected to a double cleavage with the restriction endonuclease EcoRI and Pvul, and a DNA fragment of 123 base pairs (Bp) was cleaved by gel electrophoresis. This fragment comprises part of the amino-terminal coding sequence for 3-galactosidase.

To isolate the carboxy-terminal parts of p<->galactosidase to the natural EcoRI cut sites, 20 µg of the plasmids pUR270 (Rtither and Mtiller-Hill, EMBO J.. 2 (1983) 1791-1794 ) are then digested with EcoRI and then with the enzyme Pvul. By electrophoresis on a 5% polyacrylamide gel, a DNA fragment of approx. 1200 Bp separated and isolated.

The amino-terminal and carboxy-terminal DNA fragments of β-galactosidase are ligated within 6 hours at 16° C and the ligation product is precipitated with ethanol. The precipitated and resuspended DNA is cleaved with EcoRI and then again fractionated on a 5% lg polyacrylamide gel. The DNA fragment with a length of approx. 1320 Bp is isolated by electroelution out of the gel and then ligated into the Eco RI cleavage site of the plasmids pBR322. The hybrid plasmid thus provided is designated pLZPWB.

The reaction steps described above are listed in Figure 2. The individual steps were carried out in a known manner (Maniatis et al., Molecular Cloning, Cold Spring Harbor 1982).

Plasmid pLZPWB is transformed into E. coli and amplified and again isolated. By cleavage with EcoRI, the amino- and carboxy-terminal shortened ε-galactosidase gene fragment can be cleaved out and preparatively isolated.

At the known restriction enzyme cleavage sites, other abbreviations can be constructed and then introduced into suitable expression plasmids. Figure IA shows such abbreviations^ This is how vector pLZHP is provided, when on the one hand pUR270 is cleaved with EcoRI and HpaI, and on the other hand pUR270 is cleaved with EcoRI and PvuII in an analogous manner as described above. This vector contains fragment A. 5 according to Figure 1.

Similar constructions are possible in analogous ways. Both the construction in Figure 1, part A and Figure I B, are chosen in such a way that the reading frame of the abbreviated β<->galactosidase continuously transitions into the reading frame of the desired carboxy-terminal polypeptides. The chemical-synthetic linker in figure 1, part-fi can be prepared as desired, but must of course include the stop codon-free reading frame and must also exhibit the desired restriction enzyme cleavage sites.

The gene for the variable domain of the light chain corresponds to known protein sequence (Kabat, E.A. et. al., Sequences of Proteins of Immunological Interest, Nat. Inst. of Health

(1983)) by chemical synthesis meets the desired oligonucleotide. The complete DNA sequence is given in Table 1. In an analogous manner, the variable domain of the heavy chain was prepared (Table 2).

The oligonucleotides were prepared with a DNA synthesizer (model 380A from Applied Biosystems) using the phosphoamidite method (Sinha, N.D. et. al., Nucleic Acid Res. 12 (1984) 4539). The oligonucleotides were purified by polyacrylamide gel electrophoresis and then treated with polynucleotide kinase, hybridized and ligated with T4 ligase (Dorper, T. and "Winnacker, E.L., Nucleic Acid Res. 11 (1983) 2575 and Rommens, J. et al., Nucleic Acid Rest. 11 (1983) 5921). A fragment of the expected size was preparatively purified by agarose gel electrophoresis and then ligated into vector pUC12, which had been digested with EcoRI and HindIII. The following DNA sequence was determined, and thus was verified to be correct, both for the light and heavy chain genes (Tables 1 to 4).

Two oligonucleotides were prepared for the construction of the fusion proteins consisting of the truncated β-galactosidase and the VL gene containing the recognition sequence for protease factor Xa (Ile-Glu-Gly-Arg) by extending from the EcoRI site 1 lacZ gene to the vectors and all the way to the first unique cleavage site in the v^ gene, ie a Pst-I site (Tables 3 and 4). Thus, the starting met ionine, which is not present in mature antibody molecules, becomes redundant. Both of the above-mentioned oligonucleotides were then cloned into vector pUC12 between the EcoRI site and the Pst-I site of the polylinkers and it was verified by DNA sequencing as error-free (pUC12/Xa/VL') (Fig. 3). This vector was then cleaved with PstI and HindIII and the remainder of the Vadel until the above-mentioned p"UC12/VL plasmid was inserted as isolated PstI-HindIII fragment, yielding plasmid pUC12/Xa/VL. This plasmid (pUC12/Xa /VL) was cleaved with EcoRI and HindIII and the thus obtained isolated fragment was inserted into vector pLZPWB which had just been cleaved with EcoRI and HindIII. In this way the target plasmid pLZPWB/Xa-/Vl was obtained. That the construction was correct was confirmed by verified using DNA sequencing and the exact fusions are shown in Table 3.

In an analogous manner, a fusion between vector pLZHP and the gene for the VL region was provided. Thus, the above-mentioned Pstl-HindIII fragment (corresponding to the main part of the Vj^ gene) was also ligated in the presence of a HindIII-EcoRI-HindIII Adapter A (Fig. 3 and the continuation, Fig. 3A) which provided vector pUC12/ Xa/Vl/A• Thus, the gene with the factor Xa recognition sequence could also be cleaved with EcoRI and the EcoRI site could be ligated with alkaline phosphatase-treated vector pLZHP providing plasmid pLZHP/Xa/Vl/A.

In an analogous manner, fusions with the synthetic variable domain of the heavy chain of the same antibodies were also performed, both in vector pLZPWB and in pLZHP. The sequence of the fusion region, beginning with the EcoRI site in the LacZ gene and the synthetic factor Xa recognition site, is listed in Table 4.

Example 2

Preparation, isolation and cleavage of the fusion proteins

Each of the plasmids thus provided was transformed into E. coli strain W3110. In a typical preparation, the cells were placed in 400 ml of LB medium and at an O.D.550 of 1.0 they were induced with IPTG (final concentration 1 mM) and incubated with agitation overnight. The cells were centrifuged (SORVALL rotor GSA, 4000 rpm, 15 min., 4°C) and resuspended in 40 ml of sodium phosphate buffer (50 mM, pH 7.0). Then the cells were homogenized by running the cells twice through a French Press.

A solution (160 ml) consisting of 400 g of sucrose per liter of said sodium phosphate buffer was mixed with lysate and centrifuged (SORVALL Rotor GSA, 4000 rpm. 4°C, 30 min.). The precipitate was thoroughly resuspended in 1 to 2 ml tris buffer (10 mM, pH 8.3) and then added dropwise with stirring to a 40 ml urea solution (10 mM tris-HCl. pH 8.3. 8 M urea) . Thus the fusion protein went into solution. This solution was again centrifuged (SORVALL centrifuge, Rotor SS34, 20000 rpm, 1 hour, 4°C). The supernatant was dialyzed at 4°C against tris buffer (10 mM tris-HCl, pH 8.3). The solution became clear; and the fusion protein remained in solution.

The fusion protein thus obtained was cleaved with factor XA from ox blood. Factor X was purified as described by Fujikawa et al. (Fujikawa K., et. al. Biochemistry 11 (1972) 4882), and activated to factor Xa (Fujikawa K. et. al., Biochemistry 11 (1972) 4982). Cleavage of the fusion protein with factor Xa was performed at 4°C in a tris buffer (50 mM tris-HCl, pH 8.0, 100 mM NaCl, 1 mM CaCl2), whereupon the fusion protein also becomes soluble. The cleavage product was purified to homogeneity chromatographically. Correct factor Xa cleavage was confirmed by sequencing the N-terminus of the purified products.

Claims (9)

1. Process for the production of genetically coded polypeptides by linking the structural genes for this polypeptide in the correct reading frame for a gene for β-galactosidase at a regulatory region, placing this gene structure 1 in a bacterium, expressing an insoluble fusion protein, isolating the fusion protein after cell termination and recovering the the desired polypeptides using chemical or enzymatic cleavage, characterized by dissolving the insoluble fusion protein with urea and removing the urea from the solution by dialysis and performing cleavage on the thus provided solution of the fusion protein. 2. Method according to claim 1, characterized in that the dissolution is carried out at a temperature from 0°C to room temperature, preferably at 4 to 10°C. 3. Method according to claim 1 or 2, characterized in that the urea concentration is approximately 6 to approximately 8 M. 4. Method according to one or more of the above-mentioned claims, characterized in that the concentration of the fusion proteins in the urea solution is up to 10 mg/ml. 5 . Method according to one or more of the above-mentioned claims, characterized in that the dialysis is carried out at a temperature from 0 to 10°C, preferably at approximately 4°C. 6. Method according to one or more of the above-mentioned claims, characterized in that the pH value of the dialysis solution is in the range of from 8 to 9, preferably 8.3 to 8.6. 7. Method according to one or more of the above-mentioned claims, characterized in that the dialysis-enriched solution of the fusion proteins is subjected to chemical or preferably enzymatic cleavage without further purification. 8. Method according to one or more of the above-mentioned claims, characterized in that the gene for 3-galactosidase codes for a 3-galactosidase fragment of over 250 amino acids, but which is significantly smaller than the entire ε-galactosidase sequence, preferably to about 800, especially to about 650 amino acids. 9. Method according to claim 8, characterized in that the β-galactosidase fragment consists of an amino-terminal and/or carboxy-terminal partial sequence.