US 5,710,027 AGrant

Process and Vector for Expressing Alpha-Interferon in E. Coli

Issue Date:1998-01-20

•16 Claims

•7 Drawing Sheets

Abstract

Methods and vectors for expressing interferon-alpha (IFN-.alpha.) proteins in E. coli are provided. Use of a vector comprising an IFN-.alpha. sequence fused to an E. coli heat-stable enterotoxin signal sequence (STII) under the control of the E. coli phosphatase (phoA) promoter affords high levels of correctly-folded and -processed, biologically active IFN-.alpha. polypeptides.

Metadata

Assignee

Boehringer Ingelheim International GmbH

Inventors

Rudolf Hauptmann
Edgar Falkner
Gerhard Bodo
Tilman Voss
Ingrid Maurer-Fogy

Application Information

Application Number:US 2496712

Filing Date:1994-05-26

Priority Date:1993-05-26

Art Unit:182

Classifications

IPC:

C12N 1570C12N 1521

Field of Search:

43553053669.51;320.1;252.33;69.8412;415;41623.4;23.52;24.1

Patent Drawings (7 sheets)

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention relates to a process for preparing interferon-alpha (herein "IFN.alpha.") by bacterial expression and subsequent isolation, an expression vector for this purpose and a process for purifying IFN.alpha..

2. Description of Background Art

Processes for preparing IFN.alpha. by bacterial expression are known. The conventional process is based on cytoplasmic expression of the protein in Escherichia coli, in which the expressed IFN.alpha. is either present in the cell in insoluble form in so-called inclusion bodies or is found in the soluble fraction after the cell wall has been permeabilized or lysed (Thatcher & Panayotatos, Methods Enzymol. 119:166-177 (1986); Goeddel et al., Nature 287:411-416 (1980); Dworkin-Rastl et al., Gene 21:237-248 (1983)). Cytoplasmic expression does have some disadvantages, however. The synthesized protein is not correctly folded because reducing conditions prevail in the cytoplasm and the protein does not form the necessary disulfide bridges. Therefore, IFN.alpha. formed by cytoplasmic expression has to be oxidized and re-folded during preparation. This re-folding process is inefficient and leads to unwanted by-products, such as wholly or partially reduced forms, oligomers produced by intermolecular disulfide bridge building, and wrongly folded molecules formed by the cross-linking of improper disulfide bridges. These by-products are difficult to separate. A further problem with cytoplasmic expression is that the N-terminal methionine which is forms during translation is only partly cleaved from the IFN.alpha. synthesized intracellularly, and this form is undesirable. The resulting N-Met-IFN.alpha. is almost impossible to remove from the native IFN.alpha..

A further disadvantage of the synthesis processes currently used is the use of promoters which, in the non-induced state, are not completely switched off. Moreover, such promoters have to be induced by the addition of chemicals and demonstrate an inefficient expression rates even in the induced state. An example of a commonly used inducible promotor is the trp-promotor from Serratia marcescens.

In order to overcome some of the above mentioned disadvantages and still employ the economical E. coli expression system, Breitling et al. attempted (Breitling et al., Mol. Gen. Genet. 217:384-391 (1989)) to express IFN.alpha.1 and an IFN.alpha.1/2 hybrid using a vector which enabled secretion of the IFN through the cell membrane into the periplasmic space. Breitling et al. used a promotor, ribosome binding site (RBS) and signal sequence from a bacterial staphylokinase gene (sak42D). They observed that 60-80% of the IFN.alpha. thus produced was secreted into the periplasmic space. However, the protein contained N-terminal amino acids as a result of the vector construction. These N-terminal amino acids do not occur in the corresponding native IFN.alpha.. A serious drawback of this expression system was the fact that the strains transformed with this construct did not remain genetically stable; the expression cassette was inactivated by the spontaneous insertion of an IS1 insertion element. The objective of providing an expression/secretion system in E. coli for preparing human IFN.alpha. has thus not been previously achieved.

A known expression/secretion cassette which has been successful in the expression of human growth factor receptor in E. coli was a construct from the promotor of alkaline phosphatase (phoA) and the signal sequence of the heat-stable enterotoxin II (STII) (Fuh et al., J. Biol. Chem. 265:3111-3115 (1990)).

Another problem in the production of recombinant IFN.alpha. in E. coli is the purification of the protein from the bacterial lysate. A number of processes are known (for example, Thatcher & Panayotatos, Meth. Enzymol. 119:166-177 (1986); European Patent No. EP-A 203,382)). In order to obtain the native folded protein it is preferable to use processes which do not require denaturation and precipitation steps. Such a process is described in European Patent No. EP-A 396,555. The process disclosed in EP-A 396,555 consists of the steps of performing immunoaffinity chromatography, reverse phase chromatography (RPC), cation exchange chromatography, concentration by ultrafiltration and gel filtration chromatography. This process, like other known processes, is based on the high selectivity of immunoaffinity chromatography in the first step. Prior to the present invention there was no known process for preparing highly purified IFN.alpha., particularly IFN.alpha.2, which dispenses with both the denaturation/precipitation steps and immunoaffinity chromatography. However, a process of this kind is desirable for economic and technical reasons. There is a need for monoclonal antibodies for immunoaffinity chromatography and the cost of these reagents is high. Further, since the life of the antibody-coupled matrices is limited, a continuous supply of these antibodies is required. Moreover, there is a demand for IFN.alpha. proteins and simplifying the purification of this class of proteins will make purification less costly and less difficult.

SUMMARY OF THE INVENTION

The invention is directed to a more economical and efficient process for preparing IFN.alpha., particularly IFN.alpha.2, by recombinant expression in E. coli. To achieve this goal, the problem of establishing an efficient and stable system for the expression/secretion of the protein into the periplasmic space or the culture medium has been solved. Moreover, the inventors have developed a process for highly purifying the expressed protein gently, without any denaturation/precipitation steps and without the need for immunoaffinity chromatography.

The inventors have solved these problems by means of the present invention. The establishment of a stable expression/secretion system for IFN.alpha. in E. coli was achieved by constructing a vector which contains the signal sequence (leader sequence) of heat stable enterotoxin II (STII) from E. coli, linked to the coding sequence for a mature human IFN.alpha., preferably IFN.alpha.2. It is preferred that expression control is effected by means of the promotor of alkaline phosphatase from E. coli (phoA). It is further preferred that the construct comprise the ribosome binding site of the STII gene. Surprising results were achieved by providing a purification process for IFN.alpha. which consists of the steps of performing adsorption chromatography on silica gel, performing hydrophobic interaction chromatography (HIC), performing cation exchange chromatography and performing anion exchange chromatography.

Accordingly, one object of the invention is a process for preparing IFN.alpha. by expression in E. coli, comprising the steps of: expressing IFN.alpha. in cells comprising a vector in which the signal sequence of the gene for the heat stable enterotoxin II (STII) from E. coli is linked to a sequence which codes for mature human IFN.alpha.; and isolating the expressed IFN.alpha..

Another object of the invention is a bacterial expression vector for expressing IFN.alpha. in E. coli, comprising a signal sequence of the STII gene operably linked to a sequence which codes for IFN.alpha.. It is preferred that the IFN.alpha. be mature human IFN.alpha..

Yet another object of the invention is a process of purifying IFN.alpha., comprising the steps of: performing chromatography on silica gel; performing hydrophobic interaction chromatography; performing cation exchange chromatography; and performing anion exchange chromatography.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1A) depicts a gene map of pCF2. The EcoRI-BamHI fragment of pAT153 was replaced by the expression cassette for IFN.omega.1.

FIG. 1B) depicts a sequence of the EcoRI (destroyed)-BamHI part, which contains the phoA-promotor, STII leader and an IFN.omega.1 gene (SEQ ID NO:8-9).

FIG. 2A) depicts a gene map of the plasmid pDH13. The SspI-PstI fragment of pAT153 was replaced by the IFN.alpha.2c expression cassette (EcoRI-PstI fragment of 2B). The .beta.-lactamase gene is destroyed.

FIG. 2B) depicts a nucleotide sequence of the EcoRI-HindIII insert of pDH13 (SEQ ID NO:10-11).

FIGS. 3A-D depict graphs showing the chromatographic purification of IFN.alpha.2c, extracted from bacterial biomass. Optical absorbance is indicated.

FIG. 3A) depicts an elution curve of adsorption chromatography on silica gel. The arrow indicates elution with 800 mM tetramethylammonium chloride.

FIG. 3B) depicts an elution curve of hydrophobic interaction chromatography on phenyl-Sepharose.TM.. Elution was carried out with a linear gradient of 0 to 100% of solvent B as indicated (----).

FIG. 3C) depicts an elution curve of sulphopropyl cation exchange chromatography. Elution was carried out with a gradient of 0 to 100% solvent B as indicated (----).

FIG. 3D) depicts an elution curve of anion exchange chromatography on DEAE Sepharose Fast Flow.TM.. Elution was carried out with a gradient of 0 to 100% solvent B as indicated (----).

The bars under the main peaks in each chromatogram indicate the pools which contain IFN.alpha.2, which were collected and used for the subsequent steps.

FIG. 4 depicts an SDS-PAGE of purified IFN.alpha.2c, stained with Coomassie blue. The numbers in the left-hand margin indicate the molecular weights of the standard proteins.

Lane 1: IFN.alpha.2c standard

Lane 2: 3 .mu.g IFN.alpha.2c

Lane 3: 6 .mu.g IFN.alpha.2c

Lane M: molecular weight standard

FIGS. 5A-B depict chromatograms of purified IFN.alpha.2c separated by Reversed Phase HPLC (RP-HPLC). Optical absorbance was measured at 214 nm.

FIG. 5A) depicts an elution curve of IFN.alpha.2c with a linear gradient of 20-68% solvent B in 24 minutes.

FIG. 5B) depicts an elution curve of IFN.alpha.2c with a linear gradient of 45-53% solvent B in 30 minutes.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The starting point for the construction of the vector may be a plasmid which is replicable in E. coli, such as the plasmid pAT153 (Twigg & Slierratt, Nature 283:216-218 (1980)), which is highly suitable for this purpose. Skilled artisans know of and may use other vectors and plasmids for gene expression in prokaryotes (see, for example, Sambrook et al., Molecular Cloning--A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989), for expression vectors and plasmids). A nucleotide sequence which codes for the signal peptide of the STII gene is known (Picken et al., Infection and Immunity 42:269-275 (1983); Lee et al., Infection and Immunity 42:264-268 (1983)). The skilled artisan will readily understand how to prepare variants of this sequence using methods known in the art such as, for example, by mutation (substitution, deletion, insertion, addition) without changing the basic properties thereof, and particularly to prepare nucleotide sequences which code for the same amino acid sequence of the signal peptide owing to the degeneracy of the genetic code (Sambrook et al., Molecular Cloning--A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989), especially Chapter 15). A whole series of sequences which code for members of IFN.alpha. family are known (Mantei et al., Gene 10:110 (1980); Streuli et al., Science 209:1343-1347 (1980); Goeddel et al., Nature 290:20-26 (1981)); the homology of the genes which code them is more than 70%. Other variants of these sequences can be found in nature or prepared from the known sequences by methods known in the art, e.g., by mutagenesis (see Sambrook et al., Molecular Cloning--A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989), especially Chapter 15).

As used herein the term "IFN.alpha." refers generally to any IFN.alpha. amino acid or nucleic acid sequence including, for example, known sequences and those variants whose genes are characterized by a high degree of homology with the known sequence and which code for biologically active IFN.alpha. and compounds having substantially the same biologically activity as known forms of IFN.alpha..

Herein the terms "process" and "method" are used interchangeably.

It is preferred in the invention that the sequence which codes for IFN.alpha.2, particularly IFN.alpha.2c (Dworkin-Rastl et al., Gene 21:237-248 (1983); Bodo & Maurer-Fogy, in: The Biology of the Interferon System 1985 (Stewart & Scliellekens), 59-64, Elsevier Scientific Publishing Co., Amsterdam (1985) be used. It is more preferred in the methods of the invention that the IFN.alpha.2 is IFN.alpha.2 having an amino acid sequence: ##STR1##

It is most preferred that in the constructs of the invention that the nucleotide sequence encoding IFN.alpha.2 encodes IFN.alpha.2 having the amino acid sequence of SEQ ID NO:5.

It is also preferred that in the methods and constructs of the invention the IFN.alpha.2 is IFN.alpha.2 encoded by a polynucleotide having the sequence: ##STR2##

It is more preferred that in the methods and constructs of the invention the IFN.alpha.2 is IFN.alpha.2 encoded by the nucleotide sequence of SEQ ID NO:6.

Skilled artisans will readily understand that IFN.alpha. species that are encoded by DNA sequences having homology to the sequences in SEQ ID NOS:6 and 7 will be useful IFN molecules since certain of these molecules will have IFN functions. It is a matter of routine practice for skilled artisans to mutate or select DNA sequences that are homologous to those in SEQ ID NOS:6 and 7. Techniques are known in the art for determining the degree of homology between DNA sequences (Beltz et al., Meth. Enzymol. 100:266-285 (1983)). Nucleic acid hybridization, such as filter hybridization is a common technique used to determine homology. Beltz et al., Meth. Enzymol. 100:266-285 (1983) teach washing conditions, probe length and guanidine/cytosine content, ionic strength of the wash and wash temperatures useful for filter hybridization homology determination. Moreover, skilled artisans can determine the level of DNA-DNA hybridization using methods known in the art (Owen et al., Chem. Meth. Bact. Systemat, pp. 67-93 (1985)). Methods for determining nucleic acid homology are also known in the art (Kafatos et al., Nucl. Acids Res. 7(6):1541-1552 (1979). Sequencing of DNA fragments followed by direct comparison of the homology between the fragments can also be performed.

Accordingly, it is further preferred that in the methods and constructs of the invention the IFN.alpha.2 is encoded by a nucleotide sequence that is at least about 70% homologous with the nucleotide sequence of SEQ ID NO:6, particularly a nucleotide sequence which codes for a protein having an IFN.alpha. activity.

It is also preferred that in the methods and constructs of the invention the IFN.alpha.2 is IFN.alpha.2 encoded by a polynucleotide having the sequence: ##STR3##

It is more preferred that in the methods and constructs of the invention the IFN.alpha.2 is IFN.alpha.2 encoded by the nucleotide sequence of SEQ ID NO:7.

It is further preferred that in the methods and constructs of the invention the IFN.alpha.2 is encoded by a nucleotide sequence that is at least about 70% homologous with the nucleotide acid of SEQ ID NO:7, particularly a nucleotide sequence which codes for a protein having an IFN.alpha. activity.

It is also preferred in the invention that the E. coil alkaline phosphatase (phoA) promotor is used for controlling expression and it is also preferred that the ribosome binding site of the STII gene be integrated into the gene expression constructs, plasmids and vectors of the invention. The sequence of the phoA-promotor is disclosed by Chang et al., Gene 44:121-125 (1986); Shuttleworth et al., Nucl. Acids Res. 14:8689 (1986) and that of the STII ribosome binding site is disclosed by Picken et al., Infection and Immunity 42:269-275 (1983); and Lee et al., Infection and Immunity 42:264-268 (1983)). The promotor and ribosome-binding site sequences are operably linked in the expression construct and are capable of mediating expression of the IFN gene of interest. The skilled artisan can easily produce equivalent variants from the sequences in the expression vector, particularly the promotor and ribosome-binding site sequences.

Construction of the vector, transformation of suitable E. coil strains, fermentation and extraction can be carried out using methods known in the art (Sambrook et al., Molecular Cloning--A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989)). For example, the E. coli strain W3110 (E. coli K12 Wild type f.sup.-, .lambda..sup.-, IN (rrnD-rrnE)1) is suitable for expression and is preferred in the constructs and methods of the invention. The preliminary bacterial culture can be grown in LB medium and the main culture can be made, with monitoring of the supply of oxygen and nutrients, up to an OD.sub.546 of 250 to 280.

Surprisingly, by linking the STII signal sequence to the IFN.alpha. gene, it was possible to establish a stable expression/secretion system, which had not been possible with the sak42D leader/IFN.alpha. combination known in the art. It is preferred that the constructs of the invention be expressed under the control of the phoA promotor. The integration of the ribosome binding site of the STII gene is a particularly preferred construct useful for expression. Expression can reliably be controlled by, for example, monitoring the phosphate concentration in the medium (phosphate deficiency, activates the phoA promotor); in the inactivated state there is no detectable basal expression. Additional chemicals do not need to be added for activation/induction; the expression rate in the activated state is high. The synthesized protein is secreted in large amounts into the periplasmic space. The secreted protein is correctly folded, contains the authentic N-terminus and the correct disulfide bridges. The SDS gel analysis of expression in E. coil W3110 showed that about 30-50% of the synthesized IFN.alpha. had been correctly processed; this corresponds to virtually all the secreted protein.

The initial extraction methods useful in the invention can be those known in the art (see, for example, Sambrook et al., Molecular Cloning--A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989)). However, an extraction method which is preferred in the invention involves suspending acid-inactivated biomass in dilute acetic acid with the aid of a homogenizer, adding polyethyleneamine, preferably in a concentration of about 0.25% (w/v), adjusting the mixture to an alkaline pH, preferably about pH 10, stirring the mixture and then removing the bacteria by centrifuging.

The invention provides a multistep chromatographic purification process. A preferred purification process having four chromatographic steps, namely performing adsorption chromatography on silica gel, hydrophobic interaction chromatography, cation and anion exchange chromatography, is provided in the invention. It is more preferred that the four chromatographic steps be temporally ordered thusly: performing adsorption chromatography on silica gel, followed by hydrophobic interaction chromatography, followed by cation and then anion exchange chromatography.

Type 953W gel made by Grace is preferred as the gel layer for the silica chromatography. It is also preferred that the flow rate of the column be about 25 ml/min, and that the column be equilibrated to a pH of about 7.5 with buffer, particularly Tris-HCl. A buffer, especially about 500-1500 mM tetramethylammoniumchloride (herein "TMAC"), and particularly about 800 mM TMAC, is preferred as an eluant.

For the hydrophobic interaction chromatography it is preferred that the gel bed comprises phenyl Sepharose.TM.. It is further preferred that samples be applied in the presence of about 20% ammonium sulphate and that the column be equilibrated with a buffer, such as Tris-HCl, containing about 30% ammonium sulphate. The IFN.alpha. is preferably eluted with a linear gradient having a final concentration of about 30% ethylene glycol.

Cation exchange chromatography may be used to purify the above-mentioned eluted material. It is preferred that the cation exchange chromatography be carried out using a sulphopropyl ion exchange resin, such as, for example Toyopearl.TM. TSK SP 5PW (Tosohaas). The pH of the eluate can adjusted by dialysis with Na-succinate, preferably at a concentration of about 20 mM (about pH 5.0), prior to loading onto the ion exchange resin. It is preferred that the samples be applied at a pH of 3 to 5, preferably pH 3 and the column can be equilibrated to a more preferred pH of 5. IFN.alpha. is preferably eluted with a linear common salt gradient by the addition of about 10% ethylene glycol.

The preferred gel bed used for anion exchange chromatography comprises DEAE-Sepharose.TM., such as DEAE Sepharose Fast Flow.TM. (Pharmacia). The sample application and elution is preferably carried out at above about pH 5.5 to 6.0, particularly at about pH 5.8. A linear common salt gradient with the addition of about 0.1% Tween 20 is preferred for elution. It is most preferred that the elution be carried out using about 10 mM bisTris, about 500 mM NaCl, about 0.1% Tween 20 at about pH 5.8. It is preferred that the elution flow rate be about 5 mi/min.

Any one of the chromatographic steps can be modified using techniques known in the an once the disclosed invention is understood by the skilled artisan (see, for example, Thatcher & Panayotatos, Meth. Enzymol. 119:166-177 (1986); EP-A 203,382). These modifications are within the spirit of the invention. For instance, it is within the technical capabilities of the skilled artisan to replace one or more gel materials with substantially equivalent materials or substantially functionally equivalent materials based on the same separation principles, without any inventive activity, and in this way to perform the process according to the invention.

Following purification, samples may be analyzed using reversed phase HPLC. Skilled artisans will readily be able to utilize reversed phase HPLC to analyze the composition and purity of the samples. It is most preferred that the samples be analyzed using a column comprising a bed with a particle size of about 5 .mu.m, particularly a BakerBond-WP-C18 column (250.times.4.5 mm, particle size 5 .mu.m). Tryptic peptides can be generated using methods known in the art and analyzed by reversed phase HPLC. It is preferred that the tryptic peptides be analyzed using a column comprising a bed with a particle size of about 4 .mu.m, such as, for example, a Merck Supersphere.TM. 120-4 C-18 column (125.times.4.5 mm, particle size 4 .mu.m). Standard solvents known in the art can be used for chromatography. It is preferred that the samples be chromatographed using solvent A (trifluoroacetic acid in water) and solvent B (trifluoroacetic acid in acetonitrile).

Further analysis of the purified IFN.alpha. samples can be performed by gel electrophoresis, especially an SDS polyacrylamide gel under standard reducing conditions, particularly a 16% gel. Samples may be reduced with any standard protein reducing agent, such as dithiothreitol, before electrophoresis.

The IFN.alpha. content of various samples obtained during purification may be quantitated by any technique known in the art for protein quantitation, particularly immunological techniques, and especially by ELISA. Using sandwich ELISA and the monoclonal antibodies OMG-2 and MG-7 (Adolf, G. R., Virology 175:410-417 (1990) the amount of IFN.alpha. in a sample can be conveniently quantitated.

All of the references cited herein are incorporated by reference herein in their entirety.

Having now generally described this invention the same will better be understood by reference to certain specific examples which are included for the purposes of illustration and are not intended to limit the invention unless otherwise specified.

EXAMPLES

Example 1

Preparation of the pDH13 Expression Vector and the Transformation of Bacterial Cells

A. General Methods

Restriction digestion of DNA with restriction endonucleases, filling-in reactions, phenol extraction and precipitation of DNA, agarose gel electrophoresis and elution of DNA from agarose gels, ligation of DNA molecules, transformation of bacteria and plasmid isolation from bacteria are standard procedures and were carried out as described by Sambrook et al., Molecular Cloning--A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1989).

B. Plasmids

pCF2

pCF2 was prepared from the plasmid pAT153 (Twigg & Slierratt, Nature 283:216-218 (1980)). It contains the promotor of alkaline phosphatase from E. coli (phoA) (Chang et al., Gene 44:121-125 (1986); Shuttleworth et al. , Nucl. Acids Res. 14:8689 (1986)), the coding region of the STII leader peptide (Picken et al., Infection and Immunity 42:269-275 (1983); Lee et al., Infection and Immunity 42:264-268 (1983)) and the gene for human IFN.omega.1 (Hauptmann et al., Nucl. Acids Res. 13:4739-4749 (1985)). FIG. 1 shows the gene map of pCF2 and the sequence of the relevant region comprising the PhoA and STII gene expression control sequences and the IFN sequence.

pER21/1

pER21/1 is a bacterial expression vector for IFN.alpha.2c (EPO 115,613).

C. Oligonucleotides

Table 4 depicts the oligonucleotides used in PCR reactions employed in the construction of the vectors in Example 1. The oligonucleotide sequences are depicted in the 5'.fwdarw.3' orientation. In the sections which follow the oligonucleotides are referred to by their "EBI" number (see Table 4).

D. Preparation of the Expression Cassette from the phoA Promotor, IFN.alpha.2c Sequence and STII Leader Sequence in a Two-Step PCR

pER21/1 (EPO 115,613) DNA was linearized with HindIII, pCF2-DNA with PvuI. The method used hereinafter is described as SOE-PCR ("splicing by overlap extension", Ho et al., Gene 77:51-59 (1989)).

PCR 1a (Amplification of the IFN.alpha.2c gene): 100 ng of linearized pER21/1 DNA, 25 pmol EBI-2797 and 25 pmol EBI-2798 were subjected to thermocycles in 50 .mu.l of buffer which contained 50 mM KCl, 10 mM Tris-HCl, pH 8.3, 1.5 mM MgCl.sub.2, 0.01% gelatine, 0.2 mM dATP, 0.2 mM dGTP, 02. mM dCTP, 0.2 mM dTTP and 1.25 units of Taq-polymerase, in a Perkin-Elmer Cetus Thermocycler TC-1. After 3 minutes' incubation at 94.degree. C., 10 cycles were performed (stage 1: 40 seconds at 94.degree. C., stage 2: 30 seconds at 55.degree. C., stage 3: 90 seconds at 72.degree. C.).

PCR 1b (Amplification of phoA-promotor plus STII leader sequence): 100 ng of linearized pCF2-DNA, 25 pmol EBI 2787 and 25 pmol EBI 2799 were subjected to thermocycles in the same buffer and under the same conditions as described under PCR 1a.

The resulting DNA fragments of PCR 1a (540 bp) and PCR 1b (374 bp) were gel-purified (1.2% low gelling type agarose in TBE buffer, 1.times.TBE: 10.8 g Tris/l, 5.5 g boric acid/1, 0.93 g EDTA/l). The agarose section containing the DNA fragment of PCR 1a was excised and the agarose was melted by adding 100 .mu.l of H.sub.2 O and heating to 70.degree. C.

PCR 2: 5 .mu.l of each agarose/DNA solution were combined and subjected to thermocycles in 100 .mu.l of solution containing 50 pmol of EBI-2787 and EBI-2797. The buffer was the same as described under PCR 1a. The thermocycle equipment was programmed so that a delay period of 5 minutes at 94.degree. C. was followed by 20 cycles (step 1: 40 seconds at 94.degree. C., step 2: 30 seconds at 55.degree. C., step 3: 5 minutes at 72.degree. C.; step 3 was extended by 5 seconds in each new cycle). After amplification the DNA was purified by phenol/chloroform extraction and ethanol precipitation. The PCR product was dissolved and cut with HindIII and EcoRI in the corresponding buffers.

1. Cloning of the PCR Product (pDH9)

Bluescribe M13.sup.+ (Stratagene, San Diego, Calif., U.S.A.) was doubly cut with HindIII and EcoRI and the large fragment was gel-purified with a 1.2% agarose gel. 10 ng of Bluescribe.TM. M13.sup.+ DNA and 50 ng of PCR product cut with EcoRI/Hind III were ligated in 10 .mu.l of solution containing 50 mM Tris-HCl, pH 7.6, 10 mM MgCl.sub.2, 20 mM dithiothreitol, 1 mM ATP, 50 mg/ml bovine serum albumin (BSA) and 2 units of T4-DNA-ligase (NEN), for 1 hour at 0.degree. C. and for 3 hours at ambient temperature. 8 .mu.l of this solution were used for the transformation of competent E. coli cells of the strain JM 101 (E. coli K12, SupE, thi, .DELTA.(lac-proAB), (F', traD36, proAB, lacIZ.DELTA.M15!).

A clone was selected, the DNA was isolated and the expression cassette sequenced. The sequence corresponded precisely to the sequence expected theoretically (FIG. 2). The plasmid was designated pDH9.

2. Construction of the Expression plasmid pDH13

pAT153 was doubly cut with SspI and PstI and the large fragment was isolated. pDH9 was cut with EcoRI and the ends were filled using the Klenow fragment of DNA polymerase 1 and the 4 dNTPs. After phenol extraction and precipitation of the linear pDH9-DNA, this DNA was cut with PstI and the fragment containing the phoA-promotor, the STII leader sequence and the IFN.alpha.2c gene was isolated from a 1% agarose gel.

10 ng of pAT153.times.SspI.times.PstI and 30 ng of the fragment containing the expression cassette were ligated in 10 .mu.l of solution for 5 hours at ambient temperature. 5 .mu.l of this mixture were used to transform competent E. coli bacteria of the strain HB101. The selection of the transformed bacteria was carried out on LB agar plates (10 g tryptone/l, 5 g yeast extract/l, 5 g NaCl/l, 15 g bacto-agar/l), containing 10 .mu.g/ml tetracycline. A gene map of pDH13 and the sequence of the relevant region is shown in FIGS. 2A-B.

Plasmid DNA from various colonies thus obtained was isolated and checked for correct composition by restriction analysis. A plasmid was selected and designated pDH13. The plasmid pDH13 was used for transforming E. coli W3110 (E. coli K12 . Wild type, f.sup.-, .gamma..sup.-, IN (rrnD-rrnE)1).

Example 2

Fermentation

A. Preliminary Culture

700 ml of autoclaved LB medium (10 g bacto-tryptone/l, 5 g bacto-yeast extract/l, 10 g NaCl/l, pH 7.0), containing 5 mg/l tetracycline, were inoculated in a 2 liter glass vessel from a stock culture so as to obtain an OD.sub.546 of 0.01. The culture was incubated for 10 hours at 37.degree. C. with vigorous stirring (800 rpm) and aeration (5 fermented volumes per minute (vvm)).

B. Main Culture

Composition of medium

In the fermenter:

1.21 g/l (NH.sub.4).sub.2 HPO.sub.4

3.96 g/l (NH.sub.4).sub.2 SO.sub.4

6.53 g/l K.sub.2 HPO.sub.4

1.23 g/l MgSO.sub.4 .times.7 H.sub.2 O

0.32 g/l NaCl

0.25 g/l NH.sub.4 Cl

1.0 g/l Na.sub.3 -citrate.times.2 H.sub.2 O

1.0 ml/l Trace element concentrate

12.5 g/l Glucose

20 mg/l Thiamine-HCl

50 mg/l L-tryptophan

100 mg/l L-leucine

50 mg/l L-methionine

5 mg/l Tetracycline

Trace element concentrate (amounts per 100 ml):

3.35 g FeCl.sub.3 .times.6 H.sub.2 O

1.09 g ZnSO.sub.4 .times.7 H.sub.2 O

0.267 g CoCl.sub.2 .times.6 H.sub.2 O

0.267 g Na.sub.2 MnO.sub.4 .times.2 H.sub.2 O

0.221 g CuSO.sub.4 .times.5 H.sub.2 O

0.333 g H.sub.3 BO.sub.3

1.37 g MnSO.sub.4 .times.H.sub.2 O

10 ml HCl conc. H.sub.2 O add 100 ml

Feeding during fermentation (amounts based on volume of fermenter):

350 g/l Glucose

3.70 g/l MgSO.sub.4 .times.7 H.sub.2 O

175 mg/l Thiamine-HCl

0.50 g/l L-tryptophan

4.0 g/l L-leucine

2.0 g/l L-methionine

Metered addition of antifoamers during fermentation (based on fermenter volume):

1.0 ml/l UCON LB625

Salts ((NH.sub.4).sub.2 PO.sub.4, (NH.sub.4).sub.2 SO.sub.4, K.sub.2 HPO.sub.4, NaCl, NH.sub.4 Cl and Na-citrate) were sterilized in a fermenter. Trace elements, MgSO.sub.4 glucose, thiamine, L-tryptophan, L-leucine, L-methionine and tetracycline were added aseptically after cooling so as to obtain a starting volume of 7 liters. 600 ml of the preliminary culture were automatically inoculated into the fermenter. The fermentation conditions were: stirring at 1000 rpm, aeration of 1 vvm, 0.3 bar backpressure, a temperature of 37.0.degree..+-.0.1.degree. C., the pH being maintained at 6.7.+-.0.1 using NH.sub.3 and H.sub.2 SO.sub.4. The concentration of dissolved oxygen was kept above 15% air saturation by aerating with oxygen-enriched air as necessary (at 0.3 bar backpressure). After the glucose initially present had been consumed, a feeding process was started up which was automatically triggered by the oxygen concentration and contained glucose, thiamine, MgSO.sub.4, L-tryptophan, L-leucine and L-methionine. The feeding rate started at 2.5 g/l/h and was increased continuously to 5.0 g/l/h within 24 hours and kept constant until the end of the fermentation process.

Fermentation was ended when a total quantity of 350 g/l of glucose had been added. At this time, a typical optical density of 250 to 280 was achieved at 546 nm.

To inactive the biomass the mixture was cooled to about 10.degree. C. and at the same time the pH was adjusted to 2.0 using H.sub.2 SO.sub.4. The biomass was separated off by centrifuging and stored frozen at -70.degree. C.

Example 3

Extraction of IFN

Acid-inactivated biomass (about 0.5 kg) was suspended in 500 ml of 1% acetic acid using a Polytron homogenizer and the mixture was stirred for 1 hour at 0.degree. C. Polyethyleneimine (50% stock solution, Serva, Heidelberg) was added to give a final concentration of 0.25% (w/v). The suspension was adjusted to a pH of 10.0 using 5N NaOH and stirred for a further 2 hours at 0.degree. C. After the pH had been adjusted to 7.5 using 5N HCl, the bacteria were separated off by centrifuging at 17,000.times.g (Beckman J2-21 centrifuge). The average extraction yield was 29.3.+-.5.9% of the total content of IFN.alpha.2c.

Example 4

Chromatographic Purification of IFN

A. Adsorption Chromatography on Silica Gel

The supernatant containing IFN.alpha., after separation of the bacterial pellet in Example 3, was loaded onto a silica gel column (Grace, silica type 953W; 35 mg protein/ml column material, flow rate 25 ml/min), which had been equilibrated with 20 mM Tris-HCl, pH 7.5. The column was washed with 30 column volumes of starting buffer, then a washing step with 20 mM Tris-HCl, 100 mM tetramethylammonium chloride (TMAC), pH 7.5, was carried out. IFN.alpha.2c could be eluted by increasing the TMAC concentration to 800 mM TMAC (FIG. 3A).

B. Hydrophobic Interaction Chromatography

The material eluted from the silica gel column was adjusted to an ammonium sulphate concentration of 20% (w/v) by the addition of solid (NH.sub.4).sub.2 SO.sub.4 and loaded onto a phenyl Sepharose.TM. column (phenyl toyopearl, 650S, Tosohaas) which had been equilibrated with 20 mM Tris-HCl, 30% ammonium sulphate. IFN.alpha.2c was eluted with a linear gradient from 100% loading conditions to 100% 20 mM Tris-HCl, 30% ethyleneglycol, pH 7.5, at a flow rate of 15 ml/min. The purity of the IFN.alpha. pool was 71.+-.15%.

C. Cation Exchange Chromatography

The eluate of the hydrophobic interaction chromatography was adjusted by extensive dialysis to 20 mM Na-succinate, pH 5.0. The final pH was adjusted to 3.0 with HCl, before the sample was loaded onto a sulphopropyl-ion exchange resin (Toyopearl.TM. TSK SP 5PW, Tosohaas), equilibrated with 20 mM Na-succinate, pH 5.0. IFN.alpha.2c was eluted from the column with a linear gradient from 100% loading conditions to 100% 20 mM Na-succinate, 500 mM NaCl, 10% ethyleneglycol, pH 5.5 (solvent B) at a flow rate of 6 ml/min. The IFN.alpha.2c eluted from this column routinely had a purity higher than 95%.

D. Anion Exchange Chromatography

The IFN.alpha. pool was dialyzed against 10 mM bisTris, pH 5.8, and loaded onto a DEAE Sepharose.TM. (DEAE Sepharose FastFlow, Pharmacia) which was equilibrated with the same buffer. The elution of IFN.alpha.2c was carried out with a linear gradient on 10 mM bisTris, 500 mM NaCl, 0.1% Tween.TM. 20 (polyoxyethylene sorbitas monolaurate), pH 5.8 (solvent B), flow rate 5 ml/min.

Example 5

Analysis of the IFN.alpha.2c Preparations

A. Reversed Phase HPLC

Intact IFN.alpha.2c was analyzed at 30.degree. C. with a BakerBond-WP-C18 column (250.times.4.5 mm, particle size 5 .mu.m). A Merck Supersphere.TM. 120-4 C-18 column (125.times.4.5 mm, particle size 4 .mu.m) was used at 37.degree. C. to separate tryptic peptides. The samples were chromatographed using solvent A, 0.1% trifluoroacetic acid in water, and B, 0.1% trifluoroacetic acid in acetonitrile and using the gradients as described in the relevant legend to the FIG. 3.

B. SDS-Polyacrylamide Gel Electrophoresis

IFN.alpha.2c samples were analyzed on 16% SDS polyacrylamide gels under standard reducing conditions. Samples were reduced with dithiothreitol before electrophoresis. Protein bands were visualized with Coomassie blue staining.

C. Quantifying IFN.alpha.2c by ELISA

The IFN.alpha.2c content of various samples obtained during purification was determined by sandwich ELISA with the monoclonal antibodies OMG-2 and MG-7 (Adolf, G. R., Virology 175:410-417 (1990)).

Results

Using the extraction process described in Example 3 it was possible to extract 29.3.+-.5.9% of all the IFN.alpha.2c detectable in the biomass. This corresponded to an observed processing level of about 30-50%. The extract from the biomass contained 4.5.+-.1.8% IFN.alpha.2c, based on a measurement of total protein. Silica adsorption chromatography led to an IFN.alpha.2c pool with an average purity of 16.7.+-.4.4%. Phenyl Sepharose.TM. chromatography with a yield of 93.2.+-.7.3% yielded an IFN.alpha.2c with a purity of 71.2.+-.15.5%. Sulphopropyl ion exchange chromatography produced a yield of 70.9.+-.14.8% and a purity of 97.6.+-.4.6%. Another step, namely DEAE ion exchange chromatography, resulted in 100% pure IFN.alpha.2c, in a yield of 86.9.+-.9.2%, as detailed hereinafter. The data from 6 different purifications are shown in Tables 5 (yields) and 6 (IFN.alpha.2c content). FIG. 3 shows characteristic chromatograms of each purification step.

From 1 kg of biomass, 340.+-.100 mg of purified IFN.alpha.2c were obtained. The yield of the purification process is 56.1.+-.22.2%. The total yield, based on the IFN.alpha.2c content of the biomass, is 14.4%. These data are shown in Table 7. FIG. 4 shows a typical SDS-PAGE of purified IFN.alpha.2c, eluting in the last chromatographic step. The 18 kDa band of IFN.alpha.2c is the only visible band. No contaminating bands are observed. FIG. 5A shows a typical reversed phase HPLC chromatogram. The purified IFN.alpha.2c elutes as a homogeneous peak at 24.8 minutes. When this material was eluted with a flat acetonitrile gradient (FIG. 5B), 2 contamination peaks were observed on either side of the main peak. These shoulders, which contain approximately 1.8% of the total IFN.alpha.2c content, represent forms which are oxidized at the methionine 111 (first shoulder) or acetylated at the N-terminus (second shoulder).

Claims

What is claimed is:

1. A method for preparing correctly folded and disulfide bond-linked interferon-.alpha. by expression in E. coli, comprising the steps of: (a) expressing interferon-.alpha. in E. coli transformed with a vector comprising an E. coli alkaline phosphatase (phoA) promoter operably linked to a nucleotide sequence encoding the signal peptide for the heat stable enterotoxin II (STII) of E. coli, said nucleotide sequence encoding the signal peptide being operably linked to a nucleotide sequence which codes for mature interferon-.alpha.; and (b) isolating the expressed interferon-.alpha..

2. The method of claim 1 wherein said vector further comprises the sequence for the ribosome binding site of the STII gene.

3. The method of claim 1 wherein said isolating step comprises the steps of: (a) performing chromatography on silica gel; (b) performing hydrophobic interaction chromatography; (c) performing cation exchange chromatography; and (d) performing anion exchange chromatography.

4. The method of claim 3, wherein said hydrophobic interaction chromatography employs phenyl Sepharose.TM..

5. The method of claim 3, wherein said cation exchange chromatography employs a sulphopropyl ion exchanger.

6. The method of claim 3, wherein said anion exchange chromatography employs DEAE-Sepharose.TM..

7. The method of claim 1, wherein said interferon-.alpha. is interferon-.alpha.2.

8. The method of claim 7, wherein said interferon-.alpha.2 comprises the sequence: ##STR4##

9. The method of claim 7, wherein said interferon-.alpha.2 is encoded by a nucleotide sequence comprising the sequence of SEQ ID NO:6 or a sequence encoding interferon-.alpha. which has more than about 70% sequence identity with this sequence.

10. The method of claim 7, wherein said interferon-.alpha.2 is encoded by a nucleotide sequence comprising the sequence of SEQ ID NO:7 or a sequence encoding interferon-.alpha. which has more than about 70% sequence identity with this sequence.

11. The method of claim 1, wherein 340.+-.100 mg of said interferon-.alpha. is obtained from 1 kg of E. coli.

12. A vector for expressing interferon-.alpha. in E. coli, comprising an E. coli alkaline phosphatase (phoA) promoter operably linked to a nucleotide sequence coding for the signal peptide of the heat stable enterotoxin II (STII) of E. coli, wherein said nucleotide sequence coding for the signal peptide is operably linked to a nucleotide sequence which codes for mature human interferon-.alpha..

13. The vector of claim 12, wherein said vector further comprises a ribosome binding site of the STII gene.

14. The vector of claim 12, wherein said interferon-.alpha. is interferon-.alpha.2.

15. The vector of claim 12, wherein said nucleotide sequence which codes for interferon-.alpha. comprises the sequence: ##STR5## or a sequence encoding interferon-.alpha. which has more than about 70% sequence identity with this sequence.

16. The vector of claim 12, wherein said nucleotide sequence which codes for interferon-.alpha. comprises the sequence: ##STR6## or a sequence encoding interferon-.alpha. which has more than about 70% sequence identity with this sequence.

Patent Citations (11)

Patent	Date	Inventor
US4769327	1988-09-01	Stephens et al.
US4828990	1989-05-01	Higashi et al.
US4917887	1990-04-01	Hauptmann et al.
US4963495	1990-10-01	Chang et al.
US5066786	1991-11-01	Protasi et al.
US5196323	1993-03-01	Bodo et al.
US5258287	1993-11-01	Baxter et al.
US5367060	1994-11-01	Vandlen et al.
US5455165	1995-10-01	Capon et al.
EPX115613	1984-08-01
WOXWO89/03225	1989-04-01

Non-Patent Literature (25)

Voss, T. et al., "Periplasmic expression of human interferon-.alpha.2c in Escherichia coli results in a correctly folded molecule," Biochem. J. 298(3):719-725 (15 Mar. 1994).
Adolf, G.R., Monoclonal Antibodies and Enzyme Immunoassays Specific for Human Interferon(INF) .omega.1: Evidence that IFN-.omega.1 is a Component of Human Leukocyte IFN, Virology 175:410-417 (1990).
Beltz et al., Isolation of Multigene Families and Determination of Homologies by Filter Hybridization Methods, Meth. Enzymol. 100:266-285 (1983).
Bodo & Maurer-Fogy, Molecular Species of Recombinant Human Interferon Alpha 2 Detected in Affinity Purified Preparations. In: The Biology of the Interferon System (ed. Stewart, W.E., II Schellekens, H.), Elsevier Scientific Publishing Co., Amsterdam, pp. 59-64 (1985).
Breitling et al., Secretory expression in Escherichia coli and Bacillus subtilis of human interferon .alpha. genes directed by staphylokinase signals, Mol. Gen. Genet. 217:384-391 (1989).
Chang et al., Nucleotide sequence of the alkaline phosphatase gene of Escherichia coli, Gene 44:121-125 (1986).
Derwent World Patent Index, English language abstract of European Patent 0 115 613.
Dworkin-Rastl et al., Construction of expression plasmids producing high levels of human leukocyte-type interferon in Escherichia coli, Gene 21:237-248 (1983).
Fuh et al., The Human Growth Hormone Receptor, J. Biol. Chem. 265(6):3111-3115 (1990).
Goeddel et al., Human leukocyte interferon produced by E. coli is biologically active, Nature 287:411-416 (1980).
Goeddel et al., The structure of eight distinct cloned human leukocyte interferon cDNAs, Nature 290:20-26 (1981).
Hauptmann & Swetly, A novel class of human type I interferons, Nucl. Acid Res. 13(13):4739-4749 (1985).
Ho et al., Site-directed mutagenesis by overlap extension using the polymerase chain reaction, Gene 77:51-59 (1989).
Kafatos et al., Determination of nucleic acid sequence homologies and relative concentrations by a dot hybridization procedure, Nucl. Acids Res. 7(6):1541-1552 (1979).
Lee et al., Characterization of Gene Encoding Heat-Stable Toxin II and Preliminary Molecular Epidemiological Studies of Enterotoxigenic Escherichia coli Heat-Stable Toxin II Producers, Infection and Immunity 42(1):264-268 (1983).
Mantei et al., The nucleotide sequence of clones human leukocyte interferon cDNA, Gene 10:1-10 (1980).
Owen & Pitcher, Current Methods for Estimating DNA Base Composition and Levels of DNA-DNA Hybridization, Chemical Methods in Bacterial Systematics, pp. 67-93 (1985).
Picken et al., Nucleotide Sequence of the Gene for Heat-Stable Enterotoxin II of Escherichia coli, Infection and Immunity 42(1):269-275 (1983).
Shuttleworth et al., Sequence of the gene for alkaline phosphatase from Escherichia coli JM83, Nucl. Acid Res. 14(2):8689 (1986).
Streuli et al., At Least Three Human Type .alpha. Interferons: Structure of .alpha.2, Science 209:1343-1347 (1980).
Thatcher & Panayotatos, Purification of Recombinant Human IFN-.alpha.2, Methods Enzymol. 119:166-177 (1986).
Twigg and Sherratt, Trans-complementable copy-number mutants of plasmid ColE1, Nature 283:216-218 (1980).
Miyake, T., et al. (1985) J. Biochem. (Tokyo) 92: 1429-36.
Fuh, G., et al. (1990) J. Biol. Chem. 265: 3111-15.
Morioka-Fujimoto, K., et al. (1991) J. Biol. Chem. 266: 1728-32.

Abstract

Description