Gene encoding a novel protein H capable of binding to IgG

Abstract

A gene coding for Protein H, which is capable of binding specifically to human IgG of all subclasses, was isolated from Streptococcus sp. AP1 and expressed in host cells, E. coli to produce the Protein H.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a novel protein binding specifically to the Fc fragment of human immunoglobulin G(IgG), a gene coding for the said protein and to a process for producing the said protein.

2. Description of Related Art

It has been known that certain microorganisms produce a series of proteins known as bacterial Fc receptors which have affinity to the Fc fragment of immunoglobulin (Boyle et al., Bio/Technology 5, 697(1987)).

Typical examples of such proteins are Protein A derived from Staphylococcus aureus and Protein G derived from Streptococcus G148.

These Protein characteristically bind to the Fc fragment of immunoglobulin and are used for assay, purification and preparation of antibodies as well as clinical diagnosis and biological research.

They can also be used for treatment of cancers and autoimmune diseases in which the proteins immobilized onto insoluble carriers are used to adsorb or remove undesirable immune complexes from blood(Cancer 46, 675(1980)).

These known proteins have some undesirable properties as agents for purification of human monoclonal antibodies produced from non-human animal cells or for removal of excessive IgG from blood for the purposes of blood purification.

Protein A binds to IgGs of various animal species including human beings as well as human IgA, IgM and so on. Protein G binds only to IgG but it binds both to human IgG and to IgGs of other animal species (Fahnestock, Trends in Biotechnology 5, 79 (1987)). Thus, their binding specificities are not so narrow that they can be used for assay, purification and adsorption or removal of human IgG.

Under the circumstances, development of a protein capable of binding specifically to human IgG has been demanded.

It was suspected that a protein which binds to human IgG(IgG1, IgG2, IgG3 and IgG4) but not to IgGs of other animal species would be present in cells of group A Streptococcus strains (Bjorck, J. Immunol., 133, 969(1984)). However, no such protein has been isolated. Two types of IgG-Fc-binding proteins have been isolated from group A Streptococcus, one of which binds to human IgG(IgG1, IgG2 and IgG4), pig IgG and rabbit IgG; and the other binds specifically to human IgG3 (Boyle et al., Bio/Technology 5, 697(1987)).

It has been unknown whether or not such IgG-Fc-binding protein which binds specifically to human IgG (IgG1, IgG2, IgG3 and IgG4) but which does not bind to IgGs of most other animal species and to human IgA, IgD, IgE and IgM exists and whether or not a sufficient amount of such protein can stably be obtained.

The present inventors have extensively studied seeking for proteins which specifically binds to human IgG, and, as a result, found that Streptococcus sp. AP1 belonging to group A Streptococcus produces the protein with the above-mentioned properties. They have also obtained the gene coding for the said protein and provided a process for producing the said protein by the utilization of the gene through further extensive studies.

Thus, the present invention provides a novel protein capable of specifically binding to human IgG, and useful for assay and purification of human IgG, removal or adsorption of excessive IgG from blood and for diagnosis of autoimmune diseases. It also provide a gene coding for the said protein and a process for producing the said protcin in industrial scales.

The protein provided by the present invention, which is hereinafter referred to as Protein H, is a protein capable of binding to the Fc fragment of immunoglobulin and produced by a strain of group A Streptococcus and has the following binding specificity:

i) It binds to human IgG(IgG1, IgG2, IgG3 and IgG4) and rabbit IgG; ii) It does no& bind to IgGs of mouse, rat, bovine, sheep and goat; iii) It does not bind to human IgA, IgD, IgE and IgM; or the following binding specificity: i) It strongly binds to human IgG(IgG1, IgG2, IgG3 and IgG4), human IgGFc and rabbit IgG; ii) It weakly binds to pig IgG; iii) It does not bind to IgGs of mouse, rat, bovine, sheep, goat and horse; iv) It does not bind to human IgGFab, IgA, IgD, IgE and IgM.

The strain, Streptococcus sp. A which produces the Protein H has been deposited at the Fermentation Research Institute, Japan, under the deposit No. FERM P-10374 and also under the deposit No. FERM BP-2371 according to the BUDAPEST TREATY ON THE INTERNATIONAL RECOGNITION OF THE DEPOSIT OF MICROORGANISMS FOR THE PURPOSES OF PATENT PROCEDURE.

The Protein H can be produced by the genetic engineering technology using the gene coding for the Protein H.

The Protein H produced from the gene isolated from the Streptococcus sp. APl has been found. to be of the amino acid sequence given in the FIG. 1(A-B) by the analysis of the DNA sequences of the gene.

It is to be understood that subfragments o variants of the Protein H specifically disclosed in the present application wherein original amino acid sequence is modified or changed by insertion, addition, substitution, inversion or deletion of one or more amino acids are within the scope of the present invention as far as they retain the essential binding specificity as mentioned above.

The gene coding for the Protein H can be isolated from the chromosomal DNA of a Protein H-producing stain such as Streptococcus sp. APl based on the information on the DNA sequence of the Protein H shown in FIG. 4(A-B). The isolation of the gene can also be carried out as follows:

The chromosomal DNA can be isolated from cells of the Protein H-producing strain in accordance with a known method (Fahnestock, J. Bacteriol. 167, 870(1986)). The isolated chromosomal DNA is then segmented into fragments of adequate lengths by biochemical means such as digestion with restriction enzyme or physical means.

The resulting fragments are then inserted at an adequate restriction site into an adequate cloning vector such as .lambda.gtll (Young et al., Proc. Natl. Acad. Sci. USA 80, 1194 (1983)) or plasmid vectors such as pUC18 (Messing et al., Gene 33, 103 (1985)).

The vectors are then incorporated into adequate host cells such as E. coli.

From the resulting transformants, the clones producing the protein which binds to human IgG or the Fc region of human IgG are selected by a known method (Fahnestock et al., J. Bacteriol., 167, 870 (1986)).

After the proteins capable of binding to human IgG or the Fc region of human IgG are isolated from the resulting positive clones according to a conventional method, the binding specificities of the proteins are determined to select the clones producing the Protein H. FIG. 2 shows the binding specificity of the Protein H.

After the DNA insert of the said clone is isolated by a conventional method, the DNA sequence of the insert is determined by a known method (Sanger et al., Proc. Natl. Acad. Sci. USA 74, 5463 (1977); Choen et al., DNA 4, 165 (1985)). FIG. 4(A-B) shows the DNA sequence of the DNA insert isolated from the positive clone Fc4. FIG. 5(A-B) shows the DNA sequence of the structural gene coding for the Protein H isolated from the said clone.

It is necessary for genes to be expressed that they contain expression-controlling regions such as promoter, terminator and the like. The gene shown in FIG. 4 contains such as necessary expression-controlling regions.

The expression of genes may be effected with expression vectors having the necessary expression controlling regions in which only the structural gene is inserted. For this purpose, the structural gene shown in FIG. 5(A-B) can preferably be used. The structural gene coding for the Protein H can be obtained from the gene of FIG. 4(A-B) or synthesized by a conventional method based on the amino acid sequence given in the present specification.

As for the expression vectors, various host-vector systems have already been developed, from which the most suitable host-vector system can be selected for the expression of the gene of the present invention.

It has been known that, for each host cell, there is a particularly preferable codon usage for the expression of a given gene. In constructing a gene to be used for a given host-vector system, the codons preferable for the host should be used. Adequate sequences for the gene for the Protein H to be used in a particular host-vector system can be designed based on the amino acid sequence given in FIG. 1(A-B) and synthesized by a conventional synthetic method.

The present invention is further concerned with a process for producing the Protein H by culturing a host cell transformed with an expression vector in which the gene encoding the Protein H is inserted.

The process comprises steps of

i) inserting a gene coding for the Protein H into a vector; ii) introducing the resulting vector into a suitable host cell; iii) culturing the resulting transformant cell to produce the Protein H; and iv) recovering the Protein H from the culture.

In the first step, the gene coding for the Protein H, which is isolated from the chromosomal DNA of Streptococcus sp. API or synthesized as mentioned above, is inserted into a vector suitable for a host to be used for the expression of the Protein H. The insertion of the gene can be carried out by digesting the vector with a suitable restriction enzyme and linking thereto the gene by a conventional method.

In the second step, the resulting vector with the gene is introduced into host cells. The host cells may be Escherichia coli, Bacillus subtilis or Saccharomyces cerevisiae and the like. The introduction of the expression vector into the host cells can be effected in a conventional way.

In the third step, the resulting transformant cells are cultured in a suitable medium to produce the Protein H by the expression of the gene. The cultivation can be conducted in a conventional manner.

In the fourth step, the produced Protein H is recovered from the culture and purified, which can be conducted by a known method. For example, the cells are destroyed by a known method such as ultrasonification, enzyme treatment or grinding. The Protein H released out of the cells or secreted into the medium is recovered and purified by conventional methods usually used in the field of biochemistry such as ion-exchange chromatography, gel filtration, affinity chromatography using IgG as ligand, hydrophobic chromatography and reversed phase chromatography, which may be used solely or in suitable combinations.

As mentioned above, the protein provided by the present invention can be used for identification or separation of human IgG. For these purposes, the protein may be brought into a reagent kit or a pharmaceutical composition by mixing or combining it with suitable reagents, additives or carries.

The present invention will more precisely be described by the following examples. But, they are not intended to limit the scope of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

In the attached drawings, FIG. 1(A-B) represents the amino acid sequence of the Protein H. FIG. 2 gives autoradiograms showing binding specificities of Protein G and Protein H to various antibodies. FIG. 3 is a diagram schematically illustrative of the DNA inserts of clones Fc4 and Fc16. FIG. 4(A-B) gives the DNA sequence of the DNA insert in clone Fc4. FIG. 5(A-B) gives the DNA sequence of the structural gene of the Protein H. FIG. 6 illustrates a plasmid pPH-1 and its deletion plasmids used for determination of DNA sequence. FIG. 7 is a graph showing the binding activity of Streptococcus sp. APl to human IgG, IgGFc and mouse IgG. FIG. 8 gives autoradiograms showing binding specificities of Protein H purified from the periplasmic fraction of E. coli JM109 (pPH-1) and Protein G to various antibodies.

DETAILED DESCRIPTION OF THE INVENTION

Example 1 IgG-binding activity of Streptococcus sp. APl

To an Eppendorf tube was added 20 .mu.l of dichloromethane solution of IODO-GEN.TM. (1,3,4,6-tetrachloro-3.alpha.,6.alpha.-diphenyl glycouril; Pierce and Warriner Ltd.; 0.1 mg/ml), which was dried by blowing nitrogen gas in the tube, while the tube was inclined and rotated. To the resultant, 200 .mu.l of a buffer solution A (50mM Na-phosphate buffer pH 7.5, 0.01% Pluronic F-68 (BASF Corp.)) is added. After the mixture was allowed to stand in an ice bath for 10 minutes, the buffer was removed. To the tube were added 10 .mu.l of 0.5M Na-phosphate buffer; (pH 7.5) and 25 .mu.l of IgG solution human IgG 5 .mu.g, mouse IgG 5 mg human IgGFc 3.34 .mu.g; Cappel Laboratories), followed by 2 .mu.l of Na.sup.125 I solution (IMS 30, Carrier-free, 100 mCi/ml; Amersham Corp.). The mixture was allowed to stand for 15 minutes, while the ixture was ice-cooled and softly shaken. The reaction product was transferred to a serum tube containing 200 .mu.l of a buffer solution (10mM Na-phosphate buffer pH 7.2, 150mM NaCl) after the serum tube had been treated with the buffer solution A (5 ml) in the same way as above-mentioned. The mixture was allowed to stand in an ice bath for 5 minutes. The resulting solution was applied to a PD-10 column (Pharmacia Fine Chemicals) equilibrated with buffer solution B (30mM Na-phosphate buffer pH 7.3, 120mM NaCl, 0.1% BSA) and eluted with buffer solution B. From each fraction (0.5 ml), 2 .mu.l was sampled and measured with .gamma.-counter (Ria Gamma "QUATRO"; LKB Corp.) to recover the .sup.125 I-labeled IgG.

Thus, .sup.125 I-human IgG 2.24.times.10.sup.7 cpm/.mu.g (1.12 .times.10.sup.8 cpm/ml), .sup.125 I-human IgGFc 8.98.times.10.sup.7 cpm/.mu.g (3.times.10.sup.8 cpm/ml) and 125I-mouse IgG 2.42.times.10.sup.7 cpm/.mu.g (1.21.times.10.sup.8 cpm/ml) were obtained.

A loopful of cells of Streptococcus sp. APl was inoculated to 5 ml of Todd-Hewitt culture medium (Difco Laboratories) and incubated at 37.degree. C. for 10 hours. Of the culture solution, a 2 ml-portion was added to 100 ml of Todd-Hewitt medium, incubated at 37.degree. C. for 13 hours, and centrifuged to harvest the cells.

The cells were washed with 100 ml of buffer solution C (30 mM Na-phosphate buffer pH 7.2, 120mM NaCl, 0.05 % Tween 20, 0.02% NaN.sub.3) and diluted with buffer solution C to give suspensions of different cell concentrations between 10.sup.7 to 10.sup.10 cfu/ml. In a serum-tube were added a 200 .mu.l-portion of each suspension, followed by .sup.125 labeled IgG (human IgG 10 ng human IgGFc 5.2 ng, mouse TgG 10 ng), and the mixture was stirred and allowed to stand at 37.degree. C. still for 2 hours. After the reaction has completed, 2 ml of buffer solution C was added and centrifuged to harvest the cells. After similar retreatment of the cells with 2 ml of buffer solution C, the amount of .sup.125 I-labeled IgG bound to cells was measured with a .gamma.-counter.

As the results in FIG. 7 show, Streptococcus sp. APl cell has proved to bind to human IgG and IgGFc but not to mouse IgG.

Example 2

Preparation of chromosomal DNA of Streptococcus sp. APl

A loopful of cells of Streptococcus sp. APl was inoculated to 10 ml of Todd-Hewitt culture medium and cultivated at 37.degree. C. for 13 hours. Of the culture, a 8 ml-portion was added to 400 ml of Todd-Hewitt medium, and cultivated at 37.degree. C. for 3 hours (A.sub.660 =0.6). After 22 ml of 10% cysteine and 26 ml of 0.4M DL-Threonine were added, the culture was again incubated for one hour. Then 250 ml of 15% glycine was added and cultivation was continued additionally for 2 hours. Cells were harvested by centrifugation and washed with 0.2M sodium acetate. The washed cells were suspended in 40 ml of buffer solution D (0.15M NaCl, 0.015M Na.sub.3 -citrate pH 7.4-7.6) containing 27% sucrose and 10 mM EDTA. To the suspension, 2500 units of Mutanolysin (Sigma Chemical Co.) was added and incubated at 37.degree. C. for 3 hours. To the reaction mixture, 4 ml of 10% SDS and proteinase K (0.2 mg/ml) were added and incubated overnight at room temperature. After the extractions with phenol followed by ether, twice volume of cold ethanol were added to the removed water phase and the separated DNA was recovered by winding it around a glass rod.

The recovered DNA was dissolved in 5 ml of buffer solution D and incubated with RNase A (100 .mu.g/ml) at 37.degree. C. for 1 hour. Then phenol extraction and ethanol precipitation were carried out to recover DNA.

Yield of chromosomal DNA amounted to about 1 mg.

Example 3

Cloning of Protein H gene

The chromosomal DNA (about 100 .mu.g) obtained in Example 2 was dissolved in 200 .mu.l of a buffer solution comprising 10 mM Tris.HCl (pH 7.5) and 1 mM EDTA and passed through a meedle (27G) for use in injection to shear the DNA fragments of 2 to 10 kb. About 10 .mu.g of the obtained DNA fragments was added to a solution comprising 40 mM Tris.HCl (pH 7.5), 5 mM DTT, 10 mM MgCl.sub.2, 0.1 mg/ml BSA, 50 .mu.M dNTP (dATP, dTTP, dGTP, dCTP) and 10 units of T4DNA polymerase and allowed to react at 24.degree. C. for 2 hours to make them blunt-ended. Then phenol extraction and ethanol precipitation were carried out, and the thus-collected blunt ended DNA fragments were added to 50 .mu.l of a solution comprising 100 mM Tris.HCl (pH 8.0), 100 mM NaCl, 1 mM EDTA, 80 .mu.M S-adenosylmethionine and 200 units of EcoRI methylase and allowed to react at 37.degree. C. for 20 minutes to make them methylated. Then phenol extraction and ethanol precipitation were carried out, and the thus-collected methylated DNA fragments were reacted at 16.degree. C. for 12 hours with a commercially-available EcoRI linker which had already been phospholylated by &he use of a commercially available ligation kit (Takara Shuzo Co., Ltd., Japan). The resulting reaction product was added to a solution comprising buffer E (10 mM Tris HCl (pH 7.5), 100 mM NaCl, 10 mM MgCl.sub.2, 1 mM DTT, and 200 units of EcoRI, and allowed to react at 37.degree. C. for 12 hours.

After termination of this reaction, phenol extraction and isopropanol precipitation were carried out to collect the DNA.

The thus-obtained DNA (about 0.5 .mu.g) was reacted at 16.degree. C. for 16 hours with 1 .mu.g of .lambda.gtll DNA (Protoclone.TM. .lambda.gtll system: Promega Biotech Corp.) in 13 .mu.l of a solution comprising buffer F (66 mM Tris.HCl pH 7.6, 6.6 mM MgCl.sub.2, 10 mM DTT, 0.1 mM ATP) and 400 units of T4 DNA ligase.

The ligated DNAs were packaged into phage using in vitro packaging kit (Gigapack Gold; Stratagene Corp.) and used as gene library of Streptococcus sp. APl. The packaging efficiency, as measured with E. coli Y1090, was 3.2.times.10.sup.6 pfu/.mu.g .lambda.gtll DNA.

E. coli Y1090 was cultivated in LBM medium {LB medium (Bacto tryptone 1%, Yeast extract 0.5%, NaCl 0.5%; pH 7.2), 10 mM MgSO.sub.4, 0.2% maltose, 50 .mu.g/ml Ampicillin} to grow up to A.sub.660 =0.6. From this culture, 0.2 ml was collected and centrifuged to harvest cells.

The cells were suspended in 0.2 ml of buffer solution G (10 mM Tris.HCl pH 7.4, 10 mM MgSO.sub.4, 0.01% gelatin), mixed with 100 .mu.l of buffer solution G and 7.6 .mu.l of the gene library phage particles solution (5.times.10.sup.4 pfu), and incubated at 37.degree. C. for 20 minutes. To this reaction mixture, 7 ml of soft agar solution (LBM medium, 0.4% soft agar, 47.degree. C.) was added, stirred, and overlaid onto a LBM plate (diameter 150 mm). After 3 hours' incubation at 42.degree. C., the plate was covered with nitrocellulose filter (BA 85, diameter 142 mm; Schleicher and Shuell AG) which had been immersed in 20 mM IPTG solution for 5 minutes and dried, and incubated a 37.degree. C. for 16 hours.

After the cultivation has been completed, nitrocellulose filter was taken off. Then procedure was proceeded at room temperature with slow shaking as follows:

The nitrocellulose filter was treated in 50 ml of buffer solution H (10 mM Veronal buffer pH 7.4, 0.15 M NaCl) for 5 minutes, and incubated for 1 hour in 50 ml of buffer solution H containing 0.25% gelatin and 0.25% Tween 20. After 3 hour incubation with human IgGFc fragment (2 .mu.g/ml) (CAPPEL Corp.) in 40 ml of buffer solution H containing 0.1% gelatin, the filter was washed three times with 40 ml each of buffer solution H containing 0.1% gelatin for 10 minutes. Again 1 hour incubation with goat anti-human IgGFc (Peroxidase conjugate, affinity purified; Jackson Immunoresearch Laboratories Corp.; diluted 1,000 fold with buffer solution H containing 0.1 % gelatin) in 40 ml of buffer solution H containing 0.1 % gelatin was carried out, and the filter was washed with 40 ml of buffer solution H containing 0.1 % gelatin three times each for 10 minutes. followed by 40 ml of buffer solution K (20 mM Tris.HCl pH 7.5, 0.5 M NaCl) once for 5 minutes. This filter was immersed in a color-developing solution (20 mg 4-chloro-1-naphthol/6.6 ml methanol, 20 .mu.l H.sub.2 O.sub.2 /33.4 ml buffer solution K) for 30 minutes. Out of 70,000 plaques, 17 purple-plaques were collected and suspended in 500 .mu.l of buffer solution G. To the suspension, 10 .mu.l of chroloform was added and allowed to stand for 40 minutes, and then centrifuged (10,000 rpm, 1 minute). The same procedure with the resulting supernatant was repeated to give siable clones Fc4 and Fc16.

The IgGFc-binding proteins produced by clone Fc4 and Fc16 had the same apparent molecular weight of 45 kDa as judged by the Western-blotting method.

Phage solutions of Fc4 and Fc16 (400 .mu.l, about 10.sup.10 pfu/ml) each was mixed with culture of E. coli Y1090 (400 .mu.l). to which soft agar solution (8 ml) was added. The resulting mixture was laid over 8 sheets of LB plates. after incubation at 37.degree. C. for 16 hours, 15 ml of buffer solution M (50 mM Tris.HCl pH 7.5, 100 mM NaCl, 8.1 mM MgSO.sub.4, 0.01% gelatin) was added every plate and shaken at 4.degree. C. for 3 hours.

After the buffer solutions were collected, the plates were washed with buffer solution (2 ml/plate), which was then combined with the collected buffer solution. To the combined buffer solution, 2 ml of chloroform was added and the mixture was stirred and centrifuged at 7,000 rpm for 15 minutes.

The supernatant was again centrifuged (17,000 rpm; 3 hours). The recovered precipitate was suspended in 0.5 ml of buffer solution M, to which CsCl was added to a concentration of 0.5 mg/ml. The suspension was then centrifuged (22,000 rpm; 2 hours, 4.degree. C.) to recover phage particles. The phase particle suspension was dialyzed against a buffer solution comprising 50 mM Tris.HCl, pH 8.0, 10 mM NaCl, and 10 mM MgCl.sub.2. To the dialyzed solution, EDTA (final concentration 20 mM), SDS (final concentration 0.5%) and proteinase K (final concentration 50 .mu.g/ml) were added, and the mixture was incubated at 65.degree. C. for 1 hour, and extracted with phenol followed by chloroform.

The aqueous phase was dialyzed against a buffer solution comprising 10 mM Tris.HCl, pH 8.0, and 1 mM EDTA, and precipitated with ethanol to give DNA.

The obtained phage DNA (200 .mu.g) was dissolved in 200 .mu.l of a buffer solution comprising 10 mM Tris.HCl, pH 7.5, and 1 mM EDTA, and the restriction enzyme cleavage pattern was analyzed to find that clones Fc4 and Fc16 had respective DNA inserts as shown in FIG. 3.

E. coli Y1089 was lysogenized with phage clones Fc4 and Fc16 according to the methods described by Young et al. (Proc. Natl. Acad. Sci. USA 80, 1194 (1983)).

Example 4

Binding specificity of Protein H

E. coli Y1089 lysogenized with phage Fc4 was inoculated to 40 ml of the aforesaid LBM medium and incubated at 28.degree. C. for 16 hours. The seed culture was added to 2 liters of LBM medium and incubated at 28.degree. C. for 145 minutes.

To the culture, IPTG was added to a final concentration of 1 mM and incubated at 42.degree. C. for 45 minutes, and at 37.degree. C. for additional 1 hour. The cells were harvested by centrifugation and suspended in 100 ml of a buffer solution comprising 50 mM phosphate buffer (pH 7.2), 5 mM EDTA, 5 mM benzamidine.HCl and 5 mM iodoacetamide and subjected to 10 minute ultrasonification. The mixture was centrifugated at a low speed to remove cell debris, and at 50,000 rpm for 30 minutes. The supernatant was applied to a IgG-Sepharose (6 Fast Flow; Pharmacia.) column (10 ml) which had been successively washed with 400 ml of buffer solution N (50 mM Tris.HCl pH 7.6, 150 mM NaCl, 0.05 % Tween 20), 2.5 M NaI (pH 7.2) and buffer solution N, and equilibrated with buffer solution N. After washing the column with 300 ml of buffer solution N, elution for recovering protein H was carried out with 40 ml of 2.5 M NaI (pH 7.2).

Fractions of 0.5 ml each were collected and a small amount of sample collected from each fraction was spotted on a nitrocellulose filter. Then detection of Protein H containing fractions was carried out according to the staining method described in Example 3.

The Protein H-containing fractions were combined and dialyzed once against 1 liter of buffer solution comprising 50 mM phosphate buffer (pH 7.2), 0.15 M NaCl and 0.25% NaI and twice against 5 liters each of buffer solution comprising 50 mM phosphate buffer (pH 7.2) and 0.15 M NaCl, and then concentrated to about 1 ml with Amicon YM-5 (Amicon Corp.). The concentrated solution was applied to a gel filtration column for HPLC (diameter 7.5 mm.times.6 cm, TSK gel G-3000 SW (Toyo Soda Co., Ltd.)) equilibrated with a buffer solution comprising 50 mM phosphate buffer (pH 7.5) and 0.2 M NaCl, and fluted with the same buffer solution at a flow rate of 0.4 ml/min. The fractions collected between the 34th to 36th minute of elution, which contains Protein H, were combined and concentrated with Amicon YM-5. The concentrated solution was applied to a reversed phase HPLC column (diameter 4.6 mm.times.7.5 cm, TSK gel Phenyl-5PW RP (Toyo Soda Co., Ltd.)) equilibrated with a buffer solution comprising 0.1 M glycine/NaOH (pH 10.0) and 1 mM tetra-n-butyl ammonium hydroxide, and the Protein H was eluted with a linear gradient (0%.fwdarw.66%, 2%/min) of acetonitrile. Fractions collected near the 16th minute of elution, which contains Protein H, were combined and concentrated to about 2 ml under reduced pressure. The concentrated solution was applied to a PD-10 column (Pharmacia Corp.) equilibrated with water, and eluted with water to remove salts. Of the obtained protein, yield amounts to about 53 .mu.g and molecular weight was about 45 kDa as measured by the Western-blotting technique using the staining method described in Example 3.

About 10 .mu.g of Protein G (Genex Corp.) and about 10 .mu.g of the aforesaid Protein H were labeled with Na.sup.125 I according to the method described in Example 1 to give 1.28.times.10.sup.7 cpm/.mu.g (8.5.times.10.sup.7 cpm/ml) of .sup.125 I-Protein G and 1.68.times.10.sup.7 cpm/.mu.g (1.4.times.10.sup.8 cpm/ml) of .sup.125 I-Protein H.

Human IgG1, IgG2, IgG3 and IgG4 (all, Protogen Corp.); human IgM, IgG and serum IgA, and IgGs of sheep. rabbit, bovine, and goat (all, Cappel Corp.); human IgD and IgE (all, Serotec Corp.); rat IgG (Jackson Immunoresearch Corp.); mouse IgG (Zymed Corp.) and human monoclonal IgG were dissolved in buffer solution K, and diluted with buffer solution K to concentrations of 0.08 to 10 .mu.g/200 .mu.l. Each diluted solution (200 .mu.l) was applied to nitrocellulose filter (Schleicher and Schuell Corp.) and adsorbed on the filter with BIO-DOT (BioRad Laboratories). The filter was incubated in 40 ml of a buffer solution comprising buffer K, 0.25 % gelatin and 0.25 % Tween-20 at 42.degree. C. for 1.5 hours and washed twice in buffer solution K containing 0.1 % gelatin at room temperature for 15 minutes. The washed filter was further incubated in 40 ml of a solution comprising buffer solution K, 0.1% gelatin and 0.5 .mu.g (1.6.times. 10.sup.5 cpm/ml) of .sup.125 I-Protein G or 0.5 .mu.g (2.1.times.10.sup.5 cpm/ml) of .sup.125 I-Protein H at room temperature for 3 hours.

The filter was incubated 4 times with 40 ml of a solution comprising buffer solution K, 0.25% gelatin, 0.25% Tween-20 and 0.85 M NaCl at room temperature for 15 minutes for washing. After drying the filter, antibody-binding properties of Protein G and Protein H were analyzed by autoradiography.

The autoradiograms shown in FIG. 2 demonstrated that Protein H having the specificity of

i) binding to human IgG (IgG1, IgG2, IgG3 and IgG4) and rabbit IgG, and

ii) not binding to mouse, rat, bovine, sheep and goat IgG's and human IgA, IgD, IgE and IgM.

Example 5

The nucleotide sequence of Protein H gene

The phage DNA (about 10 .mu.g) of clone Fc4 obtained in Example 3 was incubated in 100 .mu.l of a solution comprising buffer solution P (10 mM Tris.HCl, pH 7.5, 10 mM MgCl.sub.2, 1 mM DTT), 30 units of Sac I and 42 units of KpnI at 37.degree. C. for 5 hours. After termination of the reaction, phenol extraction and ethanol precipitation were carried out to recover phage DNA. On the other hand, plasmid pUC 18 (about 8 .mu.g) was incubated in 30 .mu.l of a solution comprising buffer solution P, 20 units of SacI and 14 units of KpnI at 37.degree. C. for 10 hours, Subsequently phenol extraction and ethanol precipitation were conducted to recover DNA. The recovered DNA was dissolved in 50 .mu.l of 1 M Tris HCl (pH 8.0) and incubated with 0.36 units of Bacterial Alkaline phosphatase (BAP) at 65.degree. C. for 30 minutes. After 0.36 units of BAP was added, the reaction mixture was again incubated at 65.degree. C. for further 30 minutes. Subsequently phenol extraction and ethanol precipitation were carried out to recover DNA.

The BAP-treated pUC18 (0.5 .mu.g) and the phage DNA digested with SacI and KpnI (0.1 .mu.g) were incubated in 30 .mu.l of a solution comprising buffer E and 10 units of T4 DNA ligase at 16.degree. C. for 16 hours. With this reaction mixture, E. coli JM109 cells were transformed and amplicillin-resistant transformants were obtained, from which plasmid DNAs were prepared. By analysis of restriction enzyme cleavage pattern, a transformant containing a plasmid pPH-1 as shown in FIG. 6 was selected.

Plasmid pPH-1 (about 10 .mu.g) was incubated in 25 .mu.l of a solution comprising buffer E, 12 units of BamHI and 12 units of PstI at 37.degree. C. for 8 hours. Then the resulting DNA was recovered by phenol extraction and ethanol precipitation, and treated with Deletion kit for Kilo-Sequence (Takara Shuzo Co., Ltd.). E. coli JM109 cells were transformed with the DNA and ampicillin-resistant transformants were obtained, from which plasmid DNAs were prepared. Subsequently by the analysis of the restriction enzyme cleavage pattern, transformants containing deletion plasmids shown in FIG. 6 were selected out.

The deletion plasmids and pPH-1 (about 3 .mu.g each) were dissolved in 20 .mu.l each of a solution comprising 2 .mu.l of 2N NaOH and 2 .mu.l of 2 mM EDTA and denaturated at room temperature for 5 minutes. The DNA was recovered by ethanol precipitation and the nucleotide sequence was determined with SEOUENASE (U.S. Biochemical). [.alpha.-.sup.32 P] dCTP (800 Ci/m mole: Amersham Co.. Ltd.) and Primer M3 (Takara Shuzo Co.. Ltd.).

The nucleotide sequence of the DNA fragment derived from the chromosomal DNA of Streptococcus sp. APl is as illustrated in FIG. 4. The DNA fragment contains promoter region, SD sequence, and the Protein H-structural gene coding for the amino acid sequence consisting of 376 amino acids (including Met at start point) starting from initiation codon ATG and terminating with termination codon TAA.

The structural gene encodes the amino acid sequence consisting of 376 residues beginning with Met and terminating with Asn, as shown in FIG. 5. The N-terminal amino acid sequence consisting of 41 residues beginning with Met and terminating with Ala has common characteristics to those of the signal sequence considered to be necessary for the protein secretion of gram positive bacteria, and therefore it can be considered that mature Protein H is a protein having an amino acid sequence consisting of 335 residues beginning with Glu and terminating with Asn.

Example 6

Expression of pPH-1

The E. coli JM109 (pPH-1) obtained in Example 5, was cultivated in LB medium containing ampicillin at a concentration of 50 .mu.g/ml at 37.degree. C. for 16 hours. The culture was added to 2 liters of the same medium, incubated at 37.degree. C. for 4.5 hours, and centrifuged.

The periplasmic fraction was prepared by the cold osmotic shock procedure (Nossal et al., J. Biol. Chem. 241, 3055 (1966)). A mixture of the cytoplasmic and membrane fractions was prepared by sonicating the pellet obtained after cold osmotic shock.

In this procedure, more than 95% of the .beta.-galactosidase activity was observed in the mixture of cytoplasmic and membrane fractions, while more than 95% of the .beta.-lactamase activity was observed in the periplasmic fraction.

Both fractions were analysed by the Western-blotting method described in Example 3. The Protein H having an apparent molecular weight of 45 kDa was demonstrated only in the mixture of cytoplasmic and membrane fractions, while the Protein H having an apparent molecular weight of 42 kDa was demonstrated in the periplasmic fraction.

Example 7

Binding properties of Protein H purified from the periplasmic fraction of E. coli JM109 (pPH-1)

The Protein H having an apparent molecular weight of 42 kDa was purified from the periplasmic fraction obtained in Example 6 by the successive chromatography of IgG-Sepharose and gel filtration according to the methods described in Example 4. Yield of the Protein H amounted to about 4 mg.

The N-terminal amino acid sequence of purified protein was determined by amino acid sequence (Applied Biosystoms model 477A amino acid sequencer; Applied Biosystems Corp.) to be Glu-Gly-Ala-Lys-Ile-Asp-Trp-Gln-Glu-Glu, which was inentical to the putative N-terminal amino acid sequence of nature Protein H as described in Example 5.

The purified Protein H was radiolabeled according to the methods described in Example 1. The binding properties of radiolabeled Protein H were determined according to the methods described in Example 4. In addition to immunoglobulins described in Example 4, binding to human IgGFc and human IgGFab (all, Cappel Corp.); and horse and pig IgG (all, Cooper Corp.) were also determined.

The autoradiograms shown in FIG. 8 demonstrated that Protein H having the specificity of

i) binding strongly to human IgG (IgG1, IgG2, IgG3 and IgG4), human IgGFc and rabbit IgG; ii) binding weakly to pig IgG; iii) not binding to IgGs of mouse, rat, bovine, sheep, goat and horse; and iv) not binding to human IgGFab, IgA, IgD, IgE and IgM.

Claims

1. An isolated cDNA encoding a protein H having the following amino acid sequence: ##STR1##

2. An isolated cDNA encoding a subfragment of a protein H having the following amino acid sequence: ##STR2##

3. The cDNA as in claim 1, wherein said cDNA has the nucleotide sequence: ##STR3##

4. A cDNA as in claim 2, wherein said cDNA has the nucleotide sequence: ##STR4##