HU204089B - Process for producing muteins of bfgf - Google Patents

Abstract

A mutein of basic fibroblast growth factor (bFGF) is claimed Also claimed is a recombinant DNA having a base sequence which codes for a mutein of bFGF and a transformant harbouring a vector contg. a recombinant DNA having a base sequence which codes for a mutein of bFGF. - The prodn. of the structural gene which encodes the mutein is carried out by e.g. (a) hybridising with a mutagenic digonucleotide primer a single-stranded DNA comprising 1 strand of the structural gene of bFGF, (b) elongating the primer using DNA polymerase to form a mutational heteroduplex and (c) replicating this mutation heteroduplex. The resulting mutational heteroduplex is used to transform an infectable host organism or cell. In the replication of the heteroduplex by the host, progenies are produced from both chains. Following the replication, the mutant gene is isolated from a progeny of the mutant's chain and is inserted in an approp. vector which is used to transform an approp. host organism or cell.

Description

The present invention relates to a process for the production of muteins of basic fibroblast growth factors (hereinafter also referred to as bFGF) and to recombinant DNA having a base sequence encoding the bFGF mutein.

A basic polypeptide hormone, having about 170000 molecular weights, abFGF, which is secreted mainly by the pituitary gland, was first isolated as a factor that exhibits strong growth-promoting effects on fibrohlasts, e.g. BALB / c3T3 (D. Bospodarowicz: Nat. 249, 123 (1974)). However, it was subsequently shown to show growth promoting effects on almost all mesoblast derived cells [GospodarowiczD. et al., National Cancer Institute Monograph, 48,109 (1978)]. In particular, the angiogenic effect of bFGF, together with its cell growth promoting effect, suggests that it is a therapeutic agent for the treatment of trauma, and a prophylactic and therapeutic agent for thrombosis, arteriosclerosis, and the like. against.

Bovine bFGF was first reported in the journal Proceedings of the National Academy of Sciences of the United States of America [82]. Vol. 6507 (1985)] AScience 233,545 (1986) discloses bovine bFGF-derived amino acids derived from a clone produced by human cloning of bFGF cDNA.

As regards human bFGF, Biochemical and Biophysical Research Communications, 135, 541 (1986), reports that human bFGF has been extracted from the human brain.

The European Molecular Biology Organization (EMBO) Journal, 5,2523 (1986) and International Publication No. WO / 01728 disclose that the amino acids that make up human bFGF have been deduced from the clone was produced by cloning the bFGF cDNA using bFGF bovine as a probe.

In addition, FEBSLetters, 213,189 (1987) describes the transformation of a transformant obtained by cloning the human bFGF cDNA.

Compared to human bFGF and bovine bFGF, human bFGF has a Thr at position 112, and bovine bFGF has Ser at position 112; the HU-128 on ^two Mán bFGF has Ser-rel, and bovine bFGF Pro. (In this case, NtennalPro is called firsthydric acid.)

It has been hypothesized that modification of the amino acid sequence will improve the stabilization of bFGF in the cell, per molecule! cell proliferative activity as well as unidentified biological activities are enhanced.

Biologically active biologically active proteins produced by DNA recombination techniques are known to contain cysteine residues that are not relevant to the activity of the proteins. but which may form undesirable intermolecular or intramolecular bonds.

During the production of AhFGF by recombinant DNA technology, heterogeneous conformations were found in Escherichia coli extract containing high concentrations of bFGF. These conformations are believed to occur as a result of random intra- or inter5 molecular disulfide bridge formation and cause difficulty in purifying bFGF and reducing recovery.

Recognizing this fact, we intended to modify microbiologically produced bioactive proteins as 0 e.g. recombinant bFGF, while its activity is not adversely affected by its tendency to form such intermolecular bridges and intramolecular bonds that result in the formation or unwanted formation of unwanted tertiary structures (i.e., conformations that reduce the activity of the fe5 protein).

In our work, modified bFGF muteins were created by DNA recombination and site-directed mutagenesis, and experiments were conducted to improve their stability and increase their intracellular productivity and activity, and to identify changes in their bioactivity; as a result, we have found a mutein that fulfills these purposes, and thus the present invention provides a method for (1) a single-mutant of a basic fibroblast growth factor (bFGF), (2) a DNA having a base sequence encoding the above-mentioned (1) mutein, ) for the preparation of a transformant comprising a vector comprising a recombinant DNA having the base sequence encoding the mutein (1) mentioned above.

The above-mentioned mutein (1) is prepared by culturing a transformant comprising a vector containing a recombinant DNA having a base sequence encoding said mutein in a medium to produce and accumulate mutein in the culture broth.

AbFGF can be any bFGF if it is derived from a warm-blooded animal.

Typical examples include human, bovine and rodent bFGFs.

Especially preferred are polypeptides that include the following amino acid sequence;

Phe-Phe-Leu-Arg-He-His-Pro-Asp-Gly-Arg-Val-Assp-Gly-Val-Arg-Glu-Lys-Ser-Asp-Pro (I) More preferred are the polypeptides which have the following formula:

Pro-Ale-Leu-Pro-Glu-Asp-Gly-Gly-Ser-ŰGly-Ala-Phe-Pro-Pro-Gly-His-Phe-Lys-Asp-Proe-Lys-Ar g-Leu.-Tyr -Cys-Lys-Asn-Gly-Gly-Phe-Phe-Leu-Ar g-He-His-Pro-Asp-Gly-Arg-Val-Asp-Gly-Val-ArgGlu-Lys-Ser-Asp-Pro- His-Ee-Lys-Leu-Gln-Leu-GI n-Ala-GIu-Glu-Arg-Gly-Val-Val-Ser-Ile-Lys-GlyVal-Cys-Ala-Asn-Arg-Tyr-Leu-Ala GluAsp-Met-Lys-Gly-Arg-Leu-Leu-Ala-Ser-Lys-Cys-Val-Glu-ThrAsp-Cys-Phe-Phe-Phe-Glu-Arg-Leu-Glu-Ser-Asn-Asn-Tyr -Aos-Thr-Tyr-Arg-Ser-Arg-Lys-TyrX-Ser-Trp-Tyr-Val-Ala-Leu-Lys-Arg-Thr-Gly-Gln

HU 204089 Β

-Y-rLys-Leu-Gly-Y-Lys-Ihr-Gly-Pro-Gly-GlnLysAla-Ile-Leu-Phe-Leu-Pro-Met-Ser-Ala-Lys-Ser (Π) wherein X is Thr or Ser; when X is Thr, then Y is Ser, and when X is Ser, Y is Pro.

More particularly, the following amino acid / enca polypeptide is preferred:

Pro-Ala-Leu-Pro-Glu-Asp-Gly-Gly-Gly-Ala-Phe-Pro-Gly-His-Phe-Lys-Asp-Pro-Lys-Arg-LeuTyr-Cys-Lys-Asn- Gly-Gly-Phe-Phe-Leu-Arg-Ile-H-Pro-Asp-Gly-Arg-Val-Asp-Gly-Val-Arg-Glu-Lys-Ser-Asp-Pro-His-Val-Lys -Leu-Gln-Leu-Gln-AlaGlu-Glu-Arg-Gly-Val-Ser-Ile-Lys-Gly-Val-Cys -Ala-Asn-Arg-Tyr-Leu-Ala-Met-Lys-Glu -Asp-Gly -Arg-Leu-Leu-Ala-Ser-Lys-Cys-Val-Thr-Glu-GluCys-Phe-Phe-Phe-Glu-Arg-Leu-Glu-Ser-Asn-AsnTyr-Asn-rnr Tyr-Arg-Ser-Arg-Lys-Tyr-Ser-SerTrp-Tyr-Val-Ala-Leu-Lys-Arg-Thr-Gly-Gln-TyrLys-Leu-Gly-Ser-Lys-Thr-Gly-Pro LysAla-Gly-Gln-Ile-Leu-Phe-Leu-Pro-Met-Ser-Ala-Lys-Ser.

The mutein produced by the process of the invention has substantially the amino acid sequence of the parent peptide or protein; variations thereof include the addition of an amino acid or amino acids, the deletion of a constituent amino acid or residues and the replacement of a constituent amino acid or residues with another amino acid or residues.

Such an amino acid addition means the addition of at least one amino acid.

Such splicing of the constituent amino acids involves the deletion of at least one bFGF-forming amino acid.

Such substitution of constituent amino acids with other amino acids means the replacement of at least one amino acid comprising bFGF by another amino acid.

The at least one amino acid added to the bFGF in the mutein does not include the methionine derived from the initiation codon used for peptide ld expression and signal peptide.

The number of amino acids added is at least one, but this number can be anything but as long as the characteristics of the bFGF are lost. Preferred amino acids are some or all of the amino acid sequence of proteins that are homologous to bFGF and exhibit activity similar to bFGF.

Examples of such added amino acids include some or all of the amino acid sequences of proteins such as acidic microbial growth factors (aFGF), interleukin I-α (DL1-a), interleukin 1-β (EL1-β), and a protein. , encoded by int2 in oncogenes.

Such aFGFs include human and bovine aFGF. Such IL1-α and ΙΙ, Ι-β include human ILI-α and ILI-β.

The bovine aFGF comprises the following amino acid sequence:

NH ₂ -Phe-Asn-Leu-Pro-Leu-Gly-Asn-Tyr-Lys-Pro-Lys-Leu-Leu-Tyr-Cys-Ser-Asn - Gly-GlyTyr-Phe-Leu-Arg-De- Leu-Pro-Asp-Gly-Thr-Gly-Thr-ValÁsp-Lys-Asp-Arg-Asp-Ser-Gln-His-Hegle-Leu-Leu-Cys-Ala-Glu-Ser-Ile-Gly- GluVal-Tyr-De-Lys-Ser-Thr-Glu-Thr-Gly-Gln-PheLen-Ala-Met-Asp-Thr-Asp-Gly-Leu-Leu-Tfyr-GlySer-Gln-Thr-Pro-Asn- Glu-Glu-Cys-Leu-Phe-Glu-Arg-Glu-Glu-YLeu-Asn-His-Tyr-Asn-Tyr-Thr

-Hexane-Ser-Lys-Lys-His-Ala-Glu-Lys-His-Tyr-Gly-Leu-PHEV-Lys-Lys-Asn-Gly-Arg-Ser-Lys-Leu-Gly-Pro-Arg-Thr-His -Phe-Gly-Gln-Lys-Ala-He-Le u-Phe-Leu-Pro-Leu-Pro-Val-Ser-Ser-Asp-COOH.

AzIL-a contains the following amino acid sequence:

NH ₂ - Met-Ala-Lys-Val-Pro-Asp-Met-Phe-GluAsp-Leu-Lys-Asn-Cys-Tyr-Ser-Glu-Asn-Glu-GluAsp-Ser-Ser-Ser-He-Asp -His-Leu-Ser-Leu-Asn-Gln -Lys-Ser-Phe-Tyr-His-Val-Ser-Tyr-Gly-Pro-Leu-Hi s-Ghi-Gly-Cys-Met-Asp-Gln- Ser-Val-Ser-Leu-SerBe-Ser-Glu-Thr-Ser-Lys-Thr-Ser-Lys-Leu-Thr-Phe -Lys-Glu-Ser-Met-Val-Val-Val-Ala-lhr- Asn-GlyLys-Val-Leu-Lys-Lys-Arg-Arg-Leu-Ser-Leu-SerGln-Ser-Be-Thr-Asp-Asp-Asp-Leu-Glu-Ala-Be-Ala -Asn-Asp- Ser-Glu-Glu-Glu-Be-He-Lys Asn

Pro-Arg-Ser-Ala-Pro-Phe-Ser-Phe-Leu-Ser-As-Va

1-Lys-Tyr-Asn-Phe-Met-Arg-Be-Be-Lys-Tyr - GluPhe-Ile-Leu-Asn-Asp-Ala-Leu-Asn-Gln-Ser-Ile-Be

-Arg-Ala-Asp-Asn-Gln-Tyr-Leu-Thr-Ala-Ala-Ala

-Leu-His-Ans-I ^ u-Asp-Glu-Ala-Val-Lys-Phe-Asp

-Met-Gly-Ala-Tyr-Lys-Ser-Ser-Lys-Asp-Asp-AlaLys Thr-Val-Be-Be-Be-Arg-Leu-Ser-Lys-Thr-GlnLeu-Tyr-Val-Thr -Ala-Gln-Asp-Glu-Asp-Gln-PROV-Leu-Leu-Lys-Glu-Met-Pro-Glu-Pro-Be-Be-LysTbr-Thr-Gly-Ser-Thr-Glu-Asn-Leu PhePhe-Leu-Trp-Glu-Thr-His-Gly-Thr-Lys-Asn-Phe-Thr-Ser-Tyr-Val-Ala-His-Pro-Asn-Leu-Phe-Ala-Be-LHR

Gto-Lys-Asp-Tyr-Trp-Val-Leu-Ala-Vys-Gly-Gly-Pro-Pro-Ser-Ile-Thr-Asp-Phe-Gln-Ile-Leu-Glu-Asn

Gln-Ala-COOH.

ILI β contains the following amino acid sequence:

NH ₂ -Me-Ala-Glu-Val-Pro-Glu-Leu-Ala-SerGlu-Met-Met-Alair-Tyr-Ser-Gly-Asn-Glu-AspAsp-Leu-Phe-Phe-Glu-Ala -Asp-Gly-Pro-Lys-Gln

Met-Lys-Cys-Ser-Phe-Gln-Asp-Leu-Asp-Leu-Cys

-Pro-Leu-Asp-Gly-Gly-Be-Gln-Leu-Arg-Be-Ser-A sp-His-His-Tyr-Ser-Lys-Gly-Phe-Arg-Gln-Ala-Ala

Ser-Val-Val-Val-Ala-Met-Asp-Lys-Leu-Arg-Lys

Met-Leu-Val-Pro-Cys-Pro-Gln-LHR-Phe-Gln-Glu

-Asn-Asp-Leu-Ser-Thr-Phe-Phe-Pro-Phe-Be-Phe-Glu-Giu-Glu-Pro-He-Phe-As-Thr-Trp-Asp-Asn

-Glu-Ala-Tyr-Val-Hjs-Asp-Ala-Pro-Val-Arg-Ser-Leu-Asn-Cys-Thr-Leu-Arg-Asp-Ser-Gln-Gln-Lys-Se r-Leu Met-Val-Ser-Gly-Pro-Tyr-Glu-Leu-Lys-Ala

-Leu-His-Leu-Gln-Gly-Gln-Asp-Met-Glu-Gln-Gln

Val-Val-Phe-Ser-Met-Ser-Phe-Val-Gln-Gly-Glu-Glu-Ser-Asn-Asp-Lys-Be-Pro-Val-Ala-Leu-Gly-Leu

Lys-Glu-Lys-Asn-Leu-Tyr-Leu-Ser-Cys-Val-Leu

-Lys-Asp-Asp-Lys-Pro-Thr-Leu-Gln-Leu-Glu-Ser

-Yal-Asp-Pro-Lys-Asn-Tyr-Pro-Lys-Lys-Lys-Met

HU 204089 Β

-Ghi-Lys-Arg-Phe-Val-Phe-Asn-Lys-Ile-Glu-Asn-IIeAsn-Lys-Leu-Gln-Phe-Glu-Ser-Ala-Gln-Phe-Asn-Tyr-Trp-Ile -Ser-Thr-Ser-Gln-Ala-Glu-Asn

Met-Pro-Val-Phe-Len-Gly-Gly-LHR-Lys-Gly-Gly

-Gln-Asp-He-Thr-Asp-Phe-Thr-Met-Gln-Phe-Val-5

Ser-Ser-COOH.

The amino acid sequence encoded by int-2 is as follows: 3 ^ 2 ^^^ 7 ^^^^ - 16 ^ 16 ^ 10 ^ LeuSer-leu-Leu-Glu-Pro-Ser-Trp-Pro-Thr-Thr-GlyPro- Gly-Thr-Arg-Leu-Arg-Arg-Asp-Ala-Gly-Gly-10 Arg-Gly-Gly-Val-Tyr-Glu-Eus-Leu-Gly-Gly-AlaPro-Arg-Arg-Arg-Lys Leu-Tyr-Cys-Ala-Thr-His-LysTyr-I ^ u-Gln-Leu-His-Pro-Ser-Gly-Arg-Gly-Ser-divorce-Leu-Glu-Asn-Ser-Ala-Tyr -Ser-HeLeu-Glu-De-Ihr-Ala-Val-Glu-Val-Gly-Val-Val-Ala-15-De-Lys-Gly-Leu-Phe-Ser-Gly-Arg-Tyr-Leu-AlaMet- Asn-Lys-Arg-Gly-Arg-Leu-Tyr-Ala-Ser-Tyr-AspHis-Asn-Ala-Glu-Cys-Glu-Phe-Val-Glu-argile-His-Glu-Leu-Gly-Tyr- Asn-Thr-Tyr-Ala-Ser-Ar g-Leu-Tyr-Arg-Thr-Gly-Ser-Ser-Gly-Pro-Gly-Ala 20 -Gk-Arg-Gln-Pro-Gly-Ala-Gln- Arg-Pro-Irp-TyrVal-Ser-Val-Asn-Gly-Lys-Gly-Arg-Arg-Arg-Gly-Phe-Lys-Ikr-Arg-Arg-lbr-GIn-Lys-Ser-Ser -Leu -Phe-Leu-Pro-Arg-Val-Leu-Gly-His-Lys-Asp-HisGlu-Met-Val-Arg-Leu-Leu-Gln-Ser-Ser-Gln-Pro-A 25 rg-Ala -Fro-Gly-Ghi-Gly-Ser-Gln-Pro-Arg-Gln-Arg -Arg-Gln-Lys-Lys-Glin-Ser-Pro-Gly-Asp-His-Glylys-Met-Glu-Thr-LeunSer Thr-Arg-Ala-Thr-Thr-Gln-PROSER I ^ eu-Hi s-Thr-Gly-Gly-Leu-Ala-Val-A la-COOH.

In the artificial metal produced by the process of the invention, which lacks at least one amino acid that makes up bFGF, the number of amino acids that can be eliminated as the bFGF amino acids can be any number as long as the bFGF is not characteristic.

Such deletions of constituent amino acids include, for example, the following;

deleting amino acids from the amino-ternary or carboxyl-terminus;

10 residues at the amino terminus of human bFGF; Metprom-Ala-Leu-Pro-Glu-Asp-Gly-Gly-Ser,

-humanbFGF at the amino terminus of 14 gmMPro-Ala-Leu-Pro-Glu-Asp-Gly-Gly-Ser-Gly-Ala-P-he

-humanbFGF amino terminal residue41;

2 3 4 -41

Met-Pro-AIa-Leu -......... Val, or

from the carboxy terminal of -humanbFGF:

88 146 147

Lys-Cas ....... to eliminate Lys-Ser.

In the mutein produced by the process of the invention, wherein at least one amino acid comprising bFGF has been replaced, any of the amino acids that make up bFGF before replacement can be substituted as long as no characteristic of FGF is lost.

Exemplary amino acids to be replaced prior to substitution include cysteine but also non-systemic amino acids. Cysteine is preferred. When different from the pre-exchange amino acids, these are mainly aspartic acid, arginine, glycine, serine, valine, etc. It can be mentioned.

When the constituent amino acid is a cysteine before substitution, e. sem5 legal amino acids. Examples of such neutral amino acids include glycine, valine, alanine, leucine, isoleucine, tirrine, phenylalanine, histidine, tryptophan, threonine and methionine. Serine and threonine are particularly preferred.

When the constituent amino acid is different from the title protein prior to substitution, the substituent amino acids may be selected from those which exhibit, for example, a specific amino acid residue. they differ in hydrophilicity, hydrophobicity, or electric charge from the amino acid before substitution.

When the constituent amino acid is aspartic acid before substitution, the replacement amino acid may be asparagine, threonine, valine, phenylalanine and arginine; asparagine and arginine are preferred.

When the constituent amino acid is arginine before substitution, the replacement amino acid may be glutamine, threonine, leucine, phenylalanine and aspartic acid; glutamine is preferred.

When the constituent amino acid is glycine before substitution, the replacement amino acid may be threonine, leucine, phenylalanine, serine, glutamine and arginine; threonine is preferred.

When the constituent amino acid is serine prior to substitution, the replacement amino acid may be methionine, alanine, leucine, cysteine, glutamine, arginine and aspartic acid; methionine is preferred.

When the constituent amino acid is valine prior to substitution, the substituent amino acid is leucine, leucine, proline, glycine, lysine and aspartic acid; serine is preferred.

As a replacement amino acid before replacement, aspartic acid, arginine, glycine, serine and valine are preferred.

Asparagine, glutamine, arginine, threonine, methionine, serine and leucine are preferred as replacement amino acids.

As an embodiment of mutein exchange, rysteine exchange serine (i.e., serine replaces cis 40 teenager) is most preferred.

No less than 2 replacements may be made in these exchanges, with two or three exchanges being preferred.

The mutein produced by the process of the present invention may be combined with the addition, deletion or replacement of the 2 or 3 mentioned above.

In addition to conventional DNA recombination techniques, site-directed mutagenesis is used in the present invention to produce mutein. This technique is well known and is described in 50 Genetic Engineering (eds. Lather R.F. and LecoqJ / P.), AcademicPress (1983), 31-50. side. Oligonucleotide-directed mutagenesis is described in Genetic Engineering, Principles and Methods (eds. Smith M. and Gillam S.), Plenum 55 Press (1981), vol. side.

The production of the structural gene encoding the mutein of the present invention can now be carried out.

a) hybridizing with a mutagenic oligonucleotide primer to a single-stranded DNA containing 1 strand of the bFGF structural gene,

HU 204089 segítségével (b) amplifying it with primer DNA polymerase to form a mutational heteroduplex, and (c) replicating this mutational heteroduplex.

The size of the oligonucleotide primer will depend on the conditions required for stable hybridization of the primer to the gene triethyl to which the mutation is to be introduced and will depend on the limitations of currently available methods for oligonucleotide synthesis. Factors to consider in oligonucleotide-directed mutagenesis (e.g., the total size of the nucleotide and the size of the mismatched linkage at the mutation site) are described by Smyth M. and Gillam S., cited above. The overall length of the oligonucleotide is generally selected to be stable and unique to the mutation site, and to provide a spacing between the mutation site and the 5 'and 3' termini of a size sufficient to prevent DNA polymerase mutation attributable to exonuclease activity,

The oligonucleotides used in the mutagenesis of the invention will normally contain from 12 to 24 bases, preferably 14 to 20 bases, most preferably 14 to 18 bases. Normally, they contain at least 3 bases from the codon to be changed to the 3 'terminus.

For the purpose of e.g. to obtain a mutein with an added amino acid, a mutagenic bFGF gene is generated by synthesizing the gene encoding the amino acid sequence to be added and inserted or added to the site in the bFGF gene directly or after fragmentation by restriction enzyme digestion. When any suitable restriction enzyme recognition sites can be generated by site-directed mutagenesis.

For the purpose of e.g. to obtain a mutein lacking the bFGF allo-amino acids, a mutagenic bFGF gene is produced in 3 different ways. In one of the three different pathways, the amino terminus of bFGF is bypassed; in another, removing the central region of bFGF; and in the third, removing the carboxyl terminal.

In the case of amino-terminal deletion, the codon of the gene encoding the amino acid sequence to be deleted is replaced by site-directed mutagenesis with the ATG codon encoding Met, and this codon has the appropriate restriction enzyme recognition site on the S'-terminal side to ease ligation to the promoter. .

By deleting the central region of the amino acid sequence, site-directed mutagenesis generates a unique restriction enzyme recognition site on both the 5'-tenainal and S'-tenninase side of the gene encoding the sequence to be deleted, and the BNA region is cleaved from the gene by enzymatic digestion. and then re-ligating the remaining two gene fragments to obtain the desired gene encoding bFGF lacking the indicated amino acids. Of course, it is also necessary that digestion with the restriction enzyme does not cause the reading frame to slip.

In the case of deletion of an amino acid at the acarboxyl-terminal side, the condon of the gene encoding the amino-tennal amino acids of the sequence to be deleted is replaced by a stop codon by site-directed mutagenesis.

In order to obtain a mutein in which the constituent cysteine amino acid is replaced, a mutagenic bFGF gene is produced by deleting e.g. inducing Cys-expressing codon or site mutagenesis in the TGC or TGT cys-ld-coding codon using a synthetic nucleotide primer that modifies the codon to encode a different amino acid. A primer is hybridized with the sense chain of the FGF gene, e.g., to replace the human bFGF cline (position 26) with serine. Suitable nucleotide primers include, e.g. 5'CGTTCTTGCTGTAGAGCCGCT-3', in which the underlined triplet represents the changed codon.

A suitable primer for the replacement of cysteine (position 70) with serine is 5'-AACGATTAGCGCT'CACTC-3, in which the underlined triplet represents the altered codon.

Suitable primers for the replacement of cysteine (position 88) with serine include 5'-GTAACAGACTTAGAAGCTAGT-3'. in which the underlined triplet represents the changed codon.

Suitable primers for the replacement of cysteine (position 93) with serine include 5'-TCGAAGAAGAA4G4CTCATCC-3', in which the underlined triplet represents the altered codon.

For Cys (site 26), the cysteine is rendered like the base by a T - * A exchange of base 1; for Cys (position 70), the base 1 is replaced by T -> A and the base 2 is replaced by T C; and for Cys (position 88, position 93), the cysteine is converted to a base by a G 2 exchange of base 2.

It should be noted that in the production of bFGF mutein protein by site-directed mutagenesis, 2 or more variants can be induced in the DNA sequence, i.e., the codons corresponding to the amino acids are degenerate.

In order to obtain a mu protein in which a noncysteine allcote amino acid is replaced by another amino acid, a mutagenic bFGF gene is generated by altering a codon with an oligonucleotide primer in a manner similar to that shown in the cysteine.

The oligonucleotide primer will, of course, be altered according to the amino acid to be replaced.

The primer is hybridized to a single-stranded tree derived from the cloning of a single strand of the bFGF gene, e.g. az13 (Yanisch-Perron, Vieira C. and Messing I, Gene, 33, 103-119 (1985)); Messing J .: Meghods in Enzymology 101,20-78 (1983)], fd [HernnanR. et al., Molecular and General Genetics, 177, 231 (1980) 3 or 0x174 (Smyth M. and Gillam S .: Genetic Engmeering, Plenum Press, vol. 3, pp. 1-32). (1981). It should be noted that the phage is capable of either the sense of the gene or the opposite

HU 204089 Β (antisense) to carry the chain. When the phage carries an antisense chain, in addition to the deviations from the relevant codons defining the triplet coding for another amino acid, there may be differences in the primer due to codon degeneration in the sense chain region containing the codon where the mutation is to be induced. . Similarly, when a phage carries a sense strand, the primer must not be complementary to the region of the sense chain that contains the codon in which the mutation is to be induced, and must not contain variants that pair with the codon to be deleted. The conditions used for hybridization were SmithM. and GillamS. The temperature is normally between 0 'and 70' C, most preferably between 10 'and 50' C. After hybridization, the primer is extended on phage DNA by reaction with primer Escherichia coli DNA polymerase, T4 DNA polymerase, reverse transcriptase or other suitable polymerase. The resulting double-stranded dsDNA is closed-looped with DNA ligation, e.g. Treatment with T4DNA ligase. DNA molecules containing single-stranded regions are subjected to S1 endonuclease treatment

The resulting mutational heteroduplex is used to transform an infectious host or cell. By replicating the heteroduplex by host cells, both chains produce progeny. Following areplication, the mutant gene is isolated from the progeny of the mutant chain and inserted into a suitable vector, which is then used to transform an appropriate host or cell.

The phage DNA carrying the mutation gene is then isolated and inserted into a plasmid.

Examples of plasmids into which DNA has been introduced include Escherichia coli derived plasmids, such as plasmids. pBR322 (Gene, 1975, 2, 95), pBR325 (Gene, 4, 121 (1978)), pUC12 (Gene, 19,259 (1982)), and pUC13 (Gene). , 19, 259 (1982)], and plasmids from Bacillus subtilis, e.g. pUBHO (Biochemical and Biophysical Research Communication, 112, 678, 1983), but any other plasmid may be used as long as it is replicable and maintained in the host. ¹

Examples of methods for incorporation into plasmids include pL described in Molecular Cloning (Maniatis T. et al., ColdSpringHarborLaboratory, 1982, p. 239).

The cloned gene is suitable for expression in a medium. (vector) downstream of the promoter to give an expression vector

The vectors include the aforementioned plasmids from Escherichia coli (e.g., pBR322, pBR325, pUC12 and pUC13) and Bacillus subtilis (e.g., pUBIL10, pTP5 and pC194), yeast plasmids (e.g., pSH19 and pSH15), bacteria, e.g. phage λ and animal viruses such as retroviruses and vaccinia viruses.

Said gene is a translational initiation nebula

It may contain ATG at the 5'-terminus and may also have TAA, TGA or TAG at the 3'-terminus per translational termination nebula. A promoter is ligated upstream to express said gene. Any promoter may be used in the method of the invention provided that it is sufficiently compatible with the host used for gene expression.

When a host transformer is used to transform a bacterium belonging to the genus Escherichia, the trp promoter, Iacpromotor, recA promoter, P _L promoter, Ipp promoter, and the like. useful; when the host is a bacterium of the genus Bacillus, the SP01 promoter, the SPO2 promoter, the penP promoter, and the like. suitable when the host is a yeast, PHO5 promoter, PGK promoter, GAP promoter, ADH promoter, and the like. suitable. It is particularly preferred to use a bacterium of the genus Escherichia as the host and the trp promoter or XP _L promoter as the pro20 engine.

When the host used is an animal cell, suitable promoters include SV40 promoters and retroviral promoters; especially promoters derived from SV40 are preferred.

Using the vector thus constructed, which contains a recombinant SND having a base sequence encoding a mutein, transformants are prepared.

As the host, e. bacteria of the genus Escherichia 0, bacteria of the genus Bacillus, yeasts, animal cells, etc.

Examples of bacteria of the genus Escherichia include Escherichia coli K12DH1, Proc. NatLAcad. Sci. USA, 60,160 (1968)], JM1035 at [Nucleic Acids Research, 9,309 (1981)], JA221 (Journal of Molecular Biology, 120,517 (1978)), HB101 (Journal of Molecular Biology, 41). 459 (1969)], C600 (Genetics, 39, 440 (1954)), and MM294 (Proc. Natl. Acad. Sci. USA, 73, 4174 (1976)].

Examples of bacteria of the genus Bacillus include Bacillus subtilis MU14 (Gene, 1983, 24, 255) and 207-21 (Journal of Biochemistry, 1984, 95.87).

Examples of yeasts include Saccharomyces cerevisiae AH22R', Na 87-1 ΙΑ and DKD-5D.

Examples of animal cells include COS-7 and Beating Monkey cells, CHO Chinese Hamster ¹ ovary, azeg mouse L and human FL cell.

The transformation of the aforementioned bacteria of the genus Escherichia, e.g. Proc. Natl. Acad. Sci. USA 69,2110 (1972) and Gene 17,107 (1982) et al. irodalmihelyekenleírtmódszerekszerinthajtjukvégre.

Transformation of bacteria of the genus Bacillus e.g. Molecular and General Genetics, 168, 111 (1979) et al. according to the methods described in the literature.

The transformation of yeasts, e.g. Proc. Natl.

HU 204089 Β

Acad. Sci. USA 75, 1929 (1978).

Transformation of animal cells, e.g. Virology, 52, 456 (1973).

In this way, a transformant is obtained which comprises a vector containing a recombinant DNA having a base sequence encoding mutein.

Said transformants are cultured in a medium that allows mutein to be produced.

A liquid medium containing carbon sources, nitrogen sources, mineral salts, and other materials essential for the growth of a particular transformant is suitable for the cultivation of a transformant host bacterium of the genus Escherichia or Bacillus. Examples of carbon sources include glucose, dextrin, soluble starch, sucrose, etc .; nitrogen sources include, for example, inorganic or organic materials such as. ammonium salts, nitrates, corn jam, peptone, casein, meat extract, soybean cake and potato extract; as a mineral, e.g. calcium chloride, sodium dihydrogen phosphate and magnesium chloride. Yeast extract, growth promoters, vitamins, etc. may be added to the medium. too.

The desired pH of the medium is between about 6 and about 8.

An example of a suitable medium for culturing bacteria of the genus Escherichia is M9 medium containing glucose and casino acid (Miller, 1972, Cold Spring Harbor Laboratory, Molecular Genetics, 431-433). Chemicals such as 3β-indolyl acrylic acid may also be added to the medium when it is necessary to increase the efficiency of the promoter.

When the host is a bacterium of the genus Escherichia, culturing is normally carried out at about 15-43 ° C for about 3 to 24 hours, with aeration and agitation as necessary.

When the host is a bacterium of the genus Bacillus, culturing is normally carried out at a temperature of about 38-40 ° C for about 624 hours and when necessary aeration and agitation are used.

When the host is a bacterium of the genus Bacillus, culturing is normally carried out at about 38-40 ° C for about 624 hours and, if necessary, aeration and agitation are used.

As a medium for the cultivation of transformants whose yeast hosts are mentioned, e. Burkholder Minimum Medium [Bostian, K.L., et al., Proc. Natl. Acad. Sci. USA 77,4505 (1980)]. Preferably, the pH of the medium is adjusted to between about 5 and about 8. Cultivation is normally carried out at about 20-35 ° C for about 24-72 hours and, if necessary, aeration and agitation are used.

As a medium for the cultivation of transformants whose host is an animal cell, mention may be made of the MÉM medium containing about 5-20% of wine yolk serum (Science, 122, 501 (1951) 3), DMEM (Virology, 8, 396 (1959)). 640 (Journal of the American Medical Association, 199,519 (1967)) and 199 (Proceeding of the Society for Biological Medicine, 73, 1 (1950)). Preferably the pH is between about 6 and about 8. Cultivation is normally carried out at about 30-40 ° C for about 15-60 hours and, if necessary, agitation and aeration are used.

Isolation and purification of the mutein from the aforementioned culture is carried out e.g. the following methods can be used.

Suitable methods for extracting mutein from cultured bacterial, fungal or animal cells are, for example, a method in which bacterial, fungal or animal cells are harvested after culture in a known manner and a protein denaturing agent, e.g. suspending guanidine hydrochloride in buffer to extract the desired protein from the cells; and employing a method of centrifuging the bacterial, fungal, or animal cells after fracturing, ultrasound treatment, lysozyme treatment, and / or freeze-thawing to recover the mutein. Particularly preferred is the method of combining lysozyme treatment with ultrasound treatment.

Purification of the mutein from the supernatant thus obtained can be carried out by well known methods of isolation and purification, using appropriate combinations. Well known methods of isolation and purification include those based on solubility such as solubilization and purification. salting out and solvent precipitation; methods based mainly on molecular weight differences, such as dialysis, ultrafiltration, gel filtration, and SDS-polyacrylamide gel electrophoresis; methods based on differences in electric charge, such as ion-exchange chromatography; methods based on specific affinity, e.g. affinity chromatography; methods based on the difference in hydrophobicity, e.g. reverse phase high performance liquid chromatography; and methods based on the difference of the isoelectric point, e.g. isoelectric electrophoresis.

In particular, the above supernatant is purified by ion exchange chromatography, such as nucleic acids and acidic proteins, as carrier, e.g. DEAE cellulose, etc. applied, we can remove it. For example, one can proceed by passing the supernatant through a DEAE cellulose column which has been diluted with an almost neutral buffer solution, e.g. Tris was equilibrated and non-adsorbed fractions were collected on the column. The collected fractions were to7

Subsequent to ion exchange chromatography using CM cellulose as carrier or the like, it is possible to adsorb the mutein to the carrier and elute the mutein with saline. These eluates may be lyophilized after dialysis.

Using Heparin-Sepharose, affinity chromatography can be used appropriately to purify the bFGF mutein in the extracts. fgypl. The abFGFmutein protein can be purified by loading the above eluate onto a heparin-Sepharose column equilibrated with a nearly neutral buffer (e.g., tris or phosphate buffer), and eluting the column with a linear gradient to NaCl or the like.

Heparin columns developed for high performance liquid chromatography [e.g. Shodex AF-pak HR.-894 available from ShowaDenko (Japan)] is particularly effective.

In this case, the hFGF mutein can be recovered as a homogeneous product in the same manner as for the above-mentioned heparin-Sepharose column, namely, the sample is half-loaded onto the heparin column with nearly neutral buffer, eluted thoroughly and eluted with pLNaCl.

The product thus obtained can be converted to a dry powder by 2 dialysis and lyophilization. It is desirable to protect the product with an added carrier (pL serum albumin) as this may prevent the product from being adsorbed to the wall of the container.

In addition, it is advantageous to add a small amount of reducing agent to prevent product oxidation.

Examples of such reducing agents include β-mercaptoethanol, dithiothreitol, glutathione and the like. 3

In this way, essentially pure bFGF mutein protein can be obtained. The substantially pure bFGF mutein protein produced by the process of the present invention includes products having a bFGF mutein protein content of less than 95% (w / w), and more preferably, products having no bFGF mutein protein content. less than 98% (w / w).

The protein obtained has high fibroblast growth promoting activity, capillary endothelial cell growth stimulating activity and angiogenic activity as well as high stability and low toxicity, respectively, for wound healing accelerated tissue healing, etc., or for therapeutic 5C. Alkalmazhatóreagenskéntsejttenyésztésmeggyorsításáhozis.

Particularly preferred is a mutein in which at least one constituent rysteine is serine substituted, since ezamutein is remarkably stable.

The invented mutein, when used as a medicament, cannot be safely administered parenterally or orally to warm-blooded animals (e.g., mice, rats, hamsters, rabbits, dogs and cats) and humans directly as a powder or as a medicament. such as tablets, capsules, solutions, ointments, and pharmaceutically acceptable carriers, excipients, or diluents.

Injections are made according to routine methods, e.g. physiological saline, aqueous solution containing glucose or other excipients.

The mutein produced by the process of the invention, when used as any of the above mentioned drugs, e.g. the warm-blooded animals and humans mentioned above may be administered in an amount sufficient to be in the range of 1 ng / kg / day to 100 pg / kg, taking into account the route of administration, symptoms, etc.

The mutein produced by the process of the invention, when used as a gelling agent for cell culture, is preferably added in an amount such that it contains 0.01-10 pg, preferably 0.1-10 pgmutein per liter of culture medium.

In the present specification and figures, the abbreviations used for bases, amino acids, etc. are used. based on abbreviations defined by IUPAC-IUB 5 Commission on Biochemical Nomenclature; those commonly used in the relevant art are illustrated in the list below. Whenever it is possible to have an optical isomer in an amino acid, this is the L-version of the isomer unless otherwise stated.

DNA-DNA cDNA-complementary DNA-ribonucleic acid

A-adenin> T — timin

G-guanine

C = cytosine

RNA ribonucleic acid dATP-deoxyadenosine triphosphate ¹ dTTP-deoxythymidine triphosphate dGTP-deoxyguanosine triphosphate dCIP-deoxycytidine triphosphate ATP-adenosine triphosphate Tdr

EDTA-ethylenediaminetetraacetic acid

SDS sodium dodecyl sulfate

Gly-glycine

Ala-alanine Val-valine Leu-leucine Be-isoleucine Ser-serine Thr-threonine Cys-rysteine Meth-methionine Glyglutamic acid Asp = aspartic acid Lys-lysine Arg-arginine His-histidine

HU 204089 Β

Phe = phenylalanine

Tyr tirozm

Trp = tryptophan

Pro = proline

Asn = asparagine

Gln glutamine

In the present specification and drawings, the Met derived from the translation initiation codon is designated as amino acid 1 in the numbering of the amino acids that make up human bFGF, unless otherwise indicated.

The following transformants, which are referred to in the following Reference Examples and Examples, were deposited with the Institute for Fermentation, Osaka (IFO) and the Fermentation Research Institute, Agency of Industrial

Science and Technology, Ministry of International Trade and Industry (ERI) (both operating in Japan) under the deposit numbers and due dates set out in Table 1. (deposit dates are in brackets). As regards the FRI deposit numbers, the FERM BP number indicates the deposit number under the Budapest Treaty; and when both FERM P and FERMBP are written, this indicates that the deposit under FERM P has been converted to a deposit under the Budapest Treaty and the transformants are stored at FRI under the FERM BP number.

Table 1

transformant IFO ERI E.coliK12DHFpTB627 (Reference Example 1) IFO 14494 (March 3, 1985) FERMP-8726FERMBP-1280 (April 2, 1986) kE.coliK12MM294 / pTB 669 (Reference example 2) IFO 14532 (August 1986) EERMP-8918 FERMBP-1281 (August 21, 1986) E.coliDH1 / pTB739 (Example 2) IFO 14575 (February 18, 1987) FERMP-9216EERM-BP-1641 (February 25, 1987) E. coli DH1 / pTB 742 (Example 3) IFO 14584 (March 24, 1987) FERMP-9307 FERM-BP-1632 (March 30, 1987) E. coli DH1 / pTB 743 (Example 4) SNOW 14585 (Mar. 24, 1987) FERMP-9308 FERM-BP-1643 (March 30, 1987) E. coli DH1 / pTB 744 (Example 5) IFO 14586 (March 24, 1987) EERMP-9309 FERMBP-1644 (March 30, 1987) E. coli MM294 / pIB 762 (Example 7) IFO 14613 (May 27, 1987) FERMP-9409 EERMBP-1645 (Jun 1987) E. coli MM294 / pTB 764 (Example 9) IFO 14614 (May 27, 1987) EERMP-941QFERMBP-1646 (June 11, 1987) E. coli MM294 / pTB 765 (Example II.1) IFO 14615 (May 27, 1987) FERMP-9411 FERMBP-1647 (June 11, 1987) E. coli MM294 / pTB 795 (Example 34) IFO 14700 (January 14, 1988) FERMBP-1660 (January 20, 1988) E. coli MM294 / pTB 796 (Example 35) IFO 14701 January 14, 1988) FERMBP-1661 (January 20, 1988) E.coliMM294 / pTB797 (Example 36) IFO 14702 (January 14, 1988) EERMBP-1662 (January 20, 1988)

The following hybridomas, prepared as described in Example 35 (3) (d), were deposited with IFO on August 17, 1987 under Mouse HbF 52 cell: IFO 50143; Mouse HbF cell: IFO 50144

Below is a brief explanation of the figures:

Figure 1 shows the base sequence which is a

HU 204089 Β encodes human bFGF and the deduced amino acid sequence.

THE 2. Fig. 6A shows the synthetic oligomers used as printers for the coding of Cys coding codons to Ser coding codons. 5 A3. Figure 11A shows the base sequence coding for the CS1 human bFGF mutein; it is carried by plasmid p1B739 obtained in Example 2. The figure shows the amino acid sequence of the mutein CS1 human bFGF encoded by the above base sequence. The mutated bases 10 are underlined and surrounded by an area rectangle containing the transformed amino acid.

A4. Fig. 2A shows the abase sequence encoding the human bFGF mutein CS2; it is carried by the plasmid p1B742 obtained in Example 3. The figure shows the amino acid sequence of CS2 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and surrounded by an area rectangle containing the transformed amino acid.

Figure 5 shows the base sequence encoding the human bFGF mutein CS3; it is carried by the plasmid p1B743 obtained in Example 4. The figure shows the amino acid sequence of CS3 human hFGFmutein encoded by the above base sequence. The mutated bases are underlined and surrounded by a square rectangle containing the transformed amino acid.

Figure 6 shows the base sequence coding for the CS4 human bFGF mutein; it is carried by the plasmid p1B744 obtained in Example 5. The figure shows the amino acid sequence of the human bFGFmutein CS4 encoded by the above base sequence. The mutated bases are underlined and surrounded by an area rectangle containing the transformed amino acid.

Figure 7 shows a schematic diagram of the construction of plasmid 2B (I) TB739. 35

Figure 8 shows a schematic diagram of the construction of plasmid PIBV742 in Example 3 (1).

For Figure 9, the plasmid plasmid aplB743 is outlined in Scheme 4 (1).

Figure 10 shows a schematic outline of the construction of plasmid pTB744 in Example 5 (1).

All. Fig. 4A shows the base sequence encoding the human bFGF mutein CS23; this is carried by the plasmid p1B762p obtained in Example 7. The figure shows the amino acid sequence of CS23 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and the regions containing one of the transformed amino acids are surrounded by a rectangle.

Figure 12 shows a schematic diagram of the construction of plasmid p1B762 in Example 7 (1).

A13. Figure 11A shows the abase sequence encoding the human bFGF mutein CS123; this is carried by plasmid pIB764 obtained in Example 9. The figure shows the amino acid sequence of GS123 human bFGF mutein fe 55 encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are square.

Figure 14 shows a schematic diagram of the construction of plasmid pIB762 according to Example 9 (1).

A15. Figure 12A shows the abase sequence coding for the CS1234 human bFGF mutein; this is carried by the plasmid pTB765 obtained in Example 11. The figure shows the amino acid sequence of the human bFGF mutein GS1234 encoded by the afentibase sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are square.

A16. Figure 11A is a schematic diagram of the construction of plasmid p1B765 according to Example 11 (1).

A17. Fig. 2A shows the base sequence encoding human bFGFmntein CS12; it is carried by plasmid pB776 obtained in Example 15. The figure shows the amino acid sequence of CS12 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are surrounded by a rectangle.

A18. Figure 15A shows a schematic diagram of the construction of plasmid pTB776 according to Example 15 (1).

A19. Fig. 4A shows the base sequence encoding the CS24 human bFGF mutein; this is the shoot. The plasmid p1B778 obtained in Example 1b is carried. The figure shows the amino acid sequence of GS24 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are surrounded by a rectangle.

Figure 20 shows a schematic diagram of the construction of plasmid p1B778 in Example 16 (1).

Figure 21 shows the base sequence encoding the human bFGF mutein CS13; this is carried by the plasmid p1B779 obtained in Example 17. The figure shows the amino acid sequence of GS13 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are surrounded by a rectangle.

Figure 22 is a schematic diagram of the construction of plasmid pIB779 in Example 17 (1).

Figure 23 shows the base sequence encoding the CS14 human bFGF mutein; this is carried by the plasmid pTB763 obtained in Example 18. The figure shows the amino acid sequence of the CS14 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are surrounded by a rectangle.

Figure 24 shows a schematic diagram of the construction of plasmid pIB763 in Example 18 (1).

A25. Fig. 4A shows the base sequence encoding the CS34 human bFGF mutein; this is carried by plasmid pTB777 obtained in Example 19. The figure shows the amino acid sequence of CS34 human bFGF mutein encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are surrounded by a rectangle.

Figure 26 shows plasmid pTB777 as in Example 19 (1)

EN 204 089 Β illustrates the outline of its creation.

Figure 27 shows the base sequence encoding human bFGF mutein CS234; this is carried by the plasmid pTB782 obtained in Example 20. The figure also shows the amino acid sequence of human bFGF mutein CS234 encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are enclosed in a rectangle.

Figure 28 is a schematic diagram of the construction of plasmid pTB782 in Example 20 (1).

Figure 29 shows the base sequence encoding the CS124 human bFGF mutein; this is carried by the plasmid pTB780 obtained in Example 21. The figure also shows the amino acid sequence of the human bFGF mutein CS124 encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are enclosed in a rectangle.

Figure 30 is a schematic diagram of the construction of plasmid pTB780 in Example 21 (1).

Figure 31 shows the base sequence encoding the human bFGF mutein CS134; this is carried by the plasmid pTB781 obtained in Example 22. The figure also shows the amino acid sequence of CS134 human bFGFmutein encoded by the above base sequence. The mutated bases are underlined and the regions containing any of the transformed amino acids are surrounded by a rectangle.

Figure 32 is a schematic diagram of the construction of plasmid pTB781 in Example 22 (1).

33-35. Figures 3 to 5 show the high performance liquid chromatography elution distributions of the muteins obtained in Example 24.

Figure 36 shows the high performance liquid chromatography elution distributions of the mutein oxides obtained in Example 25.

Figure 37 shows the high performance liquid chromatography elution distributions of the mutein reduction products obtained in Example 26.

Figure 38 depicts the synthetic oligomers used as primers to generate DNA encoding mutein.

Figure 39 shows the base sequence of the M13-PN10 phage obtained in Example 27 and the amino acid sequence of the ΊΜ910 mutein encoded by it; TM910 mutein is obtained in Example 32.

Figure 40 shows the base sequence of phage M13-P14 obtained in Example 27 and the amino acid sequence of the N14 mutein encoded by it; mutein N14 is obtained in Example 34.

Figure 41 depicts the base sequence of the phage M13-PC86 obtained in Example 27 and the amino acid sequence of the C86 mutein encoded by it, and

Example 35 gives you one.

Figure 42 shows the base sequence of the M13-PDN42 phage obtained in Example 27 and the amino acid sequence of the DN42 mutein encoded by it,

This is obtained in Example 36.

Figure 43 shows the base sequence of phage M13-PRQ45 obtained in Example 27, the mutein RQ45 obtained in Example 37.

Figure 44 shows the base sequence of the phage M13-PNR42 obtained in Example 28 and the amino acid sequence of the NR42 mutein encoded by it, which is obtained in Example 38.

A45. Figure 7B shows the stability of the human bFGF muteins CS2, CS3 and CS23 produced according to Example 29.

Figure 46 is a schematic representation of the construction of plasmid pTB854 according to Example 32.

Figure 47 is a schematic representation of the construction of plasmid pTB852 according to Example 33.

Figure 48 shows the base sequence encoding the human bFGF mutein NO10; this is carried by the plasmid pTB852 obtained in Example 33. The figure also shows the amino acid sequence of the N10 mutein encoded by the above base sequence. The N10 mutein is obtained in Example 33. The mutated bases are underlined.

Figure 49 is a schematic representation of the construction of plasmid pTB795 according to Example 34.

Figure 50 is a schematic representation of the construction of plasmid pTB796 according to Example 35.

Figure 51 shows a schematic diagram of the construction of plasmid pTB797 according to Example 36.

Figure 52 is a schematic representation of the construction of plasmid pTB799 according to Example 37.

Figure 53 is a schematic representation of the construction of plasmid pTB798 according to Example 38.

Figure 54 shows the base sequence encoding the EINT human bFGF mutein; it is carried by the plasmid pTB853 obtained in Example 39, the figure also shows the amino acid sequence of FINT mutein encoded by the above base sequence. The mutated bases are underlined.

Figure 55 is a schematic diagram of the construction of plasmid pTB853 according to Example 39.

Figure 56 is a schematic diagram of the construction of plasmid pTB855 according to Example 40.

Figure 57 shows the base sequence encoding human bFGF mutein N41; this is carried by the plasmid pTB855 obtained in Example 40. The figure also shows the amino acid sequence of the N41 mutein encoded by the above base sequence. Mutated bases are underlined.

Figure 58 is a schematic representation of the construction of plasmid pTB856 according to Example 41.

Figure 59 shows the base sequence encoding human bFGF mutein C129; this is carried by the plasmid pTB856 obtained in Example 41. The figure also shows the amino acid sequence of the C129 mutein encoded by the above base sequence. The mutated bases are underlined.

Figure 60 is a schematic representation of the construction of plasmid pTB831 according to Example 42.

The invention is further illustrated by the following Reference Examples and Examples. In the methods described herein, the enzymes were used according to the manufacturer's or distributor's specifications. Operations not specified in detail are Maniatis

HU 204089 Β et al .Molecular Cloning — Alaboratory

Maual ”Clod SpringHarbor Laboratory Press, N.Y.,

We did it in the USA 1982 handbook.

Reference Example 2 (Construction of a plasmid containing the gene encoding human bFGF) (1) Plasmid isolation containing cDNA:

The cDNA, hosted by Escherichia coli x 1776 and inserted into the living pCD vector by introduction of the mRNA from a primary fortification of human foreskin (see Okayama et al., Molecular Cell Cell Biology, 1983, 5,280), Dr. It was obtained from Okayama (National Ihstitute of Child Health and Human Development, Betbesda, USA). Plasmid DNA was extracted from this DNA by the alkaline method [BimboimH.C. ésDoly J .: Nucleic Acids Research, 2.1513 1979)], and the DNA was infected into Escherichia coli DHL, thus obtaining a cDNA library was prepared from about 10.2 kiónját ^6, amelynekgazdaszervezeteEscherichiacoliDHl.

The above cDNA library using Escherichia coliDH1 was plated on 10 nitrocellulose filters (Mfllipore Inc., United States of America; HATF filters) at approximately 5.10 ⁴ clones / filter, and a total of 20 replica filters were made in pairs using the above 10 filters as sample filters. Escherichia coli cells on replica filters were lysed with 0.5 N NaOH and the denatured plasmid DNA was immobilized on the filters [Grunstein M. and HagnessD.S.:Prcc. Natl. Acad. Sci. USA 72,3961 (1975)].

In the amino acid sequence of bovine basic fibroblast growth factor (as described by Escb F. et al., Proc. Natl. Acad. Sci. USA 82: 6507-7507 (1985)]. The base sequences corresponding to amino acids (Pro-ProGly-ESs-Pbe-Lys-Asp-Pro) and 89-96 (Thr-Asp-Glu-Cys-Phe-Phe-Phe-Glu) were chemically synthesized. (For some codons, the third letter is arbitrarily fixed: '5'GG A / G TC T / C TT A / G AA A / G TGGCCAGGAGG above 5'TC A / G AA A / G AA A / G AA A / G CA T / C TCGTCGGT The underlined letters represent the fixed bases). Each oligonuldeotidokat bőmérséklten 37 ° C for 1 hour where indicated 50 μΐ following composition of the reaction liquid ^two winters 5 'terminus with ³² P: obgonukleotid 0.1 g, 50 mmol / l Tris-HCl φΗ 8.0), 0.1 µM MgCl ₂ , 10 mM mercaptoethanol, μα γ- ³² Ρ ATP (5000 Ci / mmol), 3 units T4 polynucleotide kinase (Takara Shuzo Co. Ltd., Japan).

Using the 2 oligonucleotides labeled by the above method as probes, they were independently linked to the replica filter to which the DNA had previously been immobilized. The coupling reaction is carried out at 5 to 35 ° C for 16 hours in a solution containing 5 x SSPE is [180 mmol / l NaCl, 10 mmol / l NaH ₂ PO 4 * ^mm EDTA OPL φΗ 7.4)], 5 x Denhardt's solution, 0, It contained 1% SDS and 100 pg / ml denatured salmon sperm DNA and 10 pCi assay samples. After completion of reaction 60, the filter was washed with 5xSSC [0.15 mol

NaCl, 0.015M sodium citrate] and 0.1% SDS were washed three times at room temperature for 30 minutes and twice at 45 ° C for 30 minutes (Maniatis T. et al., Molecular).

Cloning, Cold Spring Harbor Laboratory (1982), 309].

Radiograms of the washed filters were recorded and the strains corresponding to the two assay samples were screened for needles to see if the replica filter pairs overlap with the two radioautograms. By this method, 1 strain of 5x10 ⁵ clones hybridizing with both test samples was obtained {Escherichia coli K12

DH1 / pTB627 (EFO14494, FERMBP-1280)].

(2) extracting plasmid DNA from Escherichia coli K12 DH1 / pTB627 (IFO 14494, FERM BP1280) obtained in step (1) above and purifying it by the alkaline method (Nucleic Acids Research, 1979, 2, 1513). )].

)

Reference Example 2 (Expression of a gene encoding AhumanbFGF in Escherichia coli) (1) Construction of plasmid pTB669 for expression of human bFGF

Plasmid pTB627 containing the human bFGF cDNA obtained in Reference Example 1 (2) above was cleaved with the restriction enzymes Aval and Ball, resulting in a 0.44 kb DNA fragment containing the coding region for human bFGF. The Ball cleavage site (blunt end) of this DNA was ligated with the pCAGATCTGBglU switch (linker) using T4 DNA ligase and a 0.44 kb Aval-BglUDNS fragment was isolated. This 0.44 kb Aval-BglH DNA fragment was digested with T4 DNA, so that the Bgl splice sites were ligated to one another, and the Aval cleavage site was blunt-ended using DNA polymerase (Klenow fragment) in the presence of dXTPs. After phosphorylation of this DNA fragment, the 5 oligonucleotides 5AATTCTATGCCAGCATTGC3 'and 5'GCA ATGCTGGCATAG3' were ligated using T4 DNA ligase and then cleaved using EcoRI and BglU to give a DNA fragment of about 0.46 kb. Plasmid ptrp781 DNA [Kurokawa T. et al., Nucleic

Acids Research, 22, 3077-3085 (1983)], which has a trp promoter, was cleaved using PstI and blunt-ended by T4 DNA polymerase reaction. The pCAGATCTGBgin linker was ligated to the blunt end using T4DNA ligase, and then cleaved using EcoRI and BglH to give a

A 3.2 kb DNA fragment containing the trp promoter, the tetracidin resistance gene, and a plasmid replication initiation site was isolated. The aforementioned 0.46 kb EcoRI-BglH DNA fragment containing the gene region encoding human bFGF and this 3.2 kb DNA fragment were ligated together to give plasmid pTB669 which is useful for expression of human bFGF.

HU 204089 Β

Using this plasmid pTB669, Escherichia coli DH1 was transformed to give Escherichia coli DH1 / pTB669, which contains plasmid pTB669.

In addition, using pTB669, strains of Escherichia coli K12 MM294 or C600 were transformed in a manner similar to that described above, whereby Escherichia coli K12 MM294 / pTB669 (DFO 14532, FERM BP-1281) outperformed E. coli C600 / pTB669. strains.

(2) Preparation of bacterial cell extracts

All of the above transformants were cultured in M9 medium containing 1% glucose, 0.4% casinoic acid and 8 pg / ml tetracycline, and when the Klet value was approximately 200, 3 β-ß-diacyloxyacrylic acid was added at a concentration of 25 pg / ml. and continued for another 4 hours. (Clett value: KLETT-SUMMERSON photoelectric colorimeter, filter 54) After culturing, bacterial cells were harvested and suspended in 10% sucrose solution containing 1/20 of 20 mM TrisHCl ( pH 7.6). To this suspension was added phenylsulfonyl fluoride (PMSF) to a concentration of 1 mM, EDTA to a concentration of 10 mM, NaCl to a concentration of 0.1 M, spermidine hydrochloride to a concentration of 10 mM, and lysozyme at 100 µg. / ml (all numbers represent final concentration) and the mixture was maintained at 0 ° C for 45 minutes and then sonicated for 30 seconds. This solution was centrifuged at 18,000 rpm (Sorval centrifuge, SS34 rotor) for 30 minutes to provide a supernatant that could be used as a bacterial cell extract.

(3) Bacterial Cell Trains Human BFGF Activity kThe activity values of human bFGF are expressed as the weight of standard samples of purified bovine brain FGF (product of Takara Shuzo Co., Ltd.) equivalent to the growth of bacterial cell extracts on BALB / c3T3 cells.

Mouse BALB / c3T3 cells were inoculated with 2.10 ³ cells / well in DMEM medium containing 5% calf serum on a Nunc 96-well microtiter plate (flat bottom) containing 0.2 ml of medium and cultured. The following day, the medium was changed to DMEM containing 0.5% bovine serum.

After culturing for 10 days, 10 µl of bacterial cell extract, previously serially diluted in DME medium containing 0.5% BASA in 5 steps, was added to each well and cultured. Twenty hours later, 2 µl of ³ H-Tdr (5 Ci / mmol, 0.5 mCi / mL. RCC Amersham, UK) was added to each well. After 6 hours, the cells were resuspended in treatment with 0.2% trypsin and 0.02% EDTA in PBS and harvested on a glass filter using a Titertech cell harvester, and the amount of ³ HTdr absorbed by the cells was counted on a scintillation counter. Similarly, the activity of bovine brain FGF samples of known weight (Takara Shuzo product) was also determined. The resulting curve was used to calculate the amount of human bFGF in the bacterial cell extract samples. The results are shown in Table 2.

As a control, the productivity of E. coli DH1 / ptrp7881 transformant obtained by transformation of Escherichia coli DH1 using plasmid ptrp781 was determined.

Table 2

Ability to produce human bFGF

transformant Human bFGF production (per liter of culture medium) E.coliDHl / pTB669 2.95 mg E.coliMM294 / pTB669 23.15 mg E.coliC600 / pTB669 8,15mg E.coliDHl / ptrp781 1.0005

Example 1 (Preparation of recombinant DNAs containing a base sequence encoding mutein) (1) Cloning of the M1 vector of the bFGF gene

In Example 2, plasmid pTB669 was obtained

It was digested with restriction enzymes EcoRI and BamHI. The replicative (RF) DNA of phage M1mp3mp8 vector (Messing J. Methods in Enzymology 101, 20-78 (1983)) was digested with EcoRI and BamHI and mixed with the human bFGF DNA fragment from digested pTB669. . The mixture was then ligated together using T4 DNA ligase, and the ligated DNA was transformed into an infectious cell of Escherichia coli strain JM105; the resulting transformant was inoculated onto a plate which used Xgal as an indicator (Messing, J. et al., Nucleic Acids Research, 1981, 9309-321), and the recombinant phage plaque (white plaque) was harvested; the base sequence of the recombined region was determined by the dideoxynucleotide synthesis chain termination method (Messing, J. et al., Nucleic Acids Research, 9, 309-321 (1981)), thereby demonstrating that human bFGF DNA was accurately inserted.

A single-stranded phage DNA was purified from this M13-PO clone, which was then used as a template for site-directed mutagenesis using a synthetic oligonucleotide.

(2) Site-specific mutagenesis

1.1 mmol of adenosine triphosphate (ATP), 50 mM iris [(hydroxymethyl) aminomethane hydrochloride, trisHCl, pH 8.0], 10 mM MgCl ₂ , 5 mM dithiothreitol (DTT) ) and 9 units in the presence of T4 kinase in a total volume of 50 µl to 40 picomoles

5 'CGTTCTTGCTGTAGAGCCGCT 3' Synthetic OligonWeotide [26. Primary for the conversion of Cys to Ser (the recognition site for the Rsal restriction enzyme disappears) (see

Figure 2)] treated with T4 kinase at 37 ° C for 1 hour. The kinase-treated primer (12 picomoles) was 67 'C

EN 204089 Β for 25 minutes in 50 µl of a mixture of 50 mM NaCl, 1.0 mM TrisHCl pH 8.0, 10 mM MgCl ₂ and 10 mM Contains 1 β-mercaptoethanol, hybridized to 5 µg single-stranded (ss) M13-PO DNA. The brazed mixture was then chilled on ice and added to 50 µl of a reaction mixture containing 0.5-0.5 mM deoxynuldeothide phosphates (dNTP), 80 mM TrisHC (pK = 7.4), 8 mM MgCl _2. 100 mM] / lNaCl, 9 units DNA polymerase Klenow fragment, 0.5 mM ATP and 2 units T4 DNA gas, and the reaction at 37 ° C for 3 hours and then at 25 ° C It was continued for 2 hours then quenched by addition of 2 µl of 0.2 mM EDTA. The reaction mixture was used to transform infectious JM105 cells, and the resulting transformant cells were grown overnight, and ssDNA was then isolated from the culture supernatant. This ssDNA was used as a template for primer extension2. cycle and agel purified RE-type DNA was transformed into infectious JM105 cells; the resulting transformant cells were plated on the surface of an agar plate and cultured overnight to obtain phage plaques.

(3) Site-directed mutagenesis

The procedure described in section (2) above was repeated, but

The 5 'AACGATTAGCGCTCACTCC 3' oligonucleotide primer was used, which modifies the codon encoding cysteine 70 to be encoded (a recognition sequence of the restriction enzyme AHaeH was prepared) / see Fig. 2. Figure /.

(4) Site-directed mutagenesis

The sessions described in Chapter (2) above were repeated, but ^;

A 5 'GTAACAGACTTAGAAGCTAGT 3' oligonucleotide primer was used which modifies the codon encoding cysteine 88 to encode it (We generated a recognition sequence for the restriction enzyme Alul) (see Figure 2). ² (5) Site-directed mutagenesis

The procedure described in section (2) above was repeated, deaz

The 5 'TCGAAGAAGAAAGACTCATCC 3' oligonucleotide primer was used, which modifies the codon encoding cysteine 93 to encode it (a recognition sequence of the restriction enzyme Hinfl was prepared) (see Figs. 2).

(6) Screening and identification of amputated plaques S

Plates containing the mutated M13-PO0 plaques [Chapter 1 above] and two plates containing the non-mutated M13-POphage plaque were cooled to 4 ° C and transferred from each plate to two round nitrocellulose filters by placing a 5 dry filter. on the agar plate for 5 minutes for the L filter and 15 minutes for the 2 filters. The filters were then immersed in a solution of 0.2 N NaOH, 1.5 M NaCl, and 0.2% Triton Χ-100 for 5 minutes, and then sieved so that for a further 5 minutes, 0 Was placed on a filter paper immersed in a solution containing 5M Tris-HCl (having a pH of 7.5) and 1.5M NaCl. The filters were washed twice with 2xSSC (standard sodium citrate) dipped filters in the same manner and allowed to dry, then dried at 80 'for 2 hours in a vacuum oven. The superimposed filters were prehybridized at 55 ° C for 4 hours at 10 ml / filter

DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and bovine serum albumin, 1x4), 02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM / l of sodium phosphate buffered saline (which has a nekpH value of 7.0) and 100 pg / ml of denatured salmon sperm DNA. Hybridization was carried out at 42 ° C for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. The filters were individually washed with wash buffer containing 0.1% SDS and reduced amount of SSC at 50 ° C for 30 minutes. The filters were then washed first with 2xSSC buffer; the control filters, which contained unmutated M13P0 plaques were examined for radioactivity, Geiger counter ^Type: eluant. While decreasing the SSC concentration step by step, the control filters were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to air dry and autoradiograms were taken with 2-3 days exposure at -70 ° C. Screening was performed with 10,000 mutated Μ13Ρ0 holders and 100 non-mutated control plaques using a kinase-treated oligonucleotide probe. None of the control plaques, while 3-10 mutated M13-P0 plaques hybridized to the assay sample.

One of the mutated M13-PO plaques was taken and inoculated into JM105 culture medium. SsDNA was prepared from the supernatant and double-stranded (ds) DNA was prepared from the bacterial cell pellet. Analysis of the base sequences was performed using the appropriate oligonucleotide primers and ssDNAs.

As a result, the TGC (26.Cys) codon was found to be AGC (Ser); TGT (70.Cys) codon to AGC (Ser) codon; the TGT (88 Cys) codon to the TCT (Ser) codon; and the TGT (93.Cys) codon changed to TCT (Ser).

From the mutated M12-PO phage, the phage in which codon 26Cys became Ser coding for Ser was designated M13-P1; the phage in which the codon 7O.Cys has become a Ser coding codon, M13-P2; the phage in which codon 88 Cys has become a codon for Ser, M13-P3; and the phage in which codon 93Cys has become a codon for Ser is called M13-P4.

Example 2 (Expression of human hFGF mutein coding gene in Escherichia coli) (1) Construction of plasmid pTB739 for expression of human hFGF mutein

HU 204089 Β

The replicative form (RF) of M14-P1 obtained in Example 1 above was cleaved with the restriction enzymes EcoRI and PstI to give a DNA fragment of about 0.5 kb containing the coding region for human bFGF mutein.

Plasmid ptrp781 DNA containing a trp promoter (Kurokawa T. et al., Nucleic Acids Research, 11: 3077-3085 (1983)) was cleaved with EcoRI and PstI and separated into a

A 3.2 kb DNA fragment containing the trp promoter, a tetracycline resistance gene and a plasmid replication initiation site. The 0.5 kb EcoRI-Psü DNA fragment containing the above-mentioned human bFGF mutein coding region and this

The 3.2 kb DNA fragment was ligated together using T4 DNA ligation to create plasmid pTB739 for expression of human bFGF mutein (Figure 7).

using this plasmid pTB739, Escherichia coliDH1 was transformed to give Escherichia coli DH1 / pTB739 (EFO14575, FERMBP-1641), which contains the plasmid pTB739 containing the mutein coding gene shown in Figure 3.

(2) Preparation of bacterial cell extract

The aforementioned transformant was grown in M9 medium containing 1% glucose, 0.4% lasamic acid, and 8 pg / ml tetracycline, and when the Klett value was 200, Sp-indolyl acrylic acid was added until a concentration of 25 pg / ml was reached. Cultured for 4 hours. After culturing, the bacterial cells were harvested and suspended in 1/20 of 10% sucrose solution containing 20 mM Tris-HCl, pH 7.6. To this suspension was added phenylmethylsulfonyl fluoride to a concentration of 1 mM, EDTA to a concentration of 10 mM, NaCl to a concentration of 0.1 M, spermidine hydrochloride to a concentration of 10 mM and lysozyme to a concentration of 100 pg / ml ( each number represents the final concentration) and the mixture was allowed to stand at 0 ° C for 45 minutes and then subjected to ultrasound treatment for 30 seconds. The solution was centrifuged at 18000 rpm (Saline centrifuge, SS34 rotor) for 30 minutes to give a supernatant which was then used as bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

Mouse BALB / c3T3 cells were inoculated with 2.10 ³ cells / well in DMEM medium containing 5% fetal bovine serum on Nunc 96-well microtiter plates (flat bottom), each containing 0.2 ml of medium and cultured. The next day, the medium was changed to DMEM containing 5% bovine serum. After culturing for 3 days, 10 µl of bacterial cell extract, previously serially diluted in DME medium containing 0.5% BSA, was added to each well in 5 steps and cultured. Twenty hours later, 2 µl of ³ H-Tdr (5 Ci / mmol), 0.5 mCi / mL, RCC, Amersham were added to each well. After 6 hours, cells were resuspended in phosphate buffered saline containing 0.2% trypsin and 0.02% EDTA and harvested on a glass filter using a Titertech cell harvester, and the amount of ³ H-Tdr taken up by the cells was determined by scintillation counting.

We have determined thatE. coli DH1 / pTB739 bacterial cell extracts show FGF activity.

In this way, mutein CS1 was obtained, in which human bFGF was replaced with Cys Ser at position 26.

Example 3 (Expression of a gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB742 for expression of human bFGF mu

The replicative form (RF) of M13-P2 obtained in Example 1 was cleaved with restriction enzymes EcoRI and PstI to give a DNA fragment of about 0.5 kb containing the coding region for human bFGF mutein.

Plasmid ptrp781 DNA containing the trp promoter was cleaved with EcoRI and PstI to isolate a 3.2 kb DNA fragment containing the trp promoter, the tetracycline resistance gene and a plasmid replication initiation site. This 3.2 kb DNA fragment and the above-mentioned 0.5 kb EcoRI-Psü DNA fragment containing the gene encoding human bFGF mutein were ligated together with T4 DNA ligase to construct plasmid pTB742 for expression of human bFGF mutein ( Figure 8).

Using this plasmid pTB742, Escherichia coli DH1 was transformed to give Escherichia coli DH1 / pTB742 (EFO 14584, FERM BP1642), which contains the plasmid pTB742 containing the mutein coding gene shown in Figure 4.

(2) Preparation of bacterial cell train

The above transformant was cultured by the method described in Example 2 (2) to give a supernatant which was then used as bacterial cell extract.

GV (3) The bacterial cell extract is the activity of human bFGF

GB

The determination of AhumanbFGF activity in the bacterial cell extracts obtained in (2) above was carried out according to the method described in Example 2 (3).

E. coli DH1 / pTB742 was found to exhibit FGF activity.

In this way, the CS2 mutein was obtained in which Cys Ser at position 70 of human bFGF was replaced.

Example 4 (Expression of human bFGF mutein coding gene in Escherichia coli)

I

HU 204089 Β (1) Construction of plasmid pTB743 for expression of human hFGF mutein

The replicative form (RF) of M13-P3 obtained in Example 1 was cleaved with the restriction enzymes EcoRI and PstI, thereby obtaining a DNA fragment of about 0.5 kb containing a silyl encoding human bFGF mutein.

Plasmid ptrp781 containing a trp promoter

DNA was cleaved with EcoRI and PstI to isolate an approximately 3.2 kb DNA fragment containing the trp promoter, tetracycline resistance gene and a plasmid replication initiation site This 3.2 kb DNA fragment and the above-mentioned 0.5 kb EcoRI-PstI DNA fragment containing the gene region encoding the human hFGF mutein was ligated together with T4 DNA ligase to create plasmid pTB743 for expression of human bFGF mutein (Figure 9).

Using this plasmid pTB743, Escherichia mii DH1p was transformed to give Escherichia coli DHI / pTB743 (IFO 14585, FERM BP1643), which contains plasmid pTB743 containing the mutein coding gene shown in Figure 5.

(2) Bacterial cell extraction

The above-mentioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human hFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

E. coli DH1 / pIB743 bacterial cells were found to exhibit FGF activity.

Thus, the CS3mutein is obtained in which it is replaced by the Cys Ser at position 88 of human bFGF.

the

Example 5 (Expression of gene encoding human bFGF mutein in E.coli) (1) Construction of human pTB744plazmid bFGFmuteinkffejeződésére ^z

The replicative form (RF) of M13-P4 obtained in Example 1 was cleaved with restriction enzymes EcoRI and PstI to give a DNA fragment of about 0.5 kb containing the coding region for human bFGF mutein.

Plasmid ptrp781 DNA containing the trp promoter was digested with EcoRI and PstI, such as an approximately 3.2kb sDNA fragment containing the trp promoter, a tetracycline resistance gene and a plasmid relational initiation site. 5 This 3.2 kb DNA fragment and the above-mentioned 0.5 kb EcoRI-Psü DNA fragment containing the human bFGF mutein gene region were ligated together with T4 DNA ligase to generate pIB744 plasmidothuman bFGF mutein 6 (10 . figure).

Using this plasmid pIB744, Escherichia coli DH1 was transformed to give Escherichia coli DH1 / pTB744 (DFO 14586, FERM BP51644), which contains the strain 6.

Fig. 6B shows a plasmid pIB744 containing the mutein coding gene shown in Figure 1c.

Preparation of GV (2) Bacterial Cell Extract kThe transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as bacterial cell extract.

(3) The bacterial cell extract is the activity of human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

E. coli DH1 / pTB744 bak0 was found to exhibit FGF activity in bacillus.

This results in a modified CS4 mutein substituted with Cys Ser at position 93 of human bFGF.

Example 6 (Screening and Identification of Amutagenized Plaques) As described in Examples 1 and 2, plates containing the mutated phage M13-P2 plaques obtained with the oligonucleotide of Figures 2 (1), 2 (3) and 2 (4) and 2 plates containing the non-mutated phage M13P2 plaques obtained in Example 1 were cooled to 4 ° C and transferred from each plate to a 2 round nitrocellulose filter so that a dry filter was placed on the needle plate for 5 minutes for filter 1 and 15 minutes. In the case. Thick were placed on filter paper immersed in a solution containing The filters ezután5percre 0,2NaOH HCl, 1.5 mol / lNaCi, and 0.2% Triton-100 Χ, then neutralized with ¹ such that a further 5 minutes 0.5 mólZl triszHCI- (pH 7.5) and 1.5 M NaCl in immersion filter paper. The filters were washed twice with 2xSSC (standard sodium citrate) saved filters in the same way and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The superimposed filters were prehybridized at 55 'temperature for 4 hours with 10 ml / filter DNA hybridization buffer solution (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll and bovine serum albumin, 1x ), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered saline (pH 7.0) and 100 pg / ml denatured salmon sperm DNA. Hybridization was carried out at 42 ° C for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. The filters were individually washed with wash buffer containing 0.1% SDS and a reduced amount of SSC at 30 ° C for 30 minutes. The filters are then washed first with 2xSSC buffer; the

HU 204089 Β control filters containing non-mutated M13-P2 plaques were tested for radioactivity using a Geiger counter. While decreasing the SSC concentration step by step, the control filters containing the non-mutated M13-P2 tarfolds were washed until detectable radioactivity remained on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to air dry and autoradiograms were taken with 2-3 days exposure at -70 ° C. Screening was carried out in 1000 mutated M13-P2 plaque and 100 unmutated control plaque using the kinased oligonucleotide probe γ ³² P-ATP. None of the control plaques, while 3-10 mutated M13-P2 plaques hybridized to the assay sample.

One of the mutated M13-P2 plaques was taken up and inoculated into JM105 culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Abase sequences were analyzed using appropriate oligonucleotide primers and ssDNAs.

We found that the TGC (26.Cys) codon changed to TCT (Ser), the TGT (88Cys) codon to TCT (ser), and the TGT (93.Cys) codon to TCT (Ser).

From the mutated M13-P2 phage, the phage in which codons 26 and 70 of Cys became Ser coding for Ser was designated M13-P12; the phage in which the Cys codons 70 and 88 changed to the coding Ser for M13P23 and the phage in which the Cys codons 70 and 93 became the coding for Ser in M13-P24.

Example 7 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB762 for expression of human bFGF mutein

The replicative form (RF) of M13-P23 obtained in Example 6 above was treated in the same manner as described in Example 2 (1) to construct plasmid pTB762 for expression of human bFGF mutein (Figure 12).

Using this plasmid pTB762, Escherichia coli MM294 was transformed to give Escherichia coli MM294 / pTB762 (IFO 14613, EERM BP1645), which contains all. Figure 1 shows a plasmid pTB762 containing the mutein coding gene shown in Figure 6A.

(2) Preparation of bacterial cell extract

The above-mentioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

The cell extract of E. coli MM294 / pTB762 was found to exhibit FGF activity, thereby obtaining the CS23 mutein in which Cys at positions 70 and 88 was replaced by Ser.

Example 8 (Screening and identification of mutagenized plaques)

By the method of Examples 1 and 2, 2 (1) and 2 (1) respectively. 2 (4) plates containing the mutated phage M13-P23 phage plaques and 2 plates containing the non-mutated phage M13-P23 phages obtained in Example 7 were cooled to 4 ° C and transferred from each plate to 2 round nitrocellulose filters. by placing a dry filter on the agar plate for 5 minutes for filter 1 and 15 minutes for filter 2. The gray was then placed on a filter paper immersed in a solution of 0.2 N NaOH in 1.5 M NaCl for 5 minutes. The filters were washed twice with filters similar to 2xSSC (standard sodium citrate) and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The successive filters were pre-hybridized at 55 ° C for 4 hours with 10 ml / filter DNA hybridization solution (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and bovine serum albumin, lx = 0, 02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered saline (pH 7.0) and 100 μg / ml denatured salmon sperm DNA. Hybridization was performed at 42 'Con. 24 hours each with 10 ⁵ cpm / ml oligonucleotide primer. The filters were washed individually at 50 ° C for 30 minutes with wash buffer containing 0.1% SDS and a reduced amount of SSC. The filters were then washed first with 2xSSC buffer; control filters containing non-mutated M13-P23 plaques were assayed for radioactivity using a Geiger counter. As the SSC concentration was reduced step by step, the control filters were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to air dry and autoradiograms were taken with a 2-3 day exposure at -70 ° C. Screening was performed on 10,000 mutated M13-P2 plaques and 100 non-mutated control plaques using a kinase-treated oligonucleotide assay in the presence of γ- ³² P-ATP. None of the control plaques, while 3-10 mutated M13-P23 plaques hybridized to the assay sample.

One of the mutated M13-P23 plaques was picked and inoculated into JM105 culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Analysis of the base sequences was performed, a

I

HU 204,089 Β using appropriate oligonucleotide primers and ssDNAs.

We found that the TGC (Cys 26) codon was a TCT (Ser) codon; the TGT (93 Cys) codon changed to TCT (Ser).

The mutated M13-P23 phage snow phage, in which the Cys codons 26, 70 and 88 became Ser codons, M13-P123, and the phage where the Cys codons 70, 88, and 93 were Ser- t coding codons, called M13-P234.

Example 9 (Expression of a gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB764 for expression of human bFGF mutein

The replicative form of M13-P123, which is referred to above

Example 8 was obtained, treated in the same manner as described in Example 2 (1), to express the plasmid pTB764 plasmidothuman bFGFmutein (Figure 14).

Using this plasmid pTB764, Escherichia coli MM294 was transformed to give Escherichia coli MM294 / pTB764 (IFO 14613, FERM BP1645), which contains

The mutant encoding gene shown in Figure 13 contains the TB762p plasmid.

(2) Preparation of bacterial cell extract

The aforementioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as bacterial cell extract.

(3) BFGF activity of bacterial cell extracts

The determination of the human bFGF activity of the bacterial cell extract obtained in (2) above was performed according to the method described in Example 2 (3).

E. coli MM294 / pTB764 was found to show FGF activity in cellular extracts.

In this manner, the mutein CS123 was obtained, ^o wherein the Cys is substituted with Ser above the 26th, 70th, and 88th site.

1o. Example ² (Screening and Identification of Amutagenic Stains ² )

By the method of Examples 1 and 2, the plates containing the mutated phage M13-P123 and Figure 21 (2) containing the phage M13-P123 mutant obtained in Example 8 were cooled to 4 ° C and the plaques were transferred from each plate to 2 round nitrocellulose filters by placing a dry filter on the agar plate for 5 minutes for Filter 1 and 15 minutes for Filter 2 δ. The filters were then treated with 0.2 N NaOH for 5 minutes,

It was placed on thick filter paper saved in a solution containing 1.5 M NaCl and 0.2% Triton Χ-100, then neutralized so that for an additional 5 minutes, 0.5 M Tris-HCl (pH 7.5) and , Placed in a filter paper immersed in a solution containing 5 mM NaCl-6,

The filters were washed twice with 2xSSC (standard sodium citrate) dipped filters in the same way and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The overlapping filters were prehybridized at 55 ° C for 4 hours with 10 ml / filter DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone,

Ficoll, and bovine serum albumin (1x-0.02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered saline (pH 7.0), and 100 μg / mI contained denatured salmon sperm DNA. Hybridization was performed at 42 ° C with 15 needles for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. The filters were washed individually at 50 ° C for 30 minutes with wash buffer containing 0.1% SDS and a reduced amount of SSC. The filters were then washed first with 2xSSC buffer; Control filters that contained no mutated M13-P123 plaques were assayed for radioactivity using a Geiger counter. While decreasing the SSC concentration step by step, the control filters containing unmutated M13-P123 plaques were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to air dry and autoradiograms were taken with a 2-3 day exposure at -70 ° C. Screening was performed with 10,000 mutated M13P123 plaques and 100 non-mutated control plaques using a kinase-treated oligonucleotide probe in the presence of γ- ³² Ρ-ΑΤΡ. None of the control plaques, while 3-10 mutated M13-P123 tar5 hybridized to the assay sample.

One of the mutated M13-P123 plaques was recorded and

JM105 was inoculated into culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Base sequence analysis was performed using appropriate oligonucleotide primers and ssDNAs.

We found that the TGT (93.Cys) codon changed to a TCT (Ser) codon.

• The mutated phage snowflake M13-P123, in which codons 26, 70, 88, and 93 became Cys coding for Ser, was designated M13-P1234.

Example 11 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB765 for expression of human bFGF mutein

The replicative form (RF) of M13-P1234, obtained in Example 10 above, was started in the same manner as described in Example 2 (1) to construct plasmid pTB765 for expression of human bFGF mutein (Figure 16).

Using this plasmid pTB765, Escherichia coli MM294 was transformed to produce Escheri18

HÜ 204089 Β chia coli ΜΜ294 / ρΤΒ765 (ΠΌ 14615, FERM ΒΡ1647) was obtained, which includes

Figure 15 shows the plasmid pTB765 containing the mutein coding gene shown in Figure 15.

(2) Preparation of bacterial cell extract

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) Activation of bacterial cell extract by human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3) below.

The cell extract of E. coli MM294 / pIB765 was found to exhibit FGF activity.

In this way, CS124 mutein was obtained in which Cys at positions 26, 70, 88 and 93 was replaced by Ser.

Example 12 (Screening and identification of mutagenized plaques)

The mutation was performed as described in Example 1, using the oligonucleotides of Figure 2 as oligonucleotides.

Plates containing the mutated phage M13-P1 phages using the method of Example 1 using the synthetic oligomer [Figure 2 (3) or (4) and 2 plates containing the non-mutated phage M13-P1 phages obtained in Example 1 at 4 ° C. cooled and the plaques were transferred from each plate to the round nitrocellulose filter 2 by placing a dry filter on the agar plate for 5 minutes for filter 1 and 15 minutes for filter 2. The filters were then placed on thick filter paper immersed in a solution containing 0.2 N NaOH, 1.5 M NaCl, and 0.2% Triton Χ-100 for 5 minutes, then neutralized by incubation for 0.5 minutes. mole Tris-HCl, pH 7.5;

It was placed on a filter paper immersed in a solution containing 1.5 M NaCl. The filters were washed twice in 2xSSC (standard sodium citrate) immersed filters in the same way and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The successive filters were pre-hybridized at 55 ° C for 4 hours with 10 ml / filter DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and bovine serum albumin, lx = 0, 02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered solution (pH 7.0) and 100 pg / ml denatured salmon sperm DNA. Hybridization was performed at 42 ° C for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. The filters were individually washed with wash buffer containing 0.1% SDS and reduced SSC at 50 ° C for 30 minutes. The filters were then washed first with 2xSSC buffer; control filters containing non-mutated M13-P1 plaques were assayed for radioactivity using a Geiger counter. While step by step reducing the SSC concentration, control filters containing anemutilized M13-P1 plaques were washed until detectable radioactivity ceased on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to air dry and autoradiograms were taken with 2-3 days exposure at -70 ° C. Screening was performed with 10,000 mutated M13-P1 plaques and 100 non-mutated control plaques using a kinase-treated oligonucleotide probe in the presence of γ- ³² Ρ-ΑΤΡ. None of the control plaques hybridized to the assay while 3-10 mutated M13-P1 plaques hybridized to the assay.

All mutated M13-P1 plaques were harvested and inoculated into JM 105 culture medium. SsDNA was prepared from the supernatant thus prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Base sequences were analyzed using appropriate oligonucleotide printers and ssDNAs.

We found that the TGT (88Cys) codon changed to TCT (Ser) and the TGT (93.Cys) codon changed to TCT (Ser).

From the mutated phage M13-P1, the phage in which codons 26 and 88 of Cys have become coded for Ser is M13P13; and the phage in which codons 26 Cys and 93 Cys became Ser coding for Ser was designated M13-P14.

Example 13 (Screening and identification of mutagenized plaques)

The mutation was carried out as described in Example 1, using oligonucleotides of Figure 2 as oligonucleotide.

Example 1 shows the plates containing the mutated phage M13-P14 phage and 2 plates containing the non-mutated phage M13-P14 phage obtained in Example 1 using the synthetic oligomer [Figure 2 (2) or (3)]. After cooling to C, the plaques were transferred from each plate to a 2-round nitrocellulose filter by placing a dry filter on the agar plate for 5 minutes for Filter 1 and 15 minutes for Filter 2. The filters were then placed on thick filter paper immersed in a solution containing 0.2 N NaOH, 1.5 M NaCl, and 0.2% Triton Χ-100 for 5 minutes, then neutralized by further 0 min. , Placed on a filter paper immersed in a solution containing 5 M Tris-HCl pH 7.5 and 1.5 M NaCl. The filters were washed twice with 2xSSC (standard sodium citrate) dip filters in the same manner and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The superimposed filters were subjected to perhybridization at 55 ° C for 4 hours with 10 ml / filter DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and deer19).

EN 204089 Β bovine serum albumin, 1x = 0.02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered saline (pH 7.0) and 100 pg / ml denatured salmon sperm DNA. Hybridization was performed at 42 ° C for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. The filters were individually washed with wash buffer containing 0.1% SDS and reduced SSC at 50 ° C for 30 minutes. The filters were then washed first with 2xSSC buffer; control filters containing non-mutated M13-P14 plaques were assayed for radioactivity using a Geiger counter. While step by step reducing the SSC concentration, the control filters containing non-mutated M13P14 plaques were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is 0.1x SSC. The filters were allowed to air dry and autoradiograms were taken with a 2-3 day exposure at -70 ° C. Screening was performed with 10,000 mutated M13-P14 plaques and 100 non-mutated control plaques using a kinase-treated oligonucleotide probe in the presence of γ- ³² Ρ-ΑΤΡ.

None of the control plaques, while 3-10 mutated

M13-P14 hybridized to the assay sample.

One of the mutated M13-P14 plaques was recorded and

JM105 was inoculated into culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Abase sequences were analyzed using appropriate oligonucleotide primers and ssDNAs.

We found that the TGT (7O.Cys) is a codon

TCT (Ser) codon and TGT (93.Cys) codon changed to TCT (Ser) codon

From the mutated phages M13-P14, the phage in which codons 26, 70 and 93 became Cys coding for Ser was M13-P124, and the phage in which codons 26, 88 and 93 became Ser. Cys codon has become a codon for Ser, designated M13-P134.

Example 14 (Identification and Amutagénné tetttarföltokátvizsgálása) ²

The mutation was performed as described in Example 1, using the oligonucleotides shown in Figure 2 as oligonucleotides.

The plates containing the mutated M13-P3 phage patches using the synthetic oligomer Example 1 (Figure 2 (4)) and the plate containing the non-mutated M13-P3 phage plaques obtained in Example 1 were cooled to 4 ° C and the plaques transferred from each plate to 2 round nitrocellulose filters by placing a dry filter on 55 agar plates for 5 minutes for true filter 1 and 15 minutes for filter 2.

The filters were then placed in immersed thick filter paper in a solution of 0.2 N NaOH, 1.5 M NaCl, and 0.2% Triton Χ-100 for 5 minutes, then neutralized by further 0 min. The filters were washed twice with 2xSSC (standard sodium5 citrate) in filters in the same manner and allowed to dry for 80 '. It was dried at C for 2 hours in a vacuum oven. The superimposed filters were subjected to perhybridization at 55 ° C for 4 hours with 10ml / filter DNAhydrogenation buffer (5xSSC) having a p value of 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and bovine serum, It contained 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered solution (having a P value of 7.0), and 100 pg / ml denatured salmon sperm DNA. Hybridization was performed at 42 ° C for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. The filters were washed individually at 50 ° C for 30 minutes with wash buffer containing 0.1% SDS and reduced SSC. The filters were then washed first with 2xSSC buffer, the control filters containing non-mutated M13-P3 plaques were assayed for radioactivity using a Geiger counter. While decreasing the SSC concentration step by step, the control filters containing non-mutated M13-P3 plaques were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to air dry and autoradiograms were taken with 2-3 days exposure at -70 ° C. Screening was performed on 10,000 mutated M13-P3 plaques and 100 non-mutated control plaques using a kinase-treated oligonucleotide assay in the presence of γ- ³² P-ATP. None of the control plaques hybridized to the assay while 310 mutated M13-P3 plaques hybridized.

One of the mutated M13-P3 plaques was picked and inoculated into JM105 culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Analysis of the base sequences using the appropriate oligonucleotide primer and ssDNA was performed.

We found that the TGT (93.Cys) codon changed to a TCT (Ser) codon.

Among the mutated M13-P3 phages, the phage in which codons 88 and 93 became Cys coding for Ser was designated M13-P34.

Example 15 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB776 for expression of human bFGF mutein

The replicative form (RF) of M13-P12 obtained in Example 6 above was started in the same manner as described in Example 2 (1) to construct plasmid pTB776 for expression of human bFGF mutein (Figure 18).

Using this plasmid pTB776, Escherichia

HU 204089 Β coli MM294 was transformed to give Escherichia coli MM294 / pTB776 strain containing the plasmid pTB766 containing the mutein coding gene shown in Figure 17.

(2) Preparation of bacterial cell extract

The above-mentioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) Human bFGF activity of bacterial cell train

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

The cell extract of E. coli MM294 / pTB776 was found to exhibit FGF activity.

In this way, the CS12 mutein was obtained in which Cys is replaced by Ser at positions 26 and 70 of human bFGF.

Example 16 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB 778 for expression of human bFGF mutein

The replicative form (RF) of M13-P24 obtained in Example 6 above was treated in the same manner as described in Example 2 (1) to construct plasmid pTB778 for expression of human bFGF mutein (Figure 20).

Using this plasmid pTB778, Escherichia coli MM 294 was transformed to obtain the Escherichia coliMM294 / pTB778 strain, which contains the plasmid pTB778 containing the mutant coding gene shown in Figure 19.

(2) Preparation of bacterial cell extract

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) Human bFGF activity of bacterial cell train

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

Cell extract of E. coli MM294 / pTB778 was found to exhibit FGF activity

In this way, the CS24 mutein was obtained in which Cys is replaced by Ser at positions 70 and 93 of human bFGF.

Example 17 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB779 for expression of human bFGF mutein

The replicative form (RF) of M13-P13 obtained in Example 12 above was treated in the same manner as described in Example 2 (1) to generate plasmid pTB779 for expression of human bFGF mutein (Figure 22).

Using this plasmid pTB779, Escherichia coli MM294 was transformed to obtain Escherichia coli MM294 / pTB779, which contains plasmid pTB779 containing the mutant coding gene shown in Figure 21.

(2) Preparation of bacterial cell train

The above-mentioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

The cell extract of E. coli MM294 / pTB779 was found to exhibit FGF activity.

Thus, the CS13 mutein was obtained in which Cys is replaced by Ser at positions 26 and 88 of human bFGF.

Example 18 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pB763 for expression of human bFGF mutein

The replicative form (RF) of M13-P14 obtained in Example 12 above was treated in the same manner as described in Example 2 (1) to construct plasmid pTB763 for expression of human bFGF mutein (Figure 24).

Using this plasmid pTB763, Escherichia coli MM294 was transformed to obtain Escherichia coli MM294 / pIB763, which contains the plasmid pTB763 containing the mutant coding gene shown in Figure 23.

(2) Preparation of bacterial cell extract

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) Human bFGF activity of bacterial cell train

The human bFGF activity of the bacterial cell train obtained in (2) above was determined by the method described in Example 2 (3).

The cell extract of E. coli MM294 / pTB763 was found to exhibit FGF activity.

In this way, the CS14 mutein was obtained in which Cys is replaced by Ser at positions 26 and 93 of human bFGF.

Example 19 (Expression of gene encoding human bFGF mutein in E. coli) (1) Construction of plasmid pTB777 for expression of human bFGF mutein

The replicative form (RF) of M13-P34 obtained in Example 12 above was treated in the same manner,

204089 Β as described in Example 2 (1), was constructed to express the plasmidotha pB777 human bFGF mutein (Figure 26).

Using this plasmid pTB777, Escherichia coli MM294 was transformed to give Escherichia coliMM294 / pIB777, which contains the pIB777 plasmid containing the mutant coding gene shown in Figure 25.

(2) Preparation of bacterial cell extracts

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human hFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

E. coli MM294 / pIB777 bacterial cell extract was found to exhibit FGF activity.

In this way, the CS34 mutein was obtained in which Cys at Ser 88 and 93 was substituted for Ser h.

Example 20 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB782 for expression of human hFGF mutein

The replicative form (RF) of M13-P234 obtained in Example 8 above was treated in the same manner as described in Example 2 (1) to generate plasmid pTB782 for expression of human bFGF mutein (Figure 28).

Using this plasmid pTB782, Escherichia coli MM294 was transformed to give strain Escherichia coli MM294 / pTB777, which contains the plasmid p1B782 containing the mutein coding gene shown in Figure 27.

Preparation of bacterial cell extracts

The aforementioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) Human hFGF activity of abactor cell extract

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

It was determined that the cell extract of bacterium E. coli MM294 / pIB782 showed FGF activity.

In this way, the CS234 mutein was obtained in which Serys was substituted with Ser Rel at positions 85, 88 and 93 of Ahcb.

Example 21 (Human hFGF mutein-encoding gene expression in Escherichia coli) (1) Construction of plasmid ρΊΒ780 for expression of human bFGF mutein 60

The replicative form (RF) of M13-P124 obtained in Example 13 above was treated in the same manner as described in Example 2 (1) to construct plasmid pIB780 for the expression of human bFGF mutein (Figure 30).

Using this plasmid pIB780, Escherichia coli MM294 was transformed to give Escherichia coli MM294 / pTB780, which contains the mutein10 coding gene plasmid B780 shown in Figure 29.

(2) Preparation of bacterial cell extracts

The above-mentioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3) 20.

E. coli MM294 / pTB780 bacterial cell extract was found to show FGF activity.

This yields the CS124 mutein in which Cys is replaced by 25 Ser at arms 26, 70 and 93 of human bFGF.

Example 22 (Expression of human hFGF mutein coding gene in Escherichia coli) (1) Construction of plasmid pTB781 for expression of human bFGF mutein

The replicative form (RF) of M13-P134, obtained in Example 13 above, was started in the same manner as described in Example 2 (1) to generate 35 mega of pTB781 for expression of human hFGF mutein (Figure 3l).

Using this plasmid pIB781, Escherichia coli MM294 was transformed to give strain Escherichia coliMM294 / pTB781, which includes the mutant coding gene for plasmid TB781 shown in Figure 32.

(2) Preparation of bacterial cell extract

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

The cell extract of E. coli MM294 / pIB781 was found to exhibit FGF activity.

In this way, the CS134 mutein was obtained in which Cys was replaced by Ser at positions 26, 88 and 93 of human bFGF.

Example 23 (Mutein Productivity)

The appropriate nucleotide was determined

FGF activities of cell extracts containing HU 204089 Β by the method of Reference Example 2 (2); the mutein productivity of the corresponding transformants was calculated in this way. The value obtained is shown in Table 3.

Table 3

plasmid Mutain Productivity (PtFGF mg / 1 breeds) pTB739 esi 0.3 pTB742 CS2 10.6 pTB743 CS3 13.3 pTB744 CS4 0.8 pTB776 CS12 0.2 pTB779 CS13 0.1 p! B763 CS14 0.5 pTB762 CS23 25.8 pTB778 CS24 0.3 ΡΊΒ777 CS34 0.5 pIB764 CS123 0.3 pTB780 CS124 0.4 pTB781 CS134. 0.2 pTB782 CS234 1.4 ΡΊΒ765 CS1234 0.1 pTB669 rhbFGF 6.0 ptrp781 - 0.0

EXAMPLE 24 (Purification of hbFGF Arthritis)

Transformants producing the appropriate muteins were cultured by the method described in Reference Example 2 to obtain a bacterial cell extract. 25 mL of each extract (prepared from 500 mL culture medium) was passed through a DEAE cellulose column (02 x 10 cm; DE 52, Wattman, Inc., UK) previously treated with 20 mM Tris-HCl, pH 7.4 ) and equilibrated with 0.2 M NaCl; thus removing the nucleic acid constituents from the extract. The effluent from the column and the washings resulting from washing the column with a solution containing 20 mM Tris-HCl, pH 7.4, and 0.2 M NaCl were combined and collected (DEAE). effluent fraction, 60 ml).

This fraction was subjected to high performance liquid chromatography using a high performance liquid chromatograph (Gibson, Inc., France) packed with a Shodex AF-HR HR-894 heparin column (8 mm internal diameter x 5 cm, product of Showa Denko, Japan). The column was washed with 20 mM Tris-HCl (pH 7.4) followed by 20 mM Tris-HCl (pH 7.4) and 0.5 M NaCl, followed by linear gradient elution with a 0.5M NaCl gradient (60 mL volume, 1.0 mL / min flow rate) in buffer containing 20 mM Tris-HCl, pH 7, 4) included).

The elution curves of the corresponding muteins are shown in Figures 33-35. Figures. In these figures, the ordinates represent the OD ₂₈₀ absorbances and the NaCl concentrations. The abscissae represent time, and the gradient elution begins at the O-point of time. Peak fractions were collected and their FGF activities were determined. The specific activities of the post - cleaning are:

Table 4 shows.

Table 4

mutein Specific activity (PtFGF mg / mg protein) CS1 0.4 CS2 0.9 CS3 1.0 CS4 1.0 CSL2 0.4 CS13 0.5 CS14 0.3 CS23 1.1 CS24 0.8 CS34 0.5 CS123 0.4 CS124 0.1 CS134 0.5 CS234 0.9 CS1234 0.1 rhbFGF 1.0

Table 4 shows the specific activities based on the FGF activity (at least 95% purity) of bovine brain FGF from Takara Shuzo Co., Ltd.

In all muteins, the peak corresponding to the bFGF I tube was eluted between 15 and 25 minutes elution time. This could also be detected in the form of a single band at about 17,000 molecular weights in 17.25% SDS-polyacrylamide gel electrophoresis.

Example 25 (Oxidation of FGF mutein)

Each of the muteins eluted from the Aheparin HPLC column and the bFGF were dialyzed twice at 4 ° C for 3 hours against 1000 volumes of distilled water and then lyophilized. 200-200 (calculated as protein) dry standard sample was dissolved in 200 μΐ oxidation reaction fluid [50 mM Tris-HCl, pH 8.5; 0.2 M NaCl; 1 mg / ml BSA; 20 mM H ₂ O ₂ ] and oxidation at 37 ° C for 30 minutes.

The solution was then passed through a heparin HPLC column and the elution distribution obtained with an elution gradient of 0.6 M to 2 M salt was examined. The resulting elution distributions are shown in Figure 36. In bFGF, peaks are generally low, while in CS2, CS3, and CS23, peaks corresponding to peak I are high; the peak elevation is noticeably specific in CS3 and CS23.

This suggests that CS3 and CS23 are oxidized by eye23

HU 204089 Β ben stable. In addition, CS2, CS3 and CS3 eluted as the peak

CS23 proteins all show FGF activity.

Example 26 (FGFmutein reduction)

200-200 gg (relative to protein) of the lyophilized standard sample obtained in the same manner as in Example 25 was dissolved in 200 µl of reduction reaction fluid (50 mM Tris-HG, pH 8.5); 0.2 M NaG; 1 mg / ml BSA; 20 mmol / ϊ dithiothritol (DDT)] and reduction was carried out at 37 'for 30 minutes.

The solution was then passed through a heparin HPLC column and the distribution shown by elution with a gradient of salt from 0.6 M to 2 M. The elution distributions obtained are shown in Figure 37. For bFGF and CS2, the peak corresponding to peak I is remarkable and suggests that the remaining peaks (peaks H-IV) are due to oxidation. This also suggests that the peaks in CS3 and CS23 are not affected by oxidation. That is, it confirms the stability given in Example 25. AbFGF and the muteins eluted from these peaks were found to have FGF activity.

Example 27 (Construction of recombinant DNAs having a mutein coding base sequence) (1) Cloning of the M13 vector for the human bFGF gene

Plasmid pTB669 obtained in Reference Example 2 was digested with EcoRI and BamHI restriction enzymes. M13mp8 phage vector (Messing, J.T., Methods inEnzymology, 101, 20-78 (1983)), replicative form (RF) DNA was mixed with a human bFGF DNA fragment from pTB669 which had been previously treated with EcoRI and BamHI. The mixture was then degassed using T4 DNA ligase. The resulting ligated DNA was transformed into infectious cells of Escherichia coli strain JM105; the transformant strains were plated on a plate containing an Xgal indicator [MessingJ. and Nucleic Acids Research 9,309,321 (1981)]; plaques containing arecombinant phages (white plaques) were collected; techniques of recombinant területbáziszekvenciáját termination method was determined by the dideoxynucleotide synthesis chain [Messing, J. et munkatár- ² sai Nucleic Acids Research, 9, 309 (1981)]; in this way, human bFGF DNA was proven to be accurately inserted.

From this M13-PO Hoon, single-stranded phage DNA was purified, which was used as a template for site directed mutagenesis using the synthetic oligomer.

(2) Site-specific mutagenesis

0.1 mM adenosine triphosphate (ATP), 50 mM hydroxymethylaminomethane hydrochloride (tris-HCl) S (pH 0.8), 10 mM MgCl ₂ , 5 mM 1 dithiothreitol (DTT) and 9 units in the presence of T4 kinase, eg total volume, 40 picomoles

5'-CGGGCATGAATTCGCCGCT-3 'Synthetic OligonHeotide [Primary for θ to Recognize the EcoRI Restriction and Subsequent to Change to 14PronalMet (See Figure 38 (2)) treated with T4 Kinase at 37 ° C 1 hour. The kinase-treated pri (12 picomoles) was heated at 67 ° C for 5 minutes and then at 42 ° C for 25 minutes with 50 µl of a solution containing 50 mM NaG, 1.0 mM Tris-HG- (pH 8.0), containing 10 mM lmgG ₂ and 10 mM δ-mercaptoethanol, was hybridized to 5 pg single-stranded (ss) M13-PO0 DNA. The soldered mixture was then cooled on ice and added to 50 µl of reaction mixture containing 0.5-0.5 mM deoxynehyde triphosphates (dNTP), 80 mM Tris-HG (pH value).

7.4), 8 mM MgG ₂ , 100 mM NaG, 9 units Klenow DNA polymerase I fragment, 0.5 mM ATP and 2 units T4 DNA ligase, and the reaction was continued at 37 ° C for 3 hours and then at 25 ° C for 2 hours, after which the reaction product was used to transform infectious JM105 cells; transformant cells were grown overnight, and ssDNA was then isolated from the culture supernatant. This ssDNA was used as a template for primer extension in2.cHusa and gel purified RF-type DNA was transformed into infectious JM105 cells; the resulting transformant cells were plated on agar plates and cultured overnight to obtain phage plaques.

(3) Site-directed mutagenesis

The procedure described in (2) above was repeated, but the synthetic oligonucleotide primer used was 5'CGCCCATGGTGCCATCCTC-3 ', which generates the Ncol test restriction enzyme in the base sequence and at the same time alters Gly 9. to Thr. Ser-tMet (see Figure 38 (1)).

(4) Site-directed mutagenesis

The procedure described in (2) above was repeated, but the synthetic oligonucleotide primer used was 5 ') TAACACCTTAAGAAGCCAG-'3, which provides the AflUrestriction enzyme recognition site in the base sequence and at the same time changes the codon terminus encoded by Lys 87 (see page 38). figure].

(5) The site-directed mutagenesis procedure described in (2) above was repeated, but the synthetic oligonucleotide primer used was 5'CCGGACTCCGTTAACTCGG-3 ', which provides the Hpal restriction enzyme recognition site in the base sequence and at the same time alters Asp 42. Asp. [See Figure 38 (4)].

(6) Site-directed mutagenesis

The procedure described in (2) above was repeated, but the synthetic oligonucleotide primer used was 5'CTTCTCCTGGACTCCGTCAAC-3 ', which deletes the HpaH restriction enzyme recognition site from the base sequence and at the same time changes the Arg 45. to GIn [see Figure 38 (5)].

(7) Screening and identification of amputated plaques

Mutated M13-PO plaques [Seq

HU 204089 Β rint] and 2 plates containing non-mutated M13-PQ phage holders were cooled to 4 ° C and the plaques were transferred from each plate to 2 round nitrocellulose filters by placing a dry filter on the agar plate for 5 minutes. for filter 1 and 15 minutes for filter 2. The filters were then placed on thick filter paper immersed in a solution containing 0.2 NaOH, 1.5 M NaCl, and 0.2% Triton Χ-100 for 5 minutes, then neutralized so that for a further 5 minutes, It was placed on a filter paper immersed in a solution containing 5 M Tris-HCl pH 7.5 and 1.5 M NaCl. The filters were washed twice with 2xSSC (standard sodium Cltrate) filters in the same manner and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The superimposed filters were pre-hybridized at 55 ° C for 4 hours with 10 ml / filter DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and bovine serum albumin, lx = 0). 02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered solution (pH 7.0) and 100 µg / ml denatured salmon sperm DNA. Hybridization was performed at 42 ° C for 24 hours on 10 ⁵ cpm / ml oligonucleotide primer. Filters were each washed at 50 ° C for 30 minutes with wash buffer containing 0.1% SDS and reduced SSC. The filters were then washed first with 2xSSC buffer; control filters containing non-mutated M13-PO carrier lids were assayed for radioactivity using a Geiger counter. While reducing the SSC concentration step by step, the control filters containing the anemutated M13-PO retention limes were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is 0.1x SSC. The filters were allowed to air dry and autoradiograms were taken with 2-3 days exposure at -70 ° C. Screening was performed on 10,000 mutated M13-PO plaques and 100 non-mutated control plaques using ^3Z P-g55ATP-treated oligonucleotide samples. None of the control plaques, while 3-10 mutated M13PO plaques hybridized to the assay sample.

One of the mutated M13-P0 plaques was taken and inoculated into JM105 culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Base sequences were analyzed using the appropriate oligonucleotide primers and ssDNAs.

We found that the GGC (9.Gly) codon was an ACC (Thr) codon; the AGC (10ser) codon to the ATG (Met) codon; codon CCG (14.Pro) to codon ATG (Met), codon AAA (Lys 87) to TAA (termination); the GAC (42 Asp) codon to the AAC (42 Asn) codon; and the CGG (45, Arg) codon was changed to a CAG (45 Gin) codon.

Amputated from M13-PO phages the phage in which a

Gly codon 9 was changed to Thr codon and Ser 10 codon to Met codon, M13-PN10 (its base sequence and the amino acid sequence encoded therein are shown in Figure 39);

the phage in which Pro codon 14 has been converted to Met coding for M13-PN14 (its base sequence and the amino acid sequence encoded by it are shown in Figure 40);

the phage in which the Lys codon 87 was changed to a termination codon, M1-3-PC86 (its base sequence and the amino acid sequence encoded by it)

Figure 41);

the phage which has been transformed from Asp codon 42 to codon Asn for M13-PDN42 (its base sequence and the amino acid sequence encoded therein are shown in Figure 42); and the codon in which Arg codon 45 was transformed into a codon for Gln was designated M13-PRQ45 (its base sequence and the amino acid sequence encoded therein are shown in Figure 43).

Example 28 (Preparation of recombinant DNAs having a mutein coding base sequence), (1) Site-specific mutagenesis

0.1 mmoles / 1 of adenosine triphosphate (ATP), 50 mmóM hydroxymethyl amino-methane hydrochloride (Tris-HCl) (having a pH value of 8.0), 10 mmol / 1 _MgCl2, 5 mM / 5 mM dithiothreitol ( DTT) and 9 units in the presence of T4 kinase in a total volume of 50 μΐ, 40 picomoles

5'-CGGACTCCTCTAACTCGGC-3 'Synthetic Oligonucleotide [Primary to Create a HpaI Restriction Recognition Site and thereby Change to Asg at 42 Asn (See Figure 38 (6) /) was treated with T4 enzyme 37 At 1 ° C for 1 hour. This kinase-treated primer was heated at 67 ° C for 5 minutes and then at 42 ° C for 25 minutes in 50 μΐ of a reaction mixture containing 50 mM NaCl, 10 mM Tris-HCl (pH 8). , 0), 10 mM MgCl ₂ and 10 mM β-mercaptoethanol were hybridized to 5 pg single-stranded (ss) M3-PDN42 DNA. The soldered mixture was then chilled on ice and added to a 50 μΐ reaction mixture of 0.5-0.5 mM deoxynucleotide triphosphates (dNTP), 80 mM Tris-HCl (pH

7.4), 8 mM MgCl ₂ , 100 mM NaCl, 9 units of the DNA polymerase I Klenow fragment, 0.5 mM ATP and 2 units of T4 DNA ligase. The reaction was quenched at 37 ° C for 3 hours and then at 25 ° C for 2 hours, then quenched with 2 µL of 0.2 mM EDTA. The reaction mixture was then used to transform infectious JM105 cells; transformant cells were grown overnight, and ssDNA was then isolated from the culture supernatant. This DNA was used as a template in the 2nd cycle of primer extension, and the gel-purified RF-type DNA was transformed into infectious JM105 cells; the resulting transformant cells are agar le25

HU 204089 Β was spread on a field and bred overnight to obtain phage plaques.

(2) Screening and identification of plaques that have been amputated

Plates containing the mutated M13-PDN42 phage using the method of Example 28 (1) using the synthetic oligomer (Figure 38 (6)) and 2 plates containing non-mutated M13-PDN42 phage plaques were cooled to 4 ° C and the plaques from each plate were cooled. transferred to a 2 round nitrocellulose filter by placing a dry filter on the agar plate for 5 minutes for filter 1 and 15 minutes for filter 2. The filters were then placed on thick filter paper immersed in a solution containing 0.2 N NaOH, 1.5 M NaCl, and 0.2% Triton Χ-100 for 5 minutes, then neutralized so that the , Placed on a filter paper immersed in a solution containing 5 M Tris-HCl pH 7.5 and 1.5 M NaCl. The filters were washed twice with 2x & SC (standard sodium citrate) immersed filters in the same way and allowed to dry, then dried at 80 'for 2 hours in a vacuum oven. The superimposed filters were pre-hybridized at 55 ° C for 4 hours with 10 ml / filter DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll, and bovine serum albumin, 1x ^ 02%), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered solution (having a p value of 7.0), and 100 pg / ml of denatured salmon sperm DNA Ahibration was carried out at 42 ° C. 10 ⁵ cpm / ml of the oligonucleotide primer. The filters were washed individually at 50 ° C for 30 minutes with wash buffer containing 0.1% SDS and reduced SSC. The filters were then washed first with 2xSSC buffer; control filters containing non-mutated M13-PDN42 plaques were assayed for di-activity using a Geiger counter. While stepwise decreasing the SSC concentration, control filters containing non-mutated M13-PDN42 plaques were washed until detectable radioactivity remained on the filters. The minimum SSC concentration used is O.lxSSC. The filters were allowed to dry in air and radiogramo- ² katvettünkf 2-3-day exposures 70 'Chőmérséldeten. Screening was performed on 10,000 mutated M13-PDN42 plaques and 100 non-mutated control plaques using ³² kinase-treated oligonucleotide probe samples. The control patches were single-stranded and 3-10 mutated M13-PDN42 plaque hybridized.

One of the mutated M13-PDN42 plaques was taken and inoculated into JM105 culture medium. From this supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Base sequence analysis was performed using the appropriate oligonucleotide primers and ssDNAs.

It has been determined that AAC (42.Asn) AGA (Arg) has been converted to 60 codons.

From the mutated M13-PDN42 phage, the phage in which the Asnkodon 42 Arg codon was coded was called M13-PNR42. Its base sequence and the amino acid sequence encoded by it are shown in Figure 44.

Example 29 (Stability of CS2, CS3 and CS23 muteins)

Studies were performed on the stability of heparin HPLC-purified human bFGF and CS2, CS3 and CS23 (obtained in Example 24) as follows. The muteins and human bFGF were separately incubated at 37 ° C in 50 mM NaAcetate pH4.0 at a concentration of 1 pg / ml. The incubation time was varied from 0 to 1 hour at 10-minute intervals and calculated using the above-mentioned mouse BALB / c3T3 cell system (% activity at appropriate incubation times) for the activity of 100% non-incubated standard samples and plotted. The curves shown in Figure 45 were thus obtained. A45. FIGS. 10A-O, -Δ, -applied O indicate the results of human bFGF, CS2 mutein, CS3 mutein, and i CS23 mutein, respectively.

Human bFGF is almost completely inactivated within 10 minutes, while CS2, CS3 and CS23 muteins remain active in ascending order for longer. It retains 50% of CS23 mutein activation even after 20 minutes. As a result, these muteins appear to be more stable than human bFGF.

Example 30 (Endothelial cell proliferation activities of muteins in human umbilical cord blood)

Heparin HPLC-purified human bFGF and the human umbilical cord endothelial cell proliferation activity of the CS2, CS3 and CS23 muteins (prepared in Example 24) were determined. The procedure used for the determinations was as follows. Human vascular endothelial cells were obtained from human umbilical vein by perfusion with trypsin solution and maintained in subculture in liquid medium prepared by adding 2.5% fetal bovine serum albumin and 2.0 ng / ml human bFGF to gITag , Japan, product). A 96 well culture plate (Nunc, 1-67008) was inoculated with 100 µl of a 2.10 ³ cell suspension of blood vessel endothelial cells and cultured in a constant temperature, carbon dioxide-filled chamber. The following day, FGF was added in an amount to give a final concentration of 2 ng / ml and 100 µl of medium containing samples at various concentrations. After 3 days of culture, the medium containing the sample was aspirated and 100 µl of a 1 mg / ml MTT solution (prepared by adding 3- (4,5-dimethyl-2-thiazolyl) 2,5-diphenyl-2H-tetrazolium in the medium) -Bromide dissolved fear

HU 204089 Β was added and incubated for 4 hours. 100 μΐ 10% SDS (aqueous sodium dodecyl sulfate) solution was then added and incubated for 5-6 hours to solubilize the cells and the MTI pigment, then the OD ₅₉₀ was determined using a spectrophotometer. Percentage of cell growth rates for the respective sample concentrations were calculated, the maximum growth rates obtained with 8.0 ng / ml human bFGF were taken as 100%, and the specific activities were determined by determining the 50% growth rate. tax concentrations were compared to standard FGF concentrations. The specific activities thus determined are shown in Table 5.

Table 5 1 5

FGF Specific activity CS2 1.6 CS3 3.0 CS23 2.5 human bFGF 1.0

The results show that the specific activity of the muteins is higher than that of human bFGF.

Example 31 (The angionogenesis potential of muteins, CAM 6).

method measured)

In vivo studies on the angionogenesis potential of heparin HPLC-purified human bFGF and the muteins CS2, CS3 and CS23 (obtained in Example 24) were performed using CAM (chorio-allantoin membrane) method according to Auerbach's method (Developmental Biology, 41). 391 (1974)]. The shells of fertilized eggs of poultry origin were removed after incubation for 3 days at 37 ° C and further incubated at 37 ° C for one week using a Napco 63000 carbon dioxide incubator (CO ₂ : 5%, H ₂ O: saturated). An aqueous solution of each protein was added to 6 mm diameter polypropylene plates so that each protein was 6.25; 12.5; 25.0 and 50.0 nanograms. After air-drying on a clean surface, the plates were incubated at 37 ° C for an additional 3 days while standing on the winged chorioallantoin membrane and observations of the angiogenesis state were performed using a stereomicroscope. The results are shown in Table 6. Note: A total of 18 eggs were used for each sample and the numbers in the table represent the percentage of eggs in which angiogenesis was recorded.

FGF Angiogenesis Potential% 50.0 25.0 12.5 6.25 ng CS2 77.8 44.4 22.2 0.0 CS3 72.2 61.1 16.7 0.0 CS23 72.2 55.6 22.2 0.0 Human bFGF 94.4 50.0 33.3 0.0 Carrier 11.1 5.6 11.1 0.0

It is clear from Table 6 that the bFGF muteins have angiogenesis potential which is approximately equivalent to that of the control human bFGF.

Example 32 (Expression of the gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB854 for expression of human bFGF myelos

The replicative form (RF) of M13-PN10 obtained in Example 27 above was treated in the same manner as described in Example 2 (1) to generate plasmid pTB354 for expression of human bFGF mutein (Figure 46).

Using this plasmid pTB854, it was transformed with Escherichia coli MM294 to obtain the strain Escherichia coli MM294 / pIB854, which contains the plasmid pTB854 containing the mutant coding gene shown in Figure 39.

The above-mentioned transfanant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell line.

(3) Human bFGF activity of abactor cell extract

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

The cellular extract of E. coli MM294 / pTB854 was found to exhibit FGF activity.

In this way, TM910 mutein was obtained in which Gly at position 9 of human bFGF and Ser at position 10 of human bFGF were replaced by Ifcr or Met.

Example 33 (Expression of the gene encoding human bFGF mutein in Es55 chericbia coli) (1) Construction of plasmid pTB852 for expression of human bFGF mutein

Plasmid pTB854 DNA obtained in Example 32 above was cleaved with EcoRI and NcoI restriction enzymes to remove the DNA fragment

EN 204089 Β which encodes the 10 amino acid residues of the human hFGFN terminus. One strand of the remaining DNA fragment was blunt-ended using Escherichia coli DNA polymer in the presence of dATP, dCTP, dGTP and dTTP and ligated using the T4 DNA ligation reaction to generate plasmid pTB852 for human bFGF mutein (Fig. 47).

Using this plasmid pTB852, Escherichia coli MM294 was transformed to give Escherichia coli MM294 / pTB852, which contains the plasmid p'EB852 containing the mutant coding gene shown in Figure 48.

(2) Preparation of bacterial cell extracts

The transformants mentioned above were cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

We found that E. coli MM294 / pIB852 bacterial cell extract showed FGF activity

In this way, the N10 mutein was obtained in which the human bFGF amino acid sequence was inserted from Pro 2 to Ser 10.

Example 34 (Expression of human bFGF mutein coding gene in Escherichia coli) (1) Construction of plasmid pTB795 for expression of human bFGF mutein

The replicative form (RF) of M13-PN14 obtained in Example 27 above was treated in the same manner as described in Example 2 (1) to generate plasmid pIB795 for expression of human bFGF mutein (Figure 49).

Using this plasmid pTB795, Escherichia coliMM294 was transformed to give Escherichia coli MM294 / pTB795 (IFO 14700, FERM BP1660), which included

Figure 40 is a plasmid TB795 containing the mutated mutein coding gene.

(2) Preparation of bacterial cell extracts

The above-mentioned transformant was cultured as described in Example 2 (2) to give a supernatant which is then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method described in Example 2 (3).

E. coli MM294 / pTB795 was found to exhibit FGF activity in cell extracts.

In this way, the N14 mutein was obtained in which the amino acid sequence of human bFGF was

The section from Pro 2 to Pro 14 is eliminated.

Example 35 (Expression of the human bFGF mutein-encoding gene in Escherichia coli) (1) Construction of plasmid pTB796 for mutational expression of human bFGF 5

The replicative form (RF) of M13-PC86 obtained in Example 27 above was treated in the same manner as described in Example 2 (1) to construct plasmid pTB796 for expression of human bFGF mutein (Figure 50).

Using this plasmid pTB796, Escherichia coli MM294 was transformed to give Escherichia coli MM294 / pTB796 (IFO 14701, FERM BP1661), which contains the pTB762 plasmid containing the mutein coding gene shown in Figure 41.

(2) Preparation of bacterial cell extracts

The aforementioned transformant was cultured as described in Example 2 (2) to give 20 supernatants, which were then used as bacterial cell extracts.

(3) Human bFGF activity of abactor cell extract

The cell extract obtained in (2) above was subjected to an enzyme-im25 munassay (ELA) sandwich method using a monoclonal antibody as described in (a) to (j) below.

It was determined that FGF has a C86 mutein in the present cell extract.

In this way, a C86 mutein was obtained in which the stretch from Lys at position 87 to Ser at position 147 was eliminated.

EIA using said monoclonal antibody. measurement was performed as follows:

(a) Immunization

BALB / c mice (4-week-old males) were injected intraperitoneally with human hFGF antigen (as obtained in Reference Example 2) in 0.4 ml of complete Freund's adjuvant (Difco Laboratories, USA). After three weeks, 10 pg of hbFGF antigen was administered intraperitoneally in 0.4 ml of incomplete Freund's adjuvant. After a further three weeks, the same immunization was terminated and after a further two weeks, 10 µg of human bFGF solution (in physiological saline data) were inoculated intintraperitoneally.

(b) Cell fusion

The spleen from the immunized mice described in (a) was excised 4 days after the last antigen administration, yielding cells that can be used for 50 cell fusions. These cells are suspended in a medium obtained by mixing Isokov medium with Ham F-12 medium in a 1: 1 ratio (hereinafter referred to as IH medium).

P3-X63-Ag 8U1 mouse myeloma cells were cultured in RPMI1640 medium containing 120% fetal calf serum in 5% carbon dioxide-95% air.

Cell fusion was performed according to the method of Köhler and Miistein [Köhler G. and Miistein

HU 204089 Β

C., Nat., 256, 495 (1975)]. 2.9.10 ⁷ cells from the aforementioned myeloma cell line and 1.5.10 ⁸ immunized lymphocytes obtained by the above method were mixed and centrifuged and 45% polyethylene glycol 6000 (hereinafter referred to as PEG 6000) in 0.3 ml IH medium. of the resulting solution was added dropwise. The PEG 6000 solution was preheated at 37 ° C and added gradually. Five minutes later, IH medium preheated to 37 ° C was added at a rate of 0.5 ml / min to a volume of 10 ml. The solution was then centrifuged at room temperature at 600 rpm for 15 minutes and the supernatant removed. This cell pellet was then suspended in 200 ml IH medium containing 20% fetal bovine serum and transferred to a 24-well microplate (Linbro) at 2 ml / well. IH medium supplemented with 1 day later, HAT (1.10 ^-4 mol / 1 hypoxanthine, 4.10 ^-7 mol / 1 aminopterine, 6.10 ^{'to 5} mol / L thymidine), 20% fetal calf serum as well (this is referred to as HAT medium hereinafter) 1 ml / well was added to the microplate, and after another three days, half of the medium was replaced with HAT medium. The cells thus grown were hybrid cell genes.

(c) Detection of antibody-producing cells

The hybrid cell culture supernatant was placed in the wells of a 96-well microtiter plate made of polystyrene at 100 µl / well previously immobilized on human bFGF and incubated at room temperature for two hours. The incubation supernatant was removed and after washing, a ton of peroxidase (HRP) -labeled anti-mouse IgG goat antibody (Miles Laboratories) was added as a second antibody and incubated at room temperature for 2 hours. The second antibody was removed and, after thorough washing of the wells, a color reaction was performed in the presence of a reaction substrate (EIA method). By this method, significant values were observed in 3 wells.

(d) Hybrid cell donation

From each of these three groups of cells, cells were transferred at a volume of 0.5 cells / well into the wells of a 96-well microtiter plate with 10 ⁴ cells / well mouse thymocytol as vegetative cells and cloning. Thus, several clones were obtained, including the mouse hybridoma HbF52 (IFO 50143) and the mouse hybridoma HbF78 (BFO 50144).

(e) Introducing hybrid cells into ascites

2.10 ⁶ cells of the HbF62 mouse hybridoma or of the HbF78 mouse hybridoma were inoculated into mice to which mineral oil had been administered intraperitoneally. Ten days later, 2 ml of ascites fluid was collected from each mouse to obtain MoAb52 and MoAb78 monoclonal antibodies.

(f) The MoAb78 monoclonal antibody thus obtained was purified from the ascites fluid according to the method of Stachelin et al., 1981, The Journal of Biological Chemistry, 256.9750. The resulting antibody was concentrated to more than 2 mg / mL and dialyzed against 0.2 M phosphate buffer, pH 7.0. From the thus obtained 1.4 ml solution of AMoAb78 (concentration: 2.8 ng / ml), 50 µl was added to a solution of S-acetylmercaptic succinic anhydride in N, N'-dimethylformamide so that the concentration

11.5 mg / ml. Air in the reaction vessel was replaced with nitrogen gas. The vessel was soldered and stirred to induce the introduction of the SH groups. The unreacted S-acetylmercaptic succinic anhydride was inactivated by treatment with 130 µL of 0.2 M Tris-HCl, pH 7.0, 13 µL of 0.2 M EDTA and 130 µL for 10 minutes. with a mixture of hydroxylamine (pH 7.0). MoAb78 was recovered by gel filtration using a column packed with Sephadex G-25 (1 cm x 80 cm), Farmacia, Sweden (flow rate: 20 ml / h).

(g) 10 mg tonne peroxidase (HRP, Boehringer Mannheim, Grade I, West Germany) was dissolved in 1.4 M phosphate buffer (pH 6.8). On the other hand, 14 mg of N- (4-carboxycyclohexylmethyl) -malimide N-hydroxysuccinimide ester was dissolved in 335 µl DMF and 100 µl of the resulting solution was added to the above HRP solution. The air in the reaction vessel was replaced with nitrogen gas and the vessel was soldered. After one hour at room temperature, the MoAb78 monoclonal protein containing SH groups was recovered by gel filtration using a column packed with Sephadex G-25 as in (f) above.

(h) A 6 mL portion of the monoclonal antibody containing SH groups obtained in (f) above and a 2 mL portion of HRP containing the maleimide group obtained in (g) above were mixed and the mixture was diluted to 1 mL. were concentrated using a collodium bag (Sartorium, West Germany) under reduced pressure at 4 ° C for 20 hours. After the reaction, unreacted SHP-containing HRP was removed using an Ultrogel AcA44 column (LKB Co., diameter: 1 cm x 80 cm, Sweden) (flow rate: 10 ml / h). Of the eluate fractions, the one with the highest antibody concentration! HRP number, i.e. the fraction containing 2.4 HRP / antibody, was used in the EIA measurement according to (i) below.

(i) The MoAb52 monoclonal antibody obtained in (e) above was purified in the same manner as described in (f) above. The AMoAb52 monoclonal antibody was diluted with PBS to 10 μg / πll or 20 μβ / η, and the diluted material was added to hmnuplale (Nunc, Denmark) at 100 µg / well. The plate was allowed to stand at 4 'C overnight to adsorb the MoAb52 monoclonal antibody to the plate. After removal of the unabsorbed antibody, the plate was washed three times with PBS, then PBS containing 0.01% mertiolate and 1% bovine serum albumin (BSA) was added at 200 µl / well and left overnight for 4 hours. 'C temperature.

(j) Cell extract containing bFGF mutein obtained in (2) above with PBS containing 0.1% BSA

HU 204089 Β coli MM294 was transformed to give Escherichia coli strain MM 294 / pTB799, which contains the plasmid pTB799 containing the mutant coding gene shown in Figure 43.

(2) Preparation of Bacterial Cell Extract The above-mentioned transformant was cultured as described in Example 2 (2) to obtain a supernatant which was then used as bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

The human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the < 2 > (3) method.

E. coli MM294 / pTB799 bacterial cell extract was found to show FGF activity.

Thus, the RQ45 mutein was obtained in which Arg is replaced by Gln at position 45 of human bFGF.

Example 38 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB798 for expression of human bFGF mutein

Afenti28. The replicative form of M13-PNR42 obtained in Example 1A was treated in the same manner as in Example 2 (1) to construct the plasmidothapHB798 human hFGF mutein (Figure 53).

Using this plasmid pTB798, Escherichia coli MM294 was transformed to give Escherichia coE strain MM294 / pTB798, which contains the plasmid pTB98 containing the mutein coding gene shown in Figure 44.

(2) Preparation of bacterial cell extract

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The activity of the bacterial cell extract in human bFGF

Human bFGF activity of the bacterial cell extract obtained in (2) above was determined by the method of Example 2 (3).

E. coli MM294 / pTB798 bacterial cell extract was found to show FGF activity.

Thus, the NR42 mutein was obtained in which Asp was replaced with Arg at position 42 of human bFGF.

Example 39 (Preparation of recombinant DNAs having a mutein coding base sequence) (1) Site-specific mutagenesis 55 µL, 1 mM adenosine triphosphate (ATP), 50 mM hydroxymethylaminomethane hydride (tris) -HQ) (having a p H value of 8.0), 10 mM MgCl ₂ , 5 mM dithiothreitol (DTT), and 9 units of T4 kinase in a total volume of 50 µg, 40 picomoles of 60 5'-TCTTCTCCAGGGATCCGTT-3 ', was washed four times with PBS, and diluted C86 bFGF mutein was added to the plate at a concentration of 100 μ, / well and allowed to adsorb to the plate overnight at 4 ° C. The unreacted C86mutein was removed and the plate was washed four times with PBS. The HRP-conjugated monoclonal antibody obtained in (h) above was diluted 1/300 in PBS containing 0.1% BSA and added to the plates at 100 μΐ / well. The reaction was completed for 4 hours at 10 at room temperature. After antibody removal, the plate was washed six times with PBS, and oxidase substrate (Bio-Rad Co., USA) was added to give 100 µl / well. Quantification was performed at 415 m by absorbance measurement; and found to be small. C86 is produced.

Example 36 (Human bFGF mutein codon gene expression in Escherichia coli) (1) Construction of plasmid pTB797 for expression of human bFGF mutein

The replicative form (RF) of M13-PDN42 obtained in paragraph 27 above was treated in the same manner as described in Example 2 (1) to construct pTB797 plasmidothuman bFGF mutein (Figure 51).

Using this plasmid pTB797, Escherichia coli MM294 was transformed to obtain Escherichia coli MM294 / pTB797 (EFO 14702, EERM BP1662), which contains the pTB762pteon 35

The aforementioned transformant was cultured as described in Example 2 (2) to give a supernatant which was then used as bacterial cell extract.

(3) Abactant cell extract activity of human bFGF

The bacterial cell extract obtained in (2) above was assayed for human bFGF activity by the method of Example 2 (3).

Cell extract of bacterium E. coli MM294 / pTB797 was found to exhibit FGF activity.

In this way, the DN42 mutein is obtained in which Asp is replaced by Asn at position 42 of human bFGF.

Example 37 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pTB799 for expression of human bFGF mutein

The replicative form of M13-PRQ45 obtained in Example 27 above was treated in the same manner as described in Example 2 (1) to construct pTB799 human bFGF mutant (Figure 52).

Using this plasmid pTB799, Escherichia

EN 204,089 Β Synthetic Oligonuldeothide [Primary to Create a BamHI Restriction Recognition Site in the Base Sequence and Subsequently Convert 44 Val to Ser and 45 Arg to Leu / See 38 (7). Fig. 1] was treated with T4 kinase at 37 ° C for 1 hour. The kinase-treated primer was treated with 50 μΐ of a mixture of 50 mM NaCl, 1.0 M tris-HQ (at pH 67 ° C for 5 minutes and then at 42 ° C for 25 minutes). 8.0), 10 mM MgCl ₂ and 10 mM β-mercaptoethanol were hybridized to 5 µg of single-stranded (ss) M3-PDN42 DNA. The soldered mixture was then chilled on ice and added to 50 µl of a reaction mixture containing 0.5 mM deoxynuldeotide triphosphate (dNTP), 80 mM Tris-HQ (pH 7.4), 8 mM MgQ _{It contained 2} , 100 mM NaCl, 9 units of DNA polymerase I Klenow fragment, 0.5 mM ATP and 2 units of T4 DNA ligase and the reaction was carried out at 37 ° C for 3 hours, followed by It was continued at 2 ° C for 2 hours, then quenched by the addition of 2 μΐ EDTA. The reaction product was used to transform infectious JM105 cells; transformant cells were grown overnight and ssDNA was then isolated from the culture supernatant. This ssDNA was used as a template in cycle 2 of primer extension and gel purified RF-type DNA was transformed into infectious JM105 cells; the resulting transformant cells were plated on an agar plate and cultured overnight to obtain phage plaques.

(2) Isolation and identification of mutagenized plaques

By the method of Example 27, using said synthetic oligomer (oligonucleotide) (Figure 38 (7)), plates containing mutated phage M13-PDN42 phage plaques and 2 plates containing un mutated M13-PDN42 plaques obtained in Example 26, After cooling to 4 ° C, the plaques were transferred from each plate to a 2-round nitrocellulose filter by placing a dry filter on the agar plate for 5 minutes for Filter 1 and 15 minutes for Filter 2. The filters were then placed on thick filter paper immersed in a solution containing 0.2 N NaOH, 1.5 M NaCl, and 0.2% TritonX-100 for 5 minutes, then neutralized by 0.5 moles for 5 minutes. These were placed on filter paper immersed in a solution containing tris (1: 3) HCl (pH 7.5) and 1.5 M NaCl. The filters were washed twice on 2xSSC (standard sodium citrate) immersed filters in the same way and allowed to dry, then dried at 80 ° C for 2 hours in a vacuum oven. The superimposed filters were pre-hybridized at 55 ° C for 4 hours with 10 ml / filter DNA hybridization buffer (5xSSC) pH 7.0 containing 4xDenhardt solution (polyvinylpyrrolidone, Ficoll and bovine serum albumin, 2x = 0.02). %), 0.1% sodium dodecyl sulfate (SDS), 50 mM sodium phosphate buffered saline (pH 7.0) and 100 pg / ml denatured salmon sperm DNA. Hybridization was carried out at 42 ° C for 24 hours with 10 ⁵ cpm / ml oligonucleotide primer. Each filter was washed with a buffer solution containing 0.2% SDS and a reduced amount of SSC at 50 ° C for 30 minutes. The filters were then washed first with 2xSSC buffer; control filters containing non-mutated M13-PDN42 plaques were assayed for radioactivity using a Geiger counter. While decreasing the SSC concentration step by step, control beers that contained no mutated M13-PDN42 plaques were washed until no detectable radioactivity remained on the filters. The minimum SSC concentration used is 0.1x SSC. The filters were allowed to air dry and autoradiograms were taken with a 2-3 day exposure at -70 ° C. Screening was performed on 10,000 mutated M13-PDN42 plaques and 100 non-mutated plaques using oligonucleotide probe treated with kinase in the presence of ³² Ρ-γATP. None of the control plaques, while 3-10 of the mutated M13-PDN42 plaques hybridized to the assay pattern.

One of the mutated M13-PDN42 plaques was taken up and inoculated into JM105 culture medium. From the resulting supernatant, ssDNA was prepared and double stranded (ds) DNA was prepared from the bacterial cell pellet. Analyzes of the base sequence were made using the appropriate oligonucleotide primer and ssDNA.

We found that the GTC (44 Val) codon and the CGG (45 Arg) codon changed to TCC (Ser) and CTG (Leu) codons respectively.

Among the mutated M13-PDN42 phages, the phage in which codon 44 Val and codon 45 Arg changed to Ser codon and Leu codon was designated M13 PINT. Its base sequence and the amino acid sequence encoded by it are shown in Figure 54.

(3) Construction of plasmid pTB853 for expression of human bFGF mutein

The replicative form of M13-PINT obtained in (2) above was treated in the same manner as described in Example 2 (1) to generate plasmid pTB853 for the human bFGF mutein (Figure 55).

Using this plasmid pIB853, Escherichia coli MM294 was transformed to give strain Escherichia coli MM294 / pTB853, which contains the plasmid pTB853 containing the mutein coding gene shown in Figure 54.

(4) Preparation of bacterial cell train

The transformant mentioned above was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(5) Human bFGF activity of bacterial cell train

The human bFGF activity of the bacterial cell extract obtained in (4) above was determined by the method described in Example 2 (3).

HU 204089 Β

E. coli MM294 / pTB853 was found to exhibit FGF activity in cell extracts.

In this way, the EINT mutein was obtained in which Asp at position 42, Val at position 44, and Arg at position 45 were replaced with Asn in the human faFGF replaced by Ser over Leu-valva.

Example 40 (Human bFGF mutein-encoding gene expression in Escherichia coli) (1) Construction of plasmid pTB855 for expression of human hFGF mutein

Plasmid pTB669 DNA obtained in Reference Example 2 above was digested with HincH and ligated with the EoRI linker (5 'CATGAATTCATG 3') using T4 DNA ligase. The resulting DNA was further cleaved with the restriction enzymes EcoRI and PstI to give a DNA fragment of about 0.35 kb. This DNA fragment was ligated with approximately 3 (2 kb) DNA obtained in Example 2 (1) (this fragment obtained by plasmid ptrp 781 with EcoRI and PsI1-cleaved) to give plasmid pTB855 for expression of human bFGF mutein. (Figure 56).

Using this plasmid pTB855, Escherichia coli MM294 was transformed to obtain Escherichia coli MM294 / pTB855, which contains the pTB855 plasmid containing the mutein-encoding gene shown in Figure 57 (2).

The above-mentioned transfanant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) ^{AbaktériumsejtkivonathumánbFGFaktivitá-:} sa

The human hFGF activity of the bacterial cell extract obtained in (2) above was determined by the method of Example 2 (3).

We found that E. coli MM294 / pTB855 'cell extract showed FGF activity

Thus, the N41 mutein was obtained in which the amino acid sequences of human hFGF from 2 Pro to 41 Val were inserted.

z

Example 41 (Expression of gene encoding human bFGF mutein in Escherichia coli) (1) Construction of plasmid pIB856 for expression of human hFGF mutein.

The DNA of plasmid pTB6O9 obtained in Reference Example 2 above was partially digested with the restriction enzyme BamHI so as to obtain a BamHI recognition site in the ABFGF gene. This DNA was cleaved with the Nhel linker having the sequence p (5'-CTAGCTAGCTAG-3 '),

We ligated with T4 DNA ligase. By treatment with Nhel restriction enzyme and digestion of the cleavage site with T4 DNA θ ligase, plasmid pTB856 was constructed to express human bFGF mutein (Figure 58).

Using this plasmid pTB856, Escherichia 5 coli MM294 was transformed to obtain Escherichia coli MM294 / pTB856, which contains the mutant coding gene for plasmid TB856 shown in Figure 59.

(2) Preparation of bacterial cell extract

The above-mentioned transfectant was cultured as described in Example 2 (2) to give a supernatant which was then used as a bacterial cell extract.

(3) The cell extract obtained in part (2) above was subjected to the enzyme 'δ immunoassay (EIA) by the sandwich method as described in Example 35 (3). Thus, it was determined that bFGF mutein is present in the cell extract. Thus, we obtained the C129 mutein in which from Lys 13. to 147, Ser.

O produced amino acid sequences.

Example 42 (Expression of a gene encoding human bFGF mutein in an animal cell) (1) Construction of plasmids pTB831, pTB833 and pTB835 to express human bFGF mutein

Plasmid pTB762, obtained in Example 7 above, was cleaved with EcoR1 and BamHI to give a 0.38 kb DNA fragment containing the coding region for the human bFGF mutein CS23, which contains the 70 and 88 positions.

Replaced by Cys Ser weapon.

On the other hand, plasmid pTB504 [1. Example II (ii) in Japanese Patent Publication No. 62175182, corresponding to European Patent Publication No. 225,705, and digested with restriction enzymes EcoR1 to give a 1.7 kb DNA fragment which is the murine leukemia virus (MuLv). It contained the LIR region, the SV40 early promoter, the splice site and the human interleukin 2 (IL-2) leader sequence.

In addition, the plasmid pIB627, which was zl. In the reference example, cleaved with Clal and EcoRI restriction enzymes, a 2 kb DNA fragment was obtained and plasmid p1B389 was cleaved with Clal and Eco RI restriction enzymes to give a 2.6 kb DNA fragment. The resulting 2 kb DNA fragment and 2.6 kb DNA fragment were ligated to give plasmid pTB675, which encodes human bFGF. The plasmid pIB389 above was obtained by inserting the PstI site into the interleukin. The 2-gene 5-terminal region and the BamHI site at the 3'-terminal region of the plasmid p11010 (described in Japanese Patent Publication No. 61-63282 (corresponding to European Patent Publication No. 172,619)).

EcoR1 site to remove the interleukin-2 gene region. Plasmid pTB675 was then cleaved with the restriction enzymes Clal and BamHI to give a 3.4 kb DNA fragment containing

20 the coding region for the amino acids at the C-terminus of human bFGF, a non-coding region, the ampicillin resistance gene from plasmid pBR322, and the origin of replication in Escherichia coli.

Three DNA fragments above (i.e., the 0.38 kb DNA fragment, 1.7 kb DNA fragment, and 3.4 kb DNA fragment) were ligated with T4 DNA ligase, such as plasmid tB831 (Figure 60).

Similarly, when the plasmid pTB763 obtained in Example 18 encoding the human bFGF CS14 mutant (in which it is replaced by Cys-EkSer at positions 26 and 93 of human bFGF) or the plasmid pTB765 obtained in Example 11, the human bFGF encodes mutein CS1234 (in which the Cys at positions 26, 70, 88, and 93 of human bFGF are replaced by Ser 15) used plasmid pTB833 and pTB835 instead of pTB762 in the above procedure. .

(2) Expression in animal cells

COS7 cells were inoculated with 6 cm diameter tissue

Plate 7 was plated in plate containing DMEM medium containing 10% fetal calf serum, and after 24 hours the medium was replaced with new medium. After 4 hours, 10-10 pgplB831, pTB833 or pTB835 DNA was infected on COS7 cells according to the calcium phosphate method (Graham et al., Virology 52: 456 (1973)). The following day, the culture medium was changed to DMEM containing 1% fetal bovine serum. After 48 hours in culture, the cell-containing culture was harvested and the cells were harvested. The cells were washed twice with PBS and then suspended in PBS. The suspension was sonicated briefly and centrifuged for 15 minutes at 15,000 rpm to obtain a supernatant.

Human bFGF activity was determined from the culture fluid of the COS7 cells infected with the plasmid and extracts from the cells according to the method described in Example 2 (3). The results obtained are summarized in Table 7.

plasmid mutein FGF activity (ng / plate) Breeding fluid cell Extract pTB831 CS23 115 420 pTB833 CS14 0.2 35 pTB835 CS1234 1.1 20 - <0.02 1

Claims (10)

CLAIMS 1. A process for the production of basic fibroblast growth factor (bFGF) deletion type and replacement type muteins, which are muteins of bFGF lacking the 1-14 amino-terminal and / or 1-61 carboxy-terminal amino acids of the bFGF component amino acids. , and which muteins of the substitution type are muteins of bFGF in which at least one of the other cysteine, aspartic, aspartic, arginine, glycine, serine, and valine residues between the constituent amino acids of bFGF is replaced by (1) constructing a structural gene encoding the above muteins by (a) hybridizing with a mutagenic oligonucleotide primer a single-stranded DNA containing a strand of the bFGF structural gene, (b) mutating the heteroduplex strand mutated with the primer DNA polymerase, and (c) , (2) as the resulting structure and upstream of the structural gene The y promoter is inserted into a vector, (3) a host cell is transformed with the resulting expression vector, (4) the resulting transformant is cultured and the mutein accumulated in the culture medium is optionally recovered. (Priority: February 24, 1987) The method of claim 1 for producing a mutein wherein at least one of the amino acids comprising bFGF is absent, characterized by hybridizing the corresponding oligonucleotide primer. (Priority 35: February 24, 1987) The method for producing a mutein according to claim 1, wherein at least one of the amino acids comprising the bFGF is replaced by an amino acid residue comprising hybridizing the corresponding oligonucleotide pri40. (Priority: August 11, 1987) The method of claim 1 for the preparation of a mutein of bFGF which comprises a Phe-Phe-Leu-Arg-Ile-His-Pro-Asp-Gly-Arg-Val45 Asp-Gly-Val-Arg-Glu-Lys-Ser-Asp-Pro amino acid sequence, characterized in that the corresponding bFGF structural gene has a single strand. containing single-stranded DNA. (Priority: February 24, 1987) 50 The process for the production of a mutein according to claim 3, wherein the substituted amino acid of the bFGF which is substituted is cysteine, wherein the corresponding oligonucleotide is hybridized to a primer. (Priority: August 11, 1987) 55 ' The method for producing a mutein according to claim 3, wherein the at least one cysteine component in bFGF is replaced by a neutral amino acid, characterized by hybridizing to the corresponding oligonucleotide primer. (Priority: August 11, 1987) 60 The method of claim 1, wherein said mutein is pre-33 HÜ 204089 Β, wherein abFGF contains at least one serum of a constituent title color. substituted, characterized in that the corresponding oligonucleotide. primer hybridized. (Priority: August 11, 1987) 8. A3. The method of claim 1, wherein the substituted amino acid of bFGF is aspartic acid, arginine, glycine, serine and / or valine, characterized by hybridizing an appropriate oligonucleotide primer. (Priority: 08.08.1987) 11.) A method according to claim 2 for the production of a mutein wherein at least one amino acid is missing at the amino terminus of the amino acids constituting the abFGF, characterized in that the corresponding oligonucleotide primer is hybridized. (Priority: February 24, 1987) The method for producing a mutein according to claim 2, wherein at least one amino acid is missing from the carboxyl terminus of the amino acids of the abFGF component, characterized in that the corresponding oligonucleotide primer 10 hybridized. (Priority: February 24, 1987)