JPH0479880A - Polypeptide - Google Patents

Abstract

NEW MATERIAL:A polypeptide in which 68th proline of human TNF having the an amino acid sequence expressed by the formula is replaced by other amino acid. USE:An antiulcer agent, etc. PREPARATION:A recombinant microorganism cell transformed by a recombinant plasmid containing DNA coding a polypeptide having a sequence in which 68th proline in a human TNF protein is replaced by other amino acid is cultured in a culture medium and the aimed polypeptide having the amino acid sequence is produced and separated.

Description

Detailed Description of the Invention "Industrial Application Field" The present invention relates to providing a novel polypeptide as a human TNF converter, and relates to the polypeptide itself, a pharmaceutical composition containing it, a method for producing it, and its gene. Concerning recombinant DNA, plasmids, microbial cells, etc.

“Prior art” TNF (tumor necrosis factor) was developed by Carsw in 1975.
Bacillus Calm
It is a physiologically active substance that was found to exist in the serum of mice infected with B. antelope Guerin (BCG) and treated with endotoxin (Proc, Natl.
Acad, Sci, USA 723666 (19
75)) In 1984, Penn1ca et al. cloned the cDNA of human TNF and clarified the entire structure (amino acid sequence) of the human TNF protein (Na
ture 312.724 (1984)). T.N.F.
Although it has specific antitumor effects such as cytotoxic activity against tumor cells, hemorrhagic necrosis against transplanted tumors, and inhibition of growth, it has recently been reported that it may also cause side effects such as hyperlipidemia, decreased blood pressure, and fever. Research and development are being carried out to find better drugs in terms of medicinal efficacy, side effects, etc.

For example, JP-A-61-40221, JP-A-63-119692
, Japanese Unexamined Patent Application Publication No. 1-277488 discloses that human TNF transformants are provided by deleting specific amino acids in human TNF protein, substituting with other amino acids, or adding them using genetic engineering techniques. There is.

"Disclosure of the Invention" The present inventors substituted proline at position 68 with another amino acid in the amino acid sequence of human TNF protein, resulting in the following:
H) TN with high antitumor activity and low side effects not described in literature
It was discovered that an F-converted polypeptide could be obtained, and the invention was completed based on this knowledge.

The present invention relates to a polypeptide having an amino acid sequence represented by Ser 1 to Leu 155 shown in SEQ ID NO: 1 in the sequence listing, in which Pro at position 68 is replaced with another amino acid. Also, a recombinant plasmid containing DNA encoding the polypeptide, a recombinant microorganism cell transformed with this recombinant plasmid, a method for producing a polypeptide using a recombinant microorganism cell,
A pharmaceutical composition, a polypeptide in which methionine is bonded to the N-terminus of the amino acid sequence, and a base sequence represented by the 1st T to the 465th G shown in Sequence Listing SEQ ID NO: 2, which encodes Pro at positions 202 to 204 C
It also relates to a DNA having a sequence in which CA is replaced by a codon encoding another amino acid.

Throughout this specification, amino acids, polypeptides, bases,
The following list is used to represent these arrays.

amino acid: base: Furthermore, the abbreviations used in this specification have the following meanings.

SDS: Sodium dodecyl 5u
lfateDTT Nidithiothreitol (Dithi)
PMS F: Phenyl
Methylsulfonyl FluorideED
TA: Ethylene diamine tetraac
etate) CPG resin: Controlled-P
ore Glass resin The 68th Pr in the present invention
Other amino acids to replace o include Aj7a, Cys
, Asp, Gnu, pHe, GI! y, His
, Iffe, Lys, Leu, Met
, Asn, Gln.

Arg, Ser, Thr, Van, Trp or Tyr, or Asp, Met or Tyr
or desirable.

Next, the embodiments of the present invention will be described in detail, or
Genetic manipulation techniques according to the present invention are listed below.

T., Maniatis et al. (1982)
: Molecular Cloning.

Co1d Spring Harbor Lab
oratory, R, WLI et al (1
983): Methods in Enzymolo
gy, 100 and 101°R, Wu et al.
(1987): Methods in Enzymo
Logy.

153, 154, and 155 The polypeptides of the present invention can be produced using various methods, means, and machines, and representative production methods are described below.

(1) Obtaining the human TNF gene As mentioned above, the nucleotide sequence and amino acid sequence of the human TNF gene have been revealed by Pennca et al.
The human TNF gene is designed as shown in SEQ ID NO: 3 in the sequence listing by appropriately changing the base sequence.

At this time, it is desirable to select codons that are suitable for the host cell (E. coli, etc.), and to select codons at appropriate positions so that cloning by ligation of DNA fragments and genetic modification for the production of transformants, which will be described later, can be easily carried out. It is desirable to locate restriction enzyme cleavage sites. Of course, it is necessary to install a translation start codon (ATG) upstream of the human TNF gene and a translation stop codon (TAA, TGA, or TAG) downstream so that they match the reading frame. In addition, it is preferable to install appropriate restriction enzyme cleavage sites downstream of the translation stop codon to improve applicability to vectors and ease of cloning.

The human TNF gene can be produced by chemically synthesizing several oligonucleotides for each of the top and upper chains, and sequentially and appropriately linking each block. For example, in SEQ ID NOs: 3 and 4 of the sequence listing, each chain of the human TNF gene is divided into 10 oligonucleotides of approximately 50 bases each, and a total of 20 oligonucleotides are chemically synthesized. Its synthesis method is the diester method [G, Khorana, “So
me Recent Developments in
Chemistry of Phosphate Es
Tersof Biological □Intere
st”, John Wiley and 5ons.

Inc., New York (1961)), triester method (R,L.

Letsinger et at, J. Am.
Chem, Soc, 89. 4801(
1967)) and phosphite method CM, D, Mat
Teucciet al, Tetrabedron
Lett., 21.719 (1980)), or the phosphite method using a fully automatic DNA synthesizer is preferably used from the viewpoint of operability.

The synthesized oligonucleotide is purified by conventional purification methods such as high performance liquid chromatography using a reversed phase chromatography column and electrophoresis using polyacrylamide gel. Thereafter, the oligonucleotide is e.g.
After phosphorylation and annealing using 4 polynucleotide kinase, ligation is performed using T4 DNA ligase. Here, oligonucleotides are divided into several blocks, sequentially ligated to obtain the desired human TNF gene sequence, cut with restriction enzymes or blunted with T4 DNA polymerase, and purified by electrophoresis. Obtained DN
For A fragments, e.g. pucs, pucs 9, pucs 18, pucs 19
(J, Messing et al., Gene.

19, 259 (1982)), and cloned by transforming competent cells by a conventional method. Plasmid DNA is extracted and purified from the obtained clone according to a known method, and it is checked whether the base sequence of the DNA fragment inserted into the vector has achieved the desired gene sequence. Each part of the human TNF gene that has been achieved is excised from a plasmid vector containing the respective part using a restriction enzyme, and then ligated and integrated into the vector again to obtain a plasmid vector containing the desired full-length human TNF gene. The desired human TNF gene can be obtained by cleaving the thus obtained plasmid vector with a restriction enzyme and then separating and purifying it by gel electrophoresis.

On the other hand, the above-mentioned method may be appropriately combined with a method using cDNA for mRNA derived from human cells expressing TNF.

(2) Construction of human TNF expression vector The human TNF gene obtained in (1) above is appropriately inserted into an expression vector and cloned to construct a human TNF expression vector. The expression vector contains a transcription promoter region and a translation signal SD (
Shine-Dalgarno) sequence, and a translation stop codon (
It is necessary to have a transcription terminator region downstream of TAA, TGA or TAG). In addition, transcription promoters include trp promoter, lac promoter, tac
Promoters such as PL promoter, circle 105 promoter, ADCI promoter, etc. can be used, and as transcription terminators, trp terminator, rr mouth BTIT, etc. can be used.
You can use 2 terminators and 1 ADCl terminator. Such expression vectors include, for example, pKK22
3-3 (Pharmacia), pPL-1ambda
(Same as on the left) Although these can be easily obtained from commercial products, it is also possible to use products that have been improved to have higher expressivity or handling properties.

(3) Construction of human 1-TNF transformant expression vector human TN
Examples of methods for producing DNA encoding the F converter polypeptide include the following methods.

(i) Produced by appropriately linking oligonucleotides chemically synthesized according to the method described in (1) Acquisition of human TNF gene. According to this method, modifications such as substitution, addition, or deletion of amino acids, polypeptides, etc. can be made freely.

(ii) After cutting the human TNF gene prepared in (1) above with an appropriate restriction enzyme and removing a specific region within the gene, a synthetic oligonucleotide having a mutated base sequence (e.g., the upper strand (double-stranded DNA fragments ligated by annealing) or an appropriate gene. According to this method, as in the case (i) above, modifications can be made freely.

(ii) A synthetic oligonucleotide having a mutated base sequence is used as a primer to extend the DNA chain to produce a converter polypeptide gene (site-directed mutagenesis method (T, A, Kunkel et al.
l, Methods in Enzymology,
154.367 (1987)). Although the same modification as in (i) above can be made using this method, it is particularly suitable for substituting any amino acid.

In the present invention, the site-directed mutagenesis method (ii) is preferred, so the following description will focus on this method.

First, in performing the site-directed mutagenesis method, a template DNA is prepared. The human TNF gene obtained in (1) above,
Messing et al. 1 (methods in Enzym
ology.

153, 3 (1987)) for the preparation of single-stranded plasmid DNA (pUc
118, pUcl19) and introduced into an E. coli strain. A clone having the desired plasmid is selected from among the transformants obtained.

This plasmid was used in dut- and ung-mutant E. coli strains (
CJ236 strain, etc.), incorporate uracil into the gene, and infect E. coli strain with mutant helverphage such as M13KO7 to obtain the desired single-stranded plasmid DNA.
get.

On the other hand, an oligonucleotide primer of about 15 to 50 bases having the conversion introduction site according to the present invention and the base sequences before and after the site is chemically synthesized. After this primer and the uracil-introduced-Kigumi plasmid DNA obtained in the above step are annealed, they are made into double strands using, for example, T4 DNA polymerase and T4 DNA ligase. The double-stranded plasmid was introduced into an ung+ E. coli strain, and colony hybridization was performed using the primers used in the resulting transformant as a probe to detect the DNA encoding the desired transformant polypeptide. Select clones with plasmids containing

By cutting out the DNA encoding the amino acid sequence of the human TNF converter polypeptide from the plasmid obtained above by restriction enzyme cleavage treatment, inserting it into an expression vector and cloning in the same manner as in (2) above,
It is also possible to construct an expression vector for a human TNF converter of interest.

For introducing an expression vector into a host cell such as an E. coli strain, a known method (Mol.
ecular Cloning, T., Maniat
iset at, (1982)).

As host cells, microbial cells such as Escherichia coli, Bacillus subtilis, and yeast can be used.Among them, E. coli JM83,
Examples include mutants of E. coli K-12 strains such as JM103.

(4) Obtaining human TNF converter polypeptide In the present invention, the transformed microbial cell described in (3) above is cultured, and the desired human TNF converter polypeptide is produced and accumulated in the culture. To separate. Conventionally known methods for culturing microbial cells, especially Escherichia coli,
For example, a method can be used in which E. coli is inoculated into a culture solution containing the nutrients required by E. coli and cultured in large quantities in a short period of time, usually by shaking or stirring at 32 to 37°C for about 12 to 24 hours. As a medium, for example, L medium or M9 medium [see above Molecular Cloning] can be used, and if necessary, antibiotics such as ampicillin may be added, and lac promoter ta may be added at the start of culture or during culture.
Isopropyl-β-D-thiogalactopyranoside can also be added to increase the efficiency of transcription promoters such as the c promoter.

After culturing the human TNF converter polypeptide of the present invention, the microbial cell aggregate is usually disrupted by ultrasonication while suspended in Tris buffer, and then centrifuged to remove bacterial cell residue. Obtained by removing. Furthermore, the product thus obtained can be layered by combining treatment with a nucleic acid/endotoxin remover, filtration with a filter, anion exchange chromatography, and other conventional protein separation and purification methods.

The human TNF converter polypeptide of the present invention has an antitumor effect similar to the antitumor effect of human TNF, and
It exhibits superior antitumor activity compared to other drugs, and has fewer side effects, so it is effective as an antitumor agent and an active ingredient in medicine. In the human TNF converter polypeptide of the present invention, Pro at position 68 in the amino acid sequence is Asp, Met or T.
Those substituted with yr are preferred. When formulating the pharmaceutical composition, it can be formulated into a pharmaceutical composition together with a pharmacologically acceptable carrier or diluent. The dosage form of the pharmaceutical composition of the present invention includes external preparations, oral preparations, injection preparations, etc., and is administered by an administration method suitable for each dosage form.

Example 1 (Design of human TNF gene) Based on the amino acid sequence of the human TNF structural gene, which has already been reported by CPennca et al., supra, a sequence list is prepared for the nucleotide sequence of the human TNF gene for convenience of gene construction and conversion generation. A DNA base sequence of SEQ ID NO: 3 was designed. Here, restriction enzyme recognition sites were incorporated at appropriate intervals, and a cleavage site using the restriction enzyme EcoRI was placed upstream of the translation start codon (ATG) and downstream of the translation stop codons (TAA and TGA) for easy ligation with the plasmid vector. A cleavage site using the restriction enzyme HindI was provided in each of the nucleotides.

Example 2 (Chemical synthesis of oligonucleotides) The DNA designed in Example 1 was synthesized using an automatic DNA synthesizer (Applied Biosystems).

It was chemically synthesized by the phosphite method using Model 381A). Synthesis was carried out by dividing into 20 oligonucleotides, U-1-, 10 and L-1 to 10, each having the base sequence shown in SEQ ID NO: 4 in the sequence list. ) and removal of the protecting group according to the manual of Applied Biosystems. Separation and purification of each oligonucleotide was performed by HPLC (high performance liquid chromatography) using a reverse phase chromatography column or by polyacrylamide gel electrophoresis (gel concentration 10 to 2096) containing 7M urea.

For HPLC method, Nucleozil 5C18 column (
It was separated and purified by reverse phase chromatography using a φ4.6 X l 50 +nm (manufactured by Chemco) and eluting with triethylaminoacetic acid (100 mM) buffer (pH 7.0) containing acetonitrile.

The above elution was performed using a linear concentration gradient of acetonitrile from 5 to 35
% (30 minutes), and the peak at about 15 minutes was collected.

Regarding the polyacrylamide gel electrophoresis method, each synthetic oligonucleotide sample is separated by electrophoresis, a band portion of the desired size is cut out based on the observation results of the migration pattern by ultraviolet shadowing method, and the polyacrylamide gel fragment is divided into approximately 1 ~2 [[Cut into 1 m3 pieces, add approximately 2 mn of elution buffer (0,5M NH
40AC and 1mM EDTA) was added and shaken overnight at 37°C. The elution buffer containing each oligonucleotide was collected and extracted with phenol (50% phenol').
/ 5096 chloroform solution), isobutanol extraction was performed, and purified samples of each oligonucleotide were obtained by ethanol precipitation.

Some of the synthesized and purified oligonucleotides were subjected to the Maxam-Gilbert method (A, M, Maxam et al.
lMethods in Enzymology, 65
499 (1980)), it was confirmed that the target base sequence existed.

In the operations related to genetic recombination described below (Examples 3.4 and 5), the reaction conditions for restriction enzymes and other related enzymes were mainly in accordance with the method described in Molecular Cloning (supra). The enzymes mentioned above were mainly obtained from Takara Shuzo, and Takara Shuzo's manual was also used as reference.

Example 3 (Human T by ligation of synthetic oligonucleotides)
Construction of NF gene) (1) First, an attempt was made to construct a human TNF gene according to Figure 1.

The synthetic oligonucleotides obtained in Example 2 were divided into three groups (U and L-1 to 4, U and L-5 to 7, and U and L-8 to 10) and cloned. That is, the 5' end of each oligonucleotide (1 to 2 μg) of U-2, 3, 4, 6, 7, 9 and IO and L-1, 2, 3, 5, 6, 8 and 9 was Unit T
4 polynucleotide kinase (Takara Shuzo), each is phosphorylated separately. The phosphorylation reaction was performed in 10 μl of an aqueous solution (50 mM Tris-HCI! pH 7, 6, 1
0mM MgC12, 0, 1mM Spermidine
, 0.1mM EDTA, lomMDTT and 1mM
ATP), carried out at 37°C for 1 hour, and after the reaction was completed,
T4 polynucleotide kinase was inactivated by treatment at °C for 10 minutes. Newly added 1 to 2 μg of U-1
, 5 and 8 and L-4, 7 and IO, respectively, an aqueous solution IO having the same composition as above.
Prepare 20 μl of each oligonucleotide (U-1-10 and L-1-10) aqueous solution with the same U and L numbers (20 μl), boil for 5 minutes at 100°C, and then slowly cool. Annealed. Next, the obtained 1
0 annealed bodies (double-stranded DNA fragments) were added to each group in an aqueous solution (66mM Tri
s-HCffi pH7,6,6,6mM MgC! !
2.10mM DTT, 1mM ATP and 100μg
/mlBSA), the total liquid volume is 120-160μ,
After annealing by heating at 40°C and slow cooling, 700 units of T4 DNA ligase (Takara Shuzo) was added and 16
The ligation reaction was carried out for 15 hours at °C. After the reaction, each reaction solution was subjected to polyacrylamide gel electrophoresis (gel concentration 696
), and the desired size (176 bp,
Band portions of 150 bp and 153 bp) were cut out, and three target DNA fragments were recovered by electro-elimination. Furthermore, each collected sample was subjected to phenol extraction (using a 50% phenol-150% chloroform solution) and isobutanol extraction, and the target DNA was purified by ethanol precipitation. According to the above method, three purified double-stranded DNA fragments were separately phosphorylated at their 5' ends using T4 polynucleotide kinase, mixed in an aqueous solution for ligation reaction, and then Annealing is performed by heating to T4
DNA ligase was added for ligation. This ligated DNA was recovered by ethanol precipitation and mixed with 1mM DTT and l
50μ1 (7) Hi/Lutobuffy (50mM Tris-HCJ) containing 00 μg/ynlBsA
? pH7, 5, 100mM NaCl, 10mM M
gC12) and 15 units of the restriction enzyme Eco
RI (Takara Shuzo) and 15 unit restriction enzyme Hind
III (Takara Shuzo) was added, and the cleavage reaction was carried out at 37°C for 2 hours. After the reaction was completed, the desired DNA fragment (approximately 480 bp) was separated and purified by polyacrylamide gel electrophoresis (gel concentration: 496) according to the above method.

On the other hand, 5 μg of plasmid vector ptJc9 (distributed from General University Genetic Information Laboratory) was cut with restriction enzymes EcoRI and Hind III according to the method described above, and subjected to agarose gel electrophoresis (gel concentration 196) to approximately 2 μg. A .7 Kbp DNA fragment was isolated and purified.

The previously purified approximately 480 bp DNA fragment (containing the human TNF gene) and this pUC9 fragment were mixed in a 20 μl ligation reaction solution according to the method described above, and 350 units of T4 DNA ligase was added. The ligation reaction was carried out at 16°C for 3 hours. Calcium chloride method (Molecul
E, col produced by ar Cloning]
i Competent cells of K-12JM83 strain (distributed by General University Genetic Information Laboratory) were transformed with the above ligation reaction solution according to a conventional method (Molecular C1
oning)] A plasmid was prepared from the obtained ampicillin-resistant reference port using a known method, treated with restriction enzymes (EcoRI and HindIII) according to the method described above, and then subjected to agarose gel electrophoresis. Insertion of the human TNF gene into the ρUC9 vector was investigated by pattern analysis. As a result, it was confirmed that a gene of approximately 250 bp had been inserted, and the nucleotide sequence of the inserted gene for that clone was determined using the dideoxy method CF, Sanger, 5ci.
ence, 214.1205 (1981)), approximately 130 b downstream from the 1EcoRI site.
It was confirmed that this gene fragment had the target human TNF gene nucleotide sequence of about 90 bp upstream from the p and HindIII sites. This clone and plasmid were named ptJA41/JM83 and pUA41, respectively.

(2) Subsequently, an attempt was made to construct a human TNF gene according to FIG.

Two groups (
U and L-3 to 6 and U and L-6 to 9), each oligonucleotide was phosphorylated and annealed in the same manner as in step (1) above, and then ligated using T4 DNA ligase. According to the above method, two DNA fragments (201 bp and 200 bp) were separated and purified by polyacrylamide gel electrophoresis (gel concentration 696).
p) separately with 100 μl of 67rnM Tri
s-HCl (pH 8,8), 6, 7mM MgCl2,
16,6mM (NH4)2SO4,6,7μM ED
T.A.

ImM DTT, 200 μg/ml BSA and 330 each
μM deoxyribonucleotide triphosphate (dATP,
dGTP, dcTP and TTP) dissolved in an aqueous solution,
2-5 units of T4 DNA polymerase (Takara Shuzo)
DN by adding and reacting at 37°C for 30 minutes.
Both ends of the A fragment were blunted. After the reaction, T4 DNA polymerase was inactivated by treatment at 68°C for 100 minutes, and the two target DNAs were separated by ethanol precipitation.
Fragment A was recovered.

Meanwhile, 5 μg of ptJc9 vector was added to 50 μl of medium salt buff 7-(lomM Tris-HC) containing 1 mM DTT and 100 μg/mff1 BSA.
J', pH 7, 5, 50mM NaC! and 10mM
MgC12) and 15 units of restriction enzyme Hi
After adding ncI (Takara Shuzo) and reacting at 37°C for 2 hours, the mixture was recovered by ethanol precipitation. The two DNA fragments previously recovered after blunting were added to the resulting cut and opened pUc9 vector separately according to the method described above.
4 Integrate using DNA ligase and incubate with E. coli
K-12JM83 strain was transformed. For each clone obtained, the nucleotide sequence of the inserted DNA was examined in the same manner as in step (1) above, and it was confirmed that it was the desired nucleotide sequence. These clones were each pUA
42/JM83 and pUA43/JM83, and the plasmids were named pUA42 and ptJA43.

The pUA41 obtained in steps (1) and (2) above was treated with the restriction enzyme EcoRI (high salt buffer) and Sa
c I (Takara Shuzo: raw salt buffer), restriction enzyme Hae I [(Takara Shuzo) and HindIII (medium salt buffer) to convert pUA42 to restriction enzyme 5a.
pUA43 was digested with cI (raw salt buffer) and paI (Takara Shuzo KCl buffer), and pUA43 was digested with the restriction enzymes Hpa I and HaeI (KCl buffer), respectively, according to the method described above. Raw salt buffer (10mM Tris-HCJ'pt(7,5
and 10mM MgCj'2) and high salt buffer or KCl<Tsnor (20mM Tris-HC
jl' pH 8,5,100mM KCn and lomM
For the Mg(J'2) combination, the cleavage reaction is
This was handled by dividing the reaction into batches and performing ethanol precipitation between the reactions. EcoRI-3ac from pUA41
lDNA fragment (127bp) and HaeII-Hin
dII [DNA fragment (80 bp), S from pUA42
ac I-Hpa I DNA fragment (147bp)
and Hpa I-HaeII D from pUA43
The NA fragments (126 bp) were separated and purified by polyacrylamide gel electrophoresis (gel concentration 6%) according to the method described above.

On the other hand, 5 μg of pUc19 plasmid vector (distributed from General University Genetic Information Laboratory) was cut with restriction enzymes EcoRI and HindIII according to the method described above, and the approximately 2.7 kbp DNA fragment was subjected to agarose gel electrophoresis. (gel concentration 196). The four previously purified DNA fragments were sequentially added as shown in the diagram (Figure 2), and ligated using T4 DNA ligase according to the method described above.Finally, the purified ptJc19 vector (approximately 2 .7 Kbll fragment) and transformed the JM83 strain. Plasmid D contained in this transformant
Regarding NA, when the base sequence of the inserted DNA was examined according to the above method, it was found that the desired full length (approximately 480
A plasmid vector (bp) containing the human TNF gene (
It was confirmed that the clone had a length of approximately 3.2 Kbp). This clone was named pUA44/JM83, and the plasmid was named pUA44.

Example 4 (Construction of human TNF expression vector) (1)
Expression vector pKK with E. coli tac promoter
The following improvements were attempted to make 223-3 (obtained from Pharmacia) easier to handle.

i) Reduce the molecular weight of the expression vector.

i) The recognition site for the restriction enzyme BamH is unique.

1ii) The direction of the tac promoter is opposite to that of the ampicillin resistance gene.

The method is illustrated in FIG.

Following the method of Example 3, 5 μg of plasmid vector pBR322 (distributed from Genetic Information Experimental Facility, General University of Japan) was cut with restriction enzymes EcoRI and HindI, and both ends thereof were digested using T4 DNA polymerase. Smoothed. According to the method of Example 3, agarose gel electrophoresis (
A DNA fragment of approximately 4.4 Kbp was separated and purified using a gel concentration of 1%), and a Bgl■ linker (a 1Obp double-stranded DNA fragment containing a restriction enzyme BglII recognition site: obtained from Takara Shuzo) was added to the open ring site using T4. Insertion ligation was performed using DNA ligase. Among the transformants obtained according to the method of Example 3, the target restriction enzyme Eco
Lacks RI and Hind■ recognition sites and uses the restriction enzyme BglI
Plasmid pBR9333 (approximately 4
．． 4 Kbp) was obtained.

5 μg of the plasmid pBR9333 obtained above was dissolved in high salt buffer according to the method of Example 3, and cut with restriction enzymes BglII (Takara Shuzo) and Pvu II (Takara Shuzo) to obtain approximately 2 μg containing the origin of replication. 3Kbp DN
The A fragment was separated and purified by agarose gel electrophoresis (gel concentration 1%). On the other hand, expression vector pKK223-3
(approximately 4.6 kbp) was dissolved in the same high salt buffer as above, and added with the restriction enzyme BamHI (Takara Shuzo).
and 5cal (Takara Shuzo). For the cleavage reaction using the restriction enzyme Bam, the amount of added enzyme is approximately 1/2 of the usual amount.
Partial cutting was performed with a reaction time of 5 to 30 minutes.

After cutting, tac promoter and rrnBT+T+
Approximately 1. including terminator etc. A 1 Kbp DNA fragment was separated and purified by agarose gel electrophoresis in the same manner as above.

Approximately 2.3 kbp DNA containing the previously purified origin of replication
Approximately 1. E. coli K-12JM103 strain (Joku University Genetic Information Laboratory) was prepared by inserting and ligating 1 Kbp DNA fragments using T4 DNA ligase as described above, and using the calcium chloride method as in Example 3. were introduced into competent cells. Among the transformants obtained, a clone having an expression vector (approximately 3.1 (bp)) containing the desired tac promoter etc. was selected, and this expression vector was named IIKKIOI.

(2) The next step will be explained according to FIG.

5μ of the expression vector pKKlo1 obtained in step (1) above
g was treated with restriction enzymes EcoRI and H according to the above method.
After cutting with indI, a DNA fragment of about 3.4 kbp containing the replication origin, transcriptional regulatory region, etc. was separated and purified by agarose gel electrophoresis (gel concentration 1%).

In addition, the plasmid pUA44 (approximately 3.2 kbp) containing the human TNF gene obtained in Example 3 was similarly digested with the restriction enzymes EcoRI and HindlI to obtain approximately 480 bl containing the entire human 1-TNF gene. )
The DNA fragments were separated and purified by agarose gel electrophoresis.

This DNA fragment containing the entire human TNF gene was inserted and ligated into the approximately 3.4 kbp DNA fragment previously purified from the expression vector pKK101 using T4 DNA ligase according to the method described above. E,co
li K-12JM103 strain. The desired human TNF expression vector (approximately 3.9K
bp) was selected, and this expression vector was named pKF4102.

Example 5 (Construction of human TNF transformant expression vector) (1
) This will be explained according to FIG.

The plasmid pUA44 obtained in Example 3 was treated with restriction enzymes EcoRI and HindIII according to the method described above.
The human TNF gene (approximately 480 bp including the entire region) was cut with
) DNA fragments were separated and purified by agarose gel electrophoresis. On the other hand, Messing et al.
nEnzymology, 153, 3 (1987
)) The plasmid vector pUcl19 for Ichikigumi plasmid DNA preparation (manual from Takara Shuzo) was treated with the restriction enzymes EcoRI and Hind.
Cut with llI to create approximately 3.2 Kbp DNA including IC region.
The A fragment was separated and purified by agarose gel electrophoresis.

This IC region (interg of M13 phage DNA)
nic region), plasmid pU
After infection with helper phage M13KO7, cl19 preferentially becomes a single-stranded DNA, is encapsulated in phage particles, and is released outside the bacterial cell. The approximately 480 bp DNA fragment containing the entire human TNF gene purified above and the approximately 3 bp DNA fragment containing the IC region.
．． The 2Kbp ptJc119 fragment was inserted into T4 as described above.
The resultant was ligated using DNA ligase, and introduced into E. coli K-12JM83 strain according to the method of Example 3 above. A clone having the desired plasmid (approximately 3.7 Kbp) was selected from among the obtained transformants, and this clone was named puct 19-hTNF/JM83, and the plasmid was named pUcl 19-hTNF.

In order to incorporate and retain uracil into the DNA of the plasmid peel 19-hTNF obtained above, Escherichia coli strain CJ236 (dut-, ung-) was prepared by the calcium chloride method according to the method of Example 3 above. were introduced into competent cells. Because the CJ236 strain has a mutation (duji) in the thymidine synthase gene, it can create DNA that incorporates uracil instead of thymine.
Furthermore, the uracil can be retained in the DNA by the ung mutation. This CJ236 strain is bio-
Obtained from Rad. Clones obtained by the above introduction (
pUcl19-hTNF/CJ236) was infected with helper phage M13KO7 (obtained from Takara Shuzo Co., Ltd.).
00tt g/ml! Ampicillin, 70μg/ml
Kanamycin and 30 μg/m I! 2XYT broth containing cyclophenicol (1,6% tryptone, l96 yeast extract, 0.5% NaCl and p17
．． 6), single-stranded plasmid DNA (approximately 3.7 Kbases) containing the desired uracil
) was encapsulated in phage particles and released outside the bacterial cells. The released phage particles were collected from the culture medium, and the desired single-stranded plasmid DNA was prepared according to the method for preparing Kigumi phage DNA.

(2) Next, explanation will be given according to FIG.

Primers 4291, 4292, and 4293 shown in sequence number 5 were designed to introduce mutations into the human TNF gene using a positive-strand oligonucleotide. Chemical synthesis and purification of this oligonucleotide were performed according to the method of Example 2 above.

Site-specific mutation introduction into the human TNF gene is a biotechnology
Rad's system (Muta-GeneoMin vit
r.

mutagenesis kit). That is, about 0.5 μg of the primer prepared above
' end was phosphorylated with T4 polynucleotide kinase according to the method described above, and the previously prepared uracil-introduced heavy chain plasmid DNA (pUc 119hT
NF) and approximately 2000 g of I Oμl annealing buffer (20 mM Tris-H (J'
p) 17.4゜2mM MgCA 2 and 50mM
Annealing (heating to about 70°C and then slow cooling) was performed in NaCJ'. After annealing, add 10x synthesis buffer (5mM each deoxyribonucleotide triphosphate (dATP, dGTP, dcTP, and TTP)
), 10mM ATP, 100mM TrisHC
ffpH7, 4, 50mM MgC12 and 20mM
Add 1/10 volume of DTT) and T4 DN of the unit.
Double stranding reaction using A polymerase and 2-4 units of T4 DNA ligase (37°C and 90 minutes)
I did it. TE Batsuno-(lomM Tris-H
The reaction was stopped by adding approximately 8 volumes of CJ' (pH 7.5 and 1 mM EDTA) and freezing. E produced by the calcium chloride method according to the method of Example 3 above.
The above reaction solution was treated with the double-stranded DNA into competent cells of E. coli K-12TGI strain (ung'': Amersham).

By introducing hetero double-stranded DNA into the uB+ strain,
The DNA strand containing uracil, which is a template, is inactivated and is not a target for replication. Therefore, the mutation introduction frequency becomes more than 50% efficient. From among the obtained transformants, a plasmid (approximately 3.7 Kbp) containing the desired transformant DNA was extracted using the colony hybridization method using the primer used for mutation introduction as a probe.
A clone with the following was selected. For the selected clones, the nucleotide sequence around the mutation introduction site of the plasmid was examined by the dideoxy method CF (Sanger, supra), and it was confirmed that the transformant DNA had been mutated as designed. These plasmids were each called pUC11.
9-F4291, pUcl19-F4292 and p
It was named Ucl 19-F4293.

The target human TNF converter gene obtained by mutagenesis was transformed into an expression vector PK! having a tac promoter according to the method for constructing a human TNF expression vector in Example 4.
(A human TNF converter expression vector was constructed by integrating into 101.The human TNF converter gene (approximately 480 bp) was
Approximately 3.7 Kbp plasmid pUc11 obtained above
9-F4291, pUc119-F4292 and pUcl 19-F4293 according to the above method,
They were separated and purified after cleavage with restriction enzymes EcoRI and HindIII, respectively. Each target human T
NF converter expression vector (pKF4291, pKF
4292 and pKF4293), similar to the human TNF expression vector (pKF4102), use E. col as a host.
i I (obtained using -12JM103 strain.

The above three types of transformant expression vectors can induce the following:
A novel bioactive polypeptide with the transformations shown below is produced in E. coli.

Vector pKF4291: Polypeptide F4291 (
The 68th Pro of human TNF is converted to Asp).

Vector pKF4292: Polypeptide F4292 (
The 68th Pro of human TNF is converted to Met).

Vector I] KF4293: Polypeptide F4293
(Conversion of Pro at position 68 of human TNF to Tyr).

Example 6 (Expression and purification of human TNF and human TNF transformants by E. coli) The human TNF expression vector (pl(F) obtained in Example 4)
4102) and the human TNF transformant expression vector (pKF4291. pKF4292, pKF4292, pKF4292) obtained in Example 5.
E. coli K-12JM with KF4293)
103 strain was cultured in M9 medium (0,69
6Na2HPO<,0.3% to 82PO,,0,05%
NaCi', 0.1% NH4Cl, 2mM MgS
O4, 0.2% glucose and 0.1mM CaCl
2) It was inoculated into 20 ml and cultured at 37°C for 18 hours. 20 ml of this culture solution was added to 1 liter of the above medium, and cultured with shaking at 37°C for 12 to 3 hours. Next, isopropyl-β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 1 mM, and culture with shaking was continued at 37°C for 18 hours.

After collecting E. coli cells by centrifugation, TP buffer (10mM Tris-H (J' pH 8, 0 and 1)
The bacterial cells were washed using 00 μM PMSF).

After washing, the bacterial cells were suspended in TP buffer with a volume of 10 (rrl) per bacterial duram, and the cells were placed in an ultrasonic generator (heat generator).
The bacterial cells were subjected to ultrasonic disruption using a Microbial System, Model W-225). The resulting suspension was centrifuged to remove bacterial cell residue and collect the supernatant fraction. The purification process after this ultrasonic crushing treatment is mainly carried out at low temperatures (0°C to 4°C).
°C).

After filtering this supernatant with a 0.45 μm filter, anion exchange chromatography (column size: φ2.
5 x 1.5 cm and flow rate: 0.5 mA/min) and assayed for the presence or absence of activity according to the 5DS-polyacrylamide gel electrophoresis described below and the method of Example 7 using L929 cells. An active fraction was obtained. A TP buffer containing NaCl was used for elution, and the NaCi7 concentration was increased stepwise from 0.05M to 0.1M, 0.2M, and 0.5M. Body (F429L F4292 and F4293
) eluted with 0.5 M Na (J'), respectively.

Furthermore, in order to remove endotoxin and the like derived from E. coli cells, treatment with a nucleic acid/endotoxin removal agent C-9 (manufactured by Kurita Kogyo, sold by Dainippon Pharmaceutical) was performed according to the attached manual. Thus, one of human TNF and human TNF converters.
Step-purified samples were obtained.

Using this sample, the antitumor activity of Examples 7 and 8 described later was evaluated.

After filtering the above sample with a 0.20 μm filter, F
Things Q■ (HR10/1) under control by PLC system
0 and HR515) prepacked column (Pharmacia L
TPQ buffer containing NaCl was analyzed by anion exchange chromatography using KB Biotechnology Co., Ltd.
-(20mM Tris-HCi7 pH8,0 and 1
0MM PMSF) and fractionated with 5DS as described below.
- An active fraction was obtained by assaying for the presence or absence of activity according to the method of Example 7 using polyacrylamide gel electrophoresis and L929 cells. The elution method was performed according to the following program under the control of the FPLC system. The third step was repeated several times until a single band was obtained in 5DS-polyacrylamide gel electrophoresis to increase purification purity.

1st step: Use MonoQ■ (HR10/lo) column (flow rate: 4ml 11min) 0-5min; 0-0.2M NaCl linear concentration gradient 5-
16 min; 0.2 M NaC178~IO min; 0.
2-0.5 M NaC] linear concentration gradient 20-24 minutes;
0.5 M NaCf The fractionation results (NaC1 concentration and retention time) of the active fraction according to the above method were as follows: human TNF was 0.2 M, 5.3 minutes, and human TNF converter F4291 was 0.5 M NaCf.
．． 2M, 5.6 minutes, same F 4292 is 0.2M, 5.3
Minutes and the same F 4293 or 0.2M.

It was 5.8 minutes.

2nd step: Use MonoQ■ (HR515) column (flow rate: Lml/min) θ~2.5 min; O-0,2M NaCl! Linear concentration gradient 2.5-8 min; 0.2 M NaC178-IO min; 0.2-0.5 M Na (J' linear concentration gradient lO
~12 minutes-〇. 5M NaCn The fractionation results (NaCJ' concentration and retention time) of the active fraction according to the above method are as follows: human TNF was 0.2M, 3.7 minutes,
Human TNF converter F 4292 was 0.2M, 4.7 minutes, and human TNF converter F 4293 was 0.2M, 3.7 minutes.

3rd step: Use MonoQ@(HR515) column (flow rate: 1 ml/min) θ~6 min; O-0,15M NaC/linear concentration gradient 6
~11 min; O,l 5-0.2 M Na (J' linear concentration gradient position 1-13 min; 0.2-0.5 M NaC#
Linear concentration gradient for 3 to 15 minutes; 0.5 M NaCn The fractionation results (NaCn concentration and retention time) of the active fraction according to the above method are as follows: Body F 4291 is 0.18M, 7.1 minutes,
The same F 4292 is 0.16M, 7.3 minutes and the same F 42
It was 93 O, 16M, 6.9 minutes.

Purified samples of human TNF and human TNF converters were thus obtained. This sample was used to evaluate the antitumor activity in Example 7 described below.

The above purification process and the purified sample were subjected to SDS polyacrylamide gel electrophoresis to confirm the expression and purification of human I TNF and the converter polypeptide. Each sample was treated with Laemml i' containing 10+y+M DTT.
s sample 6 buffer 7- (62,5mM Tris
H(J?pt(6,8,2%SDS, 0.01%B
PB and 10% glycerol) plus Laemmli
(Nature, 227630 (1970))
Electrophoresis was performed using a 15% separation gel according to the method of . After the electrophoresis was completed, the proteins in the separation gel were confirmed by staining with Coomassie Brilliant Blue. Figure 7 shows some of the results. The expression levels of human TNF and converter polypeptide by each expression vector were approximately the same, and the resulting stained gel was chromatographed.
When the expression efficiency was calculated using a scanner (Shimabara, C5-920 type), it was approximately 20% of the total E. coli cell protein.
Met. Also, the final purified sample had a single band in the resulting staining cell.

The molecular weights of human TNF and converter polypeptide were calculated from the electrophoresis positions and are shown in Table 1.

In addition, as one of the physical properties of proteins, the isoelectric points of human TNF and converter polypeptides are determined by ampholine (LKB
The results are shown in Table 1.

Table 1 Isoelectric points and molecular weights of human TNF and converter polypeptides (conversion sites/amino acids) Human TNF F 4291 (""Pro+Asp) F 4292 (
' ”Pro-+Met)F 4293(6”Pro→
Tyr) Isoelectric point 5.6 5.4 5.6 5.6 Molecular weight (Kd) 17, O 17,0 17, O I7.0 Example? (Evaluation of in vitro antitumor activity) Cytotoxic activity against mouse-derived connective tissue cell L 929 (ATCCCCLI) of the one-step purified sample and purified sample of human TNF and human TNF converter polypeptide obtained in Example 6. Aggarwal et al. (J.

Biol, Chem, 260.2345 (19
85)). That is, a 96-well tissue culture microplate (manufactured by Corning)
L929 cells were seeded at 10.1 ml/well of 3XIO' cells and cultured overnight at 37°C in the presence of 5% carbon dioxide gas.

As a medium, Dulbe containing 10% fetal bovine serum was used.
Eagle's Minimum Essential Medium modified by cco (manufactured by Sigma) was used. The next day, the medium was replaced with the above medium supplemented with actinomycin D at a final concentration of 1 μg/ml, and samples serially diluted with this medium were treated in each well. After further culturing for 20 hours, the living cells attached to the plate were stained with a 0.5% crystal violet solution (0.5% crystal violet/20% methanol) (room temperature, 15 minutes). lmM C
aCl2 and 1mM MgCl! Phosphate buffer containing 2
-PBS (lomM Na ・K phospha
te, 0.8% Na (J' and 0.02% KCj7
), and then 0.01 containing 30% ethanol.
The crystal violet remaining on the plate was extracted using 0.1 mN of N hydrochloric acid solution, and its absorbance (A492) was measured.
using an EIA reader (Bio-Rad, model 2550)
) was measured. This absorbance is proportional to the number of viable cells. Therefore, the final dilution rate of the sample in the well showing the absorbance corresponding to 50% of the absorbance of the sample-untreated well is determined, and the reciprocal of this sample dilution rate is calculated as the number of units per rrl of the sample [units/m17]. Define.

The specific activity (unit/mg-p) of each sample was determined from the L929 cytotoxic activity (unit/mA) determined according to the above method.
In order to calculate protein protein (rotein), protein quantification of each sample was performed. Quantification was performed using the Bradford method (Anal.

Biochem, 72.248 (1976)), protein concentration (mg/mn) was determined using bovine serum albumin as a standard sample. Based on these results, the specific activities calculated for human TNF and human TNF converter samples are shown in Table 2.

Table 2 Human TNF and converter polypeptide L929
Cytotoxic active human TNF 2.7xlO78,0xl
O7F4291 (“Pro-+Asp) 6.2xl
O62,4xlO7F4292(”Pro-+Met)
6.2XlO' 3.8x107F429
3(”Pro-+Tyr) 3.8xlO63,9x
107 From Table 2 above, it can be seen that the human TNF converter polypeptide has the same cytotoxic activity against mouse-derived connective tissue cell L929 as human TNF.

Example 8 (Evaluation of in vivo antitumor activity) Regarding the one-step purified samples of human TNF and human TNF converters obtained in Example 6, mouse transplantable fibrosarcoma MethA
Antitumor therapeutic activity against tumors was determined. The test consisted of lXl06 cells suspended in saline/0.2 m I! MethA tumor cells were subcutaneously transplanted into the back of a BALB/c 7 mouse (male, 5 weeks old, Charles River), and after 8 days, it was confirmed that the tumor had reached a diameter of 6 to lou+m, and then treated with physiological saline. The test was carried out by administering serially diluted samples (0.2 rrl/mouse) through the tail vein. The lethal dose was taken as the highest dose, and each dose sample was prepared by dilution in several stages.

Observation of tumor growth, etc. was continued for 2 weeks using the administration device.

Regarding tumor growth, the tumor volume (major axis x minor axis of tumor mass 2)
/2), determine the tumor volume ratio to the tumor volume on the day of sample administration (day 0), and calculate the number of days from the day of sample administration (D2 or D5) at which this value becomes 2 or 5. Then, the ratio to the control group (physiological saline administration) was calculated, and the D2% control value and D596 control value were determined.

The larger the value, the lower the tumor growth ability and the higher the antitumor activity.

The results obtained according to the above are listed in Table 3.

From Table 3, the human TNF converter polypeptide is human TNF
Compared to this, the tumor volume rate was high at the maximum dose when the mortality rate was 096, which indicates that the antitumor therapeutic effect is high.

Sequence number: 2 Length per strand: 465 Sequence type - Number of nucleic acid strands Knee heavy chain topology; linear Sequence type: Other nucleic acid synthesis NA Sequence (a) TCATCTrCTCGAACCCCGAG TGAC
AAGCCT GTAGCCCATG ■Revised AGCAA
A CCCTCAAGCTGMGGGCAGCTCCA
GTGGCT GMCCGCCGGGCCMTGCC
CTCCTGGCTM TGGAGT section AG anti-slip number: 3 Sequence length: 474 Sequence type: Number of nucleic acid strands, double-stranded topology: Linear fJ type: Other nucleic acid synthesis NA kνcriB 1 ) Number: Number from the start codon ('ATG) 2] M or M: Cutting position for oligonucleotide production (m3) 1i111 (ff enzyme: Restriction enzyme sequence prepared for human TNF gene construction) -Tadate's note 4] Lower case letters: Restriction fermentation pigeon belly button base 9j provided for ligation of human TNF gene to plasmid vector Sequence number = 4 Length of string: 27 (U-1), 50 ( U-4), 48 (U-7), 53 (U-1), 50 (L-3), 49 (L-6), 51 (L-9), Type of rice: Number of nucleic acid strands Knee heavy chain topology: Straight chain type: Other nucleic acid synthesis NA 30 (U~2), 49 (U-3), 50 (U~5), 52 (U-6), 49 (U -8), 51 (U-9), 29 (L-1), 52 (L-2), 50 (L-4), 49 (L-5), 51 (L-7), 49 (L- 8), 49 (L-10) (Effect of the invention) According to the present invention, in the amino acid sequence of human TNF, 6
By converting the 8th amino acid to another amino acid, a novel human TNF converter with higher antitumor activity and lower side effects than human TNF can be provided.

[Brief explanation of drawings]

Figures 1 and 2 show the method for constructing the human TNF gene by ligation of synthetic oligonucleotides, Figures 3 and 4 show the method for constructing the human TNF expression vector, and Figures 5 and 6 show the method for constructing the human TNF gene by linking synthetic oligonucleotides. FIG. 7 is a diagram for explaining the method for producing a gene for a human TNF converter by a mutagenesis method, and FIG.
It is a figure which shows the electrophoresis result of the purified sample of F converter. Sequence number = 5 Sequence length: 21 (Blymer 4291) 21 (Blymer 4292) 21 (Blymer 4293) Chain type: Number of nucleic acid strands Knee heavy chain topology: Straight chain Type of chain: Other nucleic acids Synthetic NA Last two * Mutation-introduced base] U-+ Figure ↓ UA41 to be incorporated into pUc9 [Note] O-5゛ terminal phosphorylation (-) ・- (+) Incorporated into pL109 to be incorporated into the obtained clone ptlc9 O-n pUA4+ UA42 UA43 UA41 Fig. pKK + 01 UA44 pKF4+02 Fig. 3 EcoRl/Polinsch Fig. Protein molecular weight marker F4291/- stage purified sample F4291/ purified sample F4292/- stage purified sample F4292/ purified sample F4293/- stage purified sample 1 F4293/ purified sample ←- human TNF converter (lT, o to d) Procedure amendment (voluntary) July 4, 1991

Claims (1)

[Claims] 1. Ser 1 to 15 shown in sequence number 1 in the sequence listing
A polypeptide characterized in that, in the amino acid sequence up to the 5th Leu, Pro at the 68th position is replaced with another amino acid. 2. The polypeptide according to claim 1, wherein the other amino acids substituted at positions 2 and 68 are aspartic acid, methionine, or tyrosine. 3. 15 from the 1st Ser shown in sequence number 1 in the sequence table
1. A recombinant plasmid comprising a DNA encoding a polypeptide having an amino acid sequence up to the 5th Leu in which Pro at position 68 is replaced with another amino acid. 4. The recombinant plasmid is pKF4291, pKF4
4. The recombinant plasmid of claim 3, which is pKF4292 or pKF4293. 5. 15 from the 1st Ser shown in sequence number 1 in the sequence table
A recombinant microbial cell transformed with a recombinant plasmid containing a DNA encoding a polypeptide having an amino acid sequence up to the 5th Leu in which Pro at position 68 is replaced with another amino acid. 6. The recombinant microbial cell according to claim 5, wherein the recombinant microbial cell is Escherichia coli. 7. Ser 1 to 15 shown in sequence number 1 in the sequence table
A recombinant microbial cell transformed with a recombinant plasmid containing a DNA encoding a polypeptide having an amino acid sequence up to the 5th Leu in which Pro at position 68 is replaced with another amino acid is transformed in a culture medium. A method for producing a polypeptide, which comprises culturing, producing and isolating a polypeptide having the amino acid sequence. 8. 15 from the 1st Ser shown in sequence number 1 in the sequence table
1. A pharmaceutical composition comprising, as an active ingredient, a polypeptide having an amino acid sequence up to the 5th Leu, in which Pro at position 68 is replaced with another amino acid. 9. 15 from the 1st Ser shown in sequence number 1 in the sequence table
A polypeptide characterized in that in the amino acid sequence up to the 5th Leu, Pro at position 68 is replaced with another amino acid, and has a Met at its N-terminus. 10, 465 from the 1st T shown in sequence number 2 in the sequence listing
202 to 2 in the base sequence represented by the number G
A DNA characterized by having a sequence in which CCA encoding Pro at position 04 is replaced with a codon encoding another amino acid.