JPH03201987A - Human serum albumin fragment - Google Patents

Abstract

NEW MATERIAL:A human serum albumin fragment deficient in carbon end part of human serum albumin. EXAMPLE:A human albumin fragment having an amino acid sequence from aspartic acid at the 1-position of human serum albumin to proline at the 303-position. USE:Preparation for drug carrier. PREPARATION:By recombinant DNA technology.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a protein fragment consisting of a portion of human serum albumin. This protein fragment is expected to be used as a carrier for transporting and supplying drugs, etc.

[Conventional technology]

Human serum albumin is a molecule synthesized in the human liver166
．． 458 high-molecular plasma protein, which in vivo mainly regulates blood osmotic pressure and regulates various substances (e.g. fatty acids, Cu'', Ni
It plays important roles, such as binding to metal ions such as Ha, bile bilirubin, many drugs, some water-soluble vitamins, etc.), transporting them to target organs, and serving as a source of amino acids to tissues. Based on these effects, human serum albumin is used in large quantities to treat hypoalbuminemia and hemorrhagic shock caused by loss of albumin due to burns, nephritis, etc., and decreased albumin synthesis due to liver cirrhosis. Serum albumin also binds non-specifically to many drugs and acts as their blood 1111. It is believed that drugs bound to albumin move through the body through blood circulation, eventually separating from albumin and diffusing through capillary walls to reach the site of action. Albumin is an excellent drug carrier for drug delivery because it has low toxicity, low antigenicity, is easily degraded in vivo, and can easily form covalent bonds and complexes with drugs. In addition, many of the binding sites for various drugs have been determined or estimated, and they also have the advantage of being easy to design for formulation.

Basically, fragments of human serum albumin also contain most of the predicted binding sites for many drugs.
It is thought that it can exhibit activity as a drug carrier. When used as a carrier, etc. in a drug delivery/supply system, it is preferable to use fragments of the human serum alpine molecule rather than the whole human serum alpha molecule from the standpoint of limiting the binding properties with drugs, etc. In general, known methods for cleaving proteins and preparing fragments thereof include using chemicals such as cyanogen bromide or proteases such as trypsin and pepsin. However, in these methods, the cleavage site is inevitably determined depending on the amino acid sequence of the protein, so it is not possible to cleave at an arbitrary site, and therefore it is not possible to obtain an ideal protein fragment. Therefore, no such fragment has been obtained for human serum albumin.

[Problem to be solved by the invention]

In contrast, using recombinant DNA technology, human serum albumin fragments consisting of any part can be synthesized in suitable host cells. Therefore, the present invention aims to provide a protein fragment of human serum albumin and a means for its production by producing a DNA encoding a desired protein fragment of human serum albumin and using recombinant DNA technology based on the DNA.

More specifically, the present invention provides C-
Human serum albumin fragments with the terminal portion deleted, and fusion proteins consisting of the fragments and other polypeptides; human serum albumin fragments with the N-terminal portion of human serum albumin deleted, and the fragments and other polypeptides. A fusion protein consisting of a peptide; DNA encoding these protein fragments or fusion proteins; a plasmid containing the DNA; a host transformed with the plasmid; and culturing the host to obtain a human serum albumin protein fragment or the fusion protein. A human serum albumin protein fragment or human serum albumin protein fragment, characterized in that a fusion protein containing the fragment is expressed, and when the fusion protein is expressed, the human serum albumin fragment is excised from the fusion protein in a host cell or in a test tube, if desired. The present invention relates to a method for producing a fusion protein containing a fragment.

[Specific explanation] The cDNA encoding normal human serum albumin A has already been cloned (Japanese Patent Application No. 63-037453)
.

Therefore, using this cDNA, any fragment of normal human serum albumin A can be produced by genetic engineering techniques.

The present invention provides such fragments as (1) a serum albumin fragment lacking the C-terminus of human serum albumin, and (
2) A serum albumin fragment lacking the N-terminus of human serum albumin is provided. The present invention provides, for example, an albumin fragment having an amino acid sequence from aspartic acid at position 1 to proline at position 303 of normal human serum albumin (miniH
3A), and N-
As an example of an albumin fragment with a deleted terminal, an albumin fragment consisting of an amino acid sequence from methionine at position 123 to leucine at position 585 of normal human serum albumin (this may be referred to as abbreviated ISA) will be described.

These two types of albumin fragments of the present invention each have the following characteristics.

All albumin fragments of the present invention contain the central portion of human serum albumin. In this way, including the central part, there are currently four known drug binding positions on the human serum albumin molecule (sizes l-1 to ■) (Sj
Oholm, 1. , Ek+++an, B, E,
+ Kober.

A., Ljugstedt-Pahlman, 1. , S
eiving, B., & 5jOdin, T.

Mo1. Pharmacol, 16,767- (19
(79)), some amino acid residues that play important roles in drug binding at these sites are also known (Fehske, et al., Biochem, Pharl).
This is because most of the I+aco1.30.688-(1981)) are concentrated in this central part.

Sjdholm et al. used microspheres in which drugs were uniformly dispersed in human serum albumin to investigate the binding positions of various drugs, and found that the diazepam site (site I), the digitoxin site (site ■), and the warfarin site (site ■)
However, in addition to this, tamoxifen (site
) or bilirubin binding sites. Fe
hske et al. diazepam site, warfarin site,
Lys195, old 5146 and Arg are amino acids that play an important role in the bilirubin binding site, respectively.
145, Trp214 and Lys199, Ly
s240 is estimated. On the other hand, the binding site for long-chain fatty acids such as palmitate seems to be located in the C-terminal region (
Reed, R,G −+ Fe1dhof f +
R, C, + C1u te+0, L, & Peters
, T, Tr, Biochemistry+ 14+ 4
578-(1975); Berde, C.B.
, Hudson, B.S., , Simon i.
, R.D., &5klar, L.A.J., Bio1.
Che+s, 254, 391 (1979)), by using the C-terminally deleted human serum albumin fragment of the present invention, it is possible to create a drug carrier that cannot bind long chain fatty acids and can bind diazepam, warfarin, etc. Become.

Human serum albumin is a high-molecular protein consisting of 585 amino acids and has 35 cysteine residues within the molecule, of which the cysteine residue located at the N-terminal side (
Only Cys-34) exists with free 5HJJ, and the others form disulfide (SS) bonds with each other, resulting in a total of 17 SS bridges being formed within the molecule. At least two types of enzymes [peptidylprolyl cis-tra
Recently, it has become clear that ns isomerase and protein disulfide isomerase (PDI) are involved, and it is the latter PDI that plays an important role in S-5 bridge formation. In mammalian cells that produce serum albumin, PDI acts during the biosynthesis and intracellular transport of serum albumin protein to form S-5 bridges within the protein molecule, and the main location of PDI is microsomes including the endoplasmic reticulum. It is known that it is a fraction. When human serum albumin is biosynthesized in prokaryotic cells such as Escherichia coli, the reaction described above occurs, and the correct S
There is no guarantee that a -3 bridge will be formed, and there is a possibility that Cys -34 will undergo a thiol/disulfide exchange reaction with a cystine residue in the molecule, resulting in misplacement of the S-3 bridge and generation of isomers. In this way, the presence of cysteine residues with free SH groups reduces the efficiency of producing proteins with normal 3D structures, and there is a risk that proteins with abnormal 3D structures may become functionally abnormal. Sex becomes big. On the other hand, in the albumin fragment of the present invention in which the N-terminal portion is deleted and consists of the amino acid sequence from methionine at position 123 to leucine at position 585, Cys34
is removed along with the other six amino-terminal cysteines to reduce this risk.

In the present invention, two types of albumin fragments having specific amino acid sequence ranges are specifically described as representative examples of the two types of albumin fragments, and each of the two types of albumin fragments has the above-mentioned characteristics. All albumin fragments capable of exhibiting these characteristics are within the scope of the present invention. For example, although the range from methionine at position 123 to proline at position 303 is exemplified as the central region where drug binding sites are concentrated, the central region is not necessarily limited to this range and includes most of the drug binding sites. As long as it is within the range, it may be longer or shorter than the 123rd to 303rd positions. In addition, the range from position 304 to the C-terminus was exemplified as the C-terminal region where a long-chain fatty acid binding site exists and should therefore be removed.
The range is not limited to this, and may be a longer range or a shorter range as long as it includes the binding site of long-chain fatty acids. Furthermore, although the range from the N-terminus to the 122nd position is exemplified as the range of the N-terminus that contains a large number of cysteines and should therefore be removed, the N-terminal region containing cysteine at position 34 is the N-terminal range that should be removed. It is not limited to the range from the end to position 122, and may be a longer or shorter range.

Therefore, various albumin fragments can be designed taking into consideration the following conditions, and they fall within the scope of the present invention. An important condition for designing fragments of human serum albumin is to select a fragment that can be expected to maintain the tertiary structure required for specific drug binding.

At this time, it is important to note that (i) the S-5 bridge that exists in the structure of natural human serum Alptan should be retained as it is; (ii) therefore, having an even number of cysteine residues, (i
ii) Avoid making cuts in the middle of polypeptide chains connected by S-3 bridges to form loops, that is, the main domain structures or at least subdomains that exist in several natural human serum alphanuclease molecules. The structure must not be destroyed, etc.

1. Normal human serum albumin contains many disulfide bonds within the molecule, and in order to produce normal human serum albumin or fragments with the same three-dimensional structure as natural products using recombinant DNA methods, It is essential that these disulfide bonds are formed correctly in the production host cell. It has recently been revealed that enzymes such as protein disulfide isomerase and peptidylprolyl cis-trans isomerase are involved in the formation of normal three-dimensional structures, and proteins that have many S-3 bonds and adopt complex three-dimensional structures have recently been found to be involved. In prokaryotic cells such as Escherichia coli and Bacillus subtilis, which contain almost no molecule, it is expected that the activity of enzyme systems related to 3D structure formation (folding) will not be strong, even if they exist. On the other hand, cells of eukaryotic higher organisms, including humans, are known to secrete many proteins with complex higher-order structures (including glycoproteins and other modified proteins); Yeast, which is an isoeukaryotic microorganism, is also known to secrete proteins through a pathway very similar to that in mammalian cells (11 u ff
aker, T.C., and Robbins, P.W.
J, Biol, Che+++, 257+3203-32
10 (1982); 5nider, M.D., in
Ginsburg, V. & Robbins + P. W.
eds,) Biology of Carbohydr
ates.

Vol.2. Wiley, New York, (198
4), pp, 163-1983゜For this reason, recent experiments have been carried out to express genes (mainly cDNA) derived from foreign organisms (especially mammals) in yeast and secrete the gene product, the protein, to the outside of the cells. Many attempts have been made.

For example, human interferon α1. αt+ r (H
itzeman, R.A.

Leung+ D, W, ,Perry, L, J, ,Ko
hr+W,J,,Levine+I1. L, +Goed
del, D, V, 5science 219.620-6
25 (1983)), calf prochymosin (Sa+
i th, R, A, ,Duncan, M, J,.

Mo1r, D, T, 5science 229.1219
-1224 (1985)), human epithelial predominant factor (Br
ake, A, J, , Merryweather+
J, P, +Colt, D, G,, leberle
ir++U,A,,Masialz,F,R,,Mul
lenbach+G,T,,Urdea,M,S,,V
alenzuela+P,,Barr,P,J.

Proc, Natl, Acad, Sci, USA, 8
1.4642-4646 (1984)), mouse interleukin 2 (Miyajima, A., Bond,
M.

W,,0tsu,to,+torai+to,,Arai+N
, Gene 3, 155-161 (1985))
, human β-endorphin [Bitter, G.

A,,Chen,ni,ni,,Banks,A,R,,L
ai+1', -H, Proc, Natl.

Acad, Sci, USA, 81.5530-553
4 (1984)), extracellular secretion by yeast has been reported, but the secretion efficiency varies considerably depending on the target protein, ranging from about 80% for mouse interleukin 2 to 4 to 10% for human interferon. . Furthermore, among these proteins, human interferon has been successfully transported into cells using its own signal peptide, and the signal peptide has been successfully cleaved and secreted. Others are yeast invertase (SU
C2) signal peptide and mating factor α1 (MFα
Intracellular transport is achieved by expressing a signal sequence necessary for intracellular transport of yeast proteins, such as the leader sequence (1), directly fused to the target mature protein. Furthermore, there are few substances that are clearly processed at the correct position; in the case of human interferon, about half are processed correctly, but in human β-endorphin, the peptide is also cleaved within the peptide.

The characteristics of genetic engineering material productivity using yeast as a host include the following.

1. Fermentation production by large-scale, high-density culture is easy and economical. Furthermore, compared to cultured cell systems for animals and plants, strictly controlled culture equipment is not particularly required.

2. Much experience has been accumulated in fermentation production.

3. Molecular genetic knowledge is rapidly accumulating.

4. It is easy to incorporate foreign genetic material into cells and genomes.

5. Our understanding of the genetics and physiology of intracellular transport and extracellular secretion of proteins is rapidly increasing.

6. If an appropriate positive vector is selected, the foreign gene can be integrated into the genome of Ebisomal LJt (using YEp-based plastic plasmid (using Ylp plasmid)), and chromosomal DNA containing yeast centromeres can be generated as cells divide. It can be placed in four different states: a state in which it can replicate together with yeast (using a YCp plasmid), and a state in which it can replicate autonomously containing the yeast autonomously replicating sequence (AR3) (using a YRp plasmid).

7. Has intracellular processing functions such as signal peptides and prosequences.

8. The sugar chains found in glycoproteins synthesized by yeast are high-mannose type tj, which is different from the complex type sugar chains in glycoproteins of higher animals and plants! Although it is a chain, the addition of core sugar chains that occurs in the endoplasmic reticulum of yeast is a process common to higher animals.
The only difference between the two is the addition of the outer IJM chain.

9. Transformants can be grown in a completely synthetic medium by adding vitamins, trace factors, etc.

10. Transformants can be grown using crude sugar sources instead of pure glucose.

Based on this background, yeast is used as a host in the present invention.

(Prepro sequence) In order to express human serum albumin fragments in yeast cells and secrete them efficiently, a prepro sequence must be present at the N-terminus. Furthermore, this pre-pro sequence needs to be excised during secretion of the target protein so that the target protein is secreted in a mature form. Therefore, in the present invention, the original prepro sequence of human serum albumin is used as a prepro sequence that satisfies such conditions.

In order to enhance the expression of a protein in yeast, it is preferable to use a codon that is efficiently translated in yeast as a codon encoding the N-terminal region of the protein. Therefore, in the present invention, a synthetic NA sequence composed of codons that are frequently used in genes that are efficiently expressed in yeast is used as the DNA sequence encoding the prepro sequence. For example, the following codon is used as such a codon.

The following sequence can be used as an example of a DNA portion encoding a prepro sequence.

Gly Vat Phe Arg Arg Upstream of the Met codon at the N-terminus of the above sequence is EcoRI.
A sticky end is provided and this restriction enzyme site allows the sequence to be inserted into the vector.

Furthermore, as the Arg codon at the C-terminus of the above prepro sequence, CGC is adopted instead of the codon mentioned above as preferable for translation in yeast, and this results in human Can be linked to serum albumin fragments.

Human 'Albumin Gene (cDNA) encoding human serum albumin A
has already been cloned, and its nucleotide sequence and amino acid sequence deduced from the nucleotide sequence are disclosed in Japanese Patent Application No. 63-0.
37453. Therefore, in the present invention, the plasmid puc-
HS^-CI etc. can be used as a source of genes encoding human serum albumin fragments. In addition,
A method for producing these positive electrodes will be described later as a reference example.

It is said that the polyA sequence and the AATAAA signal, which are present downstream of the 3'-end of the polyA sequence and the AATAAA signal coding sequence, contribute to the stability of mRNA0 in eukaryotes (Bergmann and Br.
awerman Biochemistry, 16.
259 264 (1977); Huez et al., Pr.
oc, Natl, Acad, Sci, USA, 78
, 908-911 (1981)). Therefore, in a preferred embodiment of the invention, these sequences are placed downstream of the cDNA encoding human serum albumin A. As the polyA sequence and AATAAA signal, for example, these sequences naturally associated with human serum albumin AcDN^ can be used. A human serum albumin A gene containing these sequences has already been cloned and is described in Japanese Patent Application No. 63-037453.

As a source of these sequences, for example, λgtll(IIs
A18) can be used, and its production method will be described later in Reference Examples.

Promoter In the present invention, any promoter can be used as long as it functions in yeast cells. However, in the present invention it is preferred to use constitutive promoters rather than inducible promoters. When induction is performed using an inducible promoter, human serum albumin may rapidly accumulate in cells, forming intermolecular disulfide bonds and producing molecules with non-natural three-dimensional structures. This is because there is.

Examples of weakly inducible or constitutive yeast promoters with strong activity include alcohol dehydrogenase (ADHI) promoter and glyceraldehyde-3-phosphate dehydrogenase (GAP) promoter.
promoter, and glycerate phosphate kinase (PG
K) promoter, and in the present invention, ADHI
A specific explanation will be given using a promoter as an example.

The base sequence of an approximately 2.100 base pair region containing the yeast ADHI gene (ADC1) has already been determined, and MDII
In addition to the approximately 1.100 base pair sequence encoding I, 750
A 5' untranslated sequence of base pairs and a 3' untranslated sequence of 320 base pairs are known (Bennetzen + J and Hall, B, J, Rial, Chem, 257.3
018 3025 (19B2)).

The Goldberg-Hogness box (TATA box), which is considered to be a recognition sequence by RNA polymerase during transcription, is located at 128 of the translation initiation codon ATG.
It is located upstream of the Sph I recognition site (position -128), and it is said that ADHI promoter activity is not lost even if the region upstream of the Sph I recognition site located at position -410 is deleted (Be
ier and Young, Nature 300, 724
728 (1982)).

Transcripts driven by the 80II1 promoter account for at least 1% of the total poly(A) RNA in normal yeast (An
+n+erer, G, Methods Enzymol
, 101+ 192-201 (1983)).

There have been reports of cases in which the amount of gene products decreases due to read-through during transcription [
For example, Zaret, ni, S, and 5her+++en
, F, +Ce1l 28+563-573. (198
2) ]. In order to prevent this phenomenon, it is preferable to provide a terminator downstream of the structural gene to be expressed. A yeast terminator is placed downstream of the foreign gene,
An example of increasing gene expression is an experiment in which calf chymosin was expressed using Sandwich Hector, which consists of a PGK promoter/terminator, and the introduction of a terminator increased expression several to ten times. It has been reported (Mellor et al. Gene2A+ 1
-14 (1983)). Terminators derived from various genes can be used for this purpose; for example, terminators such as the TRP5 (1-lyptophan synthase) gene and the CYCI (iso-1-cytochrome C) gene are used. has been done.

In the case of transcription involving a strong promoter, it is considered more convenient to control expression if a strong terminator is placed downstream of the promoter to prevent read-through. Therefore, in the present invention, it is preferable to use, for example, ADHI terminator, 1GAP terminator, etc., which are gene terminators having a strong promoter.

Although two or more elements directly related to expression contained in the expression plasmid of the present invention have been described, the expression plasmid of the present invention must further contain a yeast replication origin and a marker gene. As the yeast replication origin, for example, the replication origin of yeast-derived 2//m plasmid DNA can be used.

As the marker gene, commonly used marker genes can be used, such as genes that confer drug resistance to the host and genes that complement the nutritional requirements of the host. Furthermore, since it is necessary to replicate the plasmid in E. coli during plasmid recombination, the plasmid of the present invention is preferably a shuttle vector containing an E. coli replication origin and a marker gene.

A commercially available plasmid pJD8207 or the like can be used as a vector having the basic requirements for a shuttle vector. This plasmid pJ[18207
The yeast marker gene inside is β-, a leucine biosynthetic enzyme.
LEI encodes isopropylmalate dehydrogenase!
2 genes.

Therefore, in a preferred expression plasmid of the present invention, a promoter, a leader sequence encoding a prepro sequence is added to a shuttle vector comprising a yeast replication origin and marker gene and an E. coli replication origin and marker gene. A gene encoding a linked human serum albumin fragment, polyA
The array and terminator are inserted in this order.

2. Transformation of the host yeast with the plasmid of the present invention can be carried out according to a conventional method, and a specific example thereof is described in Example 9.

3. Growth of human albumin The host yeast strain transformed with the expression plasmid containing the human serum albumin fragment cDN^ can be cultured using conventional yeast culture methods. For example, it can be cultured in a natural complete medium such as YPD or an incomplete synthetic medium such as SD medium to which 1% Yeast Egis is added.

Human serum albumin fragments secreted extracellularly after culture can be recovered by various methods. ethanol, acetone,
It is expected that secreted human serum albumin fragments will be highly purified by combining various chromatography and the above partial purification methods after concentration and partial purification by fractional precipitation with ammonium sulfate, isoelectric focusing precipitation, ultrafiltration, etc. can.

A method for producing cDNA encoding all or most of normal human serum albumin A is specifically described in Reference Example 1. The DNA encoding the protein fragment of interest can be chemically synthesized in its entirety by conventional methods, or can be prepared from the cDNA described above. When prepared from cDNA, cDNA encoding all or most of normal human serum albumin A is isolated at the 5' or 3' end of the cDNA region encoding the protein fragment of interest using an appropriate restriction endonuclease. It is prepared by cutting the DNA and complementing the missing terminal coding sequence with chemically synthesized DNA. Alternatively, the cDNA can be cleaved with an appropriate restriction endonuclease outside the 5' or 3' end of the cDNAsI region encoding the protein fragment of interest, and then the excess DNA portion can be removed with an exonuclease. 5' of the above two methods
It is also possible to use a combination of different methods for processing the ends and 3' ends.

In the example of the present invention, the DNA encoding the fusion protein of human serum albumin signal peptide and propeptide and mini-ISA has already been disclosed in Japanese Patent Application No. 63-26830.
Plasmid pUc containing a DNA encoding a fusion protein of the human serum albumin signal peptide and propeptide described in 2 and a full-length mature human blood albumin molecule.
-HSA-EH to signal peptide and propeptide of human serum albumin and A of human serum albumin A
The DNA encoding up to spl-Pro152 was transformed into the plasmid pUc-ISA described in Japanese Patent Application No. 63-268302.
A fusion of a DNA fragment encoding G1u153 to Pro303 excised from -1 is used. Short I
The DNA encoding H3A lacks the portion encoding the carboxyl terminal side of human serum albumin (cDNA).
The prepro sequence of human serum albumin from clone USA-H (described in application S632/22) was transformed into pAT-trp-phoA-, which contains the DNA sequence encoding the signal peptide of E. coli alkaline phosphatase and the fusion protein of truncated human serum albumin. t) ISA (Hl,
The process is completed by obtaining the sections encoding the truncated human serum albumin (Met123 to Leu585) from the 8/25 application (described in the application on 8/25) and linking them using an appropriate method.

DN encoding the normal human serum albumin fragment of the present invention
Although A can be expressed as such, it can be expressed in a state linked to DNA encoding another peptide to obtain a fusion protein. Various peptides can be used as fusion partners to obtain such fusion proteins, one of which is, for example, the leader sequence of human serum albumin. When the desired human serum albumin fragment is obtained as such a fusion protein, the leader sequence can then be removed intracellularly or in vitro to obtain the human serum albumin fragment.

For expression of a human serum albumin fragment, for example, a DNA encoding a fusion protein as described above is inserted into a suitable expression vector, such as a plasmid, and then the vector is introduced into a host. As expression hosts, animal cells, eukaryotic cells such as yeast, and prokaryotic cells such as bacteria can be used, and the vector is selected depending on the host.

(Effects of the Invention) The albumin fragment of the present invention lacking the C-terminal region is
-Since it lacks the long-chain fatty acid binding site present at the terminal end, it does not bind long-chain fatty acids, and has the characteristic that it can bind various drugs through its central region. On the other hand, the albumin fragment lacking the N-terminal region is Cys3
4 and many other cysteine residues, which is advantageous for stable folding of the protein.

Next, the production of the human serum albumin fragment of the present invention will be specifically explained using Examples.

In addition, unless otherwise specified in the examples, the enzymatic reaction for treating DNA was performed under the following conditions.

Experimental Method Enzyme Reaction Each enzyme reaction was performed under the following conditions. Restriction enzyme EcoR■
Nippon Gene; 20 units/III), former ndll
l (Takara Shuzo; 10k'-7t/1tl), 8as+
Digestion of DNA with HI (Takara Shuzo; 12 units/ha) and Xhol (Takara Shuzo; 12 units/pl):
211g DNA, 1111 enzymes and 10X EcoR
l buffer (IM Tris-HCI (pH 7,5)
, 100mM MgClz, 500mM NaCl1
) Add sterile distilled water to 2 ttl to make 20 pl.

Incubate at 37°C for 1 hour and cut. BstE
II Nippon Gene: 7.5 units/pl),
For PstI (20 units/ha), use 100mMTr instead of 10X EcoRl buffer.
is-HCjl! (pH 8,0), 70mM Mg
C1t, 1.5M NaCj! using BstE
II is incubated at 60"C and Pstl is incubated at 37°C for 1 hour to react. 5Ilal and PPONGENE; 1
0 units/pi), replace 1QXEcoRI buffer with 100 mM Tris-HCI (pl!
8.0), 70mM Mgcf2.200a+ KC/
! Use 4 ul and incubate each at 37°C for 1 hour to react. T4 ligase treatment was performed under the following conditions.

1 g of vector DNA, equimolar amount of DNA as vector DNA
A fragment, 10X ligase buffer (660mM
Tris-FICI! , (pH 7,5), 66mM
MgCl2z, 100a+M dithiothreitol,
Add sterile distilled water to IIIMATP) 2ttl and T41) NA ligase lμl (Takara Shuzo; approx. 400 units/111) and 20J1! and keep warm at 16°C overnight.

Construction of the mini-ISA expression plasmid was performed as follows.

First, from pUc-ISA-EH (Reference Example 3) to EcoR
A fragment encoding the natural H3A prepro sequence and Aspl''-''Pro152 excised with I-Hpa II, and G1u15 excised from pUc-ISA-■ (Reference Example 5) with Hpa II-Pstl.
The fragment encoding 3 to Pro303 was inserted into the EcoRIPst1 site of plasmid pHC19 to create plasmid p[Ic-mH3AEH. This p
Hc-mHsA-EH as prepro11sA cDNA
Cut at the EcoR1 site at the 5' end of the A sequence, and insert Xh internally at both ends with EcoRI sticky end sequences.

A synthetic linker with one site, - is inserted to create plasmid pU.
c-mFlsA was prepared in advance. (7) pUc-mH
sA to Xh.

1-Hlnd l [and p
Hcn II SA to old nd m-Bam1l
The 3' region of the preproH5A cDNA excised with I was inserted into the Xho IBamH1 site of plasmid pUc18X to create plasmid ptlc-taH5AA. Note that the plasmid pUc18X used here is pUcl
In the EcoR1 site of B, both ends are EcoRI as above.
It was created by inserting a synthetic linker with a sticky end sequence and an Xho 1 site inside.

In addition, pυC-r+H5A is the natural preproll5A c
Plasmid pAT-nflsA- containing the DNA sequence
A (Reference Example 8) as Xho I / Baa+ll I
A fragment containing the preproll5A cDNA portion was obtained by double digestion with pUc18X and XhoI/B.
This is a plasmid created by ligating the large fragment obtained by double digestion with amtll. Next, plasmid pUcmH
XhoI-Ban+ll containing the preproξ2ISA translation region of 5A-A, polyA signal, and polyA sequence
r fragment, pJDB-ADH-n1lsA
-A plasmid (E. coli Escherichia coli containing this plasmid)
cherichiacoli IIBIOI/pJDB
-ADH-nH5A-A was submitted to the Institute of Microbial Technology, Agency of Industrial Science and Technology, as part of the Microbiological Research Institute No. 2454 (FERMBP-24).
54) was deposited internationally on June 8, 1989 under the Budapest Treaty. ) from Xho I~Bam1l
The fragment was ligated with the larger fragment excised using E to prepare pJDB-A0+1-mHsA plasmid.

Construction of the shortened ISA expression plasmid was performed as follows. First, we deleted the part encoding the carboxyl terminus of human serum albumen short (cDNA clone +1sA・■).
Cut with coRI and insert the resulting fragment into pUc19E
Insert into coRI site and go to plus ξ de pUcHS - [8
I got it. Plasmid puc-H58-11B to Ec
A fragment containing the 5' untranslated region of ISA and the native ISA pre-pro sequence was excised using oRI-Taq I, and EcoRI-Acc of plasmid pUc1B was cut out.
1 site to create plasmid plJc-3ig. A fragment containing the 5' untranslated region of ISA and the natural H3A prepro sequence was excised from this plasmid pUc-Sig using old ncII, and the plasmid pUc-Sig was
Plasmid plJc-5ig-5ALI was created by inserting it into the Sma 1 site of ALII (Reference Example 5). The resulting plasmid pUc-3igSAL
II encodes the 5' untranslated region of H3A, the natural H3A prepro sequence and Met123 to Pro 303, but there is an amino acid G between the prepro sequence and Met123.
The codon GGATCC corresponding to ly-5er remains as an adapter sequence. Then plJc-Sig-S
Prepro sequence excised from AL■ by BstEII-PstI and Gly-Ser and Met123~P
The fragment encoding ro303 was transformed into the previously described pUc
-nH3A was ligated to the larger fragment obtained by cleaving with BstEIIPstI to create plasmid pUc-n5ALU. Bam1l l11ind from this)
The larger fragment cut out by lI [p
AT-trp-phoA-tHSA (E. coli HBIOI containing this plasmid (pAT-trp-
phoA-tHSA) was submitted to the Institute of Microbial Technology, Agency of Industrial Science and Technology, as part of the Microbiological Research Institute No. 10951 (FEMP P-105).
It has been deposited as 1). ) to Bal1l (I −
)1ind[II, the fragment encoding the shortened ISA was inserted to create the plasmid pLIc-
ntH38 was produced. This plasmid pUc = n
The prepro sequence and tHSA-encoding fragment excised from tllsA with XhoI-HlndI[I, and the fragment encoding old ndI[[Ba
The fragment containing the polyA signal excised by mHI was inserted into Xho I Bal of plasmid pυCl8X.
1alt 1 site, plasmid pUc-n
tllsA-A was created. This plasmid pUc-
The fragment excised from n tHsA-A with Xho I 5IIIa I was inserted into pJDB-ADII
-ntlsA -A plasmid to Xho I -Se
a I ligated with the larger fragment excised with I to create the expression plasmid pJDB-ADH-tl in yeast.
lsA was produced.

ISA fragment expression plasmid pJDB-ADII-m)
ISA and pJDB-ADH-t) Transformation of yeast with ISA was carried out by a modified version of the KUR method by Hideaki Hashimoto and Koki Hikaru [Fukaku to Kogyo, 43°63 (1-637 (1985))].

First, YPD medium [1% yeast extract (Dirco),
2% bactopeptone (Dirco), 2% glucose] strain AH22 (MATa, 1eu2 3.
1eu2-112. his4-519. can
Add l- of the overnight culture in YPD medium of l) and add 30
The cells were incubated at °C until the absorbance at 600 nm reached 0.5. The bacteria were collected by centrifugation at 4°C at 2000 rpm for 5 minutes, and the bacteria were collected into 5 bacterial cells. suspended in I M Li5CN. Next, 1.5 d of it was taken out and 2000 r
pn+, collect bacteria by centrifugation for 5 minutes, and dilute the bacterial cells with 1 OuI of 2
MLi5CN was suspended in 46 μl of 50% polyethylene glycol 4000, and LOttl DNA solution (containing 5-10 jrg of DNA) was added and incubated overnight at 30"C. To this suspension, 500 JIZ sterile distilled water was added. In addition, after shaking gently with a portex mixer, centrifugation was performed at 2000 rpm for 5 minutes to collect bacteria.
Resuspend in 0 μl of sterile distilled water and transfer to selective agar medium (SD
Medium: 20n/female histidine hydrochloride, 0.67% amino acid-free yeast nitrogen base (Difco),
2% curcose and 2% agar).

After culturing at 30°C for several days, the resulting colonies were
A containing a plasmid expressing each H3A fragment by detecting the expression of the H3A fragment by the method shown in Example 4.
H22 (pJDB-ADH-mH5^) and Al-122
(pJDBADH-tllsA) was obtained.

Real IH female 4. ■1” One company produced the above transformant A1122 (pJDB-Ant(
-dSA) and 8022 (pJDB-ADH-tHs
A) was cultured in 5 ml of YPD medium at 30°C for 24 hours.

Detection of H3A fragments secreted extracellularly was performed as follows. After the culture solution was centrifuged at 11,000 Orp for 5 minutes, the supernatant was collected in 8001t1 fraction, ethanol 8001 was added, and the mixture was left in water for 30 minutes. This is 1200Orpm
, centrifuged for 5 minutes, dried the resulting precipitate using a centrifugal evaporator, and added SOS-PAGE sample buffer [2% S
O5, 5% 2-mercaptoethanol, 7% glycerol, 0.00625% Bromophenol Blue 0.06
25M Tris-HCf buffer pH 6,8) 20t
tl and boiled for 5 minutes. This sample LOttl was subjected to electrophoresis (Laemmli's method: Natu
re (London) 277°680 (1970))
After that, it was stained with Coomassie brilliant blue (CBB).

Furthermore, Western plotting was performed on the gel after electrophoresis performed in the same manner as shown below. That is, the gel after SO5-PAGE was filtered using a nitrocellulose filter (Bio-rad, Inc., using a plotting device (TEFCO, Mode 1: TC808)).
Trans-blotQ). After plotting, the filter was soaked in TBS solution containing 3% gelatin (20mM Tris-11cffi (p
After treatment with H7,5) and 0.5M NaCl) for 30 minutes, washing was performed twice with TTBS solution (TBS solution containing 0.05% Tween 20) for 5 minutes, changing the TTBS solution.Next. , horseradish peroxidase-labeled anti-H5
Transfer the filter to a solution in which antibody A (Cubbell) was diluted 1000 times with TTBS solution containing 1% gelatin.
Time processed. Filter twice with TTBS solution, TBS
After washing once for 5 minutes each with 0.015% I
I 20,. The filter was transferred to a TBS solution containing 0.05% HRP-Color Dehydropment Reagent (Bio-rad) and 20% methanol, and reacted for 15 minutes. After the reaction was completed, the filter was washed with water.

Detection of ISA fragments accumulated within bacterial cells was performed as follows. That is, 300 Ill of culture solution was heated to 500 Orpm.
, centrifuge for 5 minutes to collect bacteria, and add 30 μl of SDS-
It was suspended in PAGE sample buffer and boiled for 10 minutes at 100″CT.

This sample 1Ottl was subjected to electrophoresis and Western blotting in the same manner as above.

The results of Coomassie brilliant blue (CBB) staining are shown in FIG. In this figure, lane l is ISA standard;
Lanes 2 and 6 are molecular weight standards, lane 3 is At(22(
pJDB-ADH-mtlsA) (7) expression product,
Lane 4 is a culture of host AH22, and lane 5 is a culture of host AH22.
1122 (pJDB-AD)l-tH5A). Furthermore, the results of Western blotting are shown in FIG. In this figure, lane 1 is host^H22 culture, lane 2 is Al122 (pJD
B-ADH-tllSA) culture supernatant, lane 3 is A1
122 (pJDB-MDIlmlIsA), lane 4 is ISA standard, lane 5 is Al22 (pJDB-MDIlmlIsA).
Lane 6 is the protein in the cultured cells of AlI22 (pJDB-8DH-mHSA), and Lane 7 is the protein in the cultured cells of the host AI+22.
The results are shown for intracellular proteins.

As shown in the figure, mini H3A is secreted outside the bacterial cell, and 5O5
-Identified by PAGE as a band with a molecular weight of about 35,000. However, the truncated HSA is secreted in small amounts into the medium;
A large amount was accumulated within the bacterial body.

The above transformant AH22 (pJDB-ADH-mll)
sA) in YPD medium containing 5% glucose [1% yeast extract (formerly FCO), 2% Batato pegtone (Difc
o), 5% glucose]42 at 30°C for 40 hours. This culture solution is 150M! was cooled to O'C, 1500 ml of -20°C ethanol was added to it, and then the mixture was cooled to 0°C.
The mixture was stirred at C for 30 minutes. The precipitate obtained by centrifugation at 12,000 rpm for 15 minutes was diluted with 30 ml of 100 mM
100d of 10mg/ml RNaseA after dissolving in Tris-11cQ buffer pt+s,o
(heat treated) was added and treated at room temperature for 15 minutes. This is 7
After overnight dialysis against 50mM NaCff, 10mM sodium phosphate buffer pH 6,9, 18000 rp.
The supernatant was obtained by centrifugation at I11 for 10 minutes. This supernatant was transferred to a hydroxyapatite column (Tonen 1 lydroxyapatiteLeTAP) of high performance liquid chromatography.
Si52110 (φ21X 100cm)]
Flow rate 3 Inl1/min, 10a+M for 60 minutes
Elution was performed with a phosphoric acid concentration gradient of ~200mM. ξ2H
Identification of the 3A peak was determined by absorbance at 280 nm and SO5
-Performed by PAGE.

After dialyzing the obtained mini-ISA peak against water,
Freeze-dried and diluted with 3 trtl, 500mM Na
Ce, 50mM Tris-H (/! pH8,0,0
, 05% NaN.

This sample was treated with 5ephacryl S-200 (Phar
Masia, 5uper fine grade (1
, 6 x 90 cm) and eluted with the same solution as the sample solvent at a flow rate of 8.6 d/hr. Identification of mini-ISA peaks was performed in the same manner as above. Next, the obtained mini H3A peak was transferred to a high performance liquid chromatography reverse phase column (TSK gel, phenyl
-5PW RP (4,6X76mm) ), 0
．． Flow rate 1ml/win in the presence of 1% trifluoroacetic acid
, eluted with a 0% to 70% acetonitrile gradient over 60 minutes. As a result of identification by absorbance at 280 ns+, mini-ISA was detected as two peaks, and these peaks were used as the final purified sample.

After freeze-drying the purified mini-ISA sample of the N-amino configuration of mini-ISA, it was dissolved in trifluoroacetic acid and analyzed using an automatic amino acid sequence analyzer (Appl.
ied Biosystems, Protein 5
equencer 477A) to identify the N-terminal agonoic acid sequence. The N-terminal amino acid sequences of the two ξ2III SAs identified by an automatic amino acid sequence analyzer were as follows.

85p-Ala-Hys-Lys-X-Glu-Val
-Ala This sequence is the same as the N-terminal amino acid sequence of mature H3A, and it was found that the same processing as natural ISA is carried out during expression and secretion of mini-ISA.

A sample (approximately 1 nmol) of ξ2H3A purified on the C1 amino acid of mini-HSA was placed in a hydrolysis test tube and lyophilized, then 50 μl of anhydrous hydrazine (Ard Ich) was added and the mixture was vacuum-evaporated. The reaction was carried out for 15 hours at 100'C. After cooling to room temperature, excess hydrazine was removed under reduced pressure and further dried in a vacuum desiccator overnight. Amino acid analysis was performed on this sample using an automatic amino acid analyzer (JEOL, JLC-300), and the C-terminal amino acid was identified. In addition, after hydrochloric acid hydrolysis of the above sample, amino acid analysis was performed in the same manner, and the sample was quantified to determine the recovery rate of the C-terminal amino acid. As a result, the C-terminal amino acid of mini-H3A was identified as Pro in both of the two purified peaks by the hydrazine decomposition method.

This result is consistent with the constructed mini-ISA, since Pro should be present at the C-terminus.

A sample of ξ2ISA purified on mini H3A amino (approximately 100p100p was put into a sample test tube and lyophilized, then PICO-TA
G (TM) workstation reaction vial. This reaction vial contains 500 J of constant boiling point hydrochloric acid (Wako Pure Chemical, for precision analysis)! 1 and hydrolyzed at 110''C under reduced pressure.The reaction times were 24, 48 and 72 hours.

After the hydrolysis was completed, the hydrochloric acid in the sample test tube was removed under reduced pressure, and the amino acid composition of the obtained sample was analyzed using an automatic amino acid analyzer (JEOL, JLC-300).

The results are shown in the table below.

Beak 1 Hapeak amino acid Experimental value Theoretical value A l a 35.0 35A r
g 12.8 14A s x
31.9 31Cys ND
19G I x 45.5
42G Iy 7.8 7Hi
s 11.4 1011 e
4.9 5 Leu 29.6
32Lys 2B,,3 28M
e t 3.0 3P h e
16.9 17P r o
11.4 12S e r 10.
9 12Thr 11.7 1
2Trp ND ITyr
7.3 8Val 14.7
15ND=Undetermined PP Amino acid Experimental value Theoretical value A I a 35.0 35A r
g 13.4 14Asx
31.7 31Cys ND
19G l x 45.3 42
G l ) + 7.5 7His
11.0 10I 1 e
5.1 5L e u 30.0
32L)' S 28.0
28Met 2.6 3P h
e 17.0 17P r o
12.0 12S e r 1
1.7 12T h r 11.8
12T r p ND I
Tyr? ,6 8Val
14.8 15ND = Undetermined As is clear from the table above, the experimental values obtained are almost equal to the theoretical values, and when the results of the N-terminal amino acid sequence and C-terminal amino acid shown above are combined, the expression The secreted mini-t(SA) was in the as-constructed structure.

For screening of clones containing normal human serum albumin A cDN8 by plaque hybridization, a human liver cDN library constructed using λgtll from CLONTECH, USA as a vector was used. Escherichia coli Y1090 was infected with the λgtl1 recombinant phage, and a total of 5.5×10 transformed plaques were obtained.
' cells were formed on LB agar medium (Luria medium + 1.5% agar), and the recombinant DNA was transferred to a membrane filter (Ame
After transferring to rsham 11ybond -N), iz
Platform oligonucleotide labeled with p radioisotope 3
species (specific activity ≥10'cpm/μg) were screened using as probes (Ilenton & Davi
sScience 196, 180-182 (1977
)), these three types of probes were each described by Lawn et al. (Nu
cleic Ac1ds Res 9 + 6f036
114 (1981), the 5' untranslated region (the region from 12 nucleotides upstream of the translation initiation ATG codon to the nucleotide before the ATG codon) and the translated region (the amino terminal methionine codon (the part that encodes the 9th amino acid leucine from ATG) (ISA
-1), one that encodes glycine at position 248 to leucine at position 260 (HSA-2), and a part that encodes ronsine at position 585 from the carboxyl terminal from valine at position 576, and the following 6 nucleotides.
' - the same sequence as that containing the untranslated region (ISA-3).

The probe was synthesized using an automatic DNA synthesizer, and labeling was performed using [γ-''P) ATP and polynucleotide kinase. Prepare DNA from clones (Blattner et al.
ience 202.1279-1284 (197B)
), which was digested with EcoRI enzyme, and the Southern blot of the digest was hybridized with the 15A-2 probe [5southern, E., J., Mo1. Bio
I, 503-517 (1975)).

Three hybridized fragments were obtained, each 1.8 kb, 1.4 kb, and 1.3 kb.
It was the length of Of these, fragments of length 1.8 kb and 1.3 kb were subcloned into pUc19 hector.

This subclone was subjected to colony hybridization (Gru
nstein and Hogness r'roc, Na
tl, ^cad, Sci.

11SA, 3961-3965 (1975)). As a result, a clone λgtll (HSA I-A) that hybridized only to 15A-3 was obtained. Various DNA fragments of this clone were used for base sequencing of Hector phylum 13mp18 and mp19 RP-DN.
A and the guideoxynucleotide termination method [Sanger.

F., N1cklen+S., and Coulson.
A, R, Proc, Natl.

Acad, Sci, tlSA 74.5463 546
7 (1977)). On the other hand l5
Plaque hybridization of λgtll clones performed using A-2 as a probe resulted in 20 clones giving a positive signal, and 20 were subjected to plaque hybridization again using 15A-1 as a probe, resulting in 1 clone giving a positive signal. λgil
1 (IISA-II) was obtained. From now on Phage DN
A was prepared and the EcoRI digest was subjected to Southern hybridization using 115A-1 as a probe, and 6I LW was performed to show that a 1.25 kb fragment (ISA-II) hybridized with the probe. The base sequence of this fragment was determined by the guideeoxynucleotide termination method. HSA-■ is IIs^-
It did not hybridize with the 3 probe. This result l
l5A-■ lacks the part encoding the carboxyl terminal side, l5A-1~A lacks the part encoding the amino terminal side of human serum albumin, and the codon encoding serine at position 304 (TCA) is a translation stop codon. It was found that the opal codon was changed to TGA. A restriction enzyme map of these two DNA fragments is shown in FIG. The exact location of the enzyme recognition site was obtained from the final base sequence.

7''/SumiF pUc-ISA-CH containing DNA encoding the entire conversion of Boothmi' υC-15A-CH to mature normal human serum albumin was prepared as follows.

ll5A cD obtained from human liver cDN^ library
Ec from the NA-containing clone λgL11 (ISA-■)
A fragment generated by oRI and XbaI digestion was prepared, and this was digested with EcoRI and XbaI of pUc19 plasmid
The larger fragment of the double digest with a I was ligated using T4 DNA IJ gauze to construct a recombinant plasmid pUcISA-EX.

The smaller fragment resulting from the double digestion of Aham and 5ail was purified from this plasmid.

This fragment encodes the 12th Lys to 356th Thr of mature normal human serum albumin A protein. In order to construct a gene encoding mature normal human serum Alpine A protein from the amino terminus, a DNA sequence corresponding to the 5' end was created by annealing two chemically synthesized fragments. This synthetic NA sequence has a sticky end sequence CG generated by Hpa It and C1aI enzymatic cleavage at the 5' end so that it can be fused with a DNA sequence encoding the signal peptide of alkaline phosphatase, and is a 5′-terminus of the mature normal human serum albumin A protein. 1st amino acid Asp to 11th amino acid Ph
It has a sequence encoding e. This annealed DNA sequence was treated with T4 polynucleotide kinase to phosphorylate the 5' end, and pUc-)13A-
AhaIII/Sal produced from EX was mixed with the double digest, and then pAT153, a typical E. coli multicopy cloning vector, was added to this mixture.
(A+Wersham, Twigg, A, J, and 5herratt, Nature 283
216-218.1980) C1a1/5a
The large fragment of the double digest of il was mixed with the larger fragment, and the three components were ligated to 74 DN8 using gauze to obtain a recombinant plasmid pAT-11sA-CX.

On this plasmid, a DNA sequence encoding the amino acid Asp at position 1 to Phe at position 11 was linked to normal human serum albumin. pAT-115A-CX was double digested with EcoRI/XbaI to obtain a smaller fragment containing the DNA sequence encoding Aspl=Phe356 of normal human serum albumin A.

On the other hand, c encoding the carboxyl terminal side of HSII-A
The DNA was derived from a foreign cD clone λgtll()IsA I-A) obtained from a human liver cDNA library.
An EcoRI fragment into which the NA sequence had been inserted was prepared and cloned into the recombinant plasmid puc-l5A-bi by inserting it into the EcoRI site of the pUc18 plasmid. This will result in ISA-A's 35
585 from the 8th amino acid Leu to the carboxyl terminal
The untranslated region 6 on the 3′ side encodes the th Leu.
A double digest of Xbar/Hind m containing 2 nucleotides was prepared. This is pAT-113A-C
The EcoRl/Xba obtained from X was mixed with the large fragment of the double digest and the EcoRI/Hind double digest of pUc19 and ligated with T4 DNA ligase, resulting in a cDNA containing the entire cDNA of mature normal human serum albumin A. Recombinant plasmid pUc-H5
A-CH was obtained.

111 Examples 4 Preprocoding DNA Four types of oligonucleotides having the following sequences: 1, A
ATTCATGAAGTGGGTTACTTTCATC
TCTTTGTTGTT2, AGAACAAGAA
CAACAAAGAGATGAAAGTAACCCAC
TTCATG3, CTTGTTCTCTTCTGC
TTACTCTAGAGGTGTTTTCAGAG4
, CGCGTCTGAAAACACCTCTAGA
GTAAGCAGAAG, MatLeucci, M
, D., and Caruthers, M.H., ,Tetra.
hedron Letters21.719 (1980
) by the phosphoramidite method described in
Synthesis was performed using an NA synthesizer (Applied Biosystems model 380B). The oligonucleotide fragments were 5'-phosphorylated with T4 polynucleotide kinase, annealed, and then ligated with T4 DN^ ligase to obtain a single double-stranded DNA encoding the prepro sequence.

Next, the plasmid pLlc-ISA-C)I (Reference Example 2) containing the cDNA of normal human serum albumin A was double digested with restriction enzymes EcoRI and C1aI to obtain the larger fragment, which was then subjected to the above synthesis. rfi, DNA and T4DN8 ligase to create plasmid puc.
- Created HSAEl.

BamH on the 5' end! It has an attached end and Hpa near the 3' end.
If (Mspr) recognition sequence, its double-stranded portion is human serum albumin Met(123)-Ala.
A gene fragment completely encoding (151) was constructed as follows. In order to improve the efficiency of expression in E. coli, codons commonly used by genes that are expressed with high efficiency in E. coli (preferential codons)
The sequence was designed to contain as much of s) as possible. tRN^ species for these codons are generally abundant in E. coli [e.g., rkemura, T, J
, Mo! .

Old of, 151, 389-409 (1981);G
ouy, M. & Gautier, C0Nucleic.
Ac1ds Res, Kei0. 7055-707
4 (1982)) is expected to affect translation efficiency.

The following four oligonucleotides: Caruthers et al. (Matteucci, M., D.
, and Caruthers+M, 11. Tetrahe
Dron Letters 21.719 (1980)
Synthesis was performed using an automatic platform machine (Applied Biosystems model 380B) applying the phosphoramidite method developed by ). Synthesized DNA strand (
Approximately 30paoles) 50s+M Tris-tlc
j! (pH 7,6), 105M MgCj! □5-
One 5' end was phosphorylated by treatment in a solution (50Ii) containing dithiothreitol and 0.2 sM^TP in the presence of 6 units of T4 polynucleotide kinase (Takara Shuzo) at 37°C for 160 minutes.

The four phosphorylated fragments were mixed, kept warm in a 100°C water bath for 5 minutes, and then allowed to cool at room temperature for annealing. 2111 T40NA1 nogase (800 units, Takara Shuzo) was added and incubated overnight at 16°C to ligate the fragments to form double-stranded fragments.

This double-stranded fragment was then transformed into Hpa II (Msp
I) to obtain a 96 bp fragment.

Then, the larger fragment of the BamHI and Pstl double digest of pUc19 was ligated with T4 DNA ligase to obtain the pSAL I plasmid.

λgtl, which lacks the part encoding the amino-terminal side of normal human serum albumin and also contains a sequence in which the codon encoding serine at position 304 has been changed to a translation stop codon.
A human cDNA clone (ISA-IA) (Reference Example 1: Figure 6) was cut with EcoRI to extract the human serum albumin cDNA portion, which was then transformed into plasmid pUc.
Plasmid pUc-1 was inserted into the EcoR1 site of 19.
1SA-1 was produced.

pUc-ISA-1 was cleaved with Pstl, and the resulting 5'
After removing the terminal phosphate groups by treatment with bacterial alkaline phosphatase, Hpa II (Msp I)
A 750 bp fragment was excised. This 750 bp fragment was combined with the 96 bp fragment synthesized in Example 1 using T4 DNA ligase.
λgtll (USA-IA) (Reference Example 1, Fig. 6), which is bound using the pairing of cohesive ends of lpaII (Mspr) and contains the 3' region of human serum albumin A cDNA (Reference Example 1, Fig. 6), was purified using EcoRI. A DNA fragment containing cDNA of human serum albumin A was obtained by digestion, and this was cut with EcoRI to obtain plasmid pUc.
18 to yield pUc-ISA-1'.

1 tbsp 11 yu Boosmid AT-nHSA
9) Preprohuman serum albumin A cDNA 5
'-Prepro-human serum albumin A cDNA octopart coRI from plasmid ρUC-H5^-EX (Reference Example 2) containing the untranslated region and the first half of the coding region
A plasmid puc-ISA containing the second half of the coding region and the 3'-untranslated region of human serum albumin A cDNA was excised by double digestion with
! Xba I-11i cut out from '(Reference Example 6)
ndlII fragment and pAT153 vector (
Manufactured by Amersha: Twigg+A, J, and 5
heratt, D., Nature 283+ 216
218, 1980).
ndI[I fragment was ligated to create plasmid pAT
-H5A-f'll was obtained. 5 of the cDNA sequence encoding preprohuman serum albumin A to be adjacent to a strong promoter sequence derived from yeast.
'EcoRI site attached to the end and the third amino acid Trp to the fifth amino acid Th in the sequence encoding the signal peptide to preprohuman serum albumin.
The BstEI site that spans the r codon was used. (3 a) from the amino terminus of the 5'-untranslated region and the signal peptide to prepro-human serum albumin.
EcoRl-BstEII containing an acid-encoding sequence
The fragment was excised from pAT-ISA-[H. The remaining large NA fragment was attached to the 5' end [coRI
A synthetic DNA fragment having a sticky end sequence and a BstEII sticky end sequence at the 3' end and capable of encoding up to the third amino acid of the signal peptide to preprohuman serum albumin: EcoRI BstE II5'-A
ATTCATGAAGTGGGGTACTTCACCCA
It was ligated with TTG-5'. That is, this synthetic fragment was treated with T4 polynucleotide kinase to phosphorylate the 5'-end, and this ligation was performed using T4 DNA ligase to create a plasmid pAT-nH3A containing natural preprohuman serum albumin A cDNA.
was created.

pAT-nHS^ (Reference Example 7) was cut at the EcoRI site at the 5' end of the preprohuman serum albumin A cDNA sequence, and a bottom linker-EcoRI containing EcoRI sticky end sequences at both ends and an Xho I site inside was inserted.
XholEcoRI 5'-AATTCTCGAG GAGCTCTTAA-5' was inserted to create plasmid pAT-X-nH3^.

This pAT-X-nl (pAT adjacent to the 3' end of the preprohuman serum albumin A cDNA sequence in SA)
The old nd m -Ba5HI fragment derived from 153 plus Godo was excised, and the 3' of preprohuman serum albumin A cDNA excised from pUc-ISA-1' was used.
The region containing the poly A signal and poly A sequence in the side sequence and the p
The former nd I [-B
The amHr fragment was replaced with plasmid pAT-n.
H3A-A was produced.

[Brief explanation of drawings]

Figures 1-1 to 1-2 are mini ISA expression plasmid pJD.
The production process of B-8Dfl-mH5^ is shown. Figures 2-1 to 2-3 are truncated H3A expression plasmid pJD
The production process of B-ADH-tHsA is shown. FIG. 3 shows the production process of plasmid pUc-nH5A. Figure 4 shows AH22(pJDB-ADII-+5H
SA) (lane 3) and AH22 (pJDB-AD)
lt) IsA) (5D of expression product from lane 5)
This is an S-polyacrylamide gel electropherogram, in which protein bands are stained with Coomassie brilliant blue. Lane 1 is human serum albumin derived from purified human serum;
Lanes 2 and 6 are molecular weight standards, phosphorylase B
(molecular weight 94.000), bovine serum albumin (molecular weight 67.000), ovalbumin (molecular weight 43.000)
0), carbonic anhydrase (molecular weight 30,000), soybean trypsin inhibitor (molecular weight 20,000), and lactalbumin (molecular weight 14,400), and lane 4 is in host A1122 strain culture medium that does not contain the H3A fragment expression plasmid. FIG. FIG. 5 shows the culture supernatant (lane 2) and intracellular protein (lane 5) of AH22 (pJDB-ADH-tHSA),
Culture supernatant of AH22 (pJDB-ADH-mH5A) (
lane 3) and intracellular protein (lane 6), AH2
2 culture supernatant (lane 1) and intracellular protein (lane 7)
) and a Western plot of purified human serum-derived human serum albumin (lane 4), showing the protein that reacted with the anti-human serum albumin antibody. FIG. 6 shows a restriction enzyme map of the cDNA encoding human serum albumin. FIG. 7 shows the production process of plasmid ρIC-H5^-EH. FIG. 8 shows the production process of plasmid pSAL II. FIG. 9 shows the production process of plus and mid pAT-nH5A. FIG. 10 shows the production process of plasmid pAP-n)ISA-A.

Claims (1)

[Scope of Claims] 1. A human serum albumin fragment in which the C-terminal portion of human serum albumin is deleted. 2. 3 from aspartic acid at position 1 of human serum albumin
Claim 1 having the amino acid sequence up to proline at position 03
Fragment described in. 3. A fusion protein consisting of a human serum albumin fragment in which the C-terminal portion of human serum albumin is deleted and another polypeptide. 4. The fusion protein according to claim 3, comprising the signal peptide and propeptide of human serum albumin, and the amino acid sequence from aspartic acid at position 1 to proline at position 303 of human serum albumin. 5. A human serum albumin fragment in which the N-terminal portion of human serum albumin is deleted. 6. 5 from methionine at position 123 of human serum albumin
Claim 5 having the amino acid sequence up to leucine at position 85
Human serum albumin fragment described in . 7. A fusion protein consisting of a human serum albumin fragment in which the N-terminal portion of human serum albumin is deleted and another polypeptide. 8. The fusion protein according to claim 7, comprising the signal peptide and propeptide of human serum albumin and the amino acid sequence from methionine at position 123 to leucine at position 585 of human serum albumin. 9. Protein fragment according to claims 1 and 5 or claims 3 and 7
A DNA sequence encoding the fusion protein described in . 10. A plasmid containing the DNA sequence according to claim 9. 11. The plasmid according to claim 10, which is an expression plasmid containing, upstream of the DNA sequence, a control sequence for efficiently expressing the DNA sequence in a host. 12. A host transformed with the plasmid according to claim 11. 13. The host according to claim 12 is cultured to express a human serum albumin protein fragment or a fusion protein containing the fragment, and when the fusion protein is expressed, the human serum albumin protein fragment is optionally extracted from the fusion protein. A method for producing a human serum albumin protein fragment or a fusion protein containing the fragment, the method comprising cutting out a human serum albumin protein fragment or a fusion protein containing the fragment.