JPH09266792A - Gene positively controlling transfer of yeast mannose-1-phosphate and production of high mannose type neutral sugar chain using defective variant of the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a gene capable of positively controlling the mannose-1- phosphate transfer of yeast coding specific amino acid sequence and giving a high mannose type neutral oligosaccharide or a glycoprotein in plenty and high purity by a genetically-engineered process of the yeast. SOLUTION: This new gene is capable of positively controlling the mannose-1- phosphate transfer of the yeast coding an amino acid sequence expressed by the formula, producing a high mannose type neutral oligosaccharide having a sugar chain structure same as that of human and a glycoprotein by adding the high mannose type neutral oligosaccharide chain to an aspartic acid residue in the protein, starting from the yeast by using the recombinant DNA technology. This new gene is obtained by extracting genomic DNA from Saccharomyces cerevisae, preparing the gene library by conventional method, cloning the gene by amplification using the primer consisting of a part of the gene coding the protein positively controlling the transfer of mannose-1- phosphate according to PCR method.

Description

Detailed Description of the Invention

[0001]

The present invention relates to a high-mannose type neutral oligosaccharide having the same sugar chain structure as a high-mannose type sugar chain produced by mammalian cells such as humans, and to a high-mannose type neutral oligosaccharide. The present invention relates to a method for producing a glycoprotein in which a sugar chain is added to an asparagine residue of a protein by genetic engineering using yeast.

[0002]

2. Description of the Related Art Many proteins having important functions in living organisms are not simple proteins but glycoproteins having sugar chains. It is known that erythropoietin (EPO) does not exhibit its original biological activity when sugar chains are removed. ) And tissue plasminogen activator (TPA). [Kibata Yo, Protein Nucleic Acid Enzyme, Vol. 36, p775 (1991); Makoto Takeuchi,
Biochemistry, Vol. 62, p1272 (1990)], suggesting that sugar chains play an important role in the expression of biological activity. There are many types of glycoproteins derived from mammals such as humans, but the sugar chains added to the proteins vary depending on the type of protein, species of organism, organs, etc., and the sugar chains are not uniform in length but constant It is heterogeneous with distribution [Kibatayo, Protein Nucleic Acid Enzyme, Vol.36, p775 (1991); Makoto Takeuchi, Biochemistry, Vol.62, p.
1272 (1990)].

[0003] For this reason, the correlation between the sugar chain biological activity and the sugar chain structure is not clear, and the type and structure of the sugar chain to be added to the protein part is currently trial and error repeated for each protein. is there. Therefore, it is necessary to develop a technology that can modify and control the structure of sugar chains (saccharide type, bonding position, chain length, etc.) to be added to the protein as desired, and has a uniform chain length and a clear chemical structure. The supply of sugar chains and glycoproteins with the sugar chains added thereto is expected not only by academic societies but also by the industry.

[0004] There are two types of sugar chains that bind to proteins, an N-linked type that binds to an asparagine residue of a protein and an O-linked type that binds to a serine or threonine residue. Among them, the biosynthetic pathway of N-linked sugar chains has been widely known and has been analyzed in detail. The biosynthesis of sugar chains starts in the endoplasmic reticulum (hereinafter referred to as ER), and then the sugar chain is further modified in the Golgi apparatus. Of these, the sugar chains produced by the ER are basically the same in yeast and mammalian cells,
It is known that they are synthesized by the route shown in Fig. 1 [Fig. 1 shows Herscovics, A. and Orlean, P., FASEB J., Vol.
7, p. 540-550 (1993)]. In the present specification, the sugar chain synthesized by the ER is hereinafter referred to as a core sugar chain, and the sugar chain is composed of eight molecules of mannose (Man) and two molecules of N-acetylglucosamine (GlcNAc). (See FIG. 1). This core type sugar chain (Man ₈ G
The protein with lcNAc ₂ ) is transported to the Golgi and undergoes various modifications, but the Golgi modification differs greatly between yeast and mammalian cells [Kukuruzinska et al, Ann.
Rev. Biochem., Vol. 56, p915 (1987)].

[0005] Mammalian cells follow the following three different pathways, which differ depending on the type of protein undergoing sugar chain modification. 1) When the above-mentioned core type sugar chain is not changed at all, 2)
The N-acetylglucosamine-1-phosphate (GlcNAc-1-P) moiety of UDP-N-acetylglucosamine (UDP-GlcNAc) is the core sugar chain.
When Man-6-P-1-GlcNAc is added at the 6-position of Man and then only this GlcNAc part is removed and converted to a glycoprotein having an acidic sugar chain containing Man-6-P , 3) Five molecules of Man are sequentially removed from the core type sugar chain to become Man ₃ GlcNAc ₂ , which is preceded and followed by GlcNAc, galactose (Gal), N-acetylneuraminic acid and sialic acid (NeuNAc). Are sequentially added to produce various mixed-molded and complex-type sugar chains as a mixture [R. Kornfeld and S. Kornfe].
ld, Ann. Rev. Biochem., Vol. 54, p.631-664 (198
Five)].

On the other hand, in yeast, the above-mentioned core sugar chain (Man ₈ G
lcNAc ₂ ) to which a large number of Man are added to form a so-called sugar outer chain (outer c).
It is known that, in addition to the production of H.in, an acidic sugar chain in which mannose-1-phosphate is added to the core sugar chain portion and the sugar outer chain portion (see FIGS. 2 and 3). It has been reported that this modification, unlike animal cells, does not function as a sorting signal for glycoproteins localized in vacuoles (organelles corresponding to lysosomes of animal cells) in yeast. Therefore, the physiological function of this phosphorylated sugar chain in yeast remains unknown [Kukuruzinska e
tal, Ann. Rev. Biochem., Vol. 56, p915 (1987)].

[0007] The phosphorylation site of the mannose phosphate-containing sugar chain in yeast is α-1,3- of the core sugar chain of Man ₈ GlcNAc ₂ synthesized in the endoplasmic reticulum as shown by G in FIGS. Branch side and α-1,6-
In addition to the case of adding to the branch side, there are a total of four cases, namely, adding to the α-1,2-branch existing in a large number of mannose outer chains synthesized in the Golgi body and adding to the non-reducing end of the mannose outer chain Can be divided into types [A. Herscovics and
P. Orlean, FASEB J., Vol. 7, p540-550 (1993)]. It has been pointed out that these mannose phosphate-containing sugar chains may be involved in the localization of glycoproteins such as invertase located in the cell surface of yeast and in the intercellular aggregation, but corroborative data are available. Has not been done.

[0008] Biosynthesis of sugar outer chains in yeast is shown in Figs.
(See Ballou et al.)
al, Proc. Natl. Acad. Sci. USA, Vol.87, p3368 (199
0)). That is, following the reaction of initiation of extension in which Man is added to the core type sugar chain by α-1,6 bond (I in FIGS. 2 to 3), the reaction of sequentially extending Man by α-1,6 bond is further performed. (II in FIGS. 2 and 3), a poly-α-1,6 Man bond serving as a skeleton of the sugar outer chain is formed (FIG. 2).
~ E in Fig. 3). Man of this α-1,6 bond has α-
There is a 1,2-linked branch of Man, and this branched α
Normally, α-1,3-bonded Man is further added to the tip of -1,2-bonded Man (see G in FIGS. 2 to 3).

However, only the α-1,2 bonded Man is added to the Man at the tip of the α-1,6 bond, and the α-1,3 bond is added.
There is no additional Man in the bond (FIGS. 2-3
Natal. Aca, Gopal and Ballou, Proc. Natl. Aca
d. Sci. USA, Vol. 84, p8824 (1987)). Therefore, in order to produce a core sugar chain, a mutant strain in which this sugar outer chain is not added and sugar chain synthesis stops at the core sugar chain (A in FIGS. 2 to 3) is isolated. Is achieved by
The present inventors have already succeeded in producing a mutant strain deficient in this sugar outer chain (JP-A-6-277086). However, the glycoprotein sugar chain produced by this method has the following formula (I)

[0010]

Embedded image

(Wherein M represents mannose and GN represents N-acetylglucosamine) In addition to the neutral sugar chain represented by the following formulas (II) to (IV)

[0012]

Embedded image

[0013]

Embedded image

[0014]

[Chemical 6]

(Wherein, M and GN are the same as above)
It was found that an acidic sugar chain represented by
Annual Meeting of Biotechnology Symposium, p.153-15
7, published on October 14, 1994, published by the Biotechnology Development Technology Research Association). This acidic sugar chain has a structure that does not exist in sugar chains derived from mammals such as humans. That is, in mammalian cells, GlcNAc-1-phosphate, not mannose-1-phosphate, is added to the mannose-1-phosphate addition site of the above formulas (II) to (IV). After the addition, only the GlcNAc portion is further removed, and finally the following formulas (V) to (VII)

[0016]

Embedded image

[0017]

Embedded image

[0018]

Embedded image

(Where M and GN are the same as above)
In contrast to the acidic sugar chains represented by
Since the reaction of removing only the Man moiety from the acidic sugar chains of the formulas (II) to (IV) does not occur, the formulas (II) to (IV)
Acidic sugar chains are recognized as foreign substances in mammals and may exhibit antigenicity [Clinton E. Ballou, Methods in Enz
ymology, Vol.185, p.440-470 (1990)]. in this way,
Since the structure of the sugar chain added to the protein differs greatly between yeast and mammalian cells, the same biological activity as that of mammalian origin is detected even when yeast produces a useful glycoprotein derived from mammals such as humans using genetic engineering techniques. It has been pointed out that the differences in antigenicity due to differences in sugar chains are not observed. For this reason, it has been conventionally difficult to produce a mammalian-derived glycoprotein in yeast.

The core-type sugar chain composed of Man ₈ GlcNAc ₂ is important as an intermediate of sugar chain biosynthesis, and therefore, a small amount of mammalian-derived reagents is commercially available as a sugar chain-related research reagent. However, it is currently difficult to synthesize this sugar chain uniformly and in large quantities by organic chemistry techniques.

[0021]

The object of the present invention is to overcome the drawbacks in the production of such N-linked glycoproteins derived from humans and other mammals by using yeast by recombinant DNA technology. It is an object of the present invention to provide a method for producing a high mannose type neutral oligosaccharide chain having the same sugar chain structure as human or a glycoprotein having a high mannose type neutral oligosaccharide chain.

[0022]

Means for Solving the Problems As a result of intensive studies to solve the above problems, the present inventors have positively controlled the addition of mannose-1-phosphate, which is a yeast-specific sugar chain modification reaction. By isolating the gene (MNN4), elucidating its gene structure, and disrupting this gene, the relative amount of mannose-1-phosphate-containing acidic sugar chains having antigenicity decreases, and the formula (I ), The yeast strain having an increased content of neutral sugar chains was successfully prepared, and it was found that the use of the yeast strain enables the production of a large amount of neutral sugar chains, thereby completing the present invention. .

That is, the present invention relates to a gene which substantially encodes the amino acid sequence shown in SEQ ID NO: 1 and positively controls mannose-1-phosphate transfer in yeast. Examples of the gene include a gene represented by the nucleotide sequence shown in SEQ ID NO: 2. Examples of yeast include Saccharomyces cerevisiae. Further, the present invention is a plasmid DNA containing all or a part of the gene that positively regulates the mannose-1-phosphate transfer.

Further, the present invention provides a method for preparing a gene fragment in the above plasmid DNA which lacks the function of a gene that positively controls mannose-1-phosphate transfer, and integrating the gene fragment into a homologous region of yeast chromosome. Mannose-1
-Yeast that disrupts the function of the gene that positively controls phosphoryl transfer. Further, the present invention provides a Δmnn4 yeast in which a gene relating to mannose phosphorylation (hereinafter referred to as MNN4 gene) has been disrupted, an OCH1 gene and a sugar involved in α-1,6-mannose addition required for initiation of mannose outer chain synthesis. Yeast strain that disrupts both MNN1 genes involved in the addition of α-1,3-mannose to the non-reducing end of the chain (△ och1
Δmnn1) is a triple mutant (変 異 och1 △ mnn1 △ mnn4) yeast obtained by crossing with (mn1).

The present invention further provides the following formula (I) comprising culturing the above-mentioned yeast in a culture medium and comprising two molecules of N-acetylglucosamine and eight molecules of mannose from the culture.

[0026]

Embedded image

(Wherein, M and GN are the same as described above)
Recovering a high-mannose neutral oligosaccharide-linked glycoprotein bound to a neutral oligosaccharide represented by the formula: . Further, the present invention relates to the high mannose neutral oligosaccharide chain-linked glycoprotein from the following formula (I)

[0028]

Embedded image

(Wherein, M and GN are the same as described above)
Wherein the high-mannose neutral oligosaccharide chain represented by the formula is separated. Furthermore, the present invention is a protein comprising a polypeptide having the amino acid sequence shown in SEQ ID NO: 1 and positively controlling mannose-1-phosphotransferase. The protein according to the present invention that positively controls mannose-1-phosphate transfer has the amino acid sequence represented by SEQ ID NO: 1, and the gene that positively controls mannose-1-phosphate transfer according to the present invention. Is characterized in that it has the base sequence represented by SEQ ID NO: 2, but in addition to these, for example, as described above, such as those lacking the methionine (Met) located at the N-terminal of the amino acid, Synonyms of proteins that positively control mannose-1-phosphate transfer and whose genes have been subjected to post-translational modification or sequence modification (addition, deletion, substitution, etc.) by genetic engineering techniques and their genes are also included in the present invention. Is done. In the above, the meaning of “substantially ..” comprehensively describes the above. The genetic engineering method includes various known methods such as modification of DNA by cytospecific mutagenesis.

The gene encoding the protein of the present invention that positively controls mannose-1-phosphate transfer controls the desired mannose-1-phosphate transfer by gene recombination technology. In addition to producing the protein, it is possible to disrupt the protein-coding region of the gene in vitro and then integrate it by homologous recombination at a locus on the chromosome of Saccharomyces cerevisiae by a known method. The prepared gene-disrupted strain can be used for production of a sugar chain having a reduced content of mannose-1-phosphate and an increased content of neutral sugar chain and a glycoprotein having this sugar chain.

Further, a double mutant strain which disrupts the OCH1 gene and has the mnn1 mutation (△ och1mnn1, disclosed in Japanese Patent Application Laid-Open No.
Or OCH1 and MN described in the present specification.
By crossing with a double mutant strain (△ och1 △ mnn1) in which both N1 genes are disrupted by known methods, △ och1mn
Triple mutant strains such as n1 △ mnn4 or △ och1 △ mnn1 △ mnn4 can be prepared, and by using these strains, the content of the mannose-1-phosphate-containing acidic sugar chain is reduced, and the formula ( It is possible to produce a sugar chain having an increased content of a high-mannose-type neutral sugar chain derived from a mammal such as a human as shown in I), and a glycoprotein containing this sugar chain.

The method for producing the gene of the present invention and the specific deletion of the gene function provide mannose-1
-A method for suppressing phosphoryl transfer is described in detail below. First,
The gene of the present invention is isolated by extracting genomic DNA from Saccharomyces cerevisiae according to a general method, creating a gene library, selecting a target DNA from the gene library,
This was introduced into a host yeast strain whose mannose phosphate content was reported to be significantly reduced, and the transformed yeast cells whose mannose phosphate content had recovered to the level of wild-type cells were selected. Can be implemented. In the above, the extraction of genomic DNA from Saccharomyces cerevisiae is performed, for example, by the method of DR Cryer et al. [Methods in C
ell Biology, Vol. 12, p. 39 (1975)] and P. Philippsen
et al. [Methods in Enzymology, Vol. 194, p. 169 (19
91)].

The gene library can be prepared by using a commonly used plasmid vector, a vector derived from lambda phage, or a cosmid vector capable of cloning a large DNA fragment having both phage and plasmid properties. It can be created according to the method described above. A typical host cell into which DNA is introduced is the mnn4 mutant of Saccharomyces cerevisiae in which the content of mannose phosphate is significantly reduced, and this mutant is obtained from the Yeast Genetic Stock Center, (Department of M.
olecular and Cell Biology, University of Californi
a, Berkeley, USA) as LB6-5B and LB5-10A (Yeast Genetic Stock Center, Catalogu
e, eight edition, p.78).

The mutation point on the chromosome of the mnn4 mutation in which the mannose phosphate content is significantly reduced has already been elucidated, and the left arm of chromosome 11 of Saccharomyces cerevisiae, approximately 130 cM (centimorgan) (Yeast GeneticStock Center, Catalog,
eight edition, p.118), the US ATCC (A
It is possible to receive gene fragments containing this neighborhood from public institutions such as the American Type Culture Collection (ATCC Recombinant DNA materials, 3rdedition, 1
993). Therefore, by preparing an expression plasmid in yeast containing these fragments and introducing it into an mnn4 mutant strain, it is possible to isolate the gene of interest as a gene fragment that restores the mnn4 mutation to a wild-type phenotype. Is possible.

As a method for introducing DNA into cells and transforming it by a general method, for example, when a phage is used as a vector, a method of infecting an Escherichia coli host with the phage can be used to efficiently transfer DNA to the host. Can be incorporated to amplify the gene. In addition, as a method of transforming yeast using a plasmid, a method of treating with a lithium salt to easily take up DNA naturally,
Plasmid incorporation method or electrically DN
A method of introducing A into cells can be adopted [Becker, D.
M. and Guarente, L., Methods in Enzymology, Vol. 19
4, p.182-187 (1991)].

The isolation and purification of DNA in each of the above-described methods and in each of the subsequent operations can be performed according to a conventional method. In addition, the DNA sequence of the gene of the present invention can be determined by a conventional method such as the dideoxy method [Sanger, F. et al., Pr.
oc. Natl. Acad. Sci., USA, 74, 5463-5467 (1977)]. Further, the above-mentioned DNA base sequence can be easily determined by using a commercially available sequence kit or the like.

The PCR method uses DN in vitro.
A specific region of A was treated with a combination of sense and antisense primers at both ends of the region, a heat-resistant Taq polymerase, a DNA amplification system, etc. for 10 to 1 hour in about 2-3 hours.
Although it is a technique capable of specifically amplifying a million times, the gene of the present invention and its partial fragment are synthesized using a single-stranded DNA of 25 to 30 mer as a primer and pMN as a template.
It can be amplified by this PCR method using N4, and by utilizing the amplified DNA, mannose-1-
Various modifications and gene disruptions can be introduced into the phosphotransferase gene.

According to the present invention, the desired mannose-1-phosphate transfer can be positively regulated by culturing yeast cells transformed with a plasmid containing the above-mentioned gene that positively controls mannose-1-phosphate transfer. Produce proteins that
Can accumulate. This culture can be performed according to a conventional method commonly used for culturing yeast. For example, an amino acid (leucine in the above example) which can be supplied by a marker (for example, the LEU2 gene involved in leucine synthesis) necessary for replication and maintenance of a plasmid is added by adding various medium components supplied from Difco. A synthetic medium (including a carbon source, a nitrogen source, inorganic salts, amino acids, vitamins, etc.) which has been removed can be used [F. Sherman, Methods in Enzymology, Vol.
4, p.14 (1991)].

The desired protein that positively regulates mannose-1-phosphate transfer produced in the transformant cells by the above culture is localized in the Golgi membrane fraction based on its amino acid sequence and function. Therefore, after crushing the yeast cells, centrifugation (for example, 100,000 xg,
60 min) can be used as a protein source [Nakayama et al, EMBO J., Vol. 11, p.
2511-2519 (1992)].

Furthermore, the protein coding region (open reading fr) of a gene that positively controls mannose-1-phosphate transfer.
ame, ORF) by the aforementioned PCR method or the like, or by inserting various DNA fragments, gene fragments containing ORFs having no function can be prepared. By introducing a gene fragment lacking the activity of the protein that positively controls mannose-1-phosphate transfer into normal haploid yeast by a known method, the gene locus on the chromosome (this In this case, it is possible to incorporate the above-mentioned inactive gene (in this case, the ΔMNN4 gene in which the function of the MNN4 gene is disrupted) into the MNN4 locus [eg, R. Rothstein, Methods in Enzymo].
logy, Vol. 194, p. 281-301, (1991)]. By mutating or disrupting the MNN4 gene in this manner, a yeast mutant having no activity of positively controlling mannose-1-phosphate transfer can be produced.

Therefore, by crossing this strain with a known yeast mutant strain deficient at various stages of sugar chain synthesis by a known method, a plurality of mutations deficient at various stages of sugar chain synthesis can be obtained. It is possible to converge on two cell lines.
For example, △ och1mnn already proposed by the present inventors
Mutant strains (JP-A-6-277086) and the strains described in this example
The och1 △ mnn1 mutant is conjugated to the above-mentioned mnn4 mutant (mnn4) or mnn4 gene-disrupted strain (△ mnn4) having different mating types, and the resulting diploid cells are sporulated in a nitrogen source-deficient medium [eg, , F. Sherman, Methods in Enzymology, Vo
l. 194, p. 17 (1991)], meiosis was carried out, and the four spores produced thereby were individually separated under a microscope and their phenotypes were examined.
mnn1mnn4 or △ och1 △ mnn1 △ mnn4 triple mutants can be made [e.g., F. Sherman and J. Hicks,
Methods in Enzymology, Vol. 194, p. 21-37 (1991)].

This triple mutant contains a large amount of neutral sugar chains of Man ₈ GlcNAc _{2 represented by} the formula (I) since the production of acidic sugar chains to which mannose-1-phosphate is added is remarkably suppressed. It accumulates asparagine-linked sugar chains and can be used for producing high-mannose sugar chains derived from mammals such as humans and glycoproteins having these sugar chains.

[0043]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in detail below.
Two mutants of baker's yeast with about 90% reduction in phosphorylation at the sugar outer chain have been isolated as mnn4 and mnn6
[EM Karson and CE Ballou, J. Biol. CHem., Vo
l. 253, p. 6484-6492 (1978)]. Of these, alcian blue
Mnn4 was dominant, mnn6 in the dye-binding assay
Is recessive and both are determined to be single mutations. The phosphate group is the major negative charge on the cell wall, and the positively charged basic phthalocyanin dye alcian bl
ue is useful as a simple method to evaluate the degree of phosphorylation of mannan protein.

Since the above mnn6 mutation is a recessive mutation, a wild-type gene corresponding to the mutation can be isolated using a wild-type gene library. The present inventors have already proposed a method for cloning this gene and producing this gene product (protein), and further using this to produce a mannose-1-phosphate or a phosphate-containing acidic sugar chain ( Japanese Patent Application No. 7-299509).

On the other hand, for the mnn4 mutation, mnn4 / X2180 and
It has been reported that diploid cells of mnn4 / mnn6 do not bind dye when grown on YPD medium, but bind dye when grown on YPD medium containing 0.5 M KCl. Therefore, mnn4
Heterodiploids are dominant when grown without KCl (domi
nant), but shows a recessive (recessive) trait when grown in the presence of KCl or sorbitol (CE Ballou,
Methods in Enzymology, Vol. 185, p. 440-470 (199
0)]. Of course, haploid cells of mnn4 do not bind alcian blue dye, regardless of the presence or absence of KCl or sorbitol. Note that mnn
According to the present invention, the four mutations were found to reduce not only phosphorylation to the outer sugar chain but also phosphorylation to the core sugar chain (see Example 7 described later).

Since the mnn4 mutant has a different binding ability to the above-mentioned dyes than wild-type cells, if this property difference is utilized,
A wild-type gene corresponding to the mutation should be able to be isolated,
If this gene cloning is successful, it will be possible to analyze the function of this gene product (protein) by a biochemical technique and use it for controlling the modification of sugar chains.

Under these circumstances, the present inventors have made intensive efforts to produce a glycoprotein having the same sugar chain structure as that of mammals such as humans by using yeast, and as a result, the yeast is specific for human and human. Mannose-1-, a sugar chain modification reaction not found in
Successfully isolated the MNN4 gene involved in phosphorylation,
Using this, a yeast strain having a reduced mannose-1-phosphate content was prepared, and a technology for efficiently producing a neutral sugar chain using the yeast strain was successfully developed.

Using a yeast strain in which the content of acidic sugar chains specific to yeast has decreased and the content of neutral sugar chains has increased, the same N as the high mannose type sugar chains (Man ₈ GlcNAc ₂ ) produced by mammalian cells is used. -The method for producing a linked neutral sugar chain basically comprises the following steps. 1) Preparation of MNN4 gene-disrupted strain (nnmnn4) 2) Preparation of double mutant strain (Δoch1nnmnn1) in which OCH1 and MNN1 genes have been disrupted, and this strain and the above-mentioned △ mnn4
3) Isolation and purification of sugar chains from the cell surface glycoprotein (mannan protein) of the triple mutant strain. In addition to the above-mentioned core sugar chains, In order to produce a glycoprotein derived from a heterologous organism having a core type sugar chain, the following steps are required. 4) Using the triple mutant strain as a host, prepare a gene in which a gene (eg, cDNA) encoding a glycoprotein of interest is connected downstream of a promoter capable of expressing in yeast, and integrate the gene into the chromosome of the yeast host by homologous recombination. Alternatively, a transformant of the host is prepared by inserting the plasmid into a plasmid to transform the host, and the transformant is cultured by a known method, whereby the target sugar produced in or outside the cell is produced. Recover the protein.

Escherichia coli (JM109) containing the plasmid having the MNN4 gene developed in the process of the present invention, a double mutant (Δoch1 △ mnn1), and a triple mutant (Δoch1 △ mnn1 △ mnn4)
Has been deposited with the Institute of Biotechnology and Industrial Technology,
The deposit numbers are as follows. E. coli JM109 / pRS316-MNN4 [Δ (lac-proAB), endoAI , gryA96, hsdR17, λ -, relAI, Sup
^{E44, thi, (F ',} lacI q ZΔM15, proAB, traD36) ] FERM BP-5485 S. cerevisiae YS126-30C ( Δoch1 △ mnn1) [MATαΔoch1 :: LEU2 △ mnn1 :: URA3 △ kre2 :: TRP1 leu2
ura3 trp1 his3 lys2] FERM BP-5487 S. cerevisiae YS126-28D (Δoch1 △ mnn1 △ mnn4) [MATaΔoch1 :: LEU2 △ mnn1 :: URA3 △ mnn4 :: LYS2 △ kre
2 :: TRP1 leu2 ura3trp1 his3 lys2] FERM BP-5486 Hereinafter, the present invention will be described specifically with reference to Examples. However, the technical scope of the present invention is not limited to these examples.

[Example 1] Examination of culture conditions of a mnn4 mutant strain showing a recessive trait advantageous for isolating a gene complementary to the mnn4 mutant strain The presence or absence of a phosphate group in the sugar chain was determined by Alcian blue. (Alci
an blue) can be dyed separately. Alcian blue is a dye that is positively charged and combines with a negative charge. Therefore, the yeast cells are suspended in a buffer of pH 3 and 0.1% of Alcian Blue-8GX (Sigma, Code No. A3157).
When only is added, only those having a phosphate group in the sugar chain are dyed blue, and those having no phosphate group remain white.

The mnn4 mutant strain, LB6-5D (MATa mnn4-1 SUC2 m
al CUP1) (University of California, Berkeley, Yeast
Genetic Stock Center) or TO2-15D (MA
Ta mnn1 mnn4-1 leu2 ura3) [This strain is a YNS3-7A strain (MATα) in which the ura3 mutation has been introduced into YN3-1D, which has been deposited at the National Institute of Bioscience and Biotechnology, under the deposit number FERM P-113219.
△ och1 mnn1 ura3 his1 his3) and the above-mentioned LB6-5D were crossed and then analyzed by tetrad analysis] does not stain Alcian blue because sugar chains on the cell surface hardly have a phosphate group. . The mnn4 mutation is a dominant mutation, and the mnn4 heterodiploid strain (LB6-5D X RA1
-1B or TO2-15D X RA1-1B) does not stain Alcian blue when cultured overnight in YPD medium (1% yeast extract, 2% Bacto peptone, 2% glucose).

On the other hand, there is a report that the dominance of the mnn4 mutation changes to recessive by adding an osmotic pressure adjusting agent to the medium [Balou, CE, Methods in Enzymology, Vol.
185,440-470 (1990)]. To confirm this, use the above mnn
The 4 heterodiploid strain was cultured overnight in YPD medium supplemented with 1 M sorbitol or 0.5 M KCl, and stained with alcian blue. As a result, it was confirmed that the strain was blue. This means that under these conditions, a wild-type yeast gene library was introduced using the mnn4 mutant as a host, and a clone in which the mnn4 mutant (white) was transformed in blue was selected to respond to the mutation. This shows that a normal MNN4 gene can be isolated.

Example 2 Isolation of MNN4 Gene Using Adsorption Ability to Alcian Blue Dye The MNN4 locus is arranged on the left arm of yeast chromosome 11 in the order of URA1-TRP3-MNN4. Yeast Genetic Stock Center, Catalogue, eight ed
ition, p.141]. The yeast cosmid clone ATCC70798 containing 11.1 kb of this region was converted to American Type Culture Collectio.
n Purchased from (ATCC) [ATCC RecombinantDNA Materials,
3rd edition, p.84 (1993)], and the 11.1 kb inserted DNA fragment was converted into a single copy vector pRS316 having a URA3 marker.
[Sikorski, RS and Hieter, P., Genetics, Vol. 122,
p.19-27 (1989)] to obtain plasmid pMMN4.

The mnn4 mutant TO2-15D (MATa mnn1 mnn4-1 le
U2 ura3) 1.0 x 10 ⁸ cells were treated with pMMN4 by the lithium method.
[Ito et al., J. Bacteriol., Vol.153, p.163-168, (1
983)]. After transformation, SD-Ura (2%
Glucose, 0.67% Yeast Nitrogen Base w / o amin
o acids (manufactured by Difco), 0.3 M sorbitol, nucleobase except uracil, and amino acid mixture (20-400 mg / l)] and plated at 30 ° C for 2 days to obtain a transformant.

Alcian blue solution [0.1% in 0.02N hydrochloric acid
Cells after dissolving Alcian Blue-8GX, insert D
When it was tested whether the mnn4 mutation was complemented by the NA fragment, the transformed strain was stained in the same manner as the wild-type yeast.
It was confirmed that the MNN4 gene was present on the NN4 DNA fragment. Example 3 Determination of Base Sequence of MNN4 Gene The DNA base sequence at both ends of the inserted DNA fragment of 11.1 kb was subjected to a sequencing reaction by PCR using a fluorescent dye-bound M13 universal primer and a reverse primer (SequiThermTM Long, Leica, Inc.). -Read Cycle Sequencing
Kit-LC) and the Leica automatic sequencer (LI-COR
(Model 4000L). The obtained DNA base sequence was used to determine the total DNA base sequence data of chromosome 11 [Dujon,
B., et.al., Nature, Vol.369, p.371-377 (1995)]
In light of the above, it was revealed that the protein contained the six protein coding regions (ORF) from ykl-198 to ykl-203. Using restriction enzymes, various partial deletion plasmids were prepared, and it was examined which ORF complements MNN4.
None could be complemented alone (see a in FIG. 4).

When the nucleotide sequence of the inserted DNA fragment was carefully determined and examined in detail, an ORF encoding 3534 bp and 1178 amino acids including the ykl-200c and ykl-201c regions was found (see FIG. 4, b.). It was confirmed that this novel ORF is the MNN4 gene because the characteristic trait of the mnn4 mutation appears even if any part of this ORF is deleted. The MNN4 protein predicted from the DNA base sequence has one hydrophobic region near the N-terminus that may penetrate the membrane, and the C-terminus consists of 17 residues consisting mainly of 4 lysines and 4 residues of glutamic acid. These sequences also have MN by partial deletion experiments.
It was found to be essential for the function of N4p (see FIG. 5).

The Escherichia coli JM109 into which the plasmid pRS316-MNN4 incorporating the MNN4 gene of the present invention has been introduced is E. coli JM1
It is named 09 / pRS316-MNN4 and is deposited with the National Institute of Bioscience and Human Technology under the deposit number FERM BP-5485. [Example 4] Preparation of MNN4 gene high expression vector and Alcian blue of yeast transformant containing this plasmid
Analysis of adsorption ability to dyes Primer 1 (ATAAACTAATTAGGCATGCTTCAGCGAA) with SphI sequence added to 28 bp including ATG sequence of MNN4 gene and Pst
Primer 2 containing I site (ATAATGATGATCTGCAGTAGAATCAGT
G) and amplified by PCR. Multi-copy vector YEp51 with a galactose-inducible promoter (GAL1)
[Rothstein, R. Methods in Enzymology, Vol.194, p281
-301] SalI site downstream of the promoter
A SphI site was created and a DNA fragment (SphI-P
stI) and the downstream region of MNN4 (PstI-SphI) were ligated to obtain an MNN4p high expression plasmid YEp51MNN4.

YEp51MNN4 was transformed into wild type strains RA1-1B and mnn4 mutants, and YPD medium and YPGal medium (Yeast ext
ract 1%, Bacto popton 2%, galactose 2%) and stained with Alcian Blue as in Example 1.
Cells cultured in YPD medium showed the same trait as the parent strain, but cells cultured in YPGal medium stained bluer than the parent strain. This result indicates that the introduction of YEp51MNN4 highly expressed the MNN4 protein and added more mannose-1-phosphate to the sugar chain.

Example 5 An MNN4 gene-disrupted strain (@mnn
4) Preparation Based on the information on the DNA base sequence of the MNN4 gene obtained in Example 3, an MNN4 gene-disrupted strain (Δmnn4) in which the function of the MNN4 gene was completely deleted was prepared. Most of ORF of MNN4 (PstI
-EcoT22I, see b in FIG. 4) and inserted the LYS2 gene (mnn4-Δ1 :: LYS2). This DNA fragment was used for the G2-10 strain (MAT
α leu2 ura3 trp1 his3 ade2 lys2) and G2-9 strain (MA
Taleu2 ura3 trp1 his3 ade2 lys2) and became a lysine non-required transformant G2-10 △ 1 (MATα △ mnn
4 :: LYS2 leu2 ura3 trp1 his3 ade2 lys2) and G2-9
△ 1 (MATa △ mnn4 :: LYS2 leu2 ura3 trp1 his3 ade2 lys
2) selected. Southern blot analysis confirmed that the MNN4 gene was integrated and disrupted. This MNN4 gene-disrupted strain (@ mnn4) does not stain with Alcian Blue, so
It was presumed that no -1-phosphate addition had occurred.

Example 6 Preparation of Δoch1 Δmnn1 Double Mutant and Δoch1 Δmnn1 Δmnn4 Triple Mutant and Their Properties Baker's yeast Δoch1 mutant is a glycoprotein core type sugar chain with mannose outer chain The reaction that adds the first α-1,6-linked mannose necessary for the addition is defective. The mnn1 mutant lacks the reaction of adding α-1,3-linked mannose to the non-reducing terminals of the core sugar chain and the outer sugar chain branch. Since the mnn1 mutation may still have a small amount of activity to add α-1,3-linked mannose to the non-reducing end, the MNN11 gene-disrupted strain completely lacking the function of the MNN1 gene (△ mnn1
1) was prepared.

The MNN1 gene is located near the chromosome 5 centromere of yeast, and the DNA base sequence of the MNN1 gene is registered in GenBank database at L23753 [Yip et al. Pro.
c. Natl. Acad. Sci. USA Vol. 91, p. 2723-2727, (199
Four)]. Primer A (GGATCCGAAGAA
AACCTAATACATTGAAGT) and B (GCATGCCCTTTGGTTTAATATAAAT
CTCCGGAGTGC) and the 5 '
(GCATGCTACATAACTCCAATCAGCAGCAAATATGTC) and D (GCGGC
CGCGTGTTCTGTTCGGGTAACGTTTAAACCAAT) and amplified by PCR. These DNA fragments were converted to pJJ244 with the URA3 marker [Jones, J., & Prakaash, L., Yeast, Vol.
p.363-366 (1990)] to obtain a plasmid pJJ @ mnn1 for disrupting the MNN1 gene.

YS57-5A (MATα △ och1 :: LEU2 leu2 ura3 h
is1 his3) and Roch3-13D (MATaoch3 :: TRP1 leu2 ura
3 trp1) was transformed with pJJ △ mnn1 and the uracil marker was
Restoration of auxotrophy, tetrad analysis, and deletion of α-1,3 linked mannose at the non-reducing end confirmed that it was integrated into the chromosome of the MNN1 region, and the MNN1 gene disrupted strain YS57-5A △ mnn1 And Roch3-13D △ mnn1 were prepared.

[0063] These double gene-disrupted strains (nnoch1nnmnn
1) produces a neutral core sugar chain of Man ₈ GlcNAc ₂ as in the case of the △ och1mnn1 double mutant (JP-A-6-277086) already proposed by the inventors (see Example 7 described later). However, these double mutants contain, in addition to this neutral sugar chain, mannose
It was also found that acidic sugar chains containing -1-phosphate are by-produced (Preprints of the 12th Biotechnology Symposium, p.15)
3-157, published October 14, 1994, published by the Biotechnology Development Technology Research Association).

Therefore, in order to obtain a strain having a reduced mannose-1-phosphate content, a triple gene disruption strain of Δoch1Δmnn1Δmnn4 in which the MNN4 gene required for mannose-1-phosphate addition was disrupted was prepared. . G2-10 △ 1 produced in Example 5
Stock and YS123-3C (MATa △ och1 :: LEU2 △ mnn1 :: URA3 △ kre2 ::
After crossing with TRP1 leu2 ura3 trp1 his3 lys2) and analyzing the tetrad, double gene disrupted strain YS126-30C (MATα △ och1 :: LE
U2 △ mnn1 :: URA3 △ kre2 :: TRP1 leu2 ura3 trp1 his3 lys
2), Triple gene disrupted strain YS126-28D (MATa △ och1 :: LEU2 △ m)
nn1 :: URA3 △ m nn4 :: LYS2 △ kre2 :: TRP1 leu2 ura3 trp1 h
is3 lys2). Here, △ kre2 :: TRP1 is used to facilitate the analysis of sugar chains.
l University, Canada) is a disruption of a gene involved in o-linked glycosylation provided by 供 与 och1 △ mnn1 double mutant strain, tetrad analysis, and YS123-3C strain Produced.

The above-mentioned double mutant YS126-30C and triple mutant YS1
26-28D were S. cerevisiae YS126-30C and S. cerevi, respectively.
Named siae YS126-28D, deposited at the National Institute of Advanced Industrial Science and Technology, National Institute of Advanced Industrial Science and Technology, under the accession numbers FERM BP-5487 and FERM BP-5
Deposited at 486. Example 7 △ och1 △ mnn1 Double Mutant and △ och1 △ mn
Isolation of cell surface mannan protein from the n1 △ mnn4 triple mutant and structural analysis of its sugar chains.Mannan protein on the cell surface was isolated from cells of S. cerevisiae YS126-30C and YS126-28D [Peat, S. et. .al.,
J. Chem. Soc., P29 (1961)]. 200 ml of YPD medium supplemented with 0.3 M sorbitol was placed in a 2 l Erlenmeyer flask, and 25
Incubate at 24 ° C for 24 hours, collect cells by centrifugation, and
l of 0.02 M sodium citrate buffer (pH 7.0),
Heated in an autoclave at 121 ° C. for 1 hour. After cooling, the mixture was centrifuged to remove the supernatant, and the solid substance was again heated and centrifuged by adding 10 ml of water, and the supernatant was collected.

All the extracts were combined and poured into a three-fold amount of ethanol. The resulting white precipitate was dried, dissolved in water, and then desalted using a gel filtration column. This was lyophilized and used as a mannan protein (yield 50 mg).
The N-linked sugar chain was cut out by a hydrazine decomposition method (Hydra Club, manufactured by Honen, Code No. 80060).
0). Put 10 mg of dried mannan protein in a glass tube and inject hydrazine (Honen).
The reaction was performed at 10 ° C. for 1 hour. After removing the hydrazine, 250 ml of 0.2 M ammonium acetate and 25 ml of acetic anhydride were added, the mixture was left at room temperature for 30 minutes, and the same operation was repeated once to perform N-acetylation. This is concentrated to dryness and the sugar chain preparation (10
μg).

This sugar chain preparation was fluorescently labeled with pyridylamination (referred to as PA) with 2-aminopyridine [Akihiro Kondo et al., Biochemistry, Vol. 60, p689 (1988)]. PA conversion was performed using a carbohydrate analysis kit (Takara Shuzo, Code No. 5000).
Toy equilibrated with 10 mM ammonium acetate buffer (pH 7.0)
Gel filtration was performed on an opearl HW-40F (manufactured by Tosoh) column (1.0 × 40 cm) to remove unreacted 2-aminopyridine to obtain a PA-oligosaccharide preparation.

In HPLC using an amino column, it is possible to separate PA-oligosaccharides by their chain length and charge. The column used was ASAHIPAK NH2P-50 (4.6 x 250 mm, manufactured by Showa Denko). The solvent was a 30:70 mixture of 200 mM acetic acid-triethylamine buffer (pH 7.3) and acetonitrile (solvent A), 200 mM Acetic acid-triethylamine buffer (pH
A mixture of 7.3) and acetonitrile at 70:30 (solvent B) was prepared.

The column was previously mixed with solvent A and solvent B at 80:20.
Equilibrate by flowing the mixture at a flow rate of 1.0 ml / min, and immediately increase the ratio of solvent B to 100% over 40 minutes immediately after sample injection, then elute PA-oligosaccharide with 100% solvent B did. FIG. 6 shows the result. The N-linked sugar chains produced by the Δoch1Δmnn1 double mutant YS126-30C could be separated into four main peaks (ad). Peak a is a PA-rich high mannose sugar chain sample (Takara Shuzo, Code No. 4117, 4119, 412)
0), and Man ₈ GlcNAc ₂ shown in JP-A-6-277086.
From the comparison with the sugar chain elution position, the following formula (VIII)

[0070]

Embedded image

(Wherein, M and GN are the same as described above)
It was confirmed to be a neutral sugar chain of Man ₈ GlcNAc ₂ shown by.
Peaks b to d are acidic sugar chains and account for about 70% of the total. Peak b and peak c are composed of one residue of mannose-1-
The following formulas (IX) and (X) in which phosphoric acid is added to the α-1,3 branch side and α-1,6 branch side of the Man ₈ GlcNAc ₂ sugar chain, respectively.

[0072]

Embedded image

[0073]

Embedded image

(Wherein, M and GN are the same as described above)
Is the same as the sugar chain represented by the formula, its retention time by HPLC, conversion to neutral sugar chain by weak alkaline treatment and subsequent alkaline phosphatase treatment [Ballou, CE, Me
thods in Enzymology, Vol. 185, p. 440-470 (1990) and Japanese Patent Application No. 7-299509, Example 6]. Peak d is determined by its retention time on HPLC by NMR and T
Two residues of mannose-1-phosphate were added to both the α-1,3 branch and the α-1,6 branch because they matched the sugar chain sample whose sugar chain structure was confirmed by OF-MS. Equation (XI)

[0075]

Embedded image

(Wherein, M and GN are the same as described above)
It was confirmed that the structure was represented by Meanwhile, △ och1
In the Δmnn1 Δmnn4 triple mutant strain YS126-28D, the amounts of peaks b to d, which are acidic sugar chains, were decreased, and peak a, which was a neutral sugar chain, was decreased.
Accounted for more than 70% of the total (see FIG. 6). The result is
△ och1 △ mnn1 △ mnn4 triple mutants than △ och1 △ mnn1 double mutant, which indicates that suitable for producing neutral sugar chain of Man ₈ GlcNAc _2.

[0077]

According to the present invention, a neutral core sugar chain identical to high mannose produced by mammalian cells such as humans, or a high mannose glycoprotein having these sugar chain structures can be produced by a genetic engineering technique using yeast. It can be produced in large quantities and with high purity. This mutant is identical to the core sugar chain (Man ₈ GlcNAc ₂ ), which is a reaction product of the ER common to yeast and mammalian cells (see the compound of formula (I) and FIG. 1), in the sugar chain biosynthesis pathway. A sugar chain having the chemical structure of Therefore, it is possible to supply a stable and large amount of uniform high-mannose-type sugar chains having the same chemical structure as humans.

[0078]

[Sequence list]

 SEQ ID NO: 1 Sequence length: 1,178 Sequence type: Amino acid Topology: Linear Sequence type: Protein sequence

[0079]

SEQ ID NO: 2 Sequence length: Sequence type: Number of nucleic acid strands: Single strand Topology: Linear Sequence type: Genomic DNA Origin: Gene that positively controls mannose-1-phosphate transfer Organism name: Saccharomyces cerevisiae AB972 Sequence characteristics: Characteristic code: coding sequence Location: Sequence determination method: Sequence

[0081]

[0082]

[0083]

[0084]

[0085]

[Brief description of drawings]

FIG. 1 is a diagram showing the biosynthetic pathway of N-linked sugar chains in the endoplasmic reticulum (ER) of yeast.

FIG. 2 is a view showing a mannose addition pathway to a core sugar chain in an N-linked sugar chain of yeast.

FIG. 3 is a diagram showing a mannose addition pathway to a core sugar chain in an N-linked sugar chain of yeast.

FIG. 4 is a diagram showing isolation of the MNN4 gene and disruption of the MNN4 gene.

FIG. 5 shows an amino acid sequence of the MNN4 protein.

FIG. 6 is a diagram showing a structural analysis of a cell surface mannan protein sugar chain.

Continuation of front page (51) Int.Cl. ⁶ Identification number Office reference number FI Technical display location C12N 9/12 C12N 9/12 // (C12N 15/09 ZNA C12R 1: 865) (C12N 1/19 C12R 1 : 865) (C12N 9/12 C12R 1: 865)

Claims (8)

[Claims] 1. A gene which positively controls mannose-1-phosphotransfer in yeast, which substantially encodes the amino acid sequence shown in SEQ ID NO: 1. 2. The gene according to claim 1, which is represented by the nucleotide sequence shown in SEQ ID NO: 2 and positively controls mannose-1-phosphate transfer in yeast. 3. The mannose-1- according to claim 1 or 2.
Plasmid DNA containing all or part of a gene that positively controls phosphoryl transfer. 4. The plasmid DNA according to claim 3, wherein a gene fragment deficient in the function of a gene that positively controls mannose-1-phosphate transfer is prepared and integrated into a homologous region of a yeast chromosome. , A yeast in which the function of a gene that positively controls mannose-1-phosphate transfer has been disrupted. 5. A Δmnn4 yeast in which the MNN4 gene has been disrupted, an OCH1 gene involved in α-1,6-mannose addition required for initiation of mannose outer chain synthesis, and α-1, which is a non-reducing terminal of a sugar chain. Yeast strain disrupting both MNN1 genes involved in 3-mannose addition (@ och1 @ mnn1)
And the triple mutation (@ och1
Δmnn1 △ mnn4) Yeast. 6. The yeast according to claim 5, which is cultivated in a medium, and comprises two molecules of N-acetylglucosamine and eight molecules of mannose from the culture. (Wherein, M represents mannose and GN represents N-acetylglucosamine), wherein the high-mannose type neutral oligosaccharide chain-linked glycoprotein having a neutral oligosaccharide chain bonded thereto is recovered. A method for producing a high-mannose type neutral oligosaccharide chain-linked glycoprotein. 7. The high-mannose neutral oligosaccharide chain-linked glycoprotein according to claim 6, which is represented by the following formula (I): (Wherein M and GN are the same as described above), respectively. A method for producing a neutral oligosaccharide chain, comprising: 8. A protein comprising a polypeptide having substantially the amino acid sequence shown in SEQ ID NO: 1 and positively controlling mannose-1-phosphorylation.