JPH06277086A - Production of mammalian high mannose-type glucoprotein sugar chain by yeast - Google Patents

Abstract

PURPOSE:To provide a method for producing a sugar chain composed of two molecules of N-acetylglucosamine and eight molecules of mannose and having a uniform and same sugar chain as that of human and a glycoprotein having the same sugar chain of this chemical structure using a yeast by a recombinant DNA technique. CONSTITUTION:A sugar chain having a structure same as that of a high mannose type sugar chain produced by mammalian cell and a glycoprotein obtained by forming an adduct of this sugar chain to an asparagine residue of a protein are produced using a yeast variant having a defect in a part of biosynthesis of the sugar chain. A gene encoding for a protein derived from Mammalian is expressed by using this yeast valiant as a host. Thereby, the high mannose type glycoprotein having high purity (obtained by binding the oligosaccharide chain composed of two molecules of N-acetylglucosamine and eight molecules of mannose to the asparagine residue) same as that derived from mammalian cell can stably and abundantly be produced.

Description

Detailed Description of the Invention

TECHNICAL FIELD The present invention relates to a sugar chain having the same sugar chain structure as a high-mannose type sugar chain produced by mammalian cells, and a glycoprotein obtained by adding this sugar chain to an asparagine residue of a protein, The present invention relates to a method for producing yeast using a genetic engineering technique.

[0001]

2. Description of the Related Art Most of proteins having important functions in vivo are not simple proteins but glycoproteins having a sugar chain. Erythropoietin ( It has been revealed in various glycoproteins such as EPO) and tissue plasminogen activator (TPA) (Kihatayo, Protein Nucleic Acid Enzyme, Vol.36, p775 (199)
1); Makoto Takeuchi, Biochemistry, Vol.62, p1272 (1990)). It is suggested that the sugar chain plays an important role in the expression of biological activity, but since the correlation between the sugar chain structure and biological activity is not always clear, the type and structure of the sugar chain added to the protein part should be considered. Is currently undergoing trial and error for each protein. To this end, it is necessary to develop a technique that allows modification and control of the structure of the sugar chain added to the protein (type of sugar, binding position, chain length, etc.) as desired.

There are two types of sugar chains that bind to proteins, N-linked that binds to asparagine residues of the protein and O-linked that binds to serine or threonine. this house,
There are many findings on the biosynthetic pathway of N-linked sugar chains, and they have been analyzed in detail. According to this, the biosynthesis of sugar chains first starts in the endoplasmic reticulum (ER), and then the sugar chains are further modified in the Golgi apparatus. Of these, it is known that the sugar chains produced by ER are basically the same in yeast and mammalian cells. Hereafter, this is referred to as a core type sugar chain, and it is known that the sugar chain is composed of 8 molecules of mannose (Man) and 2 molecules of N-acetylglucosamine (GlcNAc). This core type sugar chain (Man
Proteins with ₈ GlcNAc ₂ ) are transported to the Golgi apparatus and undergo various modifications, but the modifications in this Golgi apparatus differ greatly between yeast and mammalian cells (Kukuruzinska e
t al, Ann. Rev. Biochem., Vol.56, p915 (1987)).

In mammalian cells, there is no change in the core type sugar chain.
There are some glycoproteins that are not accepted, but in most cases, core type
Man of 5 molecules is removed from the sugar chain₃GlcN
Ac ₂And, before and after this, GlcNAc and Gal
(Galactose), NeuNAc (N-acetylneura)
A variety of sugars can be added by sequentially adding methanoic acid, or sialic acid
The chains form as a mixture. On the other hand, in yeast, core type sugar
Add 7-100 molecules of Man to the chain sequentially,
Produces a loose sugar outer chain (Kukuruzinska et al, Ann. Rev.
 Biochem., Vol.56, p915 (1987)). Thus yeast
In mammalian cells, the structure of sugar chains added to proteins is large
Since they are different, genetic engineering techniques are used to
Even if a useful glycoprotein derived from milk is produced, the structure of the sugar chain
Identical biological activity detected from mammal due to differences
Or the difference in antigenicity due to the difference in sugar chains
It has been pointed out. Therefore, the glycoprotein of mammalian origin is fermented.
At present, it is difficult for mothers to produce them.

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 a mammalian-derived reagent 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.

[0005]

In order to overcome the qualitative and quantitative drawbacks in producing and supplying such N-linked glycoproteins derived from humans and other mammals, the present invention uses recombinant DNA technology to solve the problems. Production of a uniform sugar chain composed of 8 mannose and 2 N-acetylglucosamine and having the same sugar chain structure as human and a glycoprotein having the sugar chain of this chemical structure using a microorganism It is possible to do.

[0006]

That is, the present invention includes the following configurations. (1) A yeast of the genus Saccharomyces, which is an oligosaccharide chain composed of 8 molecules of mannose and 2 molecules of N-acetylglucosamine, each of which has a structure of an asparagine-binding glycoprotein sugar chain produced by a yeast cell.

[0007]

[Chemical 5] (In the formula, M represents mannose and GN represents N-acetylglucosamine)

(2) A yeast mutant strain which belongs to Saccharomyces cerevisiae in which the yeast of the genus Saccharomyces described in (1) above belongs to Saccharomyces cerevisiae and which has both och1 mutation (Δoch1) and mnn1 mutation in which the OCH1 gene is disrupted. (3) The Saccharomyces cerevisiae described in (2) above is Y.
Saccharomyce which is N3-1D strain or YN4-18A strain
s cerevisiae yeast mutant. (4) The yeast according to any one of (1) to (3) above is cultured in a medium to recover yeast cells from the culture,
A method for producing an oligosaccharide chain according to Chemical formula 6, characterized in that the oligosaccharide chain represented by Chemical formula 6 is recovered from the cells.

[0009]

[Chemical 6] (In the formula, M represents mannose and GN represents N-acetylglucosamine)

(5) A host containing the oligosaccharide chain represented by Chemical formula 7 as an asparagine-linked sugar chain, which is obtained by culturing the yeast according to any one of (1) to (3) above and culturing the yeast. A method for producing a yeast-derived glycoprotein containing the oligosaccharide according to Chemical formula 7, characterized in that the yeast-derived glycoprotein is recovered.

[0011]

[Chemical 7] (In the formula, M represents mannose and GN represents N-acetylglucosamine)

(6) Using the yeast according to any one of (1) to (3) above as a host, a gene encoding a mammalian-derived asparagine-binding glycoprotein is expressed, and the expressed glycoprotein asparagine is expressed. The method for producing a mammalian-derived glycoprotein containing an oligosaccharide according to Chemical formula 8, wherein the structure of the linked sugar chain is an oligosaccharide chain represented by Chemical formula 8.

[0013]

[Chemical 8] (In the formula, M represents mannose and GN represents N-acetylglucosamine)

The present invention will be described in detail below. The biosynthesis of N-linked sugar chains by yeast first starts in the endoplasmic reticulum (ER), and then in the Golgi apparatus, modifications such as sugar chain extension and phosphorylation occur. The biosynthetic pathway of sugar chains in the ER is already known (Kukuruzinska et al, Ann. Rev. Bioche
m., Vol.56, p915 (1987)), yeast and human and animal cells are basically the same, and the sugar chain finally produced by ER is referred to as a core-type sugar chain. The sugar chain transferred to the asparagine residue (Asn) of the protein is obtained by sequentially adding monosaccharides from sugar nucleotides to dolichol pyrophosphate (Dol-PP) localized in the ER membrane to obtain 14 sugars (Glc ₃ Man ₉ GlcN).
After becoming Ac ₂ ), this is collectively transferred to Asn which is a protein that has entered the ER lumen. 3 more after this
The molecule Glc and one molecule Man are removed to form a core type sugar chain (Man ₈ GlcNAc ₂ ), which is then transported to the Golgi apparatus by being wrapped in transport vesicles derived from ER.

The modification of the glycoprotein sugar chain in the Golgi apparatus is largely different between human and animal cells and yeast. In the former case, Man is often removed and finally Man is removed.
₃ GlcNAc _{2 and} , before and after this, GlcNAc
Whereas c, Gal and NeuNAc residues are added to form a mixture of various sugar chain structures containing various sugars, in the latter, the above core-type sugar chain (Man ₈ GlcNAc ₂ ) has Ma
About 7-100 molecules of n are added sequentially to form a so-called outer chain. It is known that the number of added Mans varies depending on the type of protein and the position of the sugar chain addition site.

For example, carboxypeptidase Y (CPY), which is localized in the vacuole, adds only about 6 molecules of Man to each of four glycosylation sites, whereas invertase (INV) secreted to the cell surface layer ), Sugar chains have been added at 13 of 14 sites where sugar chains can be added, and 9 of them have the same 6 molecules of Man as CPY.
, But Man with a chain length distribution of 30 molecules or more is added to the remaining 4 positions (Trimble et al,
J. Biol. Chem., Vol.258, p2562 (1983); Reddy et a
l., J. Biol. Chem., Vol.263, 6978 (1988); Ziegler
et al., J. Biol. Chem., Vol.263, p6986 (1988)).

Biosynthesis of the sugar outer chain in Saccharomyces yeast is considered to proceed through the pathway shown in FIG. 1 (Ballou et al, Proc. Natl. Acad. Sci. USA, Vol. 8).
7, p3368 (1990)). That is, α-1,
Reaction of extension initiation in which Man is added by 6 bond (Fig. 1, I)
Then, a reaction of sequentially extending Man with α-1,6 bond (Fig. 1, II) occurs to form poly-α-1,6Man bond as a skeleton of the sugar outer chain (Fig. 1, E). This Man with α-1,6 bond has α-1,
There is a 2-linked Man branch, and this branched α-
At the tip of 1,2-bonded Man, usually α-1,3 is further added.
Bound Man is added (see FIG. 1, G). However, the Man at the tip of the α-1,6 bond is simply added with the α-1,2-bonded Man, and the M of the α-1,3 bond is M.
an does not add further (Figure 1, F or G)
(Gopal and Ballou, Proc. Natl. Acad. Sci. USA, Vo
l.84, p8824 (1987)).

From the results of such structure analysis, Man of α-1,3 bond of the sugar outer chain exists as a termination signal of the branched structure and at the tip of the skeleton α-1,6 bond. It is presumed that the Man of α-1,2 bond may function as an extension stop signal of the outer sugar chain (Ballou et al, J. Biol. Chem., Vol. 264, p11857).
(1989)). Regarding the position of the core sugar chain in which the extension of the outer sugar chain occurs, it was conventionally thought that it might be Man described in M ° in FIG. 2, but recently, various outer chain addition partial deletion mutants such as mnn9 have been recently proposed. The chemical structure of the sugar chain produced by
From the results of detailed analysis by MR, etc., α-1, 3 and α
Man with -1,2 bond has been modified to bind with alpha-1, 6 in (M where noted in the M ^o FIG 2) (Hernandez et
al, Vol. 264, p11849 (1989); Hernandez et al, Vol.
264, p13648 (1989)).

Therefore, in order to produce a core-type sugar chain using yeast, a reaction such that a large number of Man is added to the core-type sugar chain is a modification of such a glycoprotein sugar chain peculiar to yeast. It should be achieved by isolating a mutant strain having a sugar chain biosynthesis system in which the sugar chain is not added and the sugar chain synthesis is stopped at the core type sugar chain (Fig. 1, A). Is. As a mutant strain of yeast having a deficiency in the addition of sugar chain,
Various mnn mutants have already been reported. However, the additional deletion site varies as shown in FIG. For example, the mnn2 mutant strain has an α-1,6 skeleton of the sugar outer chain and α-
There is a defect in the branching step that produces the 1,2 bond and m
The nn1 mutant strain is defective in the step of generating α-1,3 linked Man at the branching tip. However, since these mutants have no defect in the α-1,6Man bond, which is the backbone of the outer sugar chain, all of them produce a long outer sugar chain.

On the other hand, in the mnn9 mutant, α-of the outer sugar chain
Most of the 1,6 bonds are deleted, and the outer sugar chain is stopped in a structure in which 2 to 5 molecules of Man are added to the core sugar chain (Fig. 1, C or D). Of these, one molecule of Man is added to the core sugar chain as an α-1,6 bond necessary for initiation of extension of the outer sugar chain, but this extension in the α-1,6 direction is inhibited by some influence of the mutation. Therefore, it is considered that Man of α-1,2 bond, which functions to stop elongation, was added (He
rnandez et al., J. Biol. Chem., Vol.264, p11849 (1
989)).

Other addition mutations of the sugar outer chain include α
-1,6 Man bond has only about 4 to 15 molecules m
Mutants such as nn7, 8, 10 have also been isolated,
These mutants also only shorten the outer sugar chain and do not terminate the sugar chain in the core type (see Fig. 2) (Balloue
t al., J. Biol. Chem., Vol. 255, p5986 (1980); Ball
ou et al., J. Biol. Chem., Vol.264, p11857 (1989)
). In addition, in the mnn9 mutant strain in which most of the outer sugar chain is deficient, the activity level of the enzyme (α-1,6-mannosyltransferase) that catalyzes the initiation of elongation of the outer sugar chain and the subsequent extension of addition is different from that of the wild strain. Since this mutation is not present, it is speculated that this mutation was not a mutation in the gene encoding the glycosyltransferase that transfers mannose, but a secondary effect caused by a mutation in another gene also caused a defect in the extension of the sugar chain. Gopal and
Ballou, Proc. Natl. Acad. Sci. USA, Vol.84, p8824
(1987)).

Thus, as an example in which the outer sugar chain is affected by a mutation not directly related to sugar chain addition, a mutation called ssc1 or pmr1 has been reported (Smit).
h et al, Science, Vol.229, p1219 (1985)). This mutant strain has a mutation in the gene encoding a protein (Ca ²⁺ -ATPase) that is localized in the membrane of the ER or Golgi apparatus and is involved in the membrane permeation of Ca ²⁺ ions. Since the chain length of the INV sugar chain generated by this mutant strain is almost the same as that of mnn9, it is considered that the sugar chain addition is secondary to the mutation and the normal addition of the outer sugar chain does not occur. (Rudolph et al, Cell, Vol.58, p133 (1989)).

The above-mentioned addition defect of the sugar outer chain is due to the temperature sensitivity of the protein transport from the ER to the Golgi apparatus.
It is also observed in secretory mutant strains such as 18. But sec
The mutant strain is not suitable for the purpose such as secretory production of glycoprotein because the secretion of protein itself is inhibited at high temperature. On the other hand, the mnn9 and pmr1 mutants have normal protein secretion and have potential utility, but as described above, the sugar chain structure added to the protein has a core type sugar chain (Man ₈ It is larger than GlcNAc ₂ ) and further contains 2-5 molecules of Man. Of these, 1 to 2 molecules of Man are those in which α-1,3 linked Man is added to the core type sugar chain (see FIG. 1, D).

This reaction is unique to yeast and does not exist in mammals such as humans. In the mnn9 mutant strain, the core-like sugar chain (Man ₁₀ GlcNAc ₂ ) having no α-1,3 linked Man at the non-reducing end of the core sugar chain (Man ₁₀ GlcNAc ₂ ) was introduced by further introducing the above-mentioned mnn1 mutation (Fig. 1, C) has been successfully produced (Hernandez et al, J. Bio
l. Chem., Vol.264, p11849 (1989)), which still produces a core-type sugar chain produced by mammalian cells such as humans (Man.
₈ GlcNAc ₂ ) (FIG. 1, A), α-1,
An extra Man of 2 molecules in total of 6-bonded Man and α-1,2-bonded Man is added.

As described above, there are many kinds of glycoproteins derived from mammals such as humans, and the structure of the sugar chain added to the protein differs depending on the kind of protein, species of organism, organ, etc., and also the chain length of the sugar chain. It is not uniform but heterogeneous with a constant chain length distribution (Kibatayo, Protein Nucleic Acid Enzyme, Vol.36, p7
75 (1991) ； Makoto Takeuchi, Biochemistry, Vol.62, p1272 (199
0)). Therefore, it has been pointed out that the correlation between the biological activity of the sugar chain and the sugar chain structure cannot be clarified, and a sugar chain having a uniform chain length and a clear chemical structure and this sugar chain were added. The supply of glycoproteins is expected not only from academia but also from industry.

Further, as in the above-mentioned mnn1mnn9 double mutant strain, a sugar chain structure that is similar to human but not identical is not the same as that of human. Therefore, in the case of glycoprotein intended for pharmaceuticals, etc. When administered in vivo, it is recognized as a foreign antigen and may cause various side effects such as allergic reaction.

Under these circumstances, the present inventors have
As a result of diligent efforts to produce a glycoprotein having the same sugar chain structure as that of mammals such as human using yeast, it is thought that it is an addition-deficient mutant with an outer sugar chain different from mnn9 and can be used for the above purpose. The mutant strain och1 was successfully isolated. Growth of this mutant strain is temperature-sensitive, but at high temperature, the sugar chain added to glycoprotein (INV) is a core-type sugar chain (Man ₈ GlcNA) accumulated by secretory mutant sec18.
c ₂ ) (Fig. 1, A), longer than that of mnn9 (Ma
n _10-13 GlcNAc ₂ ) (Fig. 1, C and D) has already been reported (Nagasu et al, YEAST, Vol.
8, p535 (1992)).

Furthermore, it was revealed from the isolation of the gene and the analysis of the encoded protein that this mutation was due to the mutation of the gene of glycosyltransferase that recognizes a specific sugar chain structure and transfers mannose. Nakayama et al, EMBO J.,
Vol.11, p2511 (1992)). However, the detailed chain length distribution and chemical structure of the sugar chains produced by this mutant strain remained unclear. Therefore, a mutant strain (Δoch1) in which the OCH1 gene was disrupted was prepared, and an N-linked sugar chain added to the glycoprotein (INV) secreted on the cell surface of this mutant strain was isolated and produced. Chemical structure is HPLC, NM
Analysis was performed by making full use of instrumental analysis such as R and FAB-MS. The isolated sugar chain was further digested with an enzyme (mannosidase) that recognizes and hydrolyzes only a specific mannose bond, and the structure of the product was similarly analyzed.

As a result, this gene-disrupted strain (Δoch
The N-linked sugar chain generated in 1) includes the aforementioned core sugar chain (Man ₈ GlcNAc ₂ ) (FIG. 1, A), and one molecule of Man having α-1,3 linked thereto is 1 or Sugar chain structure added at two positions (Fig. 1, H and I) (Man ₉ G
It was found to consist of a mixture of lcNAc ₂ and Man ₁₀ GlcNAc ₂ ). Unlike the mnn9 mutant strain, the sugar chain of the OCH1 gene-disrupted strain has an extension from the core type in the α-1,6 direction and the extended Man
There is no Man branching or the like at the α-1,2 bond. Therefore, based on this finding, as a result of extensive studies on a method for producing the same core type sugar chain as that of mammals, the OCH1 gene-disrupted strain (Δoch1) has a defect in addition of Man at α-1,3 bond. By further introducing the mnn1 mutation,
A mutant strain (Δoch1mnn1) that produces only a uniform desired core sugar chain (Man ₈ GlcNAc ₂ ) (FIG. 1, A) that does not have α-1,3-linked Man at the non-reducing end.
I thought it could be created.

Therefore, 2 were produced by crossing haploid yeasts each having the Δoch1 and mnn1 mutations alone.
Among the haploid yeasts obtained by sporulation of polyploid yeasts by meiosis, those showing the trait of Δoch1mnn1 double mutant strain were selected, and the N-linked glycoprotein produced by this double mutant strain was selected. When the chemical structure of the sugar chain was analyzed, it was confirmed that only the desired core type sugar chain (Man ₈ GlcNAc ₂ ) (Fig. 1, A) was produced as intended, and the present invention was completed. . That is, the same N as the N-linked glycoprotein having a high mannose type sugar chain produced by mammalian cells is produced.
-The method for producing a linked sugar chain in yeast basically comprises the following steps.

1) OCH1 gene-disrupted strain (Δoch1)
Of a double mutant strain (Δoch1mnn1) by crossing with a mnn1 gene mutant strain (mnn1) 2) Isolation / purification of sugar chain from cell surface glycoprotein (mannan protein) of Δoch1mnn1 double mutant strain In order to produce a high-mannose type glycoprotein derived from mammals such as human having a core type sugar chain in addition to the chain, the following steps are further required. 3) Using the above double mutant strain as a host, a gene in which a gene (cDNA, etc.) encoding a target glycoprotein is connected downstream of a promoter capable of expressing in yeast is prepared, and homologous recombination is carried out to the chromosome of the above yeast host. Incorporate or
Alternatively, a transformant of the above-mentioned host is prepared by inserting it into a plasmid and transforming the above-mentioned host, and by culturing this by a known method, the target glycoprotein produced intracellularly or extracellularly. Collect.

[0032]

INDUSTRIAL APPLICABILITY According to the present invention, a large amount of a high-mannose glycoprotein having the same core-type sugar chain as that of high-mannose produced by mammalian cells such as human, or a high-mannose-type glycoprotein having this sugar chain structure can be produced by a genetic engineering technique using yeast. It can be produced with high purity. This mutant strain has the same chemical structure as the core type sugar chain (Man ₈ GlcNAc ₂ ) (Fig. 1, A), which is a reaction product of ER common to yeast and mammalian cells, in the sugar chain biosynthesis pathway. It produces sugar chains. Therefore, it is possible to stably and in large quantity supply a uniform high-mannose type sugar chain having the same chemical structure as human, which is in great demand as a sugar chain analysis reagent.

Also, using this core type sugar chain as an intermediate material,
It is also possible to act on various glycosyltransferases to convert into various sugar chains having different chemical structures. Further, the present invention provides
Using this yeast mutant as a host, a mammalian-derived glycoprotein having a high mannose type sugar chain can be produced by a gene recombination technique. As glycoproteins containing high mannose type sugar chains produced by mammalian cells, bovine pancreatic ribonuclease B, human IgM, bovine and rat kidney γ-glutamyl transferase, etc. have been reported (Junko Amano, Laboratory for New Chemistry). 3, sugar I,
Glycoprotein (above), p324, Tokyo Kagaku Dojin (199
0)). It has been clarified that the high-mannose type sugar chain is not only important as an intermediate of sugar chain biosynthesis, but also importantly contributes to the expression of various biological activities.

For example, homing of lymphocytes (Stoolman et
al, J. Cell Biol., Vol.99, 1535 (1984)) and TGF-
Activation of β1 (transforming growth factor β1) precursor (Miyazono and Heodin, Nature (London), Vol.338, p1
58 (1989)). In addition, it has been speculated that the high mannose type sugar chain of Fc receptor (FcrRIII) of neutrophil may affect the binding ability to IgG in a three-dimensional structure, or may serve as a receptor for bacterial mannose lectin (Kimberly et al, J. Immunol., Vol. 142, p3923
(1989)). Therefore, the above-mentioned Δoch1mnn1 double mutant strain can also be used as a host for producing a mammalian-derived glycoprotein having these useful high-mannose type sugar chains by genetic engineering technology.

The Δoch1mnn1 double mutant strains obtained here have been deposited at the Institute of Microbial Technology, Institute of Industrial Science. The deposit number is as follows. S. S. cerevisiae YN3-1D was produced by Microindustrial Research Institute No. 13219 (FERM P-13219) and S. cerevisiae YN4-18A is Microbiology Research Institute No. 13220 (FERM P-13220).

Hereinafter, the present invention will be described in detail with reference to Examples, but the present invention is not limited to the following Examples. Example 1 Isolation and purification of sugar chain of invertase (INV) produced by OCH1 gene disrupted strain (Δoch1) (1) Purification of invertase (INV) OCH1 gene disrupted strain Saccharomyces cerevisiae YS52-
1-1B (Nakayama et al., EMBO J., Vol.11, p2511 (199
2)) was used to purify INV. The cell culture is YEPD
Medium (1% yeast extract, 2% bacto-peptone, 2% D
-Glucose) plus 0.3 M sorbitol in medium 3
1 liter was placed in a 5 liter jar fermenter and the treatment was carried out at 25 ° C. After confirming that glucose was consumed using Glucose CII-Test Wako (manufactured by Wako Pure Chemical Industries, Ltd.), sucrose was added to 0.5%, and the mixture was further incubated for 4 hours to obtain INV. Was induced to produce. Cells were collected from 12 l of the culture solution and suspended in a solution containing 10 mM potassium-phosphate buffer (pH 6.5) (buffer solution A) containing 1 mM phenylmethylsulfonyl fluoride (PMSF).

The cells were crushed using a French press to obtain a cell-free extract. Ammonium sulfate was added to 75% saturation and the resulting precipitate was removed to remove INV having no sugar chain and most of the contaminating proteins. The resulting supernatant was dialyzed against buffer A, and 10 g of DEAE-Sephadex A-50 carrier (Pharmacia) equilibrated with buffer A was added. After confirming that the INV activity of the supernatant was 10% or less of the whole, the supernatant was removed. The carrier on which INV was adsorbed was poured into a column (2.6 × 35 cm), washed with 2 times the volume of buffer A containing 0.1 M sodium chloride, and then the INV was eluted with the same buffer containing 0.2 M sodium chloride. did. The active fractions were collected, dialyzed against MiliQ water, desalted, and lyophilized. Dissolve this in a small amount of MiliQ water,
Gel filtration was performed using a column (2.0 × 95 cm) packed with Sephadex G-200SF (Pharmacia). 0.
Elution was performed with buffer A containing 2M sodium chloride, and active fractions were collected. INV (2 mg) purified in this manner was electrophoretically uniform.

(2) Cleavage of sugar chain and fluorescent labeling of sugar chain INV purified from the culture solution with a total volume of 48 l by the above method
From (about 8 mg), hydrazine was decomposed using Hydra Club S-204 (manufactured by Honen Corporation, Code No. 800600) to cut out sugar chains. After N-acetylation, pyridyl amination with 2-aminopyridine (PA
Fluorescent labeling (Akihiro Kondo et al., Biochemistry, Vol. 60, p6
89 (1988)). PA conversion is a sugar analysis kit (Takara Shuzo,
Code No. 5000). Toyopeal HW- equilibrated with 10 mM ammonium acetate buffer (pH 7.0)
Gel filtration was carried out using a 40F (manufactured by Tosoh Corporation) column (1.0 × 40 cm) to remove unreacted 2-aminopyridine to obtain PA-oligosaccharide.

(3) Purification of Neutral Sugar Chain INV sugar chains produced by Saccharomyces cerevisiae include phosphorylated ones. This was separated from neutral sugar chains using an anion exchange column. The columns are
Using COSMOGEL QA (8 × 75 mm) (manufactured by Nacalai Tesque), 10 mM ammonium acetate buffer (p
H8.0) was flown at a flow rate of 0.5 ml / min to equilibrate.
Elute with 10 mM ammonium acetate buffer for up to 20 minutes after sample injection, then increase buffer concentration to 45 mM, 1
It was raised linearly in 5 minutes. The unadsorbed sugar chains that had been eluted up to 17 minutes after the sample injection were separated as neutral sugar chains.

Example 2 OCH1 gene-disrupted strain (Δoc
Structure analysis of sugar chain of invertase produced by h1) (1) Separation and identification of sugar chain by high performance liquid chromatography (HPLC) (1) -1 Separation of oligosaccharide by amino column It is possible to separate sugars by their size (chain length). Therefore, using an amino column, INV produced by Δoch1
Sugar chains were separated according to size. Column is ASAHIPAK
NH2P-50 (4.6 × 250 mm) (manufactured by Asahi Kasei Corporation) was used, and the solvent was a mixture of 200 mM acetic acid-triethylamine buffer (pH 7.3) and acetonitrile at 35:36 (solvent A), 200 mM acetic acid- A 50:50 mixture of triethylamine buffer (pH 7.3) and acetonitrile (solvent B) was prepared.

The solvent A was flown through the column in advance at a flow rate of 1.0 m.
Immediately after the injection of the sample, the proportion of solvent B was linearly increased to 100% over 50 minutes immediately after the injection of the sample to elute the oligosaccharide. The results are shown in Fig. 3. Δoc
INV sugar chain produced by h1 has three main peaks (a
~ C) and two minor peaks (d, e) could be separated. PA-modified glucose oligomer preparation (Honen Corporation, Code No. 800109) and PA-modified high-mannose sugar chain preparation (Takara Shuzo, Code No. 4117, 4119, 4120)
From the elution position of, these peaks are Man _8-12 GlcNAc ₂ -PA
It was found that a sugar chain having a large outer sugar chain was not produced at all.

(1) -2 Separation of oligosaccharide by reverse phase column The oligosaccharide separated in (1) -1 was subjected to H using a reverse phase column.
Further separation by PLC. Column is COSMOSIL 5C18-P
(4.6 × 150 mm) (manufactured by Nakarai Tesque),
The solvent was 10 mM sodium phosphate buffer (pH 3.
8) (Solvent C) and Solvent D prepared by adding 0.1% butanol to Solvent C were prepared. Then, a column was preliminarily equilibrated by flowing a mixed solution of solvent C: solvent D = 95: 5 at a flow rate of 1.0 ml / min, and the ratio of solvent D was adjusted to 90% after 10 minutes from sample injection.
It was raised linearly to 100% over the course of a minute. The result is shown in FIG.

The peaks a to c in FIG. 3 each had a single peak even in the reversed phase column, and were considered to have uniform components. On the other hand, peak d was suggested to consist of two or more components. The sugar chain a is Man ₈ GlcNAc _2-.
The elution position was the same as the PA standard (Fig. 5, structure A). Peaks a, b and c in FIG. 3 were subjected to the following structural analysis as sugar chains a, b and c, respectively. Since the amounts of peaks d and e were very small, structural analysis could not be performed.

(2) FAB-MS measurement FAB-MS was carried out using a mass spectrometer JMS-HX110 (manufactured by JEOL Ltd.) with a 6: 4 mixture of nitrobenzyl alcohol and glycerin as a matrix. The sample sugar chain was desalted by HPLC using an amino column (ASAHIPAK NH2P-50). Solvent is acetonitrile: water =
50:50 was used. 1-2 μg was subjected to FAB-MS measurement.

Regarding the sugar chain b, m / z = 1962,
Regarding the sugar chain c, signals capable of being assigned to protonated molecular ions were observed at m / z = 2123. From the results of amino column HPLC, sugar chains b and c were respectively
Man ₉ GlcNAc ₂ -PA and Man ₁₀ GlcNAc ₂ -PA are estimated, but their predicted molecular weights are 1960.7 and 212, respectively.
2.8. The measured value and the theoretical value are in good agreement, and the size of the sugar chain estimated from amino column HPLC is considered to be correct.

(3) Structural analysis of sugar chain by treatment with α-1,2-mannosidase α-1,2-mannosidase (derived from Aspergillus saitoi, purified by the Institute of Medical Science, University of Tokyo, Professor Kawata's laboratory) 5 μl of 2 pmol / μl PA-oligosaccharide was added to 15 μl of 0.1 M sodium acetate (pH 5.0) containing 1 μunit of the above) and incubated at 37 ° C. for 18 hours.

After the reaction, the reaction was stopped by heating for 5 minutes in a boiling water bath, and the centrifuged supernatant was used for HPLC analysis. As a result of analyzing the size of the sugar chain by amino column HPLC, the sugar chains a and b were treated with the enzyme and then Man ₈ GlcNAc _2- , respectively.
PA to Man ₅ GlcNAc ₂ -PA, Man ₉ GlcNAc ₂ -PA to Ma
It moved to the position of n ₈ GlcNAc ₂ -PA. In addition, the sugar chain c did not change in size even after the enzyme treatment. α-1,2-
Man ₅ GlcNAc _2- newly generated by mannosidase treatment
As a result of analyzing each of PA and Man ₈ GlcNAc ₂ -PA by reverse phase column HPLC, both were considered to consist of one component, and Man ₅
GlcNAc ₂ -PA is a commercially available Man ₅ GlcNAc ₂ -PA standard (Takara Shuzo,
The elution position coincided with Code No. 4117) (Fig. 5, structure B).

Therefore, the sugar chain a has the structure A and α-
Α-1, at the non-reducing end by 1,2-mannosidase
It was strongly suggested that the 2-bonded mannose was sequentially cleaved to form structure B. Also, the sugar chain b is Man ₉ GlcNAc ₂ -PA.
From the Man ₈ GlcNAc ₂ -PA position to mannose at the non-reducing end mannose of α-1,3-branched mannose in structure A (* 1 mannose shown in structure A). It is considered that the molecule is added. Regarding sugar chain c, it is considered that one molecule of mannose was added to mannose (* 2) at the non-reducing end of mannose further branched by α-1,6.

[0049] (4) ¹ H-NMR sugar chain structure by the analysis ¹ H-NMR were performed using a nuclear magnetic resonance apparatus JNM-GX400 (manufactured by Nippon Denshi). The sample sugar chain is Sephadex G
15 (Pharmacia) column (1 x 50 cm) 10 mM
It was desalted by gel filtration using ammonium bicarbonate solvent. Dissolve the desalted sugar chains in heavy water and add 50-200 μg
^It was subjected to ¹ H-NMR measurement. 18 μg of commercially available Man ₈ GlcNAc ₂ -PA (Code No. 4119, manufactured by Takara Shuzo) (FIG. 5, structure A) was dissolved in heavy water and subjected to ¹ H-NMR measurement. The results are shown in FIGS. 6 (a) to 6 (d), which are spectra of the anomeric proton region which is important for the assignment of the binding mode of mannose. The spectrum of the commercially available Man ₈ GlcNAc ₂ -PA preparation (FIG. 5, structure A) (FIG. 6 (a)) was completely in agreement with the spectrum of sugar chain a (FIG. 6 (b)).

Therefore, it was confirmed that the sugar chain a was a structure A, that is, a core type sugar chain. The spectrum of sugar chain b (FIG. 6 (c)) shows that one molecule of mannose having α-1,3 bond is added to structure A. α
From the results of the -1,2-mannosidase treatment, it is considered that this mannose was bound to * 1 mannose.
Therefore, it could be determined that the sugar chain b had the structure C (FIG. 5). The spectrum of the sugar chain c (FIG. 6 (d)) shows the structure C
On the other hand, it can be seen that it has a structure in which one molecule of mannose having α-1,3 bond is added. From the result of the α-1,2-mannosidase treatment, it is considered that this mannose is bound to * 2 mannose. Therefore, the sugar chain c has the structure D
(Fig. 5).

Example 3 Preparation of Double Mutant Strain (Δoch1mnn1) Having Both Δoch1 Mutation and mnn1 Mutation In the yeast mnn1 mutation, addition of mannose α-1,3-linked to mannose to the core sugar chain of glycoprotein was added. The reaction to do is missing. Therefore, in order to obtain a strain that produces only uniform core type sugar chains, a mutant strain having both Δoch1 and mnn1 mutations was prepared. Since the molecular weight of INV produced by the mutant having the double mutation is expected to be smaller than the molecular weight of INV produced by Δoch1, the double mutant is
It was selected by examining the molecular weight of NV. The molecular weight of INV was SDS-PAGE (4-20% SDS polyacrylamide gradient gel, manufactured by Daiichi Pure Chemicals, Code No. 12047).
After 6), it was estimated by performing Western blotting with an anti-INV antibody. Anti-INV antibodies are commercially available I
NV (manufactured by Sigma) was treated with endo-β-N-acetylglucosaminidase H to remove most of the N-linked sugar chains, and this was prepared using rabbits as an antigen (Japan Biotest Institute). Creation commission).

Saccharomyces cerevisiae YS57-5C (MAT
α och1 :: LEU2 leu2 ura3 trp1 his1 his3) (Nakaya
ma et al., EMBO J., Vol. 11, p2511 (1992), and prepared by the present inventors) and S. cerevisiae LB1-10B.
(MATa mnn1) (University of California, Berkeley,
When obtained from the Yeast Genetic Stock Center), the spore formation rate was about 10%, but no och1 mutation-bearing strain was obtained from the grown spores. Therefore, strains YN1-28A and YN1-28B having a mnn1 mutation were selected from this grown strain, and it was decided to recross with Δoch1. mn
Confirmation of having the n1 mutation was confirmed by anti-α-1,3-mannan antibody (CEBallou, Methods Enzymol., Vol.185, p44.
6 (1990), University of California, Berkeley,
(Provided by Professor Schekman's laboratory).

YN1-28A (MATa mnn1) and YS57-
5A (MAT α och1 :: LEU2 leu2 ura3 his1 his3) was crossed and meiotically obtained, and the target double mutant strain YN3-1D (MAT α och1 :: LEU2 mnn1 his1 and / or his
3) was obtained (Fig. 7). In addition, YN1-28B and YS57
-2C (MATa och1 :: LEU2 leu2 ura3 trp1 his1 his3)
The double mutant strain YN4-18A (MAT α och
1 :: LEU2 mnn1) was obtained.

Example 4 Isolation and Purification of Mannan Protein Sugar Chain Produced by Δoch1mnn1 Double Mutant Strain (1) Isolation of Mannan Protein Produced by Δoch1mnn1 Double Mutant Strains of S. cerevisiae YN3-1D Was isolated from mannan protein (S. Peat et al., J. Chem. Soc.,
p29 (1961)). 50 ml of YEPD medium plus 0.3 M sorbitol was placed in a 500 ml Erlenmeyer flask, cultured at 25 ° C. for 24 hours, the cells were collected by centrifugation, and 10 ml of 0.1 M sodium succinate buffer (pH 7. 0) and suspended in an autoclave for 121
Heated at ℃ for 2 hours. After cooling, the solid was collected by centrifugation, 10 ml of water was added again, and the mixture was similarly heated.

All extracts were combined and poured into 3 volumes of ethanol. The resulting white precipitate was dried, dissolved in water and desalted. A column (1 × 10 cm) packed with agarose-immobilized ConA (Honen Corporation, Code No. 800259) was loaded with 0.2 M sodium chloride, 1 mM.
Equilibration was performed with 10 mM Tris-HCl buffer (pH 7.3) containing manganese chloride and 1 mM calcium chloride, and the desalted sample was subjected to this. Adsorbed mannan protein to 0.1M α
Elution with a 0.1 M sodium chloride solution containing -methyl-D-mannoside, desalting, and lyophilization were used as mannan protein (yield 5 mg).

(2) Extraction of sugar chain by treatment with glycopeptidase A 10 munit of almond-derived glycopeptidase A (Seikagaku Corporation, Code No. 100676) was dissolved in 500 μl of 0.1 M sodium acetate buffer (pH 5.0). Was prepared as an enzyme solution. 10 mg of 1 mg mannan protein
It was dissolved in μl of the same buffer, and 3 μl of the enzyme solution was added thereto. After incubation at 37 ° C. for 18 hours, the reaction was stopped by heating in a boiling water bath for 5 minutes, and the centrifuged supernatant was freeze-dried. The excised sugar chain was labeled with PA by the method shown in Example 2- (3).

Example 5 Structural Analysis of Mannan Protein Sugar Chain Produced by Δoch1mnn1 Double Mutant (1) Separation and Identification of Sugar Chain by High Performance Liquid Chromatography (HPLC) HPLC analysis was carried out in the same manner as in Example 2. (1) -1 Analysis of oligosaccharides by amino column The results of the amino column analysis are shown in FIG. Δoch
The sugar chain of the mannan protein produced by the 1mnn1 double mutant strain had mainly one peak on the amino column. This peak is a Man ₈ GlcNAc ₂ -PA standard (Takara Shuzo, Code No. 411
9) (FIG. 5, structure A).

(1) -2 Separation of oligosaccharides by reverse phase column The oligosaccharides detected in Fig. 8 were analyzed by HPLC using a reverse phase column. As a result, as shown in FIG. 9, a single peak was obtained even on the reversed phase column, which coincided with the elution position of the Man ₈ GlcNAc ₂ -PA standard (FIG. 5, structure A).

(2) Structural analysis of sugar chain by α-1,2-mannosidase treatment α-1,2-mannosidase treatment was carried out in the same manner as in Example 2. As a result of HPLC analysis using an amino column, it was moved from Man ₈ GlcNAc ₂ -PA to Man ₅ GlcNAc ₂ -PA by enzymatic treatment. In reversed-phase column HPLC, the elution position of the reaction product was labeled with Man ₅ GlcNAc ₂ -PA (Takara Shuzo, Code No. 4117).
(FIG. 5, structure B). Therefore, Δoch1
The asparagine-linked sugar chain of the mannan protein produced by the double mutant of mnn1 S. cerevisiae YN3-1D is mainly the structure of the core sugar chain of Man ₈ GlcNAc ₂ and is α-specific to the sugar chain produced by yeast. It was found that there was no mannose at the end bound by 1,3.

(3) Sugar chain structure analysis by ¹ H-NMR In the same manner as in Example 2 (4), Δoch1 by ¹ H-NMR.
The structure of the sugar chain of the manna protein produced by the mnn1 double mutant strain was analyzed. The spectrum of the commercially available Man ₈ ClcNAc ₂ -PA (FIG. 5, structure A) and the spectrum of the Asn-linked sugar chain of the mannan protein produced by the Δoch1mnn1 double mutant strain were completely in agreement. Therefore, it was confirmed that the sugar chain is a core type sugar chain.

[Brief description of drawings]

FIG. 1 is a biosynthetic pathway of an outer sugar chain in yeast.

FIG. 2 Structure and m of sugar chain of mannan protein of baker's yeast
It is a deletion site in the nn mutant strain.

FIG. 3 Separation of sugar chains of invertase produced by Δoch1 by amino column (arrow indicates elution position of glucose oligomer).

FIG. 4 Separation of sugar chains of invertase produced by Δoch1 by reversed-phase column (ad: peak a in FIG. 3)
Corresponds to each of ~ d).

FIG. 5 shows structures of sugar chains of invertase produced by Δoch1 and standard sugar chains.

FIG. 6 is a ¹ H-NMR spectrum of a sugar chain of invertase produced by Δoch1 and a sugar chain of a standard product.

FIG. 7 is an SDS-PAGE pattern of invertase produced by a Saccharomyces cerevisiae mutant strain (western blotting using an anti-invertase antibody).

FIG. 8: Separation of mannan protein sugar chains produced by Δoch1mnn1 by an amination column (arrows indicate Man ₈ GlcNAc ₂ −
PA elution position).

FIG. 9: Separation of sugar chain of mannan protein produced by Δoch1mnn1 by reverse phase HPLC (arrow indicates Man8GlcNAc2-
Elution position of PA standard (Takara Shuzo Code No. 4119).

Continuation of front page (51) Int.Cl. ⁵ Identification number Office reference number FI technical display location C12R 1: 865) (C12P 21/00 C12R 1: 865) (72) Inventor Yoko Nakanishi 1st East, Tsukuba City, Ibaraki Prefecture No. 1 Industrial Technology Institute, Institute of Chemical Research (72) Inventor Kenichi Nakayama 1-1, Higashi Tsukuba, Ibaraki Prefecture Industrial Technology Institute, Institute of Chemical Technology (72) Inventor Atsushi Tanaka 1 Asahi Kasei Industrial Co., Ltd., Samejima, Fuji City, Shizuoka Prefecture In the company

Claims (6)

[Claims] 1. A yeast of the genus Saccharomyces, which is an oligosaccharide chain consisting of 8 molecules of mannose and 2 molecules of N-acetylglucosamine represented by the formula 1 in which the structure of an asparagine-binding glycoprotein sugar chain produced by a yeast cell is shown. [Chemical 1] (In the formula, M represents mannose and GN represents N-acetylglucosamine) 2. The yeast of the genus Saccharomyces according to claim 1.
A yeast mutant strain that belongs to Saccharomyces cerevisiae and that has both och1 mutation (Δoch1) and mnn1 mutation, which are mutants of OCH1 gene, and belong to Saccharomyces cerevisiae. 3. The Saccharomyces cerevisi according to claim 2.
ae is YN3-1D strain or YN4-18A strain Saccha
A yeast mutant belonging to romyces cerevisiae. 4. The yeast according to any one of claims 1 to 3 is cultivated in a medium, yeast cells are recovered from the culture, and the oligosaccharide chain represented by Chemical formula 2 is recovered from the cells. A method for producing an oligosaccharide chain of Chemical formula 2, which is characterized by recovering [Chemical 2] (In the formula, M represents mannose and GN represents N-acetylglucosamine) 5. A yeast derived from a host yeast containing the oligosaccharide chain represented by Chemical formula 3 as an asparagine-linked sugar chain, which is obtained by culturing the yeast according to any one of claims 1 to 3. A method for producing a yeast-derived glycoprotein containing the oligosaccharide of Chemical formula 3, which comprises recovering the glycoprotein. [Chemical 3] (In the formula, M represents mannose and GN represents N-acetylglucosamine) 6. Using the yeast according to any one of claims 1 to 3 as a host, a gene encoding a mammalian-derived asparagine-binding glycoprotein is expressed, and the expressed glycoprotein asparagine-binding sugar is expressed. A method for producing a mammalian-derived glycoprotein containing an oligosaccharide according to Chemical formula 4, wherein the chain structure is an oligosaccharide chain represented by Chemical formula 4. [Chemical 4] (In the formula, M represents mannose and GN represents N-acetylglucosamine)