JP2003334075A - Method for readily producing variant antibody - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for efficiently obtaining a variant antibody having improved integrity with a material having antigenicity. <P>SOLUTION: The method for obtaining the variant antibody involves using a variant Fv library which is a domain included in one CDR of an antibody or present at the vicinity thereof, designed so that k (k is an integer satisfying the relation of 2≤k≤20) kinds of amino acid mutation may be introduced to m (m is an integer satisfying the relation of 2≤m≤n-1) residues of amino acids selected from n (n is an integer satisfying the relation of 3≤n) residues of amino acids, and provides<SB>n</SB>C<SB>m</SB>×k<SP>m</SP>kinds of limited Fv repertories as amino acid sequences. Preferably, the variant Fv library in which k is 20, which is designed so that NNK (N is A, G, C or T; and K is G or T) mutation may be introduced as a nucleotide residue corresponding to the amino acid residue to which the mutation is introduced, and which provides the limited Fv gene repertories of<SB>n</SB>C<SB>m</SB>×32<SP>m</SP>kinds as the base sequences, is used. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a method for modifying an antibody. More specifically, it relates to a method for efficiently obtaining an antibody having improved binding to an antigen. INDUSTRIAL APPLICABILITY The present invention provides a useful strategy when modifying an antibody useful as a diagnostic agent, a measuring reagent, a therapeutic agent, etc. Useful in modifying catalytic antibodies.

[0002]

2. Description of the Related Art At present, antibody proteins as biofunctional materials are widely applied as diagnostic agents, measuring reagents, therapeutic agents, etc. due to their antigen-specific binding properties. Practical antibody is a lead (molecular species) that binds more strongly and specifically to the target antigen
Is required.

As an example, in the case of a catalytic antibody, the transition state theory (the enzyme binds to the transition state of the substrate and reduces the activation energy to promote the chemical reaction) is used to mimic the transition state of the substrate. The hapten thus prepared is obtained by immunizing a mammal such as a mouse. A problem of the catalytic antibody in practical use is its low catalytic activity.
In order to increase the catalytic activity, it is considered effective to obtain an antibody that binds more strongly to the transition-state-like hapten.

[0004] In general, an antibody has three CDRs (complementarity determining regi
on; CDR1, CDR2, CDR3; antigen complementarity determining regions) exist, and their specificity determines specific binding to an antigen. That is, the ability of an antibody to specifically bind to various antigens is due to the diversity of amino acid sequences constituting these CDRs.

Therefore, in order to obtain an antibody having enhanced binding activity, it is effective to carry out mutation on the amino acids constituting CDR. At least one amino acid in any of CDR1, CDR2 and CDR3 can be mutated to various other amino acids. Mutation may be carried out on either the L chain or the H chain, or both. Then, a scFv (single chain
Fv) A library can be prepared, displayed on a phage (phage display method), and biopanned with respect to a hapten to select a clone having a stronger binding property. Particularly important for antigen binding of the antibody of interest
If the CDR is known in advance, the mutation may be performed on the amino acid of the CDR.

[0006]

Since each CDR has a maximum of only about 19 amino acids, it is theoretically possible to introduce 20 kinds of amino acid mutations into all of the constituent amino acids. What is important here is what kind of mutation can be arbitrarily introduced into a library to efficiently obtain a clone with improved binding.

We have developed a V122 antibody that decomposes several prodrugs with vitamin B6 as a common protecting group (Japanese Patent Application No. 11-305
The antibody described in Example 3 of 864 (JP 2000-198800).
In the present specification, it may be referred to as “prodrug-degrading antibody V122”. ) Was developed. This antibody was obtained by immunizing mice with the transition-state-like hapten shown in FIG. Taking this antibody as an example, the CDR1 of the H chain of the V122 antibody is
There are 5 amino acids in. Assuming that 20 kinds of amino acid mutations are introduced at all 5 sites, 20 ⁵ (=
There will be 3.2 × 10 ⁶ ) types of amino acid repertoire. At the nucleotide sequence level, NNK mutations (N = A, G,
C, T; K = G, T) is introduced, 32 ⁵ (= 3.4 × 10 ⁷ ).
It will include a variety of base sequence repertoires.

In a library consisting of a large number of variants,
The more diverse the number of variants, the more likely it is that clones with optimal strong binding will appear, but the number of variants showing the same degree of binding as the parent strain also increases.
When biopanning is performed using this library, a small number of clones with strong binding compete with a large number of clones with medium binding. In biopanning, it is extremely difficult to set up selection conditions that strictly sort the differences in binding properties.Therefore, when this library was used for selection, a large number of clones with moderate binding properties were unspecified. There is a high probability that they will be selected, and the clones that have truly strong binding may be eliminated. Furthermore, due to the large number of original repertoires, it is highly likely that a large number of non-reproducible amino acid sequence patterns will be found among the selected clones, and among them, clones with stronger binding properties should be identified. Becomes difficult. Therefore, it can be said that this strategy results in low reproducibility that relies on randomness, making it difficult to perform experiments aiming at reasonably improved binding.

[0009]

[Means for Solving the Problems] Therefore, as an alternative method, we constructed a library in which the number of repertoires was suppressed, and examined the amino acid mutation pattern of each clone selected from among them, which was important for binding. We devised a strategy to determine the dominant mutant by probabilistically determining the amino acid of the dominant mutation that seems to be.

In detail, this time, as a library in which the number of repertoires was suppressed, we used 5 amino acids of V122 H chain CDR1 with 2 amino acid sites at 2 amino acid sites each.
By introducing 0 kinds of amino acid mutations, a 2-amino acid mutation library covering all combinations was prepared (Fig.
3). This library has a repertoire of 4000 (= 20 × 20 × 10) types of combinations, and at the nucleotide sequence level, it contains 10240 (= 32 × 32 × 10) types of nucleotide sequences. It was calculated that the proportion of clones having the exact same sequence as the original strain was suppressed to about 0.25% (1/20 x 1/20).

As a method for determining a dominant mutation, the concept of adaptive walking was used in which the evolutionary process of a protein is compared to mountain climbing in a protein sequence space of a pedestrian. Generally, the fitness landscape of a protein is considered to be a single-peak mountain of Mt. Fuji type, and it is considered that the substitution effect of amino acid residues has statistical additivity. The simplest walking method is to repeat the mutation and selection process for each amino acid, climb a mountain, and find the optimal sequence.
We consider this additivity concept more extensively, and
The effect of substituting 5 amino acids of R was evaluated at the same time, and a strategy for determining the optimal sequence was adopted. More specifically, a large number of clones selected by biopanning were selected, the amino acid sequence of the CDR1 region was examined for each clone, and the amino acid appearance rate was calculated for 5 amino acid sites. In this library, at the site where the mutation was introduced,
The appearance frequency of the original amino acid is about 60% (6/10), and the appearance frequency of the replaced amino acid is about 2% (4/4).
It was calculated as 10 × 1/20). Then, by probabilistically comparing the appearance frequency of the original amino acid and the appearance frequency of the newly selected amino acid in the clones after selection at each site, the dominant amino acid at each site can be estimated. It was Then, by applying this estimation to all 5 amino acid sites, the combination pattern of optimal dominant amino acids in this CDR site was estimated at one time. In the present specification, the term “dominant” with respect to an amino acid or an amino acid sequence refers to a case where a substitution with the amino acid gives a more advantageous effect than that having the original amino acid, except for a special case. For example, when a plurality of clones that bind more strongly to the hapten are selected by panning, the amino acid sequences of the clones after panning are compared to calculate the frequency of appearance of each amino acid sequence at a specific position, and the probability is predicted. An amino acid sequence that appears more frequently than the above value is judged to be dominant at that position.
In this case, when the frequency of occurrence of the original amino acid decreased at that particular position, but another amino acid occurred at a frequency higher than the stochastically expected value, the other amino acid sequence It is judged to be a preferred dominant amino acid.

As a result of intensive studies in this way, the present invention has been completed. That is, the present invention provides a method for designing a mutant library in which the number of repertoires is suppressed, a library produced by such a design method, and a method for determining a dominant mutant associated with the library. The present invention also provides a method for designing a mutant antibody, a method for manufacturing a mutant antibody using the method for designing, and a mutant antibody obtained by the method for designing or manufacturing the method. The details are as follows.

The present invention is a region contained in or near the CDR of one antibody, which is selected from a region consisting of n (n is an integer of 3 ≦ n) amino acid residues. m (m
Is k (k is an integer of 2 ≦ m ≦ n−1) amino acid residues.
Provides a mutant Fv library that is designed to introduce 2 ≦ k ≦ 20 amino acid mutations and can provide a limited Fv repertoire of _n C _m × k ^m amino acid sequences. . For example, assuming that 20 kinds of amino acid mutations are introduced at any two positions in a region consisting of 5 amino acid residues, theoretically, ₅ C ₂ × 20 ² = _{5 in the} library of the present invention.
P _2/2! × a ^{20 2 = (5 × 4)} / (2 × 1) to × 20 × 20 = 4000 different repertoire of amino acid combinations are present. The term "given" in the present invention means "theoretical". In the present specification, when referring to “mutation” in relation to an amino acid sequence or a base sequence, it means that the amino acid residue or nucleotide residue is changed by deletion, substitution, addition or insertion. Means that the amino acid residue or nucleotide residue is the same as one amino acid residue or nucleotide residue (when it is another amino acid residue or nucleotide residue, It may be a thing.).

Designing so that 20 kinds of amino acid mutations are introduced into a selected amino acid (that is, k = 20) is one of the preferred embodiments of the present invention. In this case, the nucleotide residue corresponding to the amino acid residue to be mutated is NNK (N = A, G, C, T; K = G,
T) It is recommended that the mutation be introduced. In this case, the obtained library has a nucleotide sequence of _n C _m × 32 ^m.
It is possible to provide a limited type of Fv gene repertoire.

The term Fv as used herein refers to a variable region (fragment) of an antibody unless otherwise specified. Mutant Fv
The gene or Fv gene repertoire refers to a mutated version of the Fv-encoding gene (Fv gene), which includes a VL (L chain variable region) -encoding polynucleotide or a variant thereof and VH (H chain variable region). It includes a polynucleotide linked to the encoding polynucleotide or a variant thereof via a linker. Mutant Fv refers to a mutant of Fv, and includes a protein or polypeptide obtained by expression of a mutant Fv gene or Fv gene repertoire.

The library of the present invention having a reduced number of repertoires does not have much change in the amino acid sequence as compared with the original strain (before mutation), so that a clone having a significantly increased binding property as compared with the original strain can be obtained. It can be difficult. However, unlike a library with a large number of repertoires, clones having the same sequence have a high degree of overlap, so there is a high possibility that clones with good reproducibility and binding can be selected. Furthermore, since the amino acid patterns of the selected clones are reduced, it is possible to easily examine whether the patterns tend to be recognized or not by comparing the sequences between the clones.

Since the library of the present invention can provide a repertoire in which a plurality of amino acid residues are substituted, it is possible to carry out studies in consideration of the interaction of a plurality of amino acid residues. Compared with a library that can give a repertoire in which only one amino acid residue is substituted, there is also an advantage that cases in which the interaction positively acts can be selected without missing.

The method for preparing the library of the present invention includes PC
R can be used. The PCR-based mutation introduction methods generally used at present are roughly classified into two methods. One is a method of amplifying a fragment of DNA and introducing a mutation. This is because when a part of the vector is amplified by PCR, P
It is created by introducing a mutation into the CR product. In this case, the amplification product is finally cut with a restriction enzyme and integrated into the vector. When there is no suitable restriction enzyme site, a restriction enzyme recognition sequence is newly introduced into the vector. The second is a mutagenesis method that amplifies the entire vector. IP for this
A method of introducing a mutation by amplifying the entire vector using a CR (inverse PCR) method or the like is known. The IPCR method is a method for amplifying the entire vector as a result by designing primers on a circular plasmid side by side in a direction opposite to the normal direction and performing PCR (FIG. 2).
In this case, a circular vector can be prepared again by performing self-ligation of the amplified PCR product. Any of them can be used to prepare the library of the present invention.

The latter method is simple in that it is not necessary to newly introduce a restriction enzyme recognition site into the vector, but since the circular plasmid is used as a template, the amplification efficiency of PCR is generally poor, and the subsequent Due to the problem of self-ligation efficiency, it is used to obtain a single clone with site-directed mutations introduced, but it is difficult to use it to prepare a mutagenesis library that requires a large number of clones. is there.

However, the IPCR method is suitable for producing a library of the present invention with a reduced number of repertoires. In particular, several (preferably 3 to 10, particularly preferably 3 to 7)
So that an amino acid mutation is introduced into several (preferably 2 to 6, more preferably 2) amino acid residues selected from the region consisting of NNK mutation is introduced as a nucleotide residue corresponding to the amino acid residue to be introduced)
The IPCR method is suitable for preparing the library of the present invention that can be designed to give a more limited repertoire. The simple and rapid IPCR-mutagenesis method, which utilizes the IPCR method in the preparation of the library introduced with the NNK mutation as shown in FIG. 3, which was developed by us this time is also included in the scope of the present invention.

The library of the present invention is preferably a phage display library because it can simultaneously provide a mutant gene and a protein expressed by the mutant gene (mutant Fv). A conventional method known to those skilled in the art can be used to prepare the phage display library.

The library of the present invention can be used for screening mutant Fv, which includes a step of selecting using a substance having antigenicity. Such a screening method is useful for producing a mutant antibody and also for modifying the function of the antibody.

The present invention also relates to (1) a phage display mutant Fv library: a region contained in or near the CDR of one antibody, wherein n (n is
(The integer of 3 ≦ n)) A region of a plurality of clones selected from those capable of giving a mutant Fv repertoire having an amino acid mutation in a region consisting of amino acid residues, and selected by binding with a substance having antigenicity. The amino acid sequence of
(2) For each position 1 to n of the region, select one amino acid sequence according to the number of occurrences; and (3)
Also provided is a method for producing a mutant antibody having a region consisting of the optimum mutant amino acid sequence, which can be obtained by using the obtained amino acid sequence pattern as the optimum mutant amino acid sequence. "Selecting one amino acid sequence according to the number of appearances" means selecting one amino acid sequence based on the number of appearances or the number calculated by the number of appearances (for example, appearance rate, appearance frequency). In the case of carrying out the present invention for the purpose of obtaining a variant having a high binding property to a substance having antigenicity,
Choosing the amino acid with the highest number of occurrences, selecting the amino acid with the highest number of occurrences, but larger than the pre-mutation amino acid, or Refers to selecting amino acids that are not.

Designed so that k (k is an integer of 2 ≦ k ≦ 20) kinds of amino acid mutations are introduced into m (m is an integer of 2 ≦ m ≦ n−1) amino acid residues. As an amino acid sequence
_A method for producing a mutant antibody capable of giving a limited mutant Fv repertoire of _n C _m × k ^m types is one of the preferred embodiments of the present invention.

The method of the present invention for estimating an amino acid sequence pattern obtained on the basis of the number of appearances as an optimal mutant amino acid sequence makes it possible to efficiently advance adaptive walking for each position in a region where mutation is carried out. .

The present invention further provides (1) a phage display mutant Fv library: a region contained in or near the CDR of one antibody, wherein n (n is
(The integer of 3 ≦ n)) A region of a plurality of clones selected from those capable of giving a mutant Fv repertoire having an amino acid mutation in a region consisting of amino acid residues, and selected by binding with a substance having antigenicity. The amino acid sequence of
(2) For each position 1 to n of the region, select one amino acid sequence according to the number of occurrences; and (3)
The obtained amino acid sequence pattern is used as an optimum mutant amino acid sequence, and a mutant antibody or a part thereof containing a region consisting of the optimum mutant amino acid sequence that can be obtained is also provided. Phage display mutant Fv library
k for m (m is an integer of 2 ≦ m ≦ n−1) amino acid residues
(K is an integer of 2 ≦ k ≦ 20) designed to introduce amino acid mutations, which may give a limited mutant Fv repertoire of _n C _m × k ^m as an amino acid sequence. The above mutant antibody or a part thereof is one of the preferred embodiments of the present invention. Also, phage display mutation
The Fv library has 8-20 for 2-9 amino acid residues selected from a region of 3-19 amino acid residues.
Such mutant antibodies or parts thereof, which are designed to carry out a class of amino acid mutations and which may contain a limited mutant Fv repertoire, are one of the preferred embodiments of the invention.

The mutant antibody of the present invention does not necessarily have to consist of a complete antibody molecule. The general structure of the active center of an antibody (site involved in binding with a substance having antigenicity) has been sufficiently elucidated, and it exists in a portion called Fab of a pair of L and H chains formed. And Fab
The fragment itself has the same antigen recognition ability as an intact antibody. Therefore, the mutant antibody may be a portion having substantially antigen binding ability, for example, a Fab fragment. In addition, Fv, Fab '(fragment slightly longer than Fab), or F (ab') ₂ (two Fab's bonded by S-S bond) may be used. “F (ab ′) ₂ ” and “Fab ′” are produced by treating immunoglobulins (monoclonal antibodies) with proteolytic enzymes such as pepsin or papain, and between the two H chains in the hinge region. By an antibody fragment produced by digestion before and after existing disulfide bonds. For example, when IgG is treated with papain, it is cleaved upstream of the disulfide bond existing between the two H chains in the hinge region, resulting in VL (L chain variable region) and CL (L chain variable region).
L chain consisting of the chain constant region), and VH (H chain variable region) and CH
It is possible to produce two homologous antibody fragments in which an H chain fragment consisting of γ1 (γ1 region in the H chain constant region) is linked by a disulfide bond at the C-terminal region.
Each of these two homologous antibody fragments is called Fab '. When IgG is treated with pepsin, it is cleaved downstream of the disulfide bond existing between the two H chains in the hinge region to produce an antibody fragment slightly larger than that in which the two Fab's are linked at the hinge region. be able to. This antibody fragment is called F (ab ') ₂ . Further, a part of the antibody of the present invention includes scFv in which a VH chain and a VL chain are linked by a linker, and those in which only the V region is expressed by genetic engineering. It is within the scope of the present invention to combine such common general knowledge with the disclosure of the present specification to produce various mutant antibody fragments.

The antibody of the present invention or a part thereof preferably has an ability to catalyze a reaction; and a substance having an antigenicity is a hapten bound to a carrier protein, and the hapten is a transition state analogue of the reaction. Is like.
Antibodies capable of catalyzing a reaction, that is, catalytic antibodies include antibodies having hydrolysis activity such as esterase activity, as well as antibodies having transfer activity or antibodies having cyclization activity. Further, as an application example of the catalytic antibody,
Degradation of cocaine (Landry, DW et al.,
(1993) Science, 259, 1899-1901) and anticancer drugs (Campbell, DA et al., (1994) J. Am. Ch.
em. Soc., 116, 2165-2166, Wentworth, P. et al., (1
996) Proc. Natl. Acad. Sci. USA, 93, 799-803) and the like.

The antibody of the present invention or a part thereof is more preferably one having the ability to catalyze an ester hydrolysis reaction, and most preferably one having the ability to degrade a prodrug, for example, a prodrug-degrading antibody. It is V122.

The method of the present invention (in particular, NNK mutation and IPCR
In the IPCR-mutagenesis method using and, it is possible to introduce a mutation at any position. Therefore, a clone having a mutation having the dominant amino acid found this time is used as a lead sequence, and the method of the present invention (for example, IPCR-m
utagenesis method), and the following library can be prepared. A new library may be one in which a mutation is introduced into another CDR region (CDR-jumping), or one in which a mutation is introduced near the CDR (CDR-walking). Techniques relating to the production of such a library, and mutant antibodies having a further increased activity of interest obtained from such a library or antibodies having a modified function or a part thereof are also included in the scope of the present invention. . The technique disclosed in the present specification can be applied not only to antibodies but also to evaluation of mutations of various proteins and functional modification.

[0031]

[Example] 1. Materials and Methods (1) Amplification of Prodrug Degrading Antibody V122 Gene RNA was extracted from the hybridoma producing the prodrug degrading antibody V122 using ISOGEN reagent (Nippon Gene). Reverse transcription of 5 μg of RNA was performed using First-Strand cDNA Synthesi.
cDNA using s Kit (Amersham Pharmacia)
Got In order to amplify the V region of V122 H and L chains, Pfu turbo pol was used with half the amount of the obtained cDNA as a template.
PCR was performed using ymerase (Stratagene). Heavy Primer Mix and Light Primer Mix (Amersham Pharmacia) were used as primers. PCR reaction is 9
3O cycles were performed under the conditions of 4 ° C for 1 min, 55 ° C for 1 min, and 72 ° C for 1 min. After confirming the amplification of the target band by agarose electrophoresis, the obtained PCR product was used for the GENECLEAN II Kit (BIO 101, I
NC.) Was used to separate and collect.

(2) V122 gene scFv (single chain
Recombinant Phage Antibody System
Performed according to the RPAS mouse ScFv Module (Amersham Pharmacia). PC of V122 H and L chains separated and collected at the beginning
Assembly and fill to bind R products via linker
-in reaction was performed. Subsequently, a 2nd PCR reaction was performed using RS Primer Mix in order to add the SfiI site of the restriction enzyme to the 5'end of the scFv DNA and the NotI site to the 3'end. PCR
After confirming the amplification by agarose gel electrophoresis, the band of interest was separated and collected using GENECLEAN II Kit. Collected
The PCR product is cleaved with SfiI and NotI, subjected to agarose electrophoresis, separated and collected in the same manner, and cleaved with SfiI and NotI.
A ligation reaction was performed with the vector pCANTAB 5E for scFv display.

Subsequently, biopanning by phage display was performed to obtain scFv-expressing clones that bind to haptens. First, BSA for panning (bovine se
rumalbumin) -Imject SuperCarrier EDC S with BSA-hapten (Figure 1) to make hapten-bound plate
It was prepared using ystem for Peptides (PIERCE). next
A panning plate was prepared by reacting a BSA-hapten solution (2 μg / ml) dissolved in PBS (phosphate buffered saline) with an EIA / RIA 8 well strip plate (CORNING COSTAR) at room temperature for 1 hr. Next, in order to generate a recombinant phage displaying V122scFv, pCANTA
The DNA product that had undergone ligation reaction with B 5E was transformed into E. coli TG1, and the helper phage VCSM was further transformed.
13 were infected to produce recombinant phage. The recombinant phage was purified by PEG (polyethyleneglycol) precipitation and then dissolved in 160 μl of SM (Suspension medium) buffer. Furthermore, after mixing with 150 μl of Block Ace solution (Dainippon Pharmaceutical Co., Ltd.) diluted 4-fold with sterilized water, the mixture was reacted at room temperature for 1 hr on a panning plate. After washing with PBS 20 times and further with PBS + 0.1% Tween20 (PBST) 20 times, TG1 solution was added and incubated for 1 hr at room temperature, and the hapten-bound recombinant phage was infected with TG1 to recover. In order to further amplify the recombinant phage, the incubated TG1 solution was infected with VCSM13, and after overnight culture, the culture supernatant was centrifuged and collected, and PEG precipitation was performed. The purified recombinant phage was collected by repeating the panning operation once again.

Escherichia coli HB2151 for expressing scFv protein was infected with this recovered recombinant phage, and Amp (ampicillin, 100
μg / ml) -containing LB plate (LBA) was seeded to form colonies. A colony was picked up and cultured in 96 deep well to express the scFv protein in the culture supernatant. The supernatant was subjected to ELISA in a BSA-hapten-binding 96-well plate to select clones expressing scFv capable of binding to hapten. The scFv-expressing phagemid was recovered from the selected clones, and BigDye terminator cycle sequencing ready reac
PCR was performed using a rotation kit (PE Applied Biosystems), and the sequence of V122scFV was determined using a fluorescence sequencer (ABI 377) (FIG. 4 and SEQ ID NO: 1).

(3) Template for preparing scFv display library Preparation of phagemid A vector serving as a template for preparing the library was prepared. In order to prevent expression of wild-type V122scFv due to mixing of template-derived phagemid during library construction, the nucleotide sequence of Tyr in the CDR1 region of the H122 chain of V122scFv
TAT Quick Change Site-Directed Mutagenesis Kit (S
TRATAGENE) was used to change to TAA and a stop codon was introduced.
Also, a His-tag sequence was introduced at the C-terminus of 122 scFv for protein purification. The introduction of these sequences was confirmed by performing a sequence.

(4) Construction of mutation-introduced scFv library by IPCR-mutagenesis method Two NNKs (N = A, G, C, T ,;
K = G, T) introduced 2-amino acid mutation library (Fig. 3)
IPCR was used to generate Equal amounts of 10 types of each of the IR and INF primers shown in Table 1 were mixed I
PCR was performed using a combination of NF primer mixtures. The PCR reaction solution was approximately 5 ng of template V122scFv plasmid, 1 × ExTaq bu
ffer, 150 μM each dNTP, each primer 25 pmol, ExTaq 2.5U
(TaKaRa) was included, and PCR was performed in a total amount of 50 ul. The reaction cycle is 94 ℃ for 1min, 94 ℃ for 1min, 60 ℃ for 1min, 72 ℃ for 7min.
Was repeated 25 times, and finally at 72 ° C. for 10 min.
After PCR, 2 μl of the reaction product was electrophoresed to confirm the amplification of the PCR product. DpnI 20U, T4 DNA poly for the remaining PCR products
merase 2U (TaKaRa), dNTP Mixture 2.5mM each (TaKa
Ra) was added in an amount of 0.5 μl and reacted at 37 ° C. for 1 hr. The reaction product is M
After removing the protein with an icropure-EZ column (Millipore), it separated by electrophoresis and the target band was collect | recovered using RECOCHIP (TaKaRa). After passing the sample through the EZ column again, ethanol precipitation was performed. The DNA of 5 PCR tubes (50 μl × 5) was used for the preparation of the library.

Five samples after ethanol precipitation were combined into one, and phosphorylation was performed for self ligation. Dissolve the ethanol-precipitated sample in sterilized water and add 2.5 μl of 10 × ligation buffer (Nippon Gene).
l, T4 polynucleotide kinase1U (TakaRa) added
Phosphorylation was performed at 37 ° C for 1 hr in a 25 µl system. Furthermore, 10 × li
2.5 μl of gation buffer, 1.5 μ (750 U) of T4 DNA ligase (Nippon Gene), and 21 μl of sterilized water were added to the phosphorylated product solution, and the mixture was allowed to react at 16 ° C. for about 20 hours or more for self ligation. The ligation product was ethanol-precipitated, dissolved in 10 ul of sterilized water and electroporated into E. coli (GENE P
ULSER II, BIO-RAD).
As a transformant competent cell, 40 μl of Escherichia coli TG1 strain was used. E. coli after electroporation is partially diluted and LBA (LB + ampicillin 100ug / ml medium)
The plate was rolled and the size of the library was calculated. The remaining E. coli was suspended in a medium containing 15% glycerol and stored at -80 ° C.

(5) Preparation of scFv-displaying phage A stock of 1 ml of the prepared library was used for 9 ml of CG (CIRC
LEGROW, BIO101 Systems) + Amp (ampicillin sodium, Wako, 100ug / ml CG) medium was added, and the mixture was cultured at 37 ° C for 1 hr. 4 × 10 ¹⁰ CFU VCSM as helper phage in culture
13 was added, and the mixture was further incubated at 37 ° C for 1 hr. After discarding the supernatant by centrifugation, 10 ml of CG + Amp + Km (Kanamysis sulfate, Wako, 2
5 μg / ml CG) and the cells were cultured at 37 ° C. overnight. The culture solution was centrifuged, the supernatant was collected, and after removing E. coli through a 0.2 μm filter, PEG precipitation was performed to collect phage particles, and 300 μl of SM buffer (Suspension
medium). E. coli XL1-bl was prepared by making a dilution series of the phage suspension to measure phage titer.
ue, and seeded on LB-Amp plates. Using the number of Amp resistant colonies as an index, CFU (colony forming un
ite) was measured.

(6) Separation and mutation of BSA-hapten-binding phage by biopanning scFv-displaying phage solution 500 μl (5 × 10 ¹¹ CFU) 60
μl Block Ace was added, and the mixture was left at room temperature for 15 minutes. Next, this phage solution was added to 4 wells of panning wells bound with BSA-hapten, and the mixture was allowed to stand at room temperature for 2 hours to bind phage particles to BSA-hapten. Then PB
Wells were washed 20 times with S (-) and 20 times with PBST to wash away non-specifically bound phages. 50 per well
Add ul elution buffer (0.1M HCl, adjusted to pH 2.2 with glycine + BSA 1mg / ml buffer), leave at room temperature for 10 minutes, and add BS
The phage bound to the A-hapten was eluted. After neutralizing by adding 3 μl of 2 M Tris per well, the solution for 4 wells was added to 600 μl of E. coli TG1 solution and left at room temperature for 15 minutes,
The eluted phage was infected with TG1. Next, 4 × 10 ⁹ helper phage VCSM13 was added to this bacterial solution to obtain 100 μg / ml.
Amp, adjusted to 2% glucose, 37 ° C for 1 hr
Cultured. Then, centrifuge and discard the supernatant, then add 10 ml of CG + Amp + K.
The pellet was suspended in m and cultured overnight at 37 ° C to produce recombinant phage. The culture solution was centrifuged, the supernatant was collected, and passed through a 0.22 μm filter to remove E. coli. Furthermore, PEG precipitation is performed to recover the phage, and SM buffer 600μ
It was suspended in 1 l and stored at 4 ° C. 32 of the collected phage fluid
The 0 μl portion was mixed with a 4-fold diluted Block Ace solution, and then biopanned again in the same manner to use for isolating a clone having further binding property. The biopanning operation was performed four times in total.

(7) Identification of BSA-hapten-binding clones by ELISA Phages selected by biopanning were cloned and binding to haptens was confirmed. Escherichia coli XL1-BLue was infected with this selected phage and plated on an LB-Amp plate to form a colony. Several colonies were seeded on a 96-well deep plate (Whatman) containing 800 μl of GS96 (BIO101.Inc.) + Amp medium per well, and cultured at 37 ° C. with shaking. About 8 hours after the start of culture, IPTG was added so that the concentration was 1 mM, and the cells were further cultured at 30 ° C. overnight to induce the expression of scFv.
The colonies seeded on the 96 well deep plate were also seeded on the LBA plate at the same time for phagemid separation to form colonies. 96 well deep plate for ELISA reaction
Escherichia coli cultivated in 1.
Was added to a BSA-hapten-bonded 96-well plate and left at room temperature for 1 hr. The plate is then washed 5 times with PBST and then HRP
/ Anti-Etag Conjugate (Amersham Pharmacia)
Add 50ul per well of the solution diluted 8000 times with PBS,
It was left at room temperature for 30 minutes. The plate was then washed 5 times with PBST and then ABTS reagent (Kirkegarrd & Perry Laboratorie
s. Inc.) was added to each well in an amount of 100 μl, the mixture was allowed to stand at room temperature for about 20 minutes to develop color, and the absorbance was measured with a plate reader (410 nm). The value of ELISA is 1.The value obtained by similarly measuring the original strain V122scFv expressed in Escherichia coli HB2151.
It was shown as a relative value in the graph as 0 (Fig. 5). In addition, colonies of each clone were picked up from the phagemid separation plate and cultured in 1.5 ml of CGA before
The phagemid was purified using the Plasmid Mini Kit. Each phagemid contains pCANTAB5-R1 primer (5'-CCATGATTAC
GCCAAGCTTTGGAGCC-3 ') was used for partial sequencing to determine the sequence of the CDR1 region (Fig. 5).

(8) Purification of scFv and measurement of binding property by ELISA Purification of scFv was performed according to the protocol of RPAS Purification Module (Amersham Pharmacia). Prepare a 2-fold dilution series from the purified scFv concentration of 10 nM and
React with A-hapten coated ELISA plate
The binding to the cFv hapten was investigated.

2. Results (1) Amplification of entire vector by IPCR and self ligation INF in which NNK mutation was introduced to perform IPCR
Primer (I nverted N NK F oward primer ; INF) were 10 kinds (INFb1～INFb10) Synthesis (Table 1).

[0043]

[Table 1]

The mix primer and IR primers were mixed this INF primer 10 types in equal amounts (I nverted R ev
erse primer (IR) was used for IPCR. When the amplified product was confirmed by electrophoresis, a main band was confirmed at a position of about 5.3 kb, which is equal to the total length of the vector (Fig.
6). Generally, since IPCR uses a circular plasmid as a template, amplification efficiency of PCR is poor and long-distance amplification is difficult, but by using Ex taq (TaKaRa),
The entire vector could be amplified.

In the amplified PCR product, the restriction enzymes DpnI and T4 were added.
DNA polymerase was added to decompose the phagemid used as a template and blunt the ends of the amplified product. This operation was performed simultaneously in the same tube to speed up and simplify the work. This reaction product was cut out after electrophoresis and subjected to self ligation. The ligation reaction and the preceding 5'-terminal phosphorylation of the PCR product were also continuously performed in the same tube, and the work could be simplified.

(2) Preparation of library and examination of quality A self ligation product was transformed into Escherichia coli TG1 to prepare a library. To examine the quality of the library, phagemids were isolated and purified from some clones and the CDR1 portion was sequenced.

Unlike the usual site-directed mutagenesis of single base substitution, the IPCR this time introduces NNK mutation of three base substitutions at two sites. Therefore, the PCR product amplified by incorporating the INF primer has a NNK site. Mismatch occurs in (Fig. 2).
Therefore, it is possible that the frequency of appearance of clones in the final product after self ligation has increased. As a result of Seques, it was found that an average of 50% (11/22) of clones performed accurate self ligation (Table 2A), and no significant increase in deletion clones was observed. A small deletion near CDR1 was noticeable in the clones in which the deletion occurred (Table 2B). In these clones, deletions were found in the sequence within the INF primer (data not s
hown), suggesting that it may be a deletion caused by a mismatch of NNK sites.

Next, it was examined whether there was any bias in the position of introducing amino acid mutation in the library. What number of INF each clone is
It was examined whether it was an amplification product derived from the primer. As a result, there were 6 types of mutation position patterns in 11 clones whose sequences could be analyzed (rightmost column of Table 2A).

Therefore, this library was considered to have no extreme bias in the position of introduction of amino acid mutations. This is also supported by the fact that no difference was observed between the primers when IPCR was performed for each INF primer and the amounts of amplification products were compared (data not shown).

The appearance frequency of the introduced amino acid mutation was also examined. Introducing the NNK sequence theoretically covers the expression of 20 types of amino acids with 32 combinations of nucleotide sequences, and therefore the frequency of appearance varies depending on the type of amino acid. From the sequence results, it was confirmed that the library prepared this time was introduced at a frequency close to the original amino acid appearance frequency found in the NNK sequence (Table 2).
C).

[0051]

[Table 2]

Examination of the size of this library 1.
It was found to have 4 × 10 ⁵ clones. The frequency of normal self ligation in this library was about 50%, so 1.4 × 10 ⁵ × 0.5 (normal self ligation
On rate) = 7 × 10 ⁴ Normal clones are calculated to exist.
In the case of a library containing two NNK mutations, considering the nucleotide sequence level, 32 (NNK) × 32 (NNK) × 10 (INFb 1 to 1
0) = 10240 combinations exist. Thus, the library becomes a calculation that covers the entire sequence pattern at about seven times ^{(7 × 10 4/10240)} , it was found that keeping the number of full repertoire.

(3) Selection of Bound Clones by Biopanning Next, it was examined whether selection from this library could be an effective means for the strategy of obtaining dominant mutants that actually improve the binding activity. ScFv-displaying phages were prepared from this library and biopanning was performed on BSA-hapten to obtain clones that bind to hapten. After repeating the panning operation four times, clones were collected and sequenced (41 clones). As a result, a clone as shown in FIG. 5 could be obtained.

Five kinds of clones having two or more clones having the same H chain CDR1 amino acid sequence were found. In particular, 13 clones appeared for the Ser → Arg mutant at the first amino acid site (Ser), accounting for about 30% of the total. In addition, double clones in which the Val → Gly mutation at the third amino acid site (Val) has been added to this have also appeared in 7 clones.
Accounted for 49%. EL is better than wild type in these two clones.
A clone with a high ISA value was also found, and it was expected that the mutant had a high hapten-binding property.

In addition to this, Val in which only one clone appeared
→ Gly mutant, Ser → Arg + Val → Arg double mutant, Met → Val mutant at the 4th amino acid position (Met) showed higher ELISA values than the wild type. Since no clone having the same sequence as the original V122scFv was detected after selection, these mutants were presumed to be clones having high hapten-binding properties.

It should be noted that among clones with the same CDR1 sequence, ELIS
The cause of the variation in the A value was that E. coli supernatant was
Since it is used for SA, it is possible that the expression level of scFv varied among clones. It is also possible that clones containing mutations other than CDR1 may coexist due to PCR error during IPCR.

(4) Estimation of dominant amino acid mutation The number of amino acids in the CDR1 region sequenced after panning was separately considered for each of the five amino acids, and the dominant amino acid was estimated. H chain CDR1 sequence of V122 antibody (Ser-Val-Tyr
-Met-His), the occurrence frequency of 5 amino acids in this library was stochastic in each Ser about 64% [(6/10) +
(4/10) × (3/32)], Val is about 63%, and Tyr, Met, and His are about 61%. If the frequency of occurrence of these amino acids is higher than this value among the clones after panning,
It can be judged that these amino acids are the dominant amino acids selected by panning. On the other hand, when the frequency of occurrence of these original sequences decreases, and a new amino acid whose frequency of occurrence is clearly higher than the probability of occurrence is appeared, this amino acid is dominant in binding to the hapten. It can be judged to be an amino acid.

The number of occurrence of amino acids in the CDR1 region sequenced after panning is shown in Table 3 for each of the 5 amino acids. Sequence of the original strain V122scFv 2
The second Tyr, fourth Met, and fifth His were found to be frequently preserved even after panning. The frequency of each occurrence is Tyr 85% (35/41), Met 95% (39/41), His
It was 100% (41/41). Therefore, these amino acids are V122
It was presumed that the amino acid was selected as the dominant amino acid for binding the antibody to the hapten.

[0059]

[Table 3]

On the other hand, the appearance frequency of the first Ser is 20% (8 /
41). Instead, Arg (22/41, about 54%) or Amb
The er mutation (11/41, about 27%) appeared at a higher frequency than expected (Amber stop codon is scFv in ELISA).
Since supE strain XL1-Blue is used in E. coli for expression, it is considered that Glu is inserted in place of the stop codon). The appearance frequency is 46% (19/4
Only 1), Gly (20/41, about 49%) appeared more frequently than expected. From the above, it was suggested that the 1st Ser and the 3rd Val may be neutral amino acids with respect to binding properties. Furthermore, Arg was detected at high frequency in the 1st and Gly was detected in the 3rd frequency, suggesting that it may be the dominant amino acid newly selected by panning.

As clones having Arg and Gly sequences, in addition to the Ser → Arg single mutant and Val → Gly single mutant,
All three patterns of double mutants of Ser → Arg + Val → Gly appeared (Fig. 5). In each clone, the ELISA value was found to be higher than the original strain, and it was estimated from the amino acid appearance frequency.
The results support the possibility that Arg and Gly are the dominant amino acids that increase the binding to the hapten.

Therefore, the Arg mutant (Arg-Val-Tyr-Met-Hi, which was predicted to be a dominant mutant from the results of the sequencing, was used.
s), Gly mutant (Ser-Gly-Tyr-Met-His), Arg + Gly double mutant (Arg-Gly-Tyr-Met-His) scFv were purified and ELISA
I went. As a result, it was suggested that these mutants had stronger binding to the BSA-hapten than the control parent strain V122scFv (Fig. 7). In particular, the double mutant showed stronger binding than the other one amino acid mutants, and the results showed that the amino acid mutations may have additively increased binding.

3. Discussion We developed a new method to estimate the dominant amino acid by comparing the appearance frequencies of the amino acid sequences of each clone selected by panning. The newly selected Ar
The incidence of g and Gly in the original library is 3.
75% [(4/10) x (3/32)], 1.25% [(4/10) x (1/32)]
Has become. Therefore, if no enrichment occurs after panning, Arg is approximately 1.5 in the 41 clones sequenced.
It is calculated that about 0.5 and Gly will appear. After 4 times of panning, 22 Arg and 20 Gly appeared, so it can be judged that these amino acids were enriched by panning. At the same time, the frequencies of occurrence of Ser and Val, which are the original sequences, respectively decreased, so that it was determined that Arg and Gly were the dominant amino acids selected by panning.

Since Gly has only a hydrogen atom in the side chain, it has a large degree of structural freedom, and it often appears in a portion where the main chain largely changes, such as at the beginning of the turn structure, and greatly affects the protein structure. Is also known and interesting.

On the other hand, 11 clones in which the first Ser has been changed to an Amber codon have also appeared. Since a suppressor mutant strain (TG1, XL1-blue) was used in E. coli in this experiment, another amino acid (Gl
u) is introduced, but it is not complete and part of the translation ends at the amber codon. Therefore, the amount of full-length scFv expressed is reduced compared to the case of using a normal codon. Therefore, clones with the Amber mutation may have been selected as a result of reduced toxicity to E. coli rather than improved binding. In fact, the ELISA values of the clones having the Amber mutation are generally low, and it is expected that the binding activity is not so strong.

The estimation of the dominant amino acid mutation in this experiment was performed on the assumption that the process of improving the binding property by substituting amino acids is independently and additively performed at each amino acid site. On the other hand, it is considered that the interaction between multiple amino acid substitutions cannot be completely ignored. Since this library is a 2-amino acid substitution library, we will also be looking at the results of interactions between two amino acids at the same time. There is an advantage that you can choose without. Arg and Gly at the moment
It is not clear that they interact with each other, but it is considered that there is no problem to presume that Arg and Gly are dominant mutations, since at least individual mutants exist.

[0067] in the case of introducing artificially NNK mutation as of this time, it is possible to introduce a mutation in the A G C (Ser) → C G G (Arg) as seen in the mutant two places of the base at a time . This is seen in the process of antibody affinity improvement seen in vivo 1
This is an effective method that can increase the repertoire of amino acids more easily than amino acid mutations that occur due to the accumulation of mutations for each base. Therefore, the combination of the IPCR-mutagenesis method used in this experiment and the method of estimating the dominant amino acid from only the amino acid sequence information will be extremely effective in the strategy to improve the binding property stepwise.

Generally, an IP using a circular plasmid as a template
CR had poor amplification efficiency and it was difficult to amplify a long DNA region. However, this protocol using Ex taq (TaKaRa) was able to amplify the entire vector of about 5.3 kb. These amplifications are based on Pfu turboDNA polymerase (STRATAG
ENE) and KOD-Plus-polymerase (TOYOBO) did not succeed (data not shown).

In this protocol, as shown in Table 4, all steps can be completed by a standard operation for 3 days. Since the enzyme reaction for each step can also be performed in one tube, the work is greatly simplified. self
Due to the efficiency of ligation, the library size is 1.4.
Although it was about × 10 ⁵ , it is possible to further increase the size by increasing the amount of PCR product and scaling up. Therefore, it is considered that this method, which is not restricted by restriction enzyme sites, is a very useful mutagenesis library preparation method in the preparation of a library which does not require a relatively large size as described above.

[0070]

[Table 4]

According to the IPCR-mutagenesis method, it is possible to easily introduce a mutation at a desired position as long as there is information on the nucleotide sequence. Therefore, using the clones containing the dominant amino acids Arg and Gly mutation found this time as templates,
CR-mutagenesis can be performed to rapidly prepare the next library. As a new library, mutagenesis in the vicinity of the CDR1 region (CDR-walking) or mutagenesis in other CDR regions (CDR-jumping) can be considered (Fig. 8). Therefore, this method capable of rapidly introducing an NNK mutation may be a very useful method that can obtain an antibody with increased binding activity rapidly and rationally, and with high throughput in mind.

[0072]

[Sequence list]       SEQUENCE LISTING <110> Japan Tobacco Inc. <120> A simple method for producing a mutant antibody <130> 020483 <160> 2 <210> 1 <211> 852 <212> DNA <400> 1 gtg aaa aaa tta tta ttc gca att cct tta gtt gtt cct ttc tat gcg 48 Val Lys Lys Leu Leu Phe Ala Ile Pro Leu Val Val Pro Phe Tyr Ala   1 5 10 15 gcc cag ccg gcc atg gcc cag gtg aaa ctg cag cag tca gga cct gag 96 Ala Gln Pro Ala Met Ala Gln Val Lys Leu Gln Gln Ser Gly Pro Glu              20 25 30 ctg gta aag cct ggg gct tca gtg aag atg tcc tgc aag gct tct gga 144 Leu Val Lys Pro Gly Ala Ser Val Lys Met Ser Cys Lys Ala Ser Gly          35 40 45 tac aca ttc act agc tat gtt atg cac tgg gtg aag cag aag cct ggg 192 Tyr Thr Phe Thr Ser Tyr Val Met His Trp Val Lys Gln Lys Pro Gly      50 55 60 cag ggc ctt gag tgg att gga tat gtt aat cct tac aat gat ggt att 240 Gln Gly Leu Glu Trp Ile Gly Tyr Val Asn Pro Tyr Asn Asp Gly Ile 65 70 75 80 aac tac aat gag aag ttc aaa ggc aag gcc aca ctg act tca gac aaa 288 Asn Tyr Asn Glu Lys Phe Lys Gly Lys Ala Thr Leu Thr Ser Asp Lys                  85 90 95 tcc tcc agc aca gcc tac atg gag ctc agc agc ctg acc tct gac gac 336 Ser Ser Ser Thr Ala Tyr Met Glu Leu Ser Ser Leu Thr Ser Asp Asp             100 105 110 tct gcg gtc tat tac tgt gca aga aag ggt ttg gac tat tgg ggc caa 384 Ser Ala Val Tyr Tyr Cys Ala Arg Lys Gly Leu Asp Tyr Trp Gly Gln         115 120 125 ggg acc acg gtc acc gtc tcc tca ggt gga ggc gct tca ggc gga ggt 432 Gly Thr Thr Val Thr Val Ser Ser Gly Gly Gly Ala Ser Gly Gly Gly     130 135 140 ggc tct ggc ggt ggc gga tcg gac atc gag ctc act cag tct cca aaa 480 Gly Ser Gly Gly Gly Gly Ser Asp Ile Glu Leu Thr Gln Ser Pro Lys 145 150 155 160 ttc atg tcc aca tca gta gga gac agg gtc agc gtc acc tgc aag gcc 528 Phe Met Ser Thr Ser Val Gly Asp Arg Val Ser Val Thr Cys Lys Ala                 165 170 175 agt cag ttt gtg ggt act tat gta gcc tgg tat caa cag aaa cca gga 576 Ser Gln Phe Val Gly Thr Tyr Val Ala Trp Tyr Gln Gln Lys Pro Gly             180 185 190 caa tct cct aaa gca ctg att tac tcg gca tcc acc cgg cgc act gga 624 Gln Ser Pro Lys Ala Leu Ile Tyr Ser Ala Ser Thr Arg Arg Thr Gly         195 200 205 gtc cct gat cgc ttc aca ggc agt gga tct ggg aca gat ttc act ctc 672 Val Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu     210 215 220 acc att agc aat gtg cag tct gaa gac ttg gca gag tat ttc tgt gag 720 Thr Ile Ser Asn Val Gln Ser Glu Asp Leu Ala Glu Tyr Phe Cys Glu 225 230 235 240 caa tac agc agc tct ccg tat acg ttc gga tcg ggc acc aag ctg gaa 762 Gln Tyr Ser Ser Ser Pro Tyr Thr Phe Gly Ser Gly Thr Lys Leu Glu                 245 250 255 atc aaa cgg gcg gcc gca ggt gcg ccg gtg ccg tat ccg gat ccg ctg 816 Ile Lys Arg Ala Ala Ala Gly Ala Pro Val Pro Tyr Pro Asp Pro Leu             260 265 270 gaa ccg cgt gcc gca cat cac cat cac cat cac tag 852 Glu Pro Arg Ala Ala His His His His His His *         275 280 <210> 2 <211> 283 <212> PRT <400> 2 Val Lys Lys Leu Leu Phe Ala Ile Pro Leu Val Val Pro Phe Tyr Ala   1 5 10 15 Ala Gln Pro Ala Met Ala Gln Val Lys Leu Gln Gln Ser Gly Pro Glu              20 25 30 Leu Val Lys Pro Gly Ala Ser Val Lys Met Ser Cys Lys Ala Ser Gly          35 40 45 Tyr Thr Phe Thr Ser Tyr Val Met His Trp Val Lys Gln Lys Pro Gly      50 55 60 Gln Gly Leu Glu Trp Ile Gly Tyr Val Asn Pro Tyr Asn Asp Gly Ile 65 70 75 80 Asn Tyr Asn Glu Lys Phe Lys Gly Lys Ala Thr Leu Thr Ser Asp Lys                  85 90 95 Ser Ser Ser Thr Ala Tyr Met Glu Leu Ser Ser Leu Thr Ser Asp Asp             100 105 110 Ser Ala Val Tyr Tyr Cys Ala Arg Lys Gly Leu Asp Tyr Trp Gly Gln         115 120 125 Gly Thr Thr Val Thr Val Ser Ser Gly Gly Gly Ala Ser Gly Gly Gly     130 135 140 Gly Ser Gly Gly Gly Gly Ser Asp Ile Glu Leu Thr Gln Ser Pro Lys 145 150 155 160 Phe Met Ser Thr Ser Val Gly Asp Arg Val Ser Val Thr Cys Lys Ala                 165 170 175 Ser Gln Phe Val Gly Thr Tyr Val Ala Trp Tyr Gln Gln Lys Pro Gly             180 185 190 Gln Ser Pro Lys Ala Leu Ile Tyr Ser Ala Ser Thr Arg Arg Thr Gly         195 200 205 Val Pro Asp Arg Phe Thr Gly Ser Gly Ser Gly Thr Asp Phe Thr Leu     210 215 220 Thr Ile Ser Asn Val Gln Ser Glu Asp Leu Ala Glu Tyr Phe Cys Glu 225 230 235 240 Gln Tyr Ser Ser Ser Pro Tyr Thr Phe Gly Ser Gly Thr Lys Leu Glu                 245 250 255 Ile Lys Arg Ala Ala Ala Gly Ala Pro Val Pro Tyr Pro Asp Pro Leu             260 265 270 Glu Pro Arg Ala Ala His His His His His His         275 280

[Brief description of drawings]

FIG. 1: Structural formula of BSA-hapten. We break down several prodrugs with vitamin B6 as a common protecting group,
The V122 catalytic antibody was developed. This antibody immunizes mice with this transition state analog hapten (4- (N-acetylaminomethyl) -5- (hydroxymethyl) -2-methyl-3-pyridyl hydrogen 5-carboxypentylphosphonate) bound to BSA. It was obtained.

FIG. 2 Introduction of NNK mutation by IPCR-mutagenesis method.
In the examples, a library was prepared using the IPCR method. Also, unlike normal site-directed mutagenesis of single nucleotide substitution, it is an NNK (N = A, G, C, T ,; K = G, T) mutation3
Base substitutions were introduced at two places. INF primer (Inverted
PCR product amplified by incorporating NNK Forward primer) is N
A mismatch occurs at the NK site.

FIG. 3: Mutagenesis site of library. In the examples, among the 5 amino acid residues of the H chain CDR of the V122 catalytic antibody,
As shown in this figure, a 2-amino acid mutation library in which NNK was introduced at two sites was designed.

FIG. 4: Sequence of V122scFV. Hapten-binding scFv-expressing clones were selected from the library prepared in the example. The sequence of V122scFV obtained was as shown in this figure.

[Fig. 5] Sequence of CDR1 site of each clone after panning
ELISA value. The value of ELISA is based on the original strain V122scFv
The values measured in the same manner for those expressed in HB2151 were set to 1.0, and shown in the graph as relative values.

FIG. 6 is an agarose electrophoresis diagram of IPCR products. Mix primer with 10 kinds of INF primer mixed in equal amount
It was IPCR in combination; (IR I nverted R everse primer ) IR primer. When the amplified product was confirmed by electrophoresis, a main band was found at a position of about 5.3 kb, which is equal to the entire length of the vector, and amplification of the entire vector was confirmed.

FIG. 7. ELISA of clones after panning. 3 was predicted to be a dominant mutant based on the results of the sequence of Examples.
As a result of purifying scFv of the one mutant and performing ELISA, it was shown that these mutants had stronger binding to the BSA-hapten than the V122scFv of the control parent strain.

[Fig. 8] Strategy for obtaining clones with additive binding improvement by the IPCR-mutagenesis method. IPCR-mutagenesis method of the present invention (NNK mutation and utilizing the IPCR method, a method of introducing a rapid and simple mutation) using a clone having a dominant amino acid found as a template, further performing IPCR-mutagenesis, and rapidly The following library can be prepared to obtain clones with further increased binding. CDR1 as a new library
Mutation introduction in the vicinity of the region (CDR-walking), or mutation introduction in another CDR region (CDR-jumping) can be considered.

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 4B024 AA01 AA11 BA43 CA04 CA11                       HA15                 4B050 CC04 DD07 LL01 LL03                 4B064 AG27 CA19 CC24 DA01 DA13                 4H045 AA11 AA20 BA10 CA40 DA76                       EA50 FA74

Claims (16)

[Claims] 1. A region contained in or near one of CDRs of an antibody, wherein n (n is an integer of 3 ≦ n)
M selected from the region consisting of amino acid residues (m is
For 2 ≤ m ≤ n-1) amino acid residues, k (k is 2
Mutant Fv library, which is designed so that ≦ k ≦ 20 kinds of amino acid mutations are introduced, and which can give a limited Fv repertoire of _n C _m × k ^m kinds as an amino acid sequence. 2. kNK is 20; and NNK is the nucleotide residue corresponding to the amino acid residue into which a mutation is to be introduced.
(N = A, G, C, T; K = G, T) Mutant Fvs designed to introduce mutations and capable of giving a limited Fv gene repertoire of _n C _m × 32 ^m types as a nucleotide sequence Library. 3. The library according to claim 2, wherein n is 3 ≦ n ≦ 10; m is 2; and is prepared by utilizing the IPCR method. 4. The library according to any one of claims 1 to 3, which displays a mutant Fv repertoire on a phage. 5. A method for screening a mutant Fv, which comprises a step of selecting the phage display mutant Fv library according to claim 4 using a substance having antigenicity. 6. A method for producing a mutant antibody, which utilizes the screening method according to claim 5. 7. A method for modifying the function of an antibody, which utilizes the screening method according to claim 5. 8. A method for producing a mutant antibody having a region consisting of an optimum mutant amino acid sequence which can be obtained by the following procedure: (1) A phage display mutant Fv library, which is contained in one CDR of an antibody Or a region present in the vicinity thereof, which can give a mutant Fv repertoire having an amino acid mutation in a region consisting of n (n is an integer of 3 ≦ n) amino acid residues, Determining the amino acid sequence of the region of a plurality of clones selected by the binding property with a substance having (2) selecting one amino acid sequence according to the number of appearances at each position of 1 to n of the region And (3) the obtained amino acid sequence pattern is used as an optimal mutant amino acid sequence. 9. A phage display mutant Fv library having k (k is an integer of 2 ≦ k ≦ 20) kinds for m (m is an integer of 2 ≦ m ≦ n−1) amino acid residues. Is designed to introduce the amino acid mutation of _n C _m × k
^The method for producing a mutant antibody according to claim 8, which is capable of giving ^m kinds of restricted mutant Fv repertoires. 10. A mutant antibody or a part thereof comprising a region consisting of an optimum mutant amino acid sequence which can be obtained by the following procedure: (1) A phage display mutant Fv library, wherein: An antigen which is included or present in the vicinity thereof and which can give a mutant Fv repertoire having an amino acid mutation in a region consisting of n (n is an integer of 3 ≦ n) amino acid residues, Determining the amino acid sequence of the region of the plurality of clones selected by the binding property with a substance having a property; (2) for each position 1 to n of the region, one amino acid sequence depending on the number of appearances And (3) the obtained amino acid sequence pattern is used as an optimal mutant amino acid sequence. 11. A phage display mutant Fv library comprising k (k is an integer of 2 ≦ k ≦ 20) kinds for m (m is an integer of 2 ≦ m ≦ n-1) amino acid residues. designed for such amino acid mutations are introduced, _n C _m as an amino acid sequence
× those that may have a k ^m types of limited mutated Fv repertoire variant antibody or a portion thereof according to claim 10. 12. The phage display mutant Fv library is subjected to 8 to 20 amino acid mutations for 2 to 9 amino acid residues selected from the region consisting of 3 to 19 amino acid residues. Mutations designed and limited to
The mutant antibody or a part thereof according to claim 10, which may include an Fv repertoire. 13. A substance having the ability to catalyze a reaction; and having antigenicity is a hapten bound to a carrier protein, the hapten being a transition state analogue of the reaction,
The mutant antibody or a part thereof according to any one of claims 10 to 12. 14. The mutant antibody or a part thereof according to claim 13, which has the ability to catalyze an ester hydrolysis reaction. 15. The mutant antibody or a part thereof according to claim 14, which is capable of degrading a prodrug. 16. A mutant antibody or a part thereof, which has 1 to 5 dominant amino acids in CDR3 of H chain and has the ability to catalyze an ester hydrolysis reaction.