JPH0949836A - Method for evaluating structure of protein by using antibody - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate the secondary structure of a protein produced by gene manipulation or subjected to so called refolding for returning the protein to a natural type from a denaturalized type. SOLUTION: The antibody to the domain, such as the loop structure, etc., of a protein is produced by using a peptide having the amino acid sequence of the antibody as an antigen and an antibody which is bonded to a natural type protein and another antibody bonded to a denaturalized protein are selected. Then whether the domain of the protein is of a natural type of denaturalized type is discriminated by examining the bonding ability of the antibody to a protein produced by gene manipulation or subjected to refolding.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the evaluation of the secondary structure relating to the activity of a protein having physiological activity. The secondary structure of the protein produced by genetic engineering or placed under physiological conditions from denaturing conditions. A method for determining the presence or absence of structure formation in a structure is provided.

[0002]

2. Description of the Related Art Microorganisms are used to produce physiologically active substances such as insulin and growth hormone by gene manipulation techniques. However, the protein produced may be inactive or less active than naturally occurring proteins. This is because the produced protein does not have a completely natural structure due to imperfect folding or cross-linking of SS bonds. The secondary structure of such proteins is evaluated by measuring the CD spectrum and changes in fluorescence intensity.

However, although these methods can measure the ratio of helices and sheet structures, it is difficult to determine the presence or absence of two helix structures, two sheet structures, or a loop structure connecting the helices and the sheet structure. These loop structures are often binding sites for receptors of physiologically active substances, and the presence or absence of loop structure formation greatly influences physiological activity. Therefore, there is a need for a method for evaluating loop structure.

[0004] Attempts have already been made to investigate the structural changes of proteins using antibodies. James F. Collawn et al. Replaced the amino acid residue with another in cytochrome C,
Alternatively, the structural change caused by chemical modification was assayed with an anti-cytochrome monoclonal antibody, and it was shown that the method using the antibody as a probe was superior to the CD spectrum measurement (Journal of Biological Chemistry, Vol.
263, No. 18, pp8625-8634 (1988). Also, Jean-Marc
Chatel et al. Produced a monoclonal antibody that binds only to the active form of acetylcholinesterase and a monoclonal antibody that binds to the inactive and denatured forms (FEBS LETTER
S, Vol.319, No.1, pp12-15 (1993)). The position of the former epitope is in a region that is not cleaved by trypsin,
The latter epitope is in a region that is cleaved by trypsin, and the difference in the position of both epitopes causes a difference in binding specificity. Conventional antibodies detect structural changes due to substitution of amino acid residues in the epitope region, cleavage by protease, etc., that is, they have been used to detect changes in amino acid residues.

[0005]

In the above-mentioned conventional techniques for evaluating proteins using antibodies, the antibodies are used as probes for detecting changes in amino acid residues, but their secondary structures have not been evaluated yet. Although antibodies that do not bind to the natural type also appear in normal antibody production,
It was excluded from screening because it was considered to be used for secondary structure evaluation. Antibodies that bind to the native protein or denatured protein that is a requirement of the present invention are natural or denatured test proteins, such as those denatured by SDS treatment in the presence of 2-mercaptoethanol, respectively. By separately administering to experimental animals such as rabbits and mice to immunize, each antibody is obtained as an antiserum. Then, the target antibody can be obtained by selecting an antiserum positive only for each of them by an enzyme-labeled antibody measurement method (hereinafter abbreviated as ELISA) using natural and denatured proteins as antigens. However, the major epitopes of both antisera often do not match, which is not suitable for evaluating the structure of a specific region as in the present invention.

Similarly, a desired monoclonal antibody can be obtained by immunizing a mouse or rat, fusing the spleen cells and myeloma cells to establish a hybridoma cell line, and screening the antibody by ELISA. However, since a large number of antibodies are induced, it is complicated and inefficient to examine the epitope and select an antibody that binds to a natural type or a denatured type with an epitope having a common amino acid sequence.

Therefore, focusing on a certain area, for example, 2
Antibodies against the native structure and the denatured form against the loop structure of the secondary structure must be produced at the same time. On the other hand, various antibodies can be produced by the technique of producing an anti-peptide antibody by immobilizing a polypeptide (hereinafter abbreviated as peptide) on a carrier and immunizing. As a result of studying the production of an antibody against the amino acid sequence of the region forming the loop structure by this technique, an antibody capable of binding to the natural loop structure and an antibody capable of binding only to the denatured type could be produced at the same time.
Then, S- having the amino acid sequence by these antibodies
When the binding properties with proteins having different S-bonds were examined, it was suggested that the denatured antibody binds well and does not form a normal secondary structure in those with low physiological activity. The present invention has been completed based on these results.

[0008] The present invention provides a method for evaluating the structure of a protein obtained by a gene manipulation technique or a refolding manipulation using an antibody.

[0009]

A technique for producing an anti-peptide antibody for simultaneously producing an antibody that is a requirement of the present invention, that is, an antibody that binds to a certain region of a protein in a natural type or a denatured type is described. Adopted. A native protein has a structural state equivalent to that in a living body, and a denatured protein is
S of the same natural protein in the presence of 2 mercaptoethanol
It is defined as the state of DS processing.

Peptides consisting of a dozen or more amino acid residues are used, but since the peptide alone has low immunogenicity, BSA (bovine serum albumi) is used as a carrier protein.
n) or KLH (keyhole limpet hemocyanin) is used as an immunogen. When the tertiary structure of the peptide is unknown, the hydrophilic region predicted by the Hydropathy method should be selected. In addition, since antibodies may be induced in regions that are judged to be hydrophobic,
It is not always necessary to select only the hydrophilic region.

When the tertiary structure is unknown, the loop structure site in the secondary structure, that is, two helix structures or 2
The amino acid sequence of the peptide for producing an antibody for one sheet structure or a loop structure that connects the helix structure and the sheet structure is estimated by the Chou, Fasman and Rose method, etc. , Can be singled out.

[0012] Furthermore, from those of which the tertiary structure is known or homologous proteins whose tertiary structure is known,
If the next structure is predicted, it is advisable to select an amino acid sequence in a region where the antigen-binding site of the antibody molecule or the light chain or heavy chain of the antibody approaches or does not approach, for example, a concave or convex region. In the former case, an antibody that does not bind in the natural type but binds in the denatured type can be expected. Also,
The loop structure belongs to the latter, but in the case of a peptide having a convex region amino acid sequence, it is expected to induce both a natural-type binding antibody and a denatured-type binding antibody.

For example, mouse nerve growth factor Nerve Growth
The tertiary structure of Facter (hereinafter abbreviated as NGF) has been elucidated using its crystal. Brain-derived Neurotrophic Facter (hereinafter BDN)
(Abbreviated as F) has high homology in its amino acid sequence,
A tertiary structure has been speculated (Tom L. Blundel et al .; Natur
e, Vol. 354 pp411-414 (1991)). Based on the inferred structure results, BDNF also has three sets of sheet structures and four loop structures, similar to NGF. 3 out of 4 loop structures
, Amino acid sequences of sequence numbers 30 to 35-DMSGG
-, The loop structure composed of the amino acid sequence of 43 to 47 -PVSKG- and the sequence numbers 92 to 96 -DSKKAR- is considered to be a binding site to the receptor.

Therefore, the peptides having these sequences may be synthesized. At this time, it is better not to place the sequence at the N-terminal or C-terminal side in the entire sequence. In BDNF, the sequence before and after the loop structure is a recessed region, for example, the sequence numbers 39 to 42-LEKV- are located at the positions surrounded by two loop structures. Is likely to bind only to the modified form. Therefore, when the peptide of the amino acid sequence of sequence numbers 39 to 50-LEKVPSKGQLK- is used, an antibody that recognizes the loop structure and an antibody that binds only to the denatured type can be expected.

The peptide having the selected amino acid sequence may be synthesized by using a gene manipulation technique or a chemical synthesis method. In the former case, if it is synthesized by fusing with another protein, it can be directly used for immunization. In the latter, BSA and KLH
Must be chemically bound to a carrier protein such as. The bond is NHS (N-hydroxysuccinimide) or EDCI.
It can be performed using (1-ethyl-3- (3-dimethylaminopropyl) carbodiimde). The fusion product of the peptide thus prepared and the carrier protein can be used for immunization as an antigen.

The anti-peptide antibody may be polyclonal, but even if it is a peptide, antibodies against several kinds of epitopes are induced. Therefore, a monoclonal antibody is preferable to narrow down the evaluation to the amino acid sequence of the target region.
In the former case, the antigen can be prepared by immunizing an experimental animal such as rabbit or rat according to a conventional method. The obtained antiserum may be used as it is, or may be purified by affinity chromatography using protein A or protein G. Then, an ELISA is performed on the natural type and the denatured type of the target protein to select an antibody that binds to the natural type or an antibody that binds only to the denatured type. The latter involves preparing a monoclonal antibody by culturing a hybridoma cell line produced by the so-called cell fusion method, and selecting an antibody that binds to a native protein or an antibody that binds only to a denatured protein. In addition, the antibody gene of the B cell of the immunized animal may be similarly selected by preparing an antibody molecule or a molecule having an antigen-binding site by a gene manipulation technique.
Furthermore, an antibody molecule or a molecule having an antigen-binding site may be prepared from the antibody gene of the hybridoma cell by a gene manipulation technique.

In accordance with a conventional method, a myeloma cell line has been established subcutaneously in an experimental animal such as a mouse or rat,
The vein or the abdominal cavity is injected with the antigen every 2 to 3 weeks for immunization. The number of injections can be determined by measuring the change in antibody titer. For example, the antibody titer may be set to 10 ⁵ or more, but there is no particular limitation. As the antigen used for measuring the antibody titer, it is preferable to use the antigen and the carrier protein in which the peptide and the carrier protein are fused. The spleen is extracted from the test individual that has reached the antibody titer, and myeloma cells and spleen cells are fused. For example, if a mouse is used as a test individual, X63 strain, P3U1 strain, NS-1 strain or SP2 strain
A strain of myeloma cell line is available.

Cell fusion can be performed using polyethylene glycol (molecular weight 1000-4000). Hybridoma cells are selected by HAT medium. Then, primary screening is performed on the culture medium of each hybridoma cell group. ELISA is carried out using the antigen and the carrier protein in which the peptide and the carrier protein are fused, and the one that is positive in the former and negative in the latter, or the one in which the former is larger is cloned by the limiting dilution method. This EL
When using a natural type protein and a denatured type protein in ISA, it suffices to select and clone those other than the negative one.

Next, secondary screening is carried out with the culture medium of the cloned cell line. EL using the antigen and carrier protein in which the peptide and carrier protein are fused
Perform ISA and select the former positive and the latter negative. A third screening is performed on the selected strain. If you use native and denatured proteins in the primary screen,
The clone omits secondary screening. In addition, 3
The cloning may be performed again before the next screening. In the third screening, each clone is selected for its ability to bind to a native type protein and a denatured type protein by dividing it into clones producing antibody A positive in the natural type and antibody B positive only in the denatured type. The selection can be performed by ELISA, dot blotting, etc., but the binding constant, the dissociation constant, and the affinity constant in the antigen-antibody reaction can also be measured and classified. The class of antibody does not need to be particularly limited.

The antibody of the selected clone can be prepared by the following procedure. An antibody can be obtained by transplanting the clone into the abdominal cavity of an animal individual previously administered with pristane, and collecting condensate after 10 to 14 days. Also, after acclimating the clone to a serum-free medium, it is cultured in a serum-free medium to obtain an antibody from the culture solution. The antibody can be purified from the condensate or the culture broth through steps such as salting out with ammonium sulfate, ion exchange chromatography and affinity chromatography.

For these purified antibodies, the position of the epitope in the amino acid sequence used is identified. Identification is performed by shifting a peptide consisting of about 6 amino acid residues by 1 residue in order from the N-terminal side, for example -LEKVPVSKGQLK-
VPVS, KVPVSK, VPSKGQ ,. . Can be synthesized and determined by examining cross-reactivity with one end of each peptide immobilized on a carrier. Alternatively, peptides having similar amino acid sequence regions of homologous proteins may be similarly immobilized on a carrier to examine cross-reactivity. The carrier may be a plastic surface or the carrier protein. Based on the results of the examination, an antibody having the amino acid sequence of the target region as an epitope is selected.

Regarding a region R of a protein which is a target protein, which has been produced by a gene manipulation technique or has been refolded, the amino acid sequence of the region R and the amino acid sequence of the epitope are the same or similar, and a natural positive antibody A and The cross-reactivity is examined by measuring the binding constant, the dissociation constant, and the affinity constant of the protein by ELISA, dot blotting, or antigen-antibody reaction using antibody B positive only to the denatured type. As a result, antibody A
If the binding is positive and the binding of the antibody B is negative, the region R in the test protein is determined to be the native type, and if only the antibody B is positive, the region is determined to be the modified type. Further, when all the antibodies are negative, it indicates that the region R is shielded by the peptide in another region regardless of the formation of the structure. That is, it is determined that the region R is not a natural type.
Regarding the region R in which only the antibody B is obtained, the region R is determined to be denatured type when the binding of the antibody B is positive. As described above, by selecting the amino acid sequence of the peptide for evaluating the secondary structure of the protein, preparing an antibody against the peptide, selecting the antibody, determining the epitope, and assaying the binding ability to the test protein, the protein is detected by the antibody. The secondary structure of can be evaluated.

[0023]

According to the present invention, an antibody A that binds to a natural region R and an antibody B that binds only to a denatured region R are prepared for a certain region R of a protein, and are prepared by a genetic engineering technique. The denaturing state was reacted with the refolded protein, and the binding ability of each was determined by ELIS.
By determining cross-reactivity by measuring the binding constant, dissociation constant, and affinity constant in A or dot blotting or antigen-antibody reaction, it is possible to determine whether the region R of the protein is a native type or a denatured type. The secondary structure of the protein is evaluated.

[0024]

EXAMPLES The present invention will be specifically described below with reference to examples, but the present invention is not limited to these examples.

Example 1 Peptides having an amino acid sequence constituting a loop structure of BDNF -LEKVPVSKGQLK- and -ALTMDSKKRIGWRF- were prepared by the following method.
Anti-peptide monoclonal antibody against was prepared. Hereinafter, the former peptide is called W336 and the latter peptide is called W338.

(1) Preparation of Peptide Antigen Peptides were synthesized by the Fmoc method and purified by reverse phase chromatography to a purity of 80 to 90%. Each of these peptides was bound to BSA to serve as a test antigen. The method is shown below.

5 mg of the peptide was dissolved in 350 ml of PBS and 50 ml of this was added so that the weight ratio of peptide to BSA was 3:10.
Of BSA solution (30 mg / ml). And 15 mg of EDCI
(1-ethyl-3- (3-dimethylaminopropyl) carbodiimide, h
(ydrochloride) was mixed and dissolved, and the mixture was allowed to stand overnight at room temperature under dark conditions. The reaction solution was dialyzed four times with 1 L of PBS and then centrifuged (10,000 × g, 5 minutes) to remove aggregates.
The supernatant was collected as an antigen and stored at -80 ° C. Hereinafter, this antigen is referred to as peptide-BSA.

The N-terminal amino acid sequence of the peptide-BSA was examined with a protein sequencer. As a result, the amino acid residues from the N-terminals of the peptide and BSA were detected in order, and the peptide bond and its amino acid sequence were confirmed, as well as the intramolecular peptide bond and the peptide bond (aspartic acid and glutamic acid residue Connection via)
It was found that there was no abnormality caused by. And BSA
When the number of peptides per BSA molecule was calculated from the detected amount of amino acid residues in the peptide, the number was 6 to 30.

(2) Immunization Each peptide-BSA was administered to two BALB / c mice (♀, 4
Adjuvant (N-acetylmuramyl-L-alanyl-Di
Soglutamine) was injected into the abdominal cavity at intervals of 2 weeks for immunization. The first time, P antigen 100μg and adjuvant 15μg
0.5 ml of the solution mixed with BS is used for the second and subsequent times and the amount of antigen is 20 μg
And a mixed solution of 15 μg of adjuvant were injected until the serum antibody titer increased to 10 ⁵ or more. As a result, it was found that none of the four administrations of the antigen achieved a sufficient antibody titer, and the immunization was continued. And with the sixth immunization
The antibody titers of W336-BSA and W338-BSA became 10 ⁵ or more. Therefore, spleen cells were excised from the mouse and subjected to cell fusion.

(3) Cell fusion and HAT selection Spleen cells and myeloma cells (P3U1) in a cell number of 5: 1
Prepare a mixed pellet to give a molecular weight of 1500
50% polyethylene glycol solution (75mM HEPES: Boehr
Cell fusion was carried out by a standard method with inger Manheim). After fusion, HAT medium containing 20% FBS (fetal bovine serum) (ERDF medium (Far East Pharmaceutical Co., Ltd.) mixed with HAT solution (Dainippon Pharmaceutical Co., Ltd.)) had a spleen cell concentration of 2 × 10 ⁶ cells / The fused cells were suspended so as to have a volume of 100 ml, and 100 μl was dispensed to each well of a 96-well microplate. Then, the culture was started in a CO ₂ incubator at 37 ° C. Four days after that, 100 μl of HAT medium was added, and the medium was replaced (50%) every two days thereafter.

(4) Screening and Cloning Antibody screening was performed in two stages, the first was peptide-BSA
Second, dot blotting was performed using rBDNF (manufactured by PEPROTECH; production by recombinant Escherichia coli, purity 96%) by ELISA using as an antigen. The clonning was performed by limiting dilution of hybridoma cells that were positive in the primary screening. The thymus of a 4-week-old mouse was removed and HT medium (ER
DF medium mixed with HT solution (Dainippon Pharmaceutical Co., Ltd.)
Suspend the same cells in a feeder cell solution (1 x 10 ⁷ cells /
ml) was prepared. And the concentration of hybridoma cells is 2
The cells were suspended in a feeder cell solution at a concentration of × 10 ⁵ cells / ml, and 100 µl thereof was dispensed into the wells of a 96-well microplate and cultured. After 10 days of culture, wells in which a single colony appeared were selected by a microscope, and 100 μl of HT medium was added. Then, the colonies that had a size that could be visually discriminated were subjected to ELISA to select positive ones.
The positive clone was cultured in a 24-well plate, confirmed to be positive again by ELISA, and stored frozen in liquid nitrogen.

Hybridoma cells were HAT-selected on four 96-well microplates, and then the culture solution in the wells in which the hybridoma cells appeared was subjected to ELISA to obtain peptide-B.
Positive for SA and negative for BSA or p
Wells with eptide-BSA higher than BSA were selected. The appearance rate of hybridoma cells was 40% on average, and hybridoma cells in the selected well were then cloned. ELISA was performed on the wells in which one colony appeared, and pep
Those that were positive for tide-BSA and negative for BSA were selected. As a result, in W336-BSA, 7 clones (4A12, 4AA5, 4AB2, 4A
B5, 4BE1, 4BF5, 4H11) and W338-BSA established 3 clones (6B1, 6B33, 6B35). Some of these hybridomas have been deposited with the Institute of Biotechnology, Industrial Technology Institute, as follows.

Hybridoma 4AA5: FERM P-15075 Hybridoma 4A12: FERM P-15076 Hybridoma 4AB2: FERM P-15077 Hybridoma 4BE1: FERM P-15078 Hybridoma 4H11: FERM P-15079 Hybridoma 6B33: FERM P-15080 of each hybridoma cell line When the heavy chains and light chains of the produced antibodies were assayed, most of the IgM type heavy chains except the 4A12 and 6B1 antibodies and the light chains were all kappa type. 4A1
The 2 and 6B1 antibodies bound the light chain antibody but not all the heavy chain antibodies. Therefore, SDS-
PAGE was performed, transferred to a nitrocellulose membrane, and blotted with a light chain antibody. As a result, a band was detected at the position of the light chain monomer (about 25 kDa) for both antibodies, and 4A12 and 6B1 were detected.
The cells were found to produce only light chains.

(5) Preparation of antibody Each hybridoma cell line was acclimated from a serum medium to a serum-free medium (subculture by gradually decreasing the serum concentration from 10% to 0% over about 2 weeks). Each of these strains can be acclimated to serum-free medium without decreasing the antibody-producing ability,
Antibodies were produced in serum-free medium. A 400 ml tapping flask (preparation volume: 8
Suspension culture was carried out in 0 ml). Approximately 10 of the culture supernatant liquid was collected by ultrafiltration membrane (molecular weight cutoff: 200 kDa, manufactured by Toyo Advantech).
After doubling the concentration and salting out with saturated ammonium sulfate, dialysis was performed to obtain an antibody.

Example 2 The above antibodies were examined by dot blot method for cross-reactivity to commercially available rBDNF (manufactured by PEPROTECH) and rNGF (manufactured by AUSTRAL Biologicals; manufactured by recombinant CHO cells, purity of 90% or more). . The reason for conducting rNGF simultaneously was that the selected peptides were NG as shown in Table 1.
RNGF because it has an amino acid sequence homologous to F
It is thought that there is a possibility of binding to.

[0036]

[Table 1]

Table 1 shows W336 and W338 peptides derived from BDNF.
The amino acid sequences of homologous sites with N335-derived W335 and W337 peptides are shown.

RNGF, 2ME-SDS-treated rNGF, rBDNF, 2ME-
SDS-treated rBDNF and peptide-BSA antigen solution (0.1 mg / m
l) and the antibody solution to the nitrocellulose membrane (Hybo
nd-C: manufactured by Amersham), 1 ml was dropped, and the antigen was fixed by drying at room temperature. After washing with PBST and PBS (2 times under shaking conditions for 5 minutes each, abbreviated as washing hereafter), the nitrocellulose membrane was mixed with milk protein solution (1/4
It was soaked in scissors (block ace), sandwiched, cut around with scissors, and the upper and lower parafilms were pressure bonded. This was left overnight at 4 ° C. for blocking. After washing, the antibody solution (0.2 ml) and the membrane were sandwiched with parafilm as described above, and reacted at 37 ° C. for 1 hour. Then, the secondary antibody and avidin-biotinylated peroxidase were sequentially bound, and 4-chloro-1-naphthol solution (3 mg / ml-methanol solution was added to PBS in a volume ratio of 1 :).
5) and H2O2 was added at 0.015%) to develop color.

Antibodies bound to rBDNF were 4A12 and 4H11
Only antibody. Moreover, both antibodies bound to rNGF and also bound to rBDNF and rNGF denatured by 2ME-SDS treatment.

Those bound only to the denatured rBDNF treated with 2ME-SDS were the 4AA5, 4AB2 and 6B33 antibodies, both of which were 2ME-SD.
It also bound to S-treated rNGF.

4BE1, 4BF5 and 6B1 antibodies are denatured rNG
Lightly developed with F, but treated with peptide-BSA and 2ME-SDS
No color development was seen with BDNF. Further, all 4AB5 and 6B35 antibodies did not develop color. Since the color of each antibody itself was similar to that of the other antibodies described above, it was considered that the reason why the color did not develop was that the affinity was extremely weak. As a result, the antibody that cross-reacts with natural rBDNF was 4A12.
And 4H11 antibody only, 4AA5, 4AB2 and 6B33 antibodies are 2ME-S
It was found to bind only to the DS-treated modified form. In addition, each antibody binds to the natural or modified form of rNGF, indicating that the epitope (binding site) has a region homologous to NGF.

Example 3 4AA5, which has a relatively high affinity in dot blotting,
The epitopes of the 4A12, 4AB2, 4H11 and 6B33 antibodies were examined.

Since the 4AA5, 4A12, 4AB2, and 4H11 antibodies all bound to 2ME-SDS-treated rNGF, these epitopes are also present in the amino acid sequence of NGF. Therefore, the presence or absence of cross-reactivity with the NGF-derived peptides shown in Table 1 was examined. As a control, W337 and W338
Was used.

Since all of the antibodies bound to the W335 peptide derived from NGF, it was considered that W335 and W336 recognize a similar sequence -LEKV-. Furthermore, since the 4AB2 and 4H11 antibodies also bind to W338, they recognize the common sequence -SK- of W336 and W338. Therefore, the epitopes of 4AA5 and 4A12 include the region containing the sequence of -LEKV-, 4AB2 and
The epitope of the 4H11 antibody was considered to be a region containing the sequence of -LEKVPVSK-. By the way, the 4AA5 and 4AB2 antibodies
Can bind only to denatured rBDNF, but not 4A12 and 4H1
The 1 antibody also binds to the native form. That is, it was found that the 4AA5 and 4A12 antibodies and the 4AB2 and 4H11 antibodies have different binding characteristics to rBDNF, although the epitopes have similar amino acid sequences. This was considered to be due to the difference in affinity for the natural epitope.

The B33 antibody is an ELISA of W337-BSA and W338-BSA.
Then, it was considered that the binding site was that which binds only to the latter, that is, that is not common to both peptides. But 2ME-SD
In the dot blots of S-treated rBDNF and rNGF, they were bound to both, indicating that the binding site was NGF and was located at a sequence other than W337. Therefore, when the amino acid sequence of W338 was searched for a sequence common to NGF other than the -ALTMD- or -WRF- sequence, the -DSK- sequence was found to be 72 NGF.
It turned out to be the 74th. And of W336-BSA
Since 6B33 antibody was negative in the ELISA, it was considered that -DSK- was specifically recognized.

Example 4 It was found from the results so far that the binding characteristics of each antibody depended on its affinity. So 4AA5, 4A12, 4A
For the B2, 4H11, 4BE1 and 6B33 antibodies, the affinity constants (KA) for native and denatured rBDNF were determined by Fisons IAsys.
It measured using the apparatus.

As a result, ka, kd and affinity constant (KA) of each antibody are shown in Table 2.

[0048]

[Table 2]

Table 2 shows 4AA5, 4A12, 4AB2, 4BE1, 4H11 and
The binding constant, the dissociation constant, and the affinity constant in the antigen-antibody reaction of the 6B33 antibody with native and denatured BDNF are shown.

Affinity constants (KA) of 4A12 and 4H11 antibodies positive in the native rBDNF dot blot are 4.31 and 2.
17, and the negative 4AA5, 4AB2, 4BE1 and 6B33 antibodies were 0.91
It turned out to be small as below. This difference is the dissociation constant (k
There is no significant difference in d), which is due to the difference in the coupling constant (ka). That is, since the ka of the 4A12 and 4H11 antibodies is an order of magnitude larger than that of the other antibodies, it can be seen that the probability of association with each epitope is high.

The KAs of the modified BDNF were all increased as compared with the natural type. The rate of increase is 4AA5, 4A1
2,4AB2 and 6B33 antibodies are 4 to 150 times larger, but 4H
The 11 antibody was found to be as low as 1.3 times. This difference is due to the fact that ka is increased in the former and is greatly decreased in the latter. That is, it was found that the 4H11 antibody was more likely to associate with the native epitope than the denatured antibody.

The antibody which is negative by dot blotting and has the maximum KA value is the 4BE11 antibody, which has a KA of 1.35 × 10 ⁷ , and
The antibody having the smallest positive KA value by dot blotting was the 4H11 antibody, and the KA value was 2.17 × 10 ⁷ . That is, when the KA value is 1.35 × 10 ⁷ or less, dot blotting is negative, 2.1
It can be determined to be positive at 7 × 10 ^{7 or more} .

From the above results, 4AA5, 4A12, 4AB2, 4H11
It was confirmed that the binding ability of 6B33 antibody to rBDNF was obviously dependent on its affinity.

By the way, 4AA5 and 4A12 antibodies, 4AB2 and 4H
Why do the 11 antibodies have different affinities for native rBDNF, even though the amino acid sequences of the epitopes are almost the same?

KA depends on the structure of the antigen-binding site of the antibody and the complementarity of the epitope. Since the KAs of the 4AA5 and 4A12 antibodies to the denatured rBDNF are almost the same, it can be seen that the antigen binding sites of both are highly complementary to the denatured epitope. However, the KA of the 4A12 antibody against native rBDNF is about 23 times as large as that of the 4AA5 antibody. It is considered that this is because the ka of the 4A12 antibody was about 20 times larger than that of the 4AA5 antibody, that is, the probability of association with the epitope was high. From this, the epitope of -LEKV- in the natural form is 2
The 4AA5 antibody is difficult to physically approach because it is located between two loop structures, but it is presumed that the 4A12 antibody of the light chain has approached the gap. In addition, the KA of 4AB2 and 4H11 antibodies is
Since the natural type and the denatured type are different, it can be said that the difference is due to the difference in the structure of the antigen-binding site between them. It was confirmed that the former has a denatured epitope and the latter has an antigen-binding site complementary to that of the natural type.

Example 5 The structure of rBDNF prepared from recombinant E. coli was changed to 4AA5, 4A12, 4
Evaluation was attempted by immunoblotting using AB2, 4H11 and 6B33 antibodies as probes.

At this time, in the 4AA5 and 4A12 antibodies, if the former was negative and the latter was positive, the -LEKV- sequence was a natural type, but if the former was positive, it was determined to be a denatured type. If both are negative, the -LEKV- sequence is blocked.
For the 4AB2 and 4H11 antibodies, assuming that the former was negative and the latter was positive, it was determined that the -LEKVPVSK- sequence was of the native type and that of the former was the denatured type. When both were negative, it was considered that the same sequence was partially or entirely shielded. When the 6B33 antibody was positive, the -DSK- sequence was denatured, and when it was negative, it was determined to be the natural type or completely shielded. Therefore, various rBDs with different SS couplings
Dot blotting was performed on NF (E-5 to 9) with the above antibody. The antibodies positive for each rBDNF are shown in Table 3.

[0058]

[Table 3]

Table 3 shows BDNF prepared from recombinant E. coli.
4AA5, 4A12, 4AB for those with different SS coupling
Shows binding of 2, 4H11 and 6B33 antibodies.

The positions of SS bonds in E-5 and E-7A and B have not yet been determined, but it is expected that the SS bonds will be crossed differently from the results of SDS-PAGE and reverse phase chromatography. E-6, 8
Nos. 9 and 9 have known SS bond positions, and A and B have two common SS bonds. These have normal ED ₅₀ values, which are indicators of biological activity.
The ability to induce the elongation of nerve fibers is lower than that with SS bond (0.07 ng / ml), but the biological activity is not completely lost.

E-6A, E-9A and B are all 4AA5, 4AB2
-LEKVPSK-and -DSK-
The sequence of is a modified type. Therefore, the loop containing these sequences was determined not to be constructed. Still, it may have biological activity because another loop was constituted. On the other hand, since E-6B is negative for the 4AB2 antibody and positive for the 4H11 antibody, it is considered that the sequence of -LEKVPSK- is in a state close to the natural type and constitutes a loop structure. And because there is no binding of 4A12 antibody, -LEKV-
It is considered that a part of the area is covered. Also, 6B
Since the 33 antibody is negative, another loop may be constructed. E-8A and B are negative for all antibodies. The epitope of each 4AA5 and 4AB2 antibody is shielded without forming a loop structure, or 4A12 and 4H are formed by forming a loop structure.
11 It is assumed that the epitope of the antibody is shielded. If it is the former, it is close to the state of E-6A, E-9A and B. However, since the biological activity was larger than these, it was determined that the latter was likely.

From these results, rBD with different SS bonds
The magnitude of the biological activity of NF depends on the state of the loop structure.Therefore, by examining the presence or absence of binding of the 4AA5, 4A12, 4AB2, and 4H11 antibodies, the state of the loop structure can be predicted to predict the biological activity of rBDNF. It turns out that the size can be judged.

[0063]

INDUSTRIAL APPLICABILITY According to the present invention, it is possible to determine whether the structure of a specific region, for example, the loop structure is a natural type or a denatured type in a protein produced by a genetic engineering technique or a protein subjected to a refolding operation. The degree of can be evaluated. Further, for diseases caused by protein structural abnormalities, by combining with immunohistological staining of a biological sample, it can be applied to the determination of biological abnormality.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location C12N 5/10 9281-4B C12N 5/00 B

Claims (3)

[Claims] 1. An antibody A that binds to a native region of the protein and an antibody B that binds only to a denatured region of the region R of the protein are prepared, and each antibody is reacted with a test protein. If the binding of the antibody A is positive and the binding of the antibody B is negative, the region R in the test protein is judged to be a natural type, and if the binding of the antibody B is positive, or the antibody A and the antibody B are positive.
A method for evaluating a protein structure by an antibody, characterized in that the region R in the test protein is judged to be denatured when all the bindings are negative. 2. An antibody A that binds to a natural type and an antibody B that binds only to a denatured type are prepared by using a peptide having the amino acid sequence of the region R of the test protein bound to a carrier as an immunogen. How to make antibodies. 3. The test protein is derived from brain-derived nerve growth factor, and the amino acid sequence of the region R of the test protein is -ALT.
The method for producing an antibody according to claim 2, which is MDSKKRIGWRF- or -LEKVPVSKGQLK-.