JPH06502526A - binding domain - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] binding domain The present invention relates to molecules comprising binding domains, methods of their preparation and methods of use. related. The invention particularly relates to immunoglobulin families or superfamilies. A molecule comprising a domain that is a synthetic analog of a natural single-chain variable domain of a member Regarding. The present invention provides methods for designing molecules comprising said domains, and for doing so. molecules designed in the future, and their use in therapy and diagnosis. Regarding the kit and how to use it.

Antibodies and other members of the immunoglobulin superfamily, such as T cell receptors – the ability to specifically recognize molecules such as antigens and bind them with high affinity has. In naturally occurring antibodies, the antigen-binding site is the heavy (H)l and light (L) chain. It is formed by the proximal portion of both variable (V) domains. inside each of those chains includes the portion of the amino acid sequence that comprises the residues that interact with the antigen, i.e., the complementarity determining There are three regions (CDR). These three CDRs are divided into four framework areas (F R) exists alternately. Winter et al. reported that single-chain V domains have high affinity and CWard et al., Nature 3, demonstrated that antigens can be bound with high specificity. 41.544-546 (1989)), their heavy chain domain antibodies (VH ) is smaller in size than the complete antibody (rmm = 160,000) and other antibody fragments. It has been proposed that this would be beneficial for some applications due to its small size.

However, VH domains unfortunately have unique deficiencies that limit their utility. Has a point. The difficulty encountered is that at least two VH domains that may be related Reflects the nature of the inn. They are expressed in small amounts when cloned in bacteria. (approximately 200 μg no-cell supernatant compared to 10 μg of Fv fragment) and VH domain. Significant amounts of material are lost during the purification of in. For example ultrafiltration and chromatography Enrichment of VH-heavy chain domains using purification on a feed column is often performed at low efficiency. cause yield. This is probably due to non-specific binding to the surface.

This was directly observed by Ward et al. (1989) supra, who A high proportion of single-chain domain antibodies exhibiting significant (and non-specific) binding to isolated. In vivo, nonspecific binding in tissues is of interest in tumor imaging and cancer therapy research. would result in lower efficacy in such applications.

For whole antibodies or antibody fragments, such as Fv or Fab fragments, they are generally You will not encounter any difficulties. Therefore, the problem involves unpaired single-stranded domains. It appears to be specific to antibody fragments.

Thus, the present invention provides - To ameliorate these or any other problems related to chain variable domain binding members. The porpose is to do.

Applicant: - The most likely cause of the undesirable properties of heavy chain domain antibodies is The hydrophobic face of a heavy chain variable domain, e.g. a VH-heavy chain domain, is exposed to an aqueous solvent. I understand that there are many things to do. In native antibodies, this face is the proximal hydrophobic surface of the VL domain. It interacts with the sexual surface and is buried inside the antibody molecule. Exposure of this aspect is surface, e.g. the chromatographic matrix, because It will not be possible to recover material from the surface. Moreover, the hydrophobic surface to aqueous solvents exposure causes a decrease in the stability of single chain variable domains, e.g. VH-heavy chain domains. This is thought to lead to denaturation and loss of binding activity during the purification procedure. It will be done. Most importantly, this hydrophobic surface is a potent source of nonspecific binding and This significantly limits the availability of these single-chain variable domain molecules in the body and in vitro. This means that

Although the VH-heavy chain domain is very small, its activity depends on different parts of the molecule. (Chothia, C. et al., J, Mo1. Biol, 186.651-663 (1985)) In some cases, side chains that have important interactions with side chains in the antigen binding site are It is in framework work. For example, the anti-progesterone antibody DB3 It has been shown that fan 47 (framework) contacts progesterone ( Arevalo, J.H., Taussig, M., and 1Vilson, Personal communication], and - hydroxyvitamin-binding antibody rgGINEW has one human has been shown to come into contact with droxyvitamin (Twinning, S. , and Atassi, M. Z., J. Biol. Chem, 253. 5259 (1978)). In some antibodies, residue 71 (framework) and CD CCho, whose interaction with R residues has been shown to be important for maintaining antigen binding. thia, C. et al., Nature 342.877-883 (1989) ), any mutations to the basic antibody or variable domain structure are likely gives pleiotropic results to the three-dimensional structure of variable domains, and allows them to be adjusted to the desired It will be clear to those skilled in the art that specificity and affinity will preclude antigen binding. . This was demonstrated in recent experiments using anti-Suzozyme-heavy chain domain antibodies. It is proven. A single amino acid change substituting Asn 35 with His can be It was found to improve the expression level of 1000-fold. However, changes The molecule was criticized as being too weak to bind zozyme (ε, S, W ard, L., Rej-chmann, G., P. Winter, personal communication).

The interactions responsible for the assembly of the active structure, especially of the VH domain, are not completely understood. Any amino acid changes you make to the basic structure must be very carefully selected. There must be. The residues at the interface between the VH-heavy chain domain and the VL domain are C The fact that you can gain not only from DR but also from the framework area The choice of changes becomes even more complex. Anti-lysozyme antibody D 1.3 (Amit, A, G, et al. 5science 233.755-758 (1986); C hotia.

C1 et al., 5science 233.755-758 (1986)) interface framework residues located at (e.g. residues 37.39.45.47.91.93 and and 103) are highly conserved in antibodies of all species. they are unchanging or, in rare cases, generally conservative substitutions, i.e., fostering similar chemical properties. Has a substitution of an amino acid by an amino acid. For example, aliphatic hydrophobic residues are typically Substituted by a similar aliphatic residue. Thus, if in those framework residues When substitutions are made to the antibody or single chain variable domain, it disrupts the structure of the molecule and thus There is a great risk of disrupting the binding of the antigen by, for example, the VH. CDR residues are anti- It varies between bodies and determines the specificity and affinity of binding to antigen. The applicant has a high affinity with Desiring to retain the ability to bind various antigens with strength and specificity, their CD It is necessary to retain the possibility of changing the R residue.

The present invention therefore provides binding domains with specificity for particular binding members. an analogue of some or all of the natural molecules comprising, compared to the natural binding substance one or more amino acids are changed to reduce the hydrophobicity of said analogue. A molecule comprising a binding domain having a polypeptide sequence that is an analog of I will provide a.

The analogues may differ from the natural binding agent, e.g. with respect to specificity, affinity or binding activity. can have substantially the same bonding properties as the quality.

In some cases, their properties may be improved. Changes are made to sparse polypeptide sequences. Can be any amino acid change that reduces aqueousity, e.g. amino acid substitutions, Can be deletions or additions.

Molecules comprising binding domains include antibodies or other receptor molecules, as well as antibodies. or fragments and derivatives of receptor molecules. In particular, The molecule can comprise a single chain variable domain of the type present in antibody molecules. Ru. Changes may be in the complementarity determining region and/or in the framework region. can. Preferably, changes are made in the framework area. Changes that reduce hydrophobicity are If in a framework region, by amino acid substitution, deletion, addition or inversion. The complementarity-determining region is also modified to change the specificity and/or binding properties of the binding substance. can be changed.

The molecule comprising the binding domain can be modified according to any one of the modifications described in Examples 2-11. or more than one.

Thus, the present invention relates to immunoglobulin family or superfamily members. A single chain variable domain that is a synthetic analog of another single chain variable domain of one or more interfacial amino acid residues of said domain in said another domain; and the modified amino acid is such that the analogue is different from the other domain. members of the immunoglobulin family or superfamily, such that they are also hydrophilic. Provides a single chain variable domain in which a member is substituted by a residue present in a homologous position do.

Altered amino acid residues may be in the framework regions. The modified amino acid residue is It may also be in the complementarity determining region. Synthetic analogs are essentially the same as another domain. can have the same binding activity. Further amino acid substitutions in the complementarity determining region sequence, modified by deletion, addition or inversion of the analogue compared to said another domain. Specificity and/or binding properties can also be altered.

- Heavy chain variable domains are synthetic analogs of single chain variable immunoglobulin heavy domains. can be done. In this case, amino acid residues 37.39°45, 47.91.93 and 103 may be modified.

The second amino acid change can comprise one or more of the following:

i) Substitution of valine 37 by glutamine or threonine: ii) Glutamic acid Replacement of glutamine 39 by: 1ii) Replacement of leucine 45 by glutamine: iv) Replacement of tryptophan 47 by aspartic acid or glycine: ■) replacement of tyrosine 91 by threonine, serine or methionine; vi) Substitution of alanine 93 by serine or glutamic acid: vii) Glutami replacement of tryptophan 103 by phosphoric acid, tyrosine or threonine; vi) Noclin 37 with either asparagine, threonine or serine; Leucine 45, Tryptophan 47, Alanine 93 and/or Tryptophan Replacement of 103; ix) Substitution of valine 37 by threonine and glutamine by glutamic acid Substitution of 39 and substitution of tryptophan 47 by glycine: X) serine or Replacement of tyrosine 91 with methionine and replacement of alanine 93 with glutamic acid Substitution and replacement of tryptophan 103 by threonine.

The single chain variable domains of the invention can further be attached to other molecular components.

Further molecular components may include enzyme, fluorescent or radioactive labels or immunoglobulins. It can be a part of phosphorus.

The present invention also includes the above-mentioned together with one or more auxiliary reagents for performing diagnostic tests. A diagnostic kit is provided comprising a single chain variable domain of.

The present invention also provides therapeutic compositions comprising a small number of single chain variable domains as described above. provide The composition may further comprise one or more excipients.

According to one aspect of the present invention, it has the improved properties as pointed out above and various Specificity for various binding partners by substitution of CDRs with novel frameworks A single chain variable domain, e.g. a VH-heavy chain domain framework that gives rise to provided. Applicant has disclosed that a molecular framework comprising a pre-isolated single-stranded domain provides for substituting framework residues to make them more polar. changed The molecule should retain the ability to bind the desired antigen. Preferably, the change will not render the interface immunogenic when administered to humans.

The invention also relates to other members of the immunoglobulin family or superfamily. Producing single chain variable domains that are hydrophilic synthetic analogs of single chain variable domains 1. A method comprising: (i) probing the interfacial region of the -heavy chain variable domain to determine hydrophobicity; identify the amino acid residue, and (ii) one or more of the hydrophobic residues are of the immunoglobulin family or by less hydrophobic residues present in homologous positions of superfamily members. producing said analogue of said heavy chain variable domain of (i) which is substituted; Provides a method comprising:

The method includes: (a) a nucleotide encoding one or more of the identified hydrophobic residues; get the code array; (b) altering said nucleotide sequence using site-directed mutagenesis to obtain alternative introducing a triplet encoding a amino acid; (C) modified nucleotide sequence; is used in a recombinant expression system to express synthetic analogs. It can consist of:

In this method, multiple amino acid residues can be substituted. Alternative amino acids are can be derived from naturally occurring monomeric members of the epiglobulin superfamily. Wear. The natural monomeric member may be rhy-t. Synthetic analogs are as described above. It can have essentially the same binding activity as another domain.

A molecule comprising a binding domain or an immunoglobulin or a fragment of an immunoglobulin or derivatives, the amino acid sites suitable for modification are identified as follows: i) for hydrophobic amino acids expected to be on the surface of the binding substance; to investigate the molecule: ii) It is predicted that it is buried at the interface between the heavy chain domain and the light chain domain of immunoglobulin. Examine the amino acid residues in more detail with respect to what is expected; ii) Residues identified in the -heavy chain domain may be exposed to solvent when used separately. Find out.

Suitable changes to make can be made by reference to homologous amino acid sequences of members of the family of related substances. It can be identified by For example, at the change site identified as described above, The amino acid sequence of a molecule that represents its location within one or more members of a family can be changed to be homologous to

A family of related substances includes the immunoglobulin family, its fragments and derivatives. can consist of Alternatively, a family of related substances may include immunoglobulins A layer of proteins containing domains that are structurally related to the immunoglobulin can comprise a par family.

Nucleotide sequences are prepared using oligonucleotides designed to introduce the desired changes. It can be modified by site-directed mutagenesis using otide. or , the modification can also be accomplished using a technique known as polymerase chain reaction. Ru.

The invention also provides molecules comprising a binding domain as provided herein. Also included are kits with the invention. The kit is a diagnostic, purification or catalytic kit. Can be done. The invention further provides a molecule comprising a binding domain according to the invention. This includes drugs that are

It specifically binds to a particular steric and polar structure of another molecule and thus refers to the regions or cavities on the surface of a protein that are defined as complementary . A domain can exist both within itself and independently of other parts of the same protein. Folds independently of complementary binding members.

immunoglobulin It consists of two sets of heavyweight polypeptides all linked together by disulfide bonds. a layer of structurally related proteins consisting of a polypeptide and two light chain polypeptides . They are different molecules in which a given immunoglobulin binds specifically to another molecule. Contains binding domains for molecules.

The protein can be naturally occurring or partially or wholly synthetic. It can also be manufactured. This term is homologous to the binding domain of immunoglobulins. Also includes any protein that has a binding domain.

Here, the numbering of immunoglobulin amino acid residues is based on Kabat, E, A, 5equences of Proteins of Immunolo gical Interest'U, S, Department of He alth and) Iuman 5 services 1987.

antibody This refers to immunoglobulins that are naturally occurring or partially or wholly synthetic. cormorant. This term has a binding domain that is homologous to that of immunoglobulins. It also includes any protein that Those proteins must be derived from natural sources It can also be partially or totally synthetic.

Examples of antibodies are immunoglobulin isotypes and Fab, F(ab')z, 5c Fv.

Fv, dAb, and Fd fragments.

immunoglobulin superfamily This is because the members are related to the structure of the variable or constant domains of immunoglobulin molecules. A layer of a polypeptide that has at least one domain with a structure similar to that of a polypeptide.

The domain contains two β-sheets and usually one conserved disulfide bond. (A, F, Williams and A, N, Barclay 1988 (See Ann, Rev. Immunol, 6381-450).

Examples of members of the immunoglobulin superfamily are CD4, platelet-derived growth factor receptor (PDGFR) and intercellular adhesion molecule (ICAM).

Unless the context specifies otherwise, immunoglobulins and immunoglobulins in this application References to globulin analogs refer to the immunoglobulin superfamily and their analogs. Includes members of the body.

interface This refers to the specific heavy or Refers to the region on the light chain.

Framework The heavy chain of an immunoglobulin has a constant (C) region and a variable (V) region. Each V area is , three complementarity-determining regions separated by four framework regions (FRs) ( CDR). CDRs are variable regions of amino acid sequence and other provides the ability to bind to other molecules. provides various binding specificities to immunoglobulins. There is some variability. FR is a section of substantially constant amino acid sequence between CDRs. Ru.

In order to understand the invention more fully, firstly, a general overview, and secondly, simply By reference to specific examples, which are intended to be illustrative and not limiting, I will now explain the details in more detail. The following description refers to the drawings.

Figure 1 shows the VH of anti-lysozyme antibody D1.3 cloned into pUc119. Figure 2 shows the nucleotide and amino acid sequences of the domains: Mutant oligonucleotides for substitution of residues found in epiglobulin heavy compounds are shown. Figure 3 shows the mutant oligonucleotide for substitution with homologous residues from rhy-t. Shows Chido: Figure 4 shows the mutant sequence obtained by substitution of the Thy-1 residue in VHDl,3. death.

Figure 5 shows the random placement of asparagine, serine or threonine into VHDl,3. Oligonucleotides for conversion; Figure 6 shows TGI (control), VHDl , 3, VHTHY-1 and VHTHY-2 (7) Glue representing lysozyme binding activity Rough; Figure 7 shows p[Jc119 (control), VHDl, 3, VHMutTrp, V This is a graph showing the lysozyme binding activity of HMutLeu and vH'rhy-3. Figure 8 shows pUc119 (control), V) ID1.3, vH’rhy-i, Thy-2, V) IThy-1, Thy-3 and VHMutWD lysozyme FIG. 9 is a graph representing binding activity; FIG. The nucleotide sequence of Ps/Bs is shown and the Pstl and BstE11 restriction regions are shown. and Figure 10c, vector fdPs/Bs (control’), rctv H’rhytThy2, fdVHDl, 3 and phage antibody D1.3 lyso It is a graph showing team binding activity.

The applicant has developed three related strategies for selecting changes to make to the framework. I made it clear. This invention has improved properties for in vivo and in vitro applications. This allows for the production of antibodies and single chain variable domains such as VH domains.

There is considerable amino acid sequence homology between different immunoglobulins. homology is different Line up the arrays one by one and reach the maximum level of matching between different arrays determined by sliding the strands with respect to each other until . Their analysis is Usually done on a computer. As mentioned above, framework residues are highly that is, the presence of a particular amino acid at the same position in a series of different antibodies. There will be. However, since rare substitutions occur, Applicants have determined that the VH interface remains Naturally occurring substitutions of the groups were investigated. This is because readily available antibody sequences [e.g., Kabat, E. A. et al., “5 sequences of Proteins of Immunological Interest' U, S, Department of Health and Huma n5 services (1987)). This analysis , allowing the identification of natural mutations at any position. Those mutated residues in antibodies Since the groups are naturally occurring (potash, although rare), the applicant believes that they are probably domestic. It was recognized that this would not seriously destroy the internal structure. Their natural replacements are antibodies Most frequently occurring once per molecule. However, the applicant Substitutions available from several antibodies are combined together into the same molecule. natural substitution Various combinations are possible.

Z. Homology in another protein of the immunoglobulin superfamily is of particular interest. The immunoglobulin fold (antibody domains found in other proteins) It has a domain that contains a three-dimensional structure unique to the protein, but does not associate with another domain. There is no protein.

Examples of molecules containing single chain domains homologous to immunoglobulin variable domains include: Thyl, Po minirin, CD7, CD28 and CTLA-4 are mentioned. (Williams, A.F. and Barelay, A.N., An. n, Rev, [mmunol.

6, 381-405 (1988)).

Other proteins contain multiple unpaired antibody-like domains. CD4 and MRCOX2 are each has an N-terminal domain homologous to the variable domain of the antibody molecule (one in MRCOX2 , two ``V''-shaped donodomains in the CDJ, a C homologous to the constant domain of the antibody molecule, Terminal domain (MRCOX2 and CD4 each contain one °C-shaped donodomain) (Williams, A.F., and Barclay, A.N., Ann , Rev. Immunol, 6.381-405 (1988)).

the kind of three-dimensional structure between those unpaired domain proteins and the domain-containing antibodies; Similarity is partly reflected in homology at the amino acid sequence level. In this case the array Although the alignment may be uncertain, the amino acid homology is almost the same as in 1 above. given by law. For 'rhy-i, VH residues 37.39.91 and 9 3 is relatively easy to align. However, VH residues 45 and 47 are Th reported in two different alignments with y-i (A, F, Will iams and J.

If there are possible alignments, another search for permutations is necessary to identify the most appropriate one. Might happen.

The exposed modified VH interface in the VH single-chain domain may be antigenic to humans. be. This would be disadvantageous for therapeutic use in vivo. Human Ttty-1 Does substitution with naturally occurring residues in Maybe.

& Hand-sawn insertion of polar amino acids at interfacial residues This strategy is less direct. This allows for the improvement of single-stranded domains using non-conventional methods. Oligos for mutagenesis Provides ambiguity for the insertion of amino acids into triplets consisting of nucleotides. To achieve this, a mixture of bases at several positions is used. For example, the applicant , by using an ambiguous codon in the second position, the highly polar residue As paragine, serine or threonine; devised. These residues can be found, for example, at interfacial positions 37.45.47. of the VH domain. 93 and 103. Then, resulting from this mutagenesis We screened 243 possible frameworks for Those possessing the desired properties can be identified. The resulting semi-thermal randomized population was screened. The strategy for leaning is to evaluate antigen binding affinity and non-specific binding by ELISA. (see below).

This is a hands-on approach without resorting to the strategies outlined in 1) and 2) above. is one example of the many strategies that can be used to change interfacial residues in Eq. .

The present invention provides single-stranded drugs with improved properties for in vivo and in vitro applications. The binding affinity and specificity of the identified antibody can be incorporated into the main molecule. Ru.

Produced using a model system using the VH domain of antibody D1.3 (Vl(01,3)) The developed framework can be derived from antibody molecules of desired affinity and specificity. Substituting part or all of its CDRs for antibodies of any specificity. It can be used as a mouse. There are many ways to achieve this . For example, after determining the sequence of an antibody's CDRs of desired properties (e.g. binding specificity) , a frame containing the necessary nucleotide substitutions to make the antibody/antibody domain more polar. Oligonucleotides and series of oligonucleotides encoding framework regions and designated CDRs Synthesize gonucleotides. This oligonucleotide is then amplified using PCR. clone into a suitable vector, e.g. IIUC119, and transform it into a bacterial The product can be expressed inside.

Alternatively, appropriate at the interface into existing single chain domain antibodies with the desired specificity. Modifications can be introduced to improve the properties of the antibody. This can be done in various ways by, e.g., site-directed mutagenesis, or by PCR (e.g., Hem5ley , A, et al [Nuc, Ac1ds Res, 16°6545-6551 ( (1989)].

In addition to this, improved single chain domain antibodies exhibit reduced non-specificity. to this The properties of single chain variable domain antibodies provided by single-stranded domains of desired specificity and affinity. Allows selection of main antibody. The framework described in this invention can be clone into a repertoire of CDRs and replace existing CDRs with a repertoire of CDRs. A new population of single-chain domain antibody molecules can be created by Can be screened for desired binding specificity. Or for example , single chain domain antibodies as described by Ward et al. (1989, supra). Although they can be isolated, their affinity and specificity may require improvement.

Cloning the CDRs from those antibodies into the polar framework described in this invention by inserting it into the single-stranded domain cloned into fd phage. Wear. Next, we perform random mutagenesis of those CDRs and use the affinity method. can be used to select antibodies of desired affinity and specificity.

Examples 15 and 16 herein demonstrate that V with a more polar framework A derivative of HDl, 3 was added to gene ■ protein (McC) while retaining its binding activity. affinity, J. et al., 1990, Na, ture 348. p552-5 54) and can be displayed on phage. Table on phage Shown are combinations of substitutions at framework residues by deliberate mutagenesis. Preparation of the fibers (an example of which is given in Example 11) and subsequent preferred bonding characteristics Enables selection of gender.

The improved single chain domain antibodies of the present invention are superior variations of conventional single chain domain antibodies. The structure of immunoglobulins (Igs) and their molecules or fragments As with the superfamily, it can be used in a number of ways. For example, 1g The child may be diagnosed with research, therapy (e.g. cancer therapy, immune status plateaus, and pathogen-induced diagnosis (e.g. assessment of hormonal status), hormonal or growth Plates, detection of activity of factors, biosensors, catalysts, purification of other molecules, and pharmaceuticals Used in screening operations for therapeutic compounds in industry . The low non-specific binding of improved single-chain domain antibodies makes them particularly useful for the above applications. You'll understand.

Increased hydrophilicity can be achieved by affinity chromatography, especially weak affinity Chromatography (Zopf, D., and Ohlson, S.).

Nature 346.87-89.1990). This may be particularly important for the use of

Anti-idiotype-improved single chain domain antibodies can also be made.

Anti-idiotype specificity (Methods Enzymol, 178. J, J , edited by Langone, Academic Press (1989) ) is prepared in a two-step process. First, directed against a specific antigen or epitope Antibody A thus obtained is used as is to raise another antibody. Part of the anti-A antibody Antibody B binds antigen to an antibody such that the antigen binding site of B is complementary to that of It will be directed against the body part. In short, antibody B, which is an anti-idiotype antibody, The antigen, binding site constitutes the original antigen or epitope recognized by antibody A. to imitate. The original antigen may be a protein or any other compound, such as carbohydrates. can be a chemical or steroid and is used at any stage of the operation The antibody can be a modified-heavy chain domain antibody. Final anti-idiotype anti The body may also be an improved heavy chain domain antibody produced as described herein. improved anti-idiotypic determinants - heavy chain domain antibody frameworks It can also be a molecule of the immunoglobulin superfamily that is transferred into the immune system.

Such anti-idiotypic molecules are advantageous for a variety of applications (Methods Enzymol, 178. Edited by J. J. Langone, Academic Press (1989): l. These include cancer treatment, bacteria, and viruses. vaccines for the treatment of diseases caused by viruses and parasites. Ru. They block cellular receptors for the pathogens mentioned above, as well as Can be used to block cellular receptors for hormones. So They can be used as substitutes for antigens and peptides in diagnostic procedures, e.g. in ELISA. It is also beneficial. Anti-idiotype specificity is useful in the pharmaceutical industry is known (Methods Enzymol, +78. J, J, La ngone IN, Academic Press (1989)).

The present invention relates to modifications derived from the immunoglobulin ([g) superfamily of molecules. Good-heavy chain domain antibodies and receptors, methods for selecting and implementing said improvements, and said antibodies in research, therapy, diagnosis, purification, catalysis and development of new therapeutics. and the use of receptors and kits therefor.

Example 1 More polar framework by replacing valine 37 with glutamine Preparation of VHDl,3 with pUc1 pUc1 used for mutagenesis experiments. 19 VHDl, nucleotide and amino acid sequences of 3 clones are shown (War d et al., 1989. (cited above).

The amino acid residues in the VH domain that interact with VL are 37.39.45.47° 91.93 and 103 (Amit et al. (1986) supra, C hotia.

C9 et al. (1986) supra]. Compilations of immunoglobulin sequences [e.g. Kabat .

E, A, et al. “5 sequences of Proteins of rmm unological 1nterest", U, S, Department of Health and Human 5 services (1987) ) was used to investigate naturally occurring amino acids in VH domains. natural heavyweight Substitutions were found at positions 37.39.45.47.91.93 and 103 in It was. We are currently working on replacing interfacial residues by mutagenesis in the next example. Select the most polar substitutions at each of those residues among the antibodies that have been sequenced in Ta.

Residue 37 is 38'5 valine out of 434 sequences investigated. Other 48 placements In the sequence, an aliphatic amino acid was substituted. One remaining example, a major human disease In this case, it was replaced with glutamine, a significantly more polar residue. VH Oligonucleotides (described in Figure 2) for the incorporation of this mutation into Dl,3 VHMLITVAL) was designed.

Assembling this mutation and other mutations described in the Examples below in various combinations Further new derivatives can be created.

In vitro mutagenesis (1) Before in vitro mutagenesis, apply the oligonucleotides described in Figure 2 to App. Synthesized on a Lied Biosystems 391 DNA synthesizer and performed using standard techniques. Technique (Sambrook, J., et al. “Mo1ecular Cloning: a 1 Laboratory Manual (2nd Edition)”, Co1d Sp ring Harbor Laboratory Press, 11.23) was purified on a urea-acrylamide gel using

(2) Preparation of single-stranded DNA template for mutagenesis. used to illustrate the invention VHDl, 3 antibody gene (Ward et al., +989, supra) on plasmid pUc 119 (Sambrook, J. et al. Mo1ecular Clonin g:　a　laboratorymanual　(2nd edition)　”,　Col1d Spring Harbor Laboratory Press.

1.14) Supported on. Standard techniques for propagation and purification (Sambroo k, J, et al Mo1ecular Cloning: a1aborator y manual (2nd edition)”, Co1d Spring Harbor Laboratory Press, 4.46) to transform the plasmid into By staining the containing TG1 cells with MI3 KO7O7 helper phage. A single-stranded template DNA was prepared.

(3) “In vitro Mutagenesis S” according to the manufacturer’s instructions. system, Oligo-nueleotide directed I[' ” (Amersham International) Mutagenesis was performed. ampicillin-resistant colonies resulting from mutagenesis, 2YT containing 100 μg/I of ampicillin (2YT = Bacto per 1 jl of water) - grown overnight in 16 g tryptone, 10 g yeast extract, 15 g NaC) . The cultures were diluted in fresh 2YT and M2S similar to (2) above. Single-stranded template DNA was prepared by KO7 infection. 5equenase 2.0 The mutant was confirmed by DNA sequencing using a type kit.

Gluta to position 39 using logic and methods similar to those described in Example 1. Decode the ribonucleotide VHMUTGLN (Figure 2) for the introduction of mic acid residues. I signed it. This substitution is found in 1 of the 420 stalwarts investigated. Group Rutamic acid is thought to be slightly more polar than glutamine. CRose et al. , 5science 229.834-838°1985).

Using logic and methods similar to those described in Example 1, the leucine group at position 45 was Oligonucleotide VHMUTLEU (Figure 2) introducing the substitution with rutamine Designed. This change was observed in two of the 402 sequences examined ( 396 has a leucine at this position).

One of the antibodies containing glutamine at this position is anti-old, specific for 6D-galactan. It is a mouse antibody.

Example 4 More polarity due to substitution of tryptophan 47 with aspartic acid Preparation of VHD1,3 with a unique framework Similar reasoning to that described in Example 1 oligonucleotide for the introduction of aspartic acid at position 47 using Creotide Vl (M[JTWI) (Figure 2) was designed. This replacement is It is recognized as one of the 392 important figures.

Example 5 More polar frame by substitution of tyrosine 91 with threonine VHDI with work. Similar logic and method to that described in Preparation Example 1 of 3. Oligonucleotide VH for the introduction of threonine at position 91 using the method I designed MUTTYR. This substitution was one of 398 stalwarts investigated. recognized.

Example 6 More polar framework by substitution of alanine 93 with serine Preparation of VHDl, 3 with a similar logic and method as described in Example I. Oligonucleotide VHM to replace alanine at position 93 with serine using We designed UTALA (Figure 2). This substitution was found among the 410 heavyweights investigated. It is recognized in four cases. One of them is in mouse anti-B2.1 fructosan .

Preparation of VHD1,3 with a more polar framework by substitution Using similar logic and methods as described in Example 1, group to position 103. Two oligonucleotides VH for the introduction of tamic acid and tyrosine respectively We designed MIJTTRP and VHMUTWY (Figure 2). These substitutions were found once in each of the 308 heavyweights surveyed. Glutamic acid is tryptic acid It's much more polar than the fans.

Although tyrosine is more polar than tryptophan, it is a conservative substitution. Ru.

rhy-t is a single-chain domain protein of the immunoglobulin superfamily. be. Alignment of rhy-t residues with immunoglobulin heavy residues was performed by William (A, F, Williams and J, Gagnon .

5science　216.696-703. 1982: A, F, Will liams and A.N. Barclay.

Ann, Rev. Immunol, 6.381-405.1988). both In the publication, residues 37.39.91 and 93 of the VH domain are identical to Thy-1. residues, but the residues at positions 45 and 47 are aligned with different residues. , reflecting the low homology of the adjacent amino acids at those positions. Will Using the alignment published by Iams & Gagnon (1982) supra, Introducing the most polar residues at positions 37, 39 and 47 found in the homologous Thy- Oligonucleotide VHTHY-1 (Figure 3) was designed to do so. Example Mutagenesis was performed in the same manner as in 1. The resulting amino acid substitutions are shown in Figure 4.

Substitution of residues at positions 91, 93 and 103 to create a more polar spectrum Preparation of VHDl, 3 with a framework using the method described in Example 8. Incorporating maximum polar substitutions of VH residues 91°93 and 103 at the by-1 residue We designed the oligonucleotide VHTHY-2 (Figure 3) for this purpose (Kabat , E, A, et al., “5 sequences of proteins of [mm unological Interest", U, S, Department of Health and Human 5 services (+987) ) o At residues homologous to the VH interface, place a label that is considered to be the most polar overall. Another oligonucleotide was used to incorporate residues found at the same position in the rat brain’rhy-+. Reotide VHT) IY-3 (Figure 3) was designed. Mutagenesis is described in Example 1. As described. The resulting amino acid changes are shown in Figure 4.

Example 10 Homologous residues from immunoglobulin family protein Thy-t more polar by substitution of residues at positions 37, 39, 47, 91, 93 and 103 of Preparation of VHD1 with a large framework of 3 mutant proteins as a template VHTHY-1 DNA sequence and mutation ori- ties for incorporating site-directed mutations Examples using the oligonucleotides VHTHY-2 and VHTHY-3 (Figure 3) By performing mutagenesis experiments as described in Example 1, the results are as detailed in Examples 8 and 9. combined amino acid changes.

Example 11 37.45.47°9 with asparagine, threonine or serine with a more polar framework due to substitution of residues at positions 3 and 103 Preparation of VHDl,3 37, 39.45.47.91.93 and 103rd position each with triplet GX T (where X is any mixture of bases C, G and T: Figure 5) Prepare the nucleotides. As a mutagenic primer as described in Example 1 Their use in Ser, T, respectively, depends on which base is incorporated. This will result in the insertion of hr and Asn. The resulting derivative is then subjected to antigen binding and and improved properties.

Example 12 VHTy-1 and V) fThy-2 mutant single chain domain antibodies Evaluation of antigen binding status VHDl, 3-interface mutant VHT produced as described in Examples 8 and 9 hy-1 and VHTy-2 were evaluated for lysozyme binding activity. mutation The antigen binding state of single chain domain antibodies is determined by EL[SA It was measured by (enzyme linked immunosorbent assay).

This covers a wide range of methods that can be used to measure antigen-antibody binding. It's just one of them. Others include Western blotting, competitive Examples include combined radioimmunoassay and fluorescence quenching.

ELISA for lysozyme binding by mutant single-stranded domains was performed as follows. did.

An overnight culture of ampicillin-resistant clones was grown with fresh 2YT (100 μg/ml aliquots). (containing ampicillin) and grown for 1 hour at 37°C. stomach Sopropyl-β-D-thiogalactopyranoside ([PTG) at a final concentration of 1 mM and cultured the cells for an additional 24-30 hours at 37°C. EL[SA directly Supernatant was prepared by centrifugation for use.

l) Plate (Falcon Microtest ■Flexible Plate), 50 2ooL of Img/ml chicken egg lysozyme in mM NaHCO+, pH 9,6 Coated overnight at room temperature by tl/well.

2) Rinse wells with 3 washes with phosphate buffered saline (PBS); 00μE/well of 2% skim milk powder in PBS for 2 hours at 37°C. Blocked.

3) Rinse the wells by washing 3 times with PBS and add 200 μl of culture supernatant. and incubated for 2 hours at room temperature.

4) 0.05%T+veen 20/PBS 2 times, PBS 3 times, U The well was washed.

5) Rabbit polyclonal against Fv fragment in 2% skim milk powder/PBS Appropriate dilution of antiserum (1:1OOO) 200μ! to each well and at 2 o'clock Incubated for a while.

6) Repeat the washing as in (4).

7) Horseradish peroxidase conjugated goat in 2% skim milk powder/PBS Appropriate dilution (1:5000) of anti-rabbit antibody (Sigma) 200μ! each wells and incubated for 1 hour at room temperature.

8) (Repeat washing as in 41.

9) 200μβ of 2,2'-azinobis(3-ethylthiazolinesulfonic acid) ) [Sigmal (contains 30% hydrogen peroxide/water of luI! per 10ml) Citrate buffer (Citrate buffer is 54ml of 50mM quench per 100d) (containing 50mM trisodium citrate of +46-) ml! ) was added to each well and allowed to develop for up to 10 minutes at room temperature.

By adding 0.05% sodium azide in 50mM citric acid pH 4.3 The reaction was further stopped. Titertek Multiskan M, C, inside Read the ELISA plate at 405 nm and take the optical density reading in each well. gave. The optical density reading is based on the primary antibody used in the ELISA (in this case, single chain domain antibody) and is proportional to the amount and affinity of the antibody.

The results shown in Figure 6 indicate that these mutants retain the ability to bind lysozyme. prove that. The vHrhy-i mutant has a higher lysovirus than the parental VHDl,3. has an affinity/amount of team-binding activity, whereas mutant VHTy-2 has slightly lower I think that the.

Given the complexity of the problem (see above), this result is surprising; As taught in the specification, VH single-chain dosing can be achieved without unduly impairing antigen binding. This strengthens the argument that the properties of the main antibody can be surprisingly improved.

Example 13 VHMutTrp, VHlliutLeu and VHTy-3 Evaluation of antigen binding status of mutant single chain domain antibodies VHDl, 3 interface mutants prepared as described in Examples 3, 7 and 9, respectively. Lysozyme binding of VHMutLeu, VHMutTrp and VHTy-3 The activity was evaluated. The antigen binding state of the mutant-heavy chain binding domain is shown in Example 12. Measured by ELISA as described. The results shown in Figure 7 indicate that those mutants Proving that it retained the ability to bind lysozyme.

Example 14 VHTy-1, Thy-2: VHTy-1, Thy-3 and Evaluation of antigen binding state of VHMutWD mutant single chain domain antibody VHDl, 3-interface mutant VHThy (, T hy-2: VHThy-1, Thy-3 and VHMutWD wo'Jzochy The membrane binding activity was evaluated. Mutations - Heavy chain binding domain Determines the antigen binding status of antibodies. Measured by EL (SA) as described in Example 12. The results shown in FIG. demonstrate that the mutant retained the ability to bind lysozyme.

In this way, the polarity of the domain can be modified without affecting its ability to bind lysozyme. A wide range of changes (6 amino acid substitutions) can be made to the VH interface that increase the . The approach used to select substitutions as taught herein does not apply to any VH It is expected that it can be applied to domains as well.

Assemble the DNA sequence of the mutant protein VHTy-1 as a template and site-directed mutagenesis. carried out by using the mutated oligonucleotide VHTy-2 to The genes encoding the derivatives VHTy-1 and Thy-2 produced in Example 10 were Gene ■Display this VH domain on the phage as a fusion with the protein For this purpose, it was subcloned into the vector fdPs/Bs. Vector fdPs/ Bs contains Pstl and BstB11 restriction sites for cloning Other than fdcATl (MeCafferty, J, et al.

1990, Nature 348. p552-554) is the same as (Figure 9 ).

pUc119 VH using standard methods (Sambrook et al., 1989, supra). A mini preparation of Thy-1, ThY-2 DNA was prepared. Primer RVHTH VHTy-1 and Thy-2 were isolated by PCR using YFOR and KSJ6. The code sequence was amplified.

RVHTHYFOR 5' TGA GGA GACGGT GACCGT GGT GCCTTG GCCAGT G 3' This is the 3' end f of VHTy-1, Thy-2 gene , :Introduces the BstE11 site.

5J6 5' AGG TGCAGCTGCAGG AGT CAG G 3' This is V A Pst site is introduced at the 5' end of the HThy-4,7hy-2 gene.

PCR was performed using 20mM Tris (pH 7 at 70°C, 3).

50mM KCl, 4mM MgClz, 0.01% gelatin, ioμ Each oligonucleotide, 1mM each dNTP, 5 units of Taq polymerase and about 50 ng of pcU119 VHTy-1, Thy-2 DNA. It was carried out. The PCR reaction product was ethanol precipitated and 20 u1. l of Resuspended in omM Tris, pH 8, 0, 0, 1 mM EDTA. lOμ Part m was mixed with PstI (20 units) and B in NEB buffer 2 in a total volume of 50μβ. Digested with stEl (2o units) at 37°C for 2 hours (restriction enzyme was Ne w England Biolabs, CP Labs, B15hopsS (obtained from tortford). After digestion, the reaction mixture was phenol-extracted and Tanol was precipitated. The product was transferred to a 1% agarose-Tris acetate-EDTA gel. Electrophoresis was performed on a Genecle gel, and a band of approximately 350 bp was cut out. using an (Bio 101, La Jolla, Ca1ifornia) The DNA was purified. using standard methods (Sambrook et al., 1989, supra). Vector DNA (fdPs/Bs RF type) was prepared. This DNA (1 , 2 μg) of Pst[ and BstEl (50 μg) in 100 μf NEB buffer 3. Unit) for 90 minutes at 37°C. The product was extracted with phenol and The nol-precipitated and resuspended DNA was prepared by Sambrook et al. (1989, Phosphatase treatment was performed as described by (supra). 0.7% T for preparation ris-phosphate-EDTA-agarose electrophoresis was performed, and a band of approximately 9 kb was detected. Excise, purify the DNA using Geneclean, and incubate in 10 ttl. Resuspend in omM Tris, pH 8, 0, 0, 1mM EDTA. Digest Use 5μβ each of the vector and insert fragment DNA to make 10μ! NEB ligase buffer Ligation was performed using 200 units of T4 DNA IJ gauze. This concatenated mixture (Prepared according to Sambrook et al. (1989, supra) using an 8 μl sample. E. coli (E. coli) MC1061 cells were transformed; Plate the mixture on 2YT agar containing 20 μg/ml tetracycline. did. Collect a colony and make one! To prepare DNA, Sambrook et al.

1989, supra), 5equenase 2.0 kit (United 5t ates Biochemical.

DNA was sequenced using C1leveland, U,S,A,). Insertion disconnection The sequence of the fragment matched VHTy-1 and Thy-2. This derivative was fdVHTh They were named y-1 and Thy-2.

Starting from pUc119VHD1.3 in TefdPs/Bs (7) VHDl, 3(7) Clones were prepared. psWl-VHD in Pstl and BStEl l, by digestion of 3-TAGI (Ward E, S, et al., 1989, supra). An insert encoding VHD1,3 was prepared. Other steps are the same as above. It was. This derivative was named fdVHDl,'3.

The fdVHTy-1, Thy-2 phages produced in Example 15 were tested by ELISA It has been proven that the antigen lysozyme is functional in binding to the antigen lysozyme. It was.

fdVHTt+y-1, Thy-2; fdVHDl, 3: Thy-antibody 01 ．． 3 (Display 5cFvD1.3 + McCafferty, J. et al., 19 90. Nature 348, p552-554) or fdPs/Bs. E. coli MC1061 cells were incubated with E. coli MC1061 cells containing 2 Prepare virus particles by growing for 16-24 hours in YT medium 50- did. 8X501nI! Centrifugation for 10 minutes at 11000 orp in rotor Culture supernatants were collected by separation. 115 volumes of 20% polyethylene glycol (PE G) / Add 2.5M NaCl and leave at 4°C for 1 hour to remove phage. The particles were allowed to settle. Phage particles were isolated by centrifugation for 15 minutes as above. Pellet the pellet and add sterile 10mM Tris containing 1% gelatin. , pH 8,0, llIIM in EDTA to 1/40 of the original volume. Resuspend.

1. Coat the ELISA plate with lysozyme as described in Example 12. and blocked with PBS containing skim milk powder.

Z wells were washed with PBS.

3. Add concentrated phages (200 μl) to each well as appropriate and incubate for 2 hours at room temperature. It was incubated.

4. Wash the wells 3 times with 0.5% Tween 20/PBS and 3 times with PBS. Ta.

Sheep anti-M13 serum (200 μm) in 1 f PBS containing 5.2% skim milk powder :1:1000) was added to each well and incubated for 1 hour.

Washing was repeated as in 684.

7. Peroxidase-conjugated rabbit anti-goat immunoglobulin (200 μl; 1:5 000: Sigma) was added and incubated for 1 hour.

Washing was repeated as in 8.4.

9. Add peroxidase substrate as described in Example 12 and perform one-time development. Closed.

Both fdVHTy-1, Thy-2 and fdVHDl, 3(7) are fdPs /Bs gave an ELISA signal 4-5 times higher than the value obtained using , when using phage antibody D1.3, the signal was obtained using fdPs/Bs. This value was approximately 8 times that of the actual value (Figure 10). Therefore, changes in VH interface residues may Does not affect the ability of the domain to bind lysozyme when displayed above Ta.

Although the invention has been described by way of example only, without departing from the scope of the invention, It will be appreciated by those skilled in the art that similar results can be achieved with substantial modifications to the method. It would be obvious to anyone.

+30 140 150 160170 1elO 10 203303403 50360 mi ← lid 1 ：Riself Mi・Eye → Bow =・U　Mi・l ;110th =03 Sha・8・8; Bow・Tongue・( Mortality g0 卜 ν耘←阜　pl・eyes indigo ELISA of mutant VH domains F also J℃go.

ELISA for VH variant domains Gene■ fdVHTy-1, 7hy-2 and control ELISA Iwl-, -No　ρ Continuation of CT/GB 91101253 front page (51) Int, C1,5 identification symbol Internal reference number C07K 13100 8517-4HC12N 15/12 //(C12P 21102 C12R1:19) CO7K 99:00 (81) Specified times EP (AT, BE, CH, DE.

DK, ES, FR, GB, GR, IT, LU, NL, SE), 0A (B F, BJ, CF, CG, CI, CM, GA, GN, ML, MR, SN, TD, T G), AT, AU, BB, BG, BR, CA, CH, C3, DE, DK.

ES, FI, GB, HU, JP, KP, KR, LK, LU, MC, MG, MN, MW, NL, No, PL, RO, SD, SE, SU, US I (72) Inventor: Jackson, Ronald Henry, UK, Poppy Bridge CP12NU, Kixton Street (72) Inventor Chiswell , David John, United Kingdom, Patskingham, M.G. 182 L.D., MI Dollar Clayton, Sand Hill House 1

Claims (18)

[Claims] 1. Another single strand of a member of the immunoglobulin family or superfamily A single chain variable domain that is a synthetic analog of a variable domain, in which the former one or more interfacial amino acid residues of said domain are altered compared to said another domain; wherein the modified amino acid is such that the analog is more specific than the other domain. Immunoglobulin family or superfamily, as they are more hydrophilic said single-stranded variable domain, wherein said single-stranded variable domain is substituted by a residue present in a homologous site of a member of hmm. 2. 2. The method of claim 1, wherein the modified amino acid residue is in a framework region. Full chain variable domain. 3. Claim 1 or Claim 2, wherein the altered amino acid residue is in a complementarity determining region. A single chain variable domain as described in. 4. Claim wherein said synthetic analog has essentially the same binding activity as said another domain. The single chain variable domain according to any one of Items 1 to 3. 5. altering the specificity and/or affinity of said analog compared to the native domain; amino acid substitutions, deletions, additions, or inversions to The single-stranded variable domain according to claim 2 or claim 3, wherein the acid sequence is further modified. in. 6. Claims 1-5 which are synthetic analogs of single chain variable immunoglobulin heavy chain domains. Single chain variable domain according to any one of. 7. one of amino acid residues 37, 39, 45, 47, 91, 93 and 103 or 7. The single chain variable domain of claim 6, wherein a plurality of has been altered. 8. Amino acid changes i) replacement of valine 37 by glutamine or threonine; ii) glutamic acid iii) replacement of leucine 45 by glutamine; iv) replacement of tryptophan 47 by aspartic acid or glycine; V) replacement of tyrosine 91 by threonine, serine or methionine; vi) substitution of alanine 93 by serine or glutamic acid; vii) glutami replacement of tryptophan 103 by phosphoric acid, tyrosine or threonine; viii) Valine 37 with either asparagine, threonine or serine , leucine 45, tributophane 47, alanine 93 and/or tryptophan Replacement of fan 103; ix) Substitution of valine 37 by threonine and glutamine by glutamic acid substitution of 39 and substitution of tributophane 47 by glycine; x) serine or Replacement of tyrosine 91 with methionine and replacement of alanine 93 with glutamic acid Substitution and replacement of tributophane 103 by threonine A single strand according to claim 6 or claim 7, comprising one or more of: variable domain. 9. The one according to any one of claims 1 to 8, further bonded to other molecular components. Chain variable domain. 10. The further other molecular component is an enzyme label, a fluorescent label or a radioactive label, or Immunoglobulin-main chain variable domain according to claim 9, which is part of an immunoglobulin. Main. 11. Claims 1-1, together with one or more auxiliary reagents for carrying out a diagnostic test. A diagnostic kit comprising a single chain variable domain according to any one of 0. 12. at least a single chain variable domain according to any one of claims 1 to 10; A therapeutic composition comprising: 13. Another single chain of a member of the immunoglobulin family or superfamily A method for producing a single chain variable domain that is a hydrophilic synthetic analog of a variable domain, the method comprising: , (i) The interfacial region of the single-chain variable domain was examined in detail to identify hydrophobic amino acid residues. identify; and (ij) one or more of said hydrophobic residues belong to an immunoglobulin family or group; is replaced by a less hydrophobic residue present in the homologous position of a member of the bar family. producing said analogue of said single chain variable domain of (i) which has been modified; A method comprising: 14. Next stage: (a) a nucleotide encoding one or more of the identified hydrophobic residues; get the code array; (b) altering said nucleotide sequence using site-directed mutagenesis to obtain alternative introducing a triplet encoding a amino acid; (c) an altered nucleotide sequence; is used in a recombinant expression system to express synthetic analogs. 14. The method of claim 13, comprising: 15. Claim 13 or Claim 14, wherein multiple amino acid residues are substituted. Method. 16. Alternative amino acids are natural monomeric members of the immunoglobulin superfamily 16. A method according to any one of claims 13 to 15, which is derived from -. 17. 17. The method of claim 16, wherein the natural monomeric member is Thy-1. 18. said synthetic analog has essentially the same binding activity as said another domain; 18. The method according to any one of claims 13 to 17.