JPH04502454A - Methods for identifying active domains and amino acid residues of polypeptides and hormone variants - 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] of polypeptide and hormone variants Method of Identification of Active Domains and Amino Acid Residues This application was filed on This is a continuation of National Patent Application No. 07/264.611.

(Technical field) The present invention relates to methods for identifying active domains and amino acid residues in polypeptides.

The invention also relates to hormone variants.

(Background technology) Polypeptides, or peptides, and proteins each have specific amino acid sequences and structures. It includes a wide range of biological molecules with structure and function. Hatondo Polypepti The polypeptide interacts with a specific substrate to carry out the function of the polypeptide. Therefore Enzymes such as butyricin, amylase, and tissue plasminogen activator are specific interacts with the substrate and causes hydrolysis at its specific cleavage site, while human Sex hormones such as long hormones and insulin interact with specific receptors. and regulate growth and metabolism. In particular, polypeptides and immunogenic receptors There are also interactions with substances that are not the primary target of the polypeptide. many polypeps molecules that interact with separate receptors or receptors that produce distinct biological effects. It is multifunctional in that it contains distinct regions. For example, human growth hormone (hGH) is related to sugar and lipid metabolism in adults, and induces long tube growth in children. guide

Efforts are being made to modify the function of natural polypeptides by modifying their amino acid sequences. It's getting worse. One method is to modify one or more amino acids in the amino acid sequence of a polypeptide. It is the substitution of one amino acid with another. In vitro mutagenesis and cloning Protein engineering through the expression of oxidative genes can improve the thermal or oxidative stability of various proteins. Examples have been reported where it has been applied to the improvement of Villafranc power (Villafranc power) a) + J, E, et al. (1983) 5science 222.782- 788. Perry + L, J. et al. (1984) Science e 226.555-557; Estell + D, ^.

(1985) J, Biol, Chew, 260-16518-652 1 Rosenberg + S. et al. (1985) Nat ure (London) 1 Sat 2, 77-80: Courtr+ey ) + M, et al. (1985) Nature (London), 1 Sat↓, 149-1 57. Furthermore, we have applied this method to generate enzymes with altered substrate specificity. It has also been reported that. Estell, D, A.

(1986) 5science 223.655-663; ik) + C, S, et al. (1985) Science, 228.291-2 97; Farsht, A, R, et al. (1985) Nat ure (London) 1 Sat., 235-238; Winther (N1nther L) J, R, (1985) Carlsberg Res, Common, 50, 273-284 i-Wealth (He11g), J, ^0 etc. (1987) Proc, Natl, Acad.

Sci, Eng↓, 1219-1223. Determination of amino acids to be corrected is mainly based on polypeptide the crystal structure of Tide, the effect of chemical modification on the function of the polypeptide, and/or Interaction of polypeptides with various substrates to confirm the mode of action of polypeptides It is done on the basis of action. In some cases, e.g. Related polypeptides such as amino acid sequence differences in the substrate binding region of subtilisins Amino acid substitutions may also be induced based on differences in specific amino acid residues.

Wells (Ilellg), J.^0 et al. (1987) Proc, Natl. , Acad, Sci, USA 8th Engineering, 5767. In other cases, the known activity of the molecule Modify the amino acid sequence of the region to match the amino acid sequence of the known active region of the second molecule It has also been reported that Wharton (Harton) R, P, etc. (19 85) Nature 316.601-605, and Wharton (l1har ton), R, P, et al. (1984) Ce1138.36m-369 (fur Replacement of one receptor by another receptor (mouse) ; Jones + P, T, et al. (1986) Nature 321.522-525 (mouse antibody of human myeloma protein variable region) (replacement by corresponding region), this method makes some predictions about the properties that change due to the replacement. Ensures evaluation of all regions and residues that can provide insight but determine specific properties Experimental estimates of energies for molecular contact forces of substituted residues, not methods at best The extent to which this is done is that Hydrogen bonding (first (Farsht) + ^, R1 etc. (1985) Nature 3 Sat ↓, 235 ; Bryan LP, et al. (1986) Proc, Natl. Acad, Sci.

USA83.3743i Wealth (llellg), J, ^2 etc. (1986 ) Philos, Trans, R, Soc, London A, 3 Sat 1. 415), electrostatic interaction (Wells (He11g), J, ^, 4 etc. (19B ? ) Proc, Natl.

Acad, Sci, USA 8↓, 1219; Cronin + C.

N, 1st class (1987) J, Am, Chew, Soc, Sat 09.2222 ) and hydrophobic and steric effects (Estell + D-^0 etc. (1986) 5science 233.659; Cken + J , T.

(1987) Bioches+1try 26.4093) strength is estimated It was done. These reports and another (Laskolaski) +■0 etc. (1987) Co1d Spring Harbor Symp, Quant, Biol.

■, 545: He1ls, J, A, etc. (1987) Pro c.

Natl, Acad, Sci, USA 84.5167; Jeans (Jo nes) + P, T, et al. (1986) Nature 321, 522; WharLon + R, P. et al. (1985) Nature 316. (601) showed that mutagenesis of known contact residues causes large effects on binding; We conclude that mutagenesis of non-contact residues has a relatively small effect.

A second way to understand the relationship between amino acid sequence and function is to analyze the relationship between related genes. Hybrids encoding hybrid polypeptides using in vivo homologous recombination This is due to the generation of red DNA sequences. Such a hybrid polypeptide are Escherichia coli and Salmonella typhimurium (typ. trpB and trpA (Schneider (5chn eider) + LP.

(1981) Proc, Na11. Acad, Sci, USA 78 -12169-2173) i alpha 1 and alpha 2 leukocyte interferon Univer + IA, and Weissma nn)+C, (1983) Nuc, Ac1ds Res, 1 Sat, 5661 ) i Escherichia coli -12 outer membrane pore proteins ompC and phoE () Masen ( Thosmassen LJ, et al. (1985) EMBO4, 1583-1 587); and Bacillus stearothermophilus (Bacillus stearothermophilus); rother+5ophilus) and Bacillus litheniformis (Bac thermophilic alpha-amylase derived from P. illus lichiniformis (Gray) + G, L, etc. (1986) J, Bacteri It has been reported that it can be obtained by homologous recombination of Ru. Such a method has been reported for hybrid alpha-amylase useful information regarding useful hybrid polypeptides in addition to amino acid sequences and functions, such as It is important to systematically study a given polypeptide and use it as one of the target selection methods. These methods are useful for measuring detailed regions and amino acid residues in polypeptides that indicate sex. method is difficult to use, which means that the DNA and amino acids of the hybrid The site of cross-recombination that defines the acid sequence is basically the DNA sequence of the target gene and host. This is due to its dependence on the recombination mechanism of the principal cell. These methods Coding one polypeptide except in the fortuitous environment where crossover occurs near the ' or 3' end. A corresponding segment from another gene of a relatively small segment of the DN^ does not provide a predetermined and methodologically necessary replacement for .

Studying the interaction of protein sex hormones and their receptors in several ways There have been reports that One method uses hormone peptide fragments. The location of the receptor binding site on the hormone was determined. otherwise neutralized monochrome Epitope mapping utilizes competition between natural antibodies and peptide fragments. The location of the receptor binding site was determined. Examples of these methods include human growth hormone There are studies reported on hGH.

Human growth hormone (hGH) plays a role in many regulations in normal human growth and development. is involved in This 22,000 Dalton pituitary hormone is responsible for linear growth (body development). ), lactation, macrophage activation, and other insulin-like and diabetogenic effects. Chawla, R, K, which exhibits many biological effects including fruits. (1983) Ann, Rev. Med, 34.519; Edwards (Ed. wards)+C,K.

(1988) 5science 239.769; Thorner )+H001 etc. (1988) J, Cl1n, Invest, B Sat, 74 See 5. Growth hormone deficiency causes dwarfism in children, but this was over a decade ago. It has become possible to successfully treat this disease by administering hGH.

We also distinguish between genetically and post-translationally modified forms of hGH, and We investigated the immunological response to H and quantified the circulating levels of this hormone. There is also interest in the antigenicity of hGH in order to u, J. (1984) Ann.

Rev, Physiol, 46.33) ahGH is placental lactogen, prola Homologous hormones, including genes and species variants of cutin and other growth hormones A member of a group. Nichol, C.S., et al. (1986) End ocrine Reviews 7.169゜hGH shows broad species specificity and Monomerically each cloned somatothetica ---/li (Leung, D. II, etc. (19g?) It is unusual among these groups in that it matches. hGH The lonning gene was expressed as a secretory form in E. coli (Chang, C. K et al. (1987) Gene 55.189), and its DNA and amino acid sequence. Columns have also been reported (Goeddel et al. (1979)). Three-dimensional folding patterns of growth hormone have been reported with moderate resolution and precision. (Abdel-Meguid, S., et al. (198 7) Proc, Natl, Acad, Sci, us^84.6434 ).

Peptide fragments derived from hGH help determine the location of hGH receptor binding sites used. Li, C, H, (1982) Growth hormone protein. After white matter degradation, a fragment of residues 96-133 was isolated. Yamasakin (Ya a+asakin) et al. (1970) Biochemistry 9.1107 oL, when a larger fragment containing residue 1133 was recombinantly produced. No detectable binding activity was observed. Krivi, G, G , etc. InternationalSys+poslus on Growth llormone; Ba5ic and Cl1nical Aspect s.

Abstract l-18, Final Program, Seven Ronosymbosia (5erono) Provided by Symposia, USA, June 14-18, 1987. These results are contradictory and suggest that proteinaceous hormones, especially their binding sites, are consisting of two or more discontinuous epitopes and/or structure-dependent epitopes The use of peptide fragments to locate receptor binding sites for It shows that he lacks cetacean nature.

Neutralizing monoclonal for receptor binding site localization by epitope mapping Similar limitations apply to the use of dual antibodies. For example, monoclonal antibodies hGH receptor binding assay for a peptide consisting of hGH residues 98-128. There are reports of it being used to compete with printers. Even if hGH hormone peptide 9 Even if only B-128 binds to the neutralizing monoclonal antibody, this region Based on competitive experiments, it was suggested that it contains a receptor binding site. Retedine (Ret) egin), L, ^, 1st prize (1982) Endocrinolog! 11 1, 668 A similar method was used in an attempt to identify the antigenic site of mhGH hormone. It has been. According to reports, 24 different monochromes against hGH due to competitive binding. In epitope muffing of local antibodies, there are only 4 types of antigenic sites on the hormone. It was classified as Surowy + T, K, et al. (1984) Mol, lm5uno1.21.345; Aston (^5ton) + R- (1985) Pharmac, Ther, 27-1403, L Kashiko Sentai The location of the epitope on the amino acid sequence of the hormone could not be determined.

Another way to localize antigenic sites is to target short linear peptides on the protein of interest. There are antibody binding tests. Geysen + H, M, et al. (1984 ) Proc, Natl, Acad, Sci, USA 81.3998 i Geysen + H, M, (1985) 11111111 101゜Today　6.364゜However, this method depends on the location of the receptor binding site. is subject to the same limitations as using linear peptide fragments. useful and This linear sequence requires a structure found in the antigen of this antibody to recognize it. Antibody structure must be preserved, further revealed by X-ray crystallography Based on the site of the epitope (5 Cheriff + S- et al. (19 87) Proc.

^, G, 1st class (1986) 5science 233.747) Fact 1 All anti- The body binding site is partially discontinuous (Barlow + D, J.

(1986) Nature 322.747) and as a result, the linear fragment It has been assumed that individuals are unable to successfully manipulate these structures. Also Peptide fragments of hGH are produced by non-covalent binding of these fragments. has been studied. Several researchers have determined that the natural amino acid sequence of human growth hormone or reported the combined analysis of relatively long fragments containing chemically modified peptides. Ru. Burstein + S, et al. (1979) J, ofEn do, Net, 48.964 (amino terminal fragment hGH-(1-13 4> and carboxyl-terminal fragment hGH-(141-191))) ;S(Li)+c,n, et al. (1982) Mol.

Ce11. Biochem, Earthwork, 31i Mills, J, B, et al. (1980) Endocrinology Earthwork 1.391 (hGH hGH-(1-1 34) and human chorionic somatomamotolobin (also called placental lactogen). The chemically modified carboxy-terminal fragment (hC3-(141-191)) was also modified. chemically modified fragments hcs-(1-133) and hGH-(141-191) Similarly, they were bound non-covalently. U.S. Patent 4,189,426, these researchers The first 134 binding determinants of growth hormone to the liver growth hormone receptor erroneously reported that it was an amino-terminal residue (Burstein). + etc. (1978) Proc, Natl, Acad, Sci, USA ■, 5391-5394), clearly these methods lead to wrong results, and By using two large fragments, this method potentially without identifying the specific residues or regions actually involved in the interaction. The function can be assigned to either of the two fragments used in the combination of It is possible. A review of the above-mentioned methods and experiments on hGH is reported. . Nichol (N1chol L C, S- et al. (1986) Endocrine R ev, 7.169-203゜ hGH 1 deletion peptide (hGH-32-46) This 7-residue peptide fragment was used to synthesize other mammalian growth hormones (hereinafter referred to as “segments”). Another method has been reported that has been modified to include amino acid residues derived from But that U.S. Pat. No. 4,699, which has not reported any effect of substitutions such as , 897 references, nevertheless disadvantages of using short peptide fragments The linear sequence of these fragments imitates the structure found in native growth hormone. It is clear that it must be able to be incorporated to be recognized both in vitro and in vivo. It is white.

Many natural hGH variants have been identified. This includes hGH-V Seeberg, P. H. (1982) On DNA.

239; U.S. Patent Nos. 4.446.235, 4,670.393 and 4,6 65,180) and 20K hGH (No. 65,180), which contains a deletion of residues 32-46 of hGH ( Kostyo + J, L-et al. (1987) Biochemi ca et Biophysica Acta 925-1314; Lewis (Le wis Lυ, J, 5 etc. (197B) J, Biol, Chew, 253. 2679) is included.

Some researchers have developed a conventional Cy reported the disruption of the disulfide bond formed between s-53 and Cys-165.

Tokunaga + T, etc. (1985) Eur, J, et al. Biochew, top] Lord, 445. This single substitution clearly changes the tertiary structure of hGH. Mutants were generated that were maintained and recognized by the receptor for hGH.

Other researchers have reported in vitro synthesis of hGH on solid resin supports. this The researchers' first report published the incorrect 188 amino acid sequence of hGH.

(Li)tc, H, et al. (1966) J, ^-, Chew, Soc, 88.2050; and U.S. Patent No. 3.853,832, second report 19 0 amino acid sequences were published. No. 3,853,833, this latter The arrangement is also incorrect. In particular, the hGH sequence has glutamine after position 68. The site 73 is not glutamine but glutamic acid, and site 106 is aspa. Aspartic acid instead of lachin, position 108 is asparagus instead of aspartic acid. It's Gin.

In addition to those listed above, hybrids with altered binding to specific monoclonal antibodies are also available. Interferon has also been reported. Cawble + R, et al., “Peptide; Structure and Function” Interferon-α2 produced from a synthetic gene Properties of analogues, Proceedings of the N1nth ＾meri can Peptide Symposium, (1985) De her) isoline, Pierce Chemical Co., Chicago, Ill., 375-384, text Amino acid residues 101-11 of α-1 interferon as published in 4 or residues 98-114 of γ-interferon to α-2 interferon has been replaced with

α-2 interferon binds NK-2 monoclonal antibody, but α-1 interferon binds NK-2 monoclonal antibody. This particular region in α-2 interferon does not bind α- Seven of the 27 amino acid differences between α-1 and α-2 interferons are in this region. It was selected because it exists throughout the region. Obtained in this way as reported The hybrid has substantially reduced activity with the NK-2 monoclonal antibody.

When tested for antiviral activity, these hybrids were found to be more susceptible to wild-type α-2 in It showed antiviral activity comparable to that of Turferon. smaller within this area Replacement of new sections was also reported. Furthermore, a series of clusters of 3 to 7 alanine residues Continuation 1 was also reported. However, only one indigestion body (Ala-30, 32, 33) rFN-α2 has been published.

Alanine Substitution and 2A within the Small Peptide Fragment of Chicken Egg White Lysozyme The effects of these substitutions on stimulation of 11 or 3A9 cells have also been reported. a Len (^1len) + P, M, et al. (1987) Nature 327. 713-715° Secondary structure including the antigen-binding loop Jones LP, T, et al. (1986) Nature 321.5 22-525> or DNA-certified ffi lysox (Wharton ) + R, P, et al. (1985) Nature 3 Upper Work, 601-605) replacement There are other reports on the manipulation of connectivity by

The above-mentioned documents are merely provided for publication before the registration date of this application, and are not related to the earlier invention. or earlier publication dates due to priority rights based on previously registered applications. Nothing herein shall be construed as an admission that there is no right in the invention.

Clarify the relationship between structure and function, given the situation indicated by the above references. There is a clear need for effective systematic analysis of polypeptides. However, The purpose here is to identify active molecules within a polypeptide that contribute to the functional activity of the polypeptide. The main purpose is to provide an identification method.

Furthermore, the present objective is to provide a method for measuring active amino acids to determine functional activity. .

A further object of the present invention is to systematically identify biologically active domains within polypeptides. The objective is to provide a method for

Additionally, the objective is to develop desirable biological, biochemical and immunological The objective is to provide hormone variants with epidemiogenic properties.

Furthermore, the objective is to have reduced activity for one biological function and for a second target. provide hormonal variants with substantial or increased activity for That's true.

Furthermore, the present objective is to demonstrate the altered consequences of somatogenic receptors for hGH. hGH variants with synthetic and/or biological activity and increased efficacy. It is to provide.

Furthermore, the present objective is to provide a method that retains one or more desirable biological properties but exhibits diabetogenic activity. The purpose of this invention is to provide a reduced hGH variant.

Furthermore, the present objective is to increase the binding activity of hGH for somatogenic receptors. The purpose of the present invention is to provide hPRL and hPL.

Furthermore, the present objective is to clone and develop polypeptide variants, including hGH variants. Provides DNA sequences, vectors, and expression hosts containing the vectors currently in use. That's true.

(Summary of the present invention) One feature of the present invention is that the present invention provides methods for influencing the activity of a polypeptide with respect to a first target substance. Systematic study of polypeptide structure and function by identifying unknown active domains Provide analytical methods. These unknown active domains contain the primary amino acids of the polypeptide. The active domain comprises at least two discontinuous amino acid segments in its sequence. Selected amino acid segments of a polypeptide (referred to as the parent polypeptide) are By substituting a similar amino acid segment derived from the l1 position of the lipeptidase. It is determined. This analog has a different activity than the parent peptide with respect to the target substance. . The activity of each segment-substituted polypeptide thus generated regarding the target substance Test gender. These activities are compared to those of the parent polypeptide. This structure Amino acid segments that image 1s are analogs that interact differently with the target substance. This activity comparison is based on the active domain in the parent polypeptide. Provide an indication of location.

The method further includes identifying active amino acids in the active domain of the parent polypeptide. This method uses each amino acid residue within the active domain of the parent polypeptide. including substitution with acid and assay for the target substance of the residue-substituted polypeptide. Ru. The activity of each residue-substituted polypeptide is compared to that of the parent polypeptide. Active a These tests are performed for various amino acids in the active domain until the amino acid residues are identified. Repeat steps.

Another feature of the invention is that different active domains and activities can be used for different target substances. Methods of identifying amino acid residues are provided. These methods involve This includes repeating the method described above.

different compared to the parent polypeptide with respect to one or more target substances according to the method described above. Identify polypeptide variants that have activity.

These variants contain active domains that determine the activity of the parent polypeptide with respect to the target substance. or based on the identification of active amino acid residues in the active domain.

Furthermore, the present invention provides a growth hormone consisting of at least three parts, prolactin and and placental lactogen variants, the first portion contains at least one of the amino acids of the parent hormone. the third portion corresponds to at least one portion of the parent hormone; and the third portion corresponds to at least one portion of the parent hormone; The second part corresponds to the amino acid sequence of the 1s form of the parent hormone, and the second part corresponds to the polypeptide. Of the amino acid residues of the parent hormone that are not included between the first and third parts of the variant! ! It is a 4gJ body.

The present invention also provides specific human, including segment substitution and residue substitution variants of hGH. Contains growth hormone, human prolactin and human placental lactogen variants.

Brief description of the drawing Figure 1 illustrates the strategy used to identify active domains.

Figure 2 shows the conservation and amino acid sequences of hGH, hPL, pGH and hPRL. and variable amino acid residues are shown.

Figure 3 shows the predicted low-resolution structure of hGH and the N-terminal beginning for each helix. A helical projection view from the residue is shown. Hydrophobic, neutral and charged residues are each ○ , ■ and .

Figure 4 shows various segment substitution hGH mutants for soluble hGH receptor. Figure 2 is a bar graph showing the relative reduction in binding.

Figure 5 shows active domains A and C that interact with somatogenic hGH receptors. Similar amino acids in and F are shown.

Figure 6 shows the relative binding positions of somatogenic receptors and 8 to hGH. Monoclonal antibodies are shown.

Figure 7 shows soluble hGH tumagenic receptors of various alanine-substituted hGH mutants. 2 is a bar graph showing the relative increase/decrease in binding to the The diagonal bar of T175 is around This indicates that serine is substituted instead of nin. The R178 knitting rod is Arani. This indicates that asparagine is substituted instead of .

Figure 8 shows the DNA and amino acid sequence of the hGH gene used in the examples. There is.

Figure 9 shows the construction of vector pB0475 containing the synthetic hGH gene.

FIG. 10 is the DNA sequence of pB0475 showing the amino acid sequence of hGH.

Figure 11 shows the construction of vector pJ1446.

Figure 12 shows the amino acid sequence of the soluble portion of the liver-derived somatogenic receptor. This is the DNA sequence of pJ1446.

Figures 13 to 20 show the results of hGH assays for each of eight different monoclonal antibodies. The epitope binding site of is shown.

Figure 21 shows the activity involved in binding to somatogenic receptors in hGH. A helical projection for amino acids and helices 1 and 4 is shown.

Figure 22 shows hGH and hGH variants administered at 50 micrograms/kg/day. The figure shows the weight gain of rats over time.

Figure 23 shows the semi-lower Kd ratio for the activity of hGH mutants compared to wild-type hGH. This is a graph plot.

Figure 24 expresses human serum (○) or plasmid phGHr (1-238) against hGH-binding protein isolated from E. coli KS330 culture (“5I ) Competition binding curve of hGH and unlabeled glhGH. Bars are standard deviations from the mean. The inset shows the difference derived from the competitive binding curve. ing. The concentrations of human serum and E. coli-derived binding proteins were 0.1 part M and 0.1 part M, respectively. and 0.08 nM.

Figure 25 is based on the structure of pGH determined from the X-ray structure at 2.8 human resolution (hGH This is a structural model of Panel A is functional contours of hGH receptor epitopes. Panel B shows the same map determined for hPRL receptor epitopes. It shows. The size of the black circle indicates the magnitude of the destructive effect regarding alanine substitution of each residue. It shows. Small circles indicate more than twice the destruction, and large circles indicate more than 10 times the destruction. show. (Panel A) 4-fold higher binding affinity for hGH receptor epitopes The position of E174A to be increased above is shown.

Figure 26 shows plasmid pBO7 used for intracellular expression of hPRL in E. coli. 60 is shown.

Figure 27 shows residues in hGH that strongly influence binding to hG H-binding proteins. It shows the position of. 10-fold or more decrease in binding affinity (○), 4-10-fold decrease (■) or an alanine substitution that causes a 4-fold or more increase (mu) (Tl 75 or more) or in the case of R178, serine or asparagine, respectively). to α- The helical projection of helix HMA shows their amphipathic properties and in helix 4 It was revealed that the most important determining factor exists on the hydrophilic surface (shaded area). There is.

Figure 28 shows hGH (-), wild type hPRL (--) and hPRL mutant D. Circular dichroism in the far UV (Panel A) or near UV (Panel B) of (----) Showing the spectrum.

Figure 29. Binding limited by homologous and alanine scanning mutagenesis. A sequence comparison of hGH and hPRL with respect to regions important for Identical residues are shadows and the numbering is based on the hGH sequence. When mutated, the binding affinity is Residues that change by 4-fold or more are circled. The asterisk above the residue indicates the parent to which the mutation binds. Sites that reduce compatibility by 2-4 times are shown.

(Detailed description of the invention) In one embodiment, the method of the invention relates to an interaction between a parent polypeptide and a target substance. human growth hormone to determine one or more active domains in a polypeptide that 0 books providing a systematic analysis of parent polypeptides such as mono or human prolactin In employing the method of the invention, the polypeptides exhibiting different activities toward the desired target substance may be used. There must be one or more llfD forms for putide.

Therefore, as used herein, "parent polypeptide" refers to the parent polypeptide. A polypeptide that has an “am” form that exhibits activity toward a target substance different from that of tide. Taste. Examples of such polypeptides, analogs and target substances are shown in Table 1.

Table 1 mgo<-f-f　l２　’! L---11---t1 target or up Seihito Growth Hormone Human Placental Lacto Somatogenic, Lacun Diene, Human Sea lion sea lion, diabetalolactin and genic, fat buying and disassembly, bibuta Growth hormonal nitrogen delay, macropheron activation and hGH Against insulin-like effects receptor; rat Weight method: Raft weight gain Yes, 0B10B mouse is insulin in dogs. Phosphorus tolerance assay, human liver, ajibose, phosphorus paocytes, thymocytes and eggs receptors on nest tissue Table 1 (continued) hPRL pGH binding to human prolactin receptor Rabbit GH Receipt Human GH Receipt Pigter binding to rabbit GH α-interf related human interferon α1 interferon leeron - feron and binding to the scepter and animal interface Felon Human tissue growth factor (H) TGF-β3 Human hematopoietic cell growth regulator (TGF-β1) or inhibit Epidermal growth factor TGF-α carotene cell proliferation (EGF) Mouse tissue necrosis Human tissue necrosis diagram Mouse TNF receptor factor (mTNF) (hTNF) Activated human granulocytic cells Mouse granulocytic cells Proliferation of human bone marrow stem cells Orophage colo, Chlophagico and differentiation knee stimulating factor, rho-stimulating factor Human CD-47 to mouse CD-4 HIV: J, 0) gp - Table 1 (continued) Receptor Receptor 120 subtilisin subtilisin succinyl-ala-ala-(Bacillus ami (Bacillus Ami Pro-glu-P-nitroallylquifei Anyformi (S) Nirid Shans) Human T-Ink Related Human Ink Human Interferon Rape 10-Fe Activation of Ron and Scepter and Animal Interface Insulin composition Insulin IGF-Rusebuter growth factor (IGF-1) Long growth regulatory receptor tissue plasmino tribusinuro plasminogen (cut ) - Gen activator Kinase Fibrin (binding) molecules (tPA) Of course, the parent polypeptides, analogs and target substances in Table 1 are merely examples. Also Parent polypeptides include, for example, succinyl coenzyme A synthetase, mitochondrial Doria ATPase, aminoacyl-tRNA synthetase, glutaminesin Cetase, glyceraldehyde-3-phosphate dehydrogenase and As Paraty + - Proteins containing one or more subunits such as transcarbamorase contains white matter (Huang et al. (1982) Ann, Rev. , Biochem, 5↓, 935-971). Such multiple subunits In the case of a parent polypeptide, the active domain comprises two or more subunits of the parent polypeptide. It spans over a wide range of sites. Therefore, the method described in more detail later is Explore each subunit of the titanium, partially contained on one subunit and partially Activity toward a specific target substance that is partially contained on one or more other subunits It can be used to define domains and active amino acids.

Parent polypeptides and analogs generally belong to one group of polypeptides with respect to function. do. Furthermore, such parent polypeptides and analogs usually have certain amino acid sequences, That is, it has conserved residues. Although these sequence homology can be over 90%, It can be as low as about 15% to 20%.

In addition to primary sequence homology, analogs of a parent polypeptide are The three-dimensional framework of analogs is also limited. Therefore, analogs are Even if the molecule shows diversity from the parent polypeptide, all of the tertiary structure of the molecule, or may have structural homology with the parent polypeptide based on some comparisons.

Chothia, C., et al. (1986) Embo, J., 5 ．． 8:23゜In general, tertiary structural analogs are those whose three-dimensional structure is similar to that of the parent polypeptide. If the structure is known to be the same as that of a putide, it can be considered the same. α-carbon By performing root mean square analysis (RMS) of the axes (e.g. Sutc (1987) Importance of regions with protein similarity is identified. Preferably for more than 60% of the test analog sequences, if α- The carbon axis overlaps the alpha-carbon axis of the parent polypeptide, or about 2 to about 3.5 carbon axes. If within the RMS, the test analog is three-dimensionally similar to the parent polypeptide. ing. This of course excludes insertions or deletions that exist between the two sequences. Ru.

Although the parent polypeptides and analogs described above are natural molecular disclosures, Parent polypeptides and analogues include naturally occurring sequences introduced by in vitro recombinant techniques. It should be understood that variants containing mutations in are included. In this way the parent polypeptide or a variant used as an analogue has 1 in its parent polypeptide or analogue. Includes variants containing substitutions, insertions and/or deletions of one or more amino acid residues. child Variants such as active domain and/or active amino acid It can be used to identify acids or to prepare polypeptide variants of the invention. Therefore h Natural mutants of GH or recombinant mutants containing the Ala substitution of Cys-165 are may be used as parent polypeptides or analogs as long as they have activity against the target. Ru. Such natural and recombinant variants include specific residues in other parent polypeptides. may include different amino acid residues equivalent to. These different amino acids are their The residues are structurally similar by primary sequence or tertiary structure, or they are functionally similar. If they are functionally equivalent, they can be said to be equivalent.

Furthermore, it is clear that many parent polypeptides and analogs can interchange their roles. It is. Therefore, each group of non-human growth hormones and their related analogues is Can be used as a parent polypeptide and equally probe its active domain . Furthermore, target substances such as the CD-/l receptor for the old V virus are also CD- As a parent polypeptide for 4 receptor analogs, active molecules that contribute to HIV binding Used for identification of main and active amino acids and preparation of CD-4 variants.

As used herein, a "target" is a substance that interacts with a parent polypeptide. It is. Target substances include proteinaceous hormones, enzyme substrates, and proteinaceous receptors. hormones, common ligands and polypeptides for proteinaceous binding proteins. These include the immune system, which can come into contact with the putido. An example of a hormone receptor is hGH somatogenic and lactogenic receptors for and hPRL Contains receptors for Other target substances include antibodies and protease inhibitors. hormones that bind to proteinaceous receptors and tissue plasminogen activity. It includes fibrin that binds to sexualizing factor (1-PA).

In general, a target substance targets a parent polypeptide by contacting the "active domain" on the parent polypeptide. Interacts with polypeptides. Generally, such active domains are associated with the polypeptide. The polypeptide is present on the surface of the polypeptide or due to a change in its three-dimensional structure. is brought to the surface of The surface of a polypeptide is defined as The native structure exists under conditions similar to those under which it was expressed in vivo or in vitro. stipulated. Its amino acid segments and amino acid residues can be It is confirmed. If the three-dimensional crystal structure is known with sufficient resolution, the point Amino acid residues on the surface of lipeptides are “surface-accessible” amino acid residues. .

These surface-accessible residues have theoretical water content that “rolls” over the surface of their three-dimensional structure. It is the residue that contacts the child.

The active domain on the surface of a polypeptide consists of the primary amino acid sequence of the polypeptide. Contained in a single segment. However, in many cases the natural polypeptide An active domain consists of two or more amino acid sequences in the primary amino acid sequence of a parent polypeptide. consists of components. For example, human growth hormone somatogenic fucreseb The active domain for the target is shown in FIG. The domain of the active domain as shown. In^, C and F each exist on a discontinuous amino acid segment of the hGH molecule. Ru. These amino acid segments are indicated by letters A, C, and F in Figure 4. Ru. The discontinuous amino acid segments that make up the active domain are separated by many amino acid residues that are not significantly involved in interactions between target substances. ing. Generally, the spacing between discontinuous amino acid segments is at least 5 amino acid residues. It's a minute.

The method of the invention aims to detect unknown active domains in the amino acid sequence of a parent polypeptide. It has been the target. If a three-dimensional crystal structure of a polypeptide for a single substance is available, If, for example, the crystal structure of the enzyme for the inhibitor or the transitional am+a form is Most active domains of many polypeptides remain unknown, except where available. There is even.

As used herein, "!1 polypeptide segment" or "llll superposition polypeptide segment" Segment 1 replaces the corresponding sequence in the parent polypeptide to create a “segment replacement polypeptide”. 41 segment refers to the amino acid sequence of the parent polypeptide. Substitutions, insertions or deletions of one or more different amino acid residues in the peptide However, the relative affinities of other residues in the selected segment that are substituted in the parent polypeptide are The amino acid sequence has a maintained sequence. Generally, the l1st segment is the parent polype The entire amino acid sequences of 114g1 and 114g1 were determined in such a way that the sequence identity between both sequences was maximized. They are identified by aligning them. Similar segments based on this sequence alignment are parent selected for replacement of the polypeptide with the corresponding sequence. Similarly, tertiary structure homology The similar segments derived from the lfR body showing - are these regions showing structural homology? It can be induced from Such similar segments are similar to the corresponding sequence in the parent polypeptide. Replaced. Furthermore, tertiary structural homologs, such as adjacent regions of structurally similar regions, Other regions are also used as similar segments.

If possible, select similar segments and include them in the segment replacement polypeptide. Introduction of stabilizing amino acid residues should be avoided.

Such substitutions include those that introduce bulky side chains or hydrophilic side chains into the hydrophobic core region. included.

Generally, the amino acid sequences of parent polypeptides and analogs are known and In some cases, its three-dimensional crystal structure is also available.

Alignment of the amino acid sequences of one or more analogs and the parent polypeptide to at least predict Easily find conserved amino acid residues in unaltered sequences by preparative analysis. can be used. Sequence alignment also involves substitutions, insertions or insertions of one or more amino acid residues. also reveals regions of sequence variation, including deletions. The region containing such changes It is determined which segments in the polypeptide have been replaced with similar segments.

Therefore, substitution of similar segments of analogs is not only a substitution of amino acid residues; Insertions and/or deletions of amino acid residues may also occur.

If 3D structure information is available, it should not be replaced with similar segments. It becomes possible to identify regions in the parent polypeptide. For example, the parent polypeptide If overcrowded regions are identified on the hydrophobic surface of the amphipathic helices in the Should not be replaced with a segment. Residues so identified can be used to modify polypeptides. should be retained in the variant and only surface residues should be replaced with similar residues .

Generally similar segments are 3 to 30 amino acid residues in length, preferably about 3 to 15 amino acid residues in length. groups, most preferably about 3 to 15 residues. If so, 11A4m segment The preferred length of the polypeptide is 1141 segments compared to the homologous or parent polypeptide. Some changes may be due to insertions and/or deletions of one or more amino acid residues. if When the three-dimensional structure of the parent polypeptide is not available, it is generally A class that covers most of the polypeptides (segment substitutions for the following segments) We need to make polypeptides. However, segment substitution of the entire amino acid sequence is always If it is not necessary, for example, a chance segmentation that covers only part of the entire amino acid sequence Substitutions provide sufficient information to identify active domains for specific targets. do. However, in most cases, approximately 15% or more of the amino acid sequence cannot be identified as an active domain. Therefore, it is necessary to undergo segment replacement.

Generally about 50%, preferably about 60%, more preferably about 75%, most preferably Or segment replacement if 100% of the amino acid sequence does not have ntiit information. It will be done.

As used herein, “similar amino acid residues” or “analogous residues” refer to parent polypeptides. Amino acid residues in similar segments that differ from the corresponding amino acid residues in the corresponding segment of refers to a noic acid residue. Therefore, if the substitution in the M4H segment is one amino acid If a substitution occurs, the amino acid residues are similar residues. once the parent polypeptide and and one or more all (if 14 were identified, similar segments from one or more analogues) segment with a selected segment in the parent polypeptide. Create a replacement polypeptide. Such substitutions can be easily made using recombinant DNA technology. be exposed. Generally, the DNA sequence encoding the parent polypeptide can be expressed in a convenient host. Cloning and other processing is performed as described below. Parent polypeptide code The DNA is derived from the mRNA of cells expressing the parent polypeptide or can be obtained from genomic libraries by synthetically constructing DNA sequences. (Maniatis +↑1st prize (1982) Mo1ecula r Cloning, Cold Spring Harbor Laboratory 1N, Y,) Then this! Insert the llDNA into a suitable plasmid or vector and insert it into host cells. used for the transformation of The cloning and expression of DNA sequences polypeptides, segment substitution polypeptides, residue substitution polypeptides and polypeptides Prokaryotes are preferred for making peptide variants, e.g. E. coli k12294. strain (ATCC No. 31446), Escherichia coli B strain, large intestine@X1116 strain (ATC CNo, 31537) and Escherichia coli C600 strain and c600hfj strain, large intestine Fungus W3110 (F-, γ-1 autotrophic type, ATCC No. 27325), withered Bacilli such as grass fungi and Salmonella typhimurium (Salwonell) a typhimurlus) or Serratia marcesans (Serratia marcesans) Other intestinal bacteria such as P. marcesans and various Pseudomonas species used. A preferred prokaryote is large 111 [W3110 (ATCC 27325 ) Thea 4. When expressed in Illkaryotes, polypeptides generally have an N-terminal Mochi containing thionine or formylmethionine and not glycosylated Of course, these examples are intended to be illustrative rather than limiting.

In addition to prokaryotes, eukaryotes such as yeast cultures and cells from multicellular organisms are also used. be done. In principle, cultures of such cells can also be used. However, the vertebral cells The most interesting thing is that the growth of vertebral cells in culture (tissue culture) can be repeated. "Tissue Culture" Academic Press Edition, Kruse and Baderlin ( Patterson) HA (1973) - These useful host cell lines Examples of columns include VERO and HeLa cells, Chinese hamster ovary (CH O) Cell lineages include W+38, BHK, CO5-7 and MDCK cell lines.

Plasmids that generally contain replication and control sequences from a species compatible with the host cell vectors are used in conjunction with these hosts.

Typically, this vector can provide phenotypic selection for the replication site and transformed cells. It has a sequence that encodes a protein. For example, E. coli is a plastic derived from E. coli. transform with pBR322 (Mandel, M, et al. (1970) J, Mol.

Biol, 53. 154). Plasmid pBR322 contains ampicillin and Contains a tracycline resistance gene and therefore provides an easy means of selection. Ru. A preferred vector is pBO475.

See Example 1. This vector allows movement between hosts, thereby allowing mutations Contains phage and E. coli replication origins to facilitate induction and expression .

"Expression vector" means a suitable control system that enables the expression of DNA in a suitable host. refers to a DNA construct comprising said DNA sequence operably linked to a target sequence. this These control sequences include a promoter that enables transcription, a promoter that controls such transcription. control operator sequence, which encodes the appropriate mRNA ribosome binding site. and sequences that control transcription and translation termination. Baek The target can be a plasmid, a phage particle, or simply a potential genomic insert. stomach. Once transformed into a suitable host, the vector becomes independent of the host genome. or, in some cases, become integrated into the genome itself.

In this specification, “plasmid” and “vector” refer to plasmid, which is currently the most commonly used term. Because they are commonly used vectors, they are often used synonymously. . However, the present invention is capable of other forms of expression that function similarly and are well known in the art. Also includes vectors.

"Functionally linked" when describing the relationship between two DNAs or polypeptides The term means that they are functionally related to each other. For example perhaps The signal is released in the process of secreting the mature protein by cleavage of the signal sequence. The sequence is operably linked to the peptide if it functions as a sequence. promo A vector must be operably linked to a coding sequence if it controls the transcription of that sequence. Ru. If the ribosome binding site is in a position that allows translation, it is part of the coding sequence. Functionally connected.

Once the polypeptide has been cloned, the cloned parent DNA Specific mutagenesis (Carter LP.

Futo mutagenesis (Illells, J. et al. (1985) Ge ne 34, 315), restriction selection mutagenesis (Hells + J , A, et al. (1986) Philog+ Trans, R, Soc, Lo ndon Ser^.

317.415) or 11 (by H segment) with other techniques A segment replacement DNA sequence is created that encodes a change in the defined amino acid sequence. It will be done. When operably linked to an appropriate expression vector, the segment replacement polypeptide A petite is obtained. Parent polypeptide or segment modified polypeptide, if any. Recovery of the DN encoding the parent polypeptide or the segment-modified polypeptide from the expression host by using an appropriate signal sequence operably linked to the A sequence. This is possible by expressing and secreting these molecules. These techniques are in our field. well known. Of course, there are also in vitro chemical synthesis of desired polypeptides. Alternative methods were also used to generate these polypeptides and segment replacement polypeptides. (Barany, G., et al. (1979) “Peptides”) (E, Gross and J, Meienhof er) &1) Academic Press Edition, N, Y, 2%, pp. 3-254).

Once various segment-substituted polypeptides are generated, they are brought into contact with the target substance. If there is an interaction between the target substance and each segment replacement polypeptide, Measure the Compare these activities with those of the parent polypeptide for the same target substance. Compare. If the am form has different activity with respect to its target substance compared to the parent polypeptide, in the parent polypeptide, those segment replacement polypeptides have It is presumed to contain similar segments that define the active domain.

If one analog has greater activity than the parent polypeptide, multiple ment-substituted polypeptides may exhibit increased activity with respect to their target substances. . In such cases, the active domain of the analog and its activity with respect to the target substance may be increased. Suitable regions of the parent polypeptide that can be modified to add to the polypeptide can be effectively identified. So The region of the 1i4H body involved in interaction with the target of Such an event is likely to occur when it exists in a segment, if It(one body If the “active domain” of “active domain” is included in a discontinuous region of the 1IlU amino acid sequence, The increase in sex is due to segment replacement polypeptides, all active domains from 11412 bodies. The activity of the segment-replacing polypeptide must be introduced at the same time. An increase in sex is unlikely.

Therefore, the lllfJ body has less activity against the target substance than the parent polypeptide. In such cases, the segment replacement polypeptide preferably has an active Loss can be seen. However, once the active domain in the parent polypeptide has been determined, , this polypeptide may sequentially or simultaneously replace such active domains. a second vL polypeptide lacking the activity of the first parent polypeptide with respect to the target substance; It can be used as an analog for

The active domain in the polypeptide is a segment-replacement polypeptide related to the target substance. is identified by comparing the activity of the parent polypeptide with that of the parent polypeptide. Many analytical measurements However, for non-catalytic binding of target substances, segment-substituted polypeptides The dissociation constant Kd of the complex formed between polypeptide and target substance amount is expressed as Kd with respect to the parent polypeptide. One of the simplest methods is to compare it with d. Similar replacement residues by segment substitution An increase or decrease in the Kd value of about 1.5, preferably about 2.0 per This indicates that this is the active domain for the interaction between the target substance and the parent polypeptide. .

In the case of catalytic interactions with target substances, appropriate parameters are used to measure the activity against crude enzymes. The meter is to compare the ratio to Kcat/.

Keat/K of about 1.5, preferably 2.0 per substituted position residue relative to the parent enzyme An increase or decrease in m value indicates displacement of the active domain.

As used herein, "scanning amino acids" refer to the activity within the parent polypeptide. An amino acid used to identify an amino acid. °Residue substituted polypep "Tide" is the least of the amino acids in the parent polypeptide for scanning amino acids. are also polypeptide variants that contain a single substitution. “Residue substitution DNA sequence” is a residue position encodes a converted polypeptide. In this way, NA and polypeptide are segmented. It is prepared by the method described for the preparation of substituted DNA and polypeptides.

“Active amino acid residues” identified through amino acid scanning generally come into direct contact with the target substance. It is something that can be touched. However, the active amino acids are other residues or subdivisions such as H,O. salts formed with ionic species such as Na'', Ca'', Mg'' or Zn+'' Some insects come into contact with the target substance indirectly through bridges.

In some cases, scanning amino acids are identified in the active domain of the parent polypeptide. is substituted with the amino acid that is Many residue-substituted polypeptides are commonly prepared. , each of which scans amino acids at various amino acid residue positions within the active domain. Contains a single acid substitution. Such residue substitution polypeptides for specific target substances Compare the activity of the peptide with that of the parent polypeptide to determine the interaction with the target substance. Determine the amino acid residues in the active domain. The skills used in such analysis The replacing amino acid is a different amino acid than the substituted one, i.e., the other 19 amino acids. It is a natural amino acid.

Straw table Ala Ser, Gly Glu Gln, Asp Gln Asn, Glu Asp Asn, Glu Asn Aln, Asp Leu Met, l1e Gly Pro, Ala Lys Met, Arg Ser Thr, Ala Val I Ie, Thr fi, rg Lys, Met, AsnThr Ser, Met Chapter ■Table (continued) I Is Met, Leu, ValMet Ile, Leu Cys Ser, Ala Trp Phe His Asn, Gln This table uses the following symbols for each amino acid.

Amino is that - Xie two-pronged master y Alanine Ala A Glutamic acid Glu E Glutamine Gin Q Asparagine i! l! Asp D Asparagine Asn N Roisine Lea L Glycine cry G Lysine Lys K Serin Ser S Valin Val V Arginine Arg R Threonine Thr T Proline Pro P Isoleucine Ile I Methionine Met M Amino is the origin of Yu Bunshu E Phenylalanine Phe F Tyrosine Tyr Y Cystine Cys C Tryptophan Trp W Histidine His H If there is an effect on the activity of the residue-substituted polypeptide, it is can be systematically attributed to changes of natural amino acid residues to amino acid residues. Most preferably, the scanning amino acids are the same for each residue in the substituted polypeptide. Delicious.

In some cases, scanning amino acid substitutions at one or more residues are Residues that are not expressed at levels that allow isolation in sufficient quantities to perform assays for activity resulting in a replacement polypeptide. In such cases, various amino acids, preferably iso- Steric amino acids are used.

The most preferred scanning amino acids are relatively small, neutral amino acids. child Amino acid residues like include alanine, glycine, serine and cysteine. It will be done. Alanine does not have a protruding side chain from the beta carbon and is a polypeptide with a bracket substitution. This is the preferred scan among this group because the main chain structure of Tido is difficult to change. It is a ning amino acid. Alanine is also the most common amino acid. are also preferable; moreover, it exists in both buried and exposed forms (crates Creighton, T, E, “Protein” (LH, Free Mann Publishing Edition, N, Y, ), Chothia, C.

(1976) J, Mo1. Biol, 150. 1) If alanine If the substitution does not result in an adequate amount of residue-substituted polypeptide, isosteric amino acids No acids are used. The following amino acids are used in order of decreasing preference: S er+ Asn and Leu.

Use of scanning amino acids confirmed by analysis of segment replacement polypeptides is not limited to the identification of active amino acids in active domains that are If one or more amino acids in the parent polypeptide are known, or if the target substance If the residue and surrounding amino acids are thought to be involved in an interaction with Scanning amino acids can be used to search for acid residues. Furthermore, what if If the parent polypeptide is a small peptide, i.e. about 3-50 amino acid residues , scanning amino acid analysis can be performed over the entire molecule.

Once an active amino acid residue is identified, it is replaced with an isosteric amino acid. . This isosteric substitution does not need to be made in all cases and is also necessary for active amino acids. This is also done before the identification.

This isosteric amino acid substitution can cause structural potential. This is done in such a way that destructive effects are minimized. Table ■ Isosteric amino acids Shown below.

Active amino acid residues alter the activity of the residue-substituted polypeptide with respect to the target substance compared to the parent polypeptide. It can be identified by comparing it with Tido. Generally, a two-fold increase or decrease in Kd value is due to the substitution of residues. This indicates that there is activity in interaction with the target substance. Similarly, contact with the target substance In the case of mediated interactions, a two-fold increase or decrease in the value of Kcat/l (−) indicates the residual activity of the parent enzyme. Indicates substitution of groups.

Putative or known active amino acid residues are subjected to scanning amino acid analysis. When searching, the amino acid residues adjacent to that residue should be scanned. the result A 3-residue substituted polypeptide is created. One is a putative or known active amino acid Site N contains a scanning amino acid, preferably alanine. The other two contains scanning amino acids at positions N+1 and N-1. Each replacement poly The peptide has an effect of approximately twice or more on the Kd or Kcat/Km value for the target substance. If you rub, the scanning amino acids are substituted at sites N+2 and N-2. There is. This means that the effect on Kd or Kcat/Km value is less than double. until one, preferably four, residues are identified in both directions, or Repeat until the end of the polypeptide is reached. one or more amino acids are the same along the contiguous amino acid sequence involved in the interaction with can be determined.

The methods of the present invention utilize active domains for one or more targets of a particular parent polypeptide. used for detection of Furthermore, active amino acids within various active domains can also be determined using this method. Identified by More than 27 active domains and active amino acid residues identified at once If a polypeptide has been identified for various target substances, its parent polypeptide Various modifications are made to the polypeptide to enhance the interaction between the polypeptide and one or more target substances. It can be fixed. For example, the two active domains on the hGH surface are The receptor and prolactin receptor were also identified. this special case have overlapping active domains. Therefore somatogenic and prolactic There are many common active amino acid residues that interact with receptors. hG Various modifications to H are made based on information described in more detail below. .

In some cases the active domains for different target substances are non-overlapping, such Modifications of active amino acids in the parent polypeptide for one receptor may vary in different situations. This is done by substituting an amino acid in The interaction will be decreased or increased and the physiology of these variants will be altered.

As used herein, an "interactive change" is one in which one or more active domains are modified. This is a polypeptide whose interaction between the target substance and its variant is changed compared to the parent polypeptide. Used for lipeptidic variants. Interaction changes are identified with the parent polypeptide The interaction of the polypeptide variant is at least twice that of the target substance. Defined by increasing or decreasing.

Target substances and parent polypeptides, polypeptide variants, and segment replacement polypeptides and/or interactions with residue-substituted polypeptides in vitro or in vivo. can be measured using a simple assay.

In vitro assays involve target substances and polypeptides, such as enzymes, substrates, and hormones. used to measure detectable interactions such as hormone receptors and antibodies and antigens. I can stay. Such detection includes color changes, radioactivity changes, solubility changes, and gel electrophoresis. and/or measurement of molecular weight changes by gel exclusion methods. in vivo Assays include physiological effects, such as changes in body weight, changes in electrolyte balance, and changes in clotting time. Includes assays to detect changes in blood smear resolution and induction of antigenic responses. However, in general, the target substance and the target polypeptide are not limited to those listed above. in vitro as long as there are variable parameters to measure interaction changes between peptides. Roansei is used.

Examples of the present invention include those related to somatogenic receptors for human growth hormone. Identification of active domains and active amino acids of hormones that determine their activity There is an embodiment in which segment substitutions and residues are used when practicing this embodiment of the invention. Growth hormone somatogenetics with human growth hormone mutants, including substitutional hGH mutants The binding interaction with the human growth hormone is different from that of natural human growth hormone. manufactured or identified. A small number of these human growth hormone variants (one is soma has a higher affinity for togenic receptors and is It has a higher ability for matogenesis. others are somatogenic Has low activity for receptors. Such hGH mutants are useful as agonists or antagonists, and their variants are somatotropic. Since there is no substantial interaction with the genetic receptor, human growth hormone and other has a higher ability to stimulate the receptors of Furthermore, these mutants h Also useful in immunoassays for hGH as a GH standard or tracer . Reactivity with human and mouse sera containing e.g. anti-hGH polyclonal antibodies Mutants with significantly lower levels have been identified.

Another has binding affinity for the same somatogenic receptors as hGH. However, some are more capable of stimulating growth.

Activity of human growth hormone interacting with liver-derived somatogenic receptors The method for measuring domains is shown in FIG. In this method, segments of hGH , a cloned hGH liver receptor and a monoclonal antibody against hGH Derived from the hcH1I position, which is known to have a very low affinity for carp. was systematically replaced with the am sequence of Such hGH414E1 body contains human placenta. ctogen (pPL), porcine growth hormone (pGH) and human prolactin (h These analogs have somatogenic hGH receptors (h GHr) and approximately 100 to 10,000 times the cloned hGH receptor. also has a small binding affinity (Llarington , ^, C0 et al. (1986) J, ClIn, Invest, 77. 1817 ; Bausann, G., et al. (1986) J., Cl1n, Endocrinol, Metab, 62.137).

Homologous proteins have similar three-dimensional structures even when they have large sequence diversity These 11 positions are used because they are known to have Chothia, C1 et al. (1986) FMBOJ, 5,823), When done in this manner, **sequence substitutions can be made without significantly destroying the original structure of the molecule. Human growth hormone and its analogues have increasing promise of being readily applicable to The amino acid sequences of PL, pGl (and hPRL) are shown in FIG.

The last three of these analogs were at levels of 85%, 68% and 23%, respectively. Shares sequence identity with hGH.

Referring to Figure 1, the somatogenic receptor for human growth hormone one or more of the human growth hormones that interact with (the “standard” for hGH) The entire strategy for identifying the active domain of is shown. As indicated, hGH is positive avidity for the target receptor, in this case the somatogenic receptor. have. However, hPRL, hPL and pGH@billions are shown with a minus sign. As shown in the figure, substances with extremely low activity with respect to the target substance are indicated by 0 to F. The six segment-substituted growth hormones contain selected amino acid segments of hGH. hpRt, is generated by substitution with a frequent amino acid segment of the II position. child Each of these selected segments is different from each other and can be selected to represent the entire hGH Explore regions expected to contain amino acid sequences or active domains. segment Regarding each hGH somatogenic receptor after preparation of substituted human growth hormone Perform an assay to measure its activity. These results compared with hGH are shown in the section of Figure 1. indicated with a + or - sign below the component-modified human growth hormone. Minutes from Figure 1 As shown, segment replacement human growth hormone C and F in this figure are somatogeats. Based on these results, growth hormone mutant C, which does not bind to nick receptors, and the lI suprasegment from the 1141J body in F in the corresponding growth hormone. This region is the active domain for hGH binding to somatogenital receptors. Identified.

As shown, if there is structural information or other data, There is no need to test the entire amino acid sequence of growth hormone or other parent polypeptides. therefore, low-resolution or high-resolution information from crystallography is selected from analogs. This can provide important information to avoid inadvertent substitutions of amino acid segments.

For example, although the X-ray coordinates of human growth hormone are not available, the low-resolution X-ray coordinates of pGH Based on the crystal structure <The helisox projection diagram from the pGH structure model shows four helices. Three of them (helices 1.3 and 4) have parents with strong hydrophobic moments. It was revealed that it is mediated. See Figure 3, Elsenbe rg), D, et al. (19B4) J, Mol Biol, 179, 125. polype Since the hydrophobic core in putidos is very densely packed (Ponder, J.

-9 etc. (1987) J, Mo1. Biol, Kamishushisu, 775), those Changes in buried amino acid residues generally lead to destabilization (Alber ) + Ding 0 etc. (1987) Biol, Chem, 26.3754; Raider Lou Olson (Reidhaar-01son) + J, F, (198B) 5 science241.53>.

Furthermore, amino acids within polypeptide groups, such as the human growth hormone group, are highly The saved area does not need to be examined, at least not first, as this This is because destruction of the existing sequence is considered to destroy the structure of the molecule. still other Data show that certain regions of the parent polypeptide are not implicated in interactions with specific target substances It may indicate that. For example, the N-terminus of hGII by mutagenesis Deletion of 3 amino acid residues (Ashkenazi (^5hkenaziL A).

(1987) Endocrinology)↓1 soil, 414) and residue 3 Neutral mutant of hGH deleted from 2 to 46 (20Kd mutant; Lewis ), u, J, et al. (1980) Bioche-, Biophys, Res, C om-un.

92.5111) significantly affects binding to somatogenic receptors. It is reported that there is no. Additionally, some of the residues between residues 134 and 149 or The production of two-chain derivatives of hGH by restricted proteolysis with all deletions is also a somatotropy. showed no significant effect on binding to genic receptors. Li (Li), C, H.

(1982) Mo1. Ce11. Bioche+s, 46. 31: Mi Mills.

J, B, et al. (1980) Endoerinology 107.391° Based on the information of 6 segments of the amino acid sequence of hGH and many hcul was chosen for replacement with the corresponding 114m1 amino acid segment from the 141J body.

These selected segments correspond to A to F in FIG. These segments disulfide bonds, secondary structure boundaries (see Figure 4), and growth hormone groups. Regions of highly conserved sequences or their binding to somatogenic receptors separated by regions previously identified not to be involved.

17 segmental substitutions that collectively replaced 85 of the 191 residues of hGH The GH mutant was tiW1. Areas allocated to A to F in Figure 2 and within each area The hGH mutants in which the segment was replaced are summarized in Table 2.

■Table hcn None 0.34 1.0 All-33hPL (12-25) N12H, F25L r, s, :/ 1. 44.1 Table m (continued) C54-74hPL (56-64) E56D, R64M Kasef) 1030 Chapter ■Table (continued) 17 Restricted Choice - Ilellg, J., et al. (1986) Phil os, Trans, [1, Soc, London Ser A 317+ 415゜27 force cent mutagenesis - Wells 01ells), J, ^0 etc. (1985) Gene 34.315°31 Recombinant mutagenesis - Gray (G ray), G, L, et al. (1986) J, Bacteriol, 166, 635° In general, segment substitution hGH variants substitute human growth hormone sequences. Determined by similar segments. However, in some cases the kind in the replaced similar segment Not all of the similar residues are retained in the construct, thus hPL (12-25) is amino acid 12-25 of human placental lactogen (hPL) Segment substitution hGH mutant in which is substituted with amino acid residues 12 to 25 of 1lhGH It is about. This I! The effect of exasperated segment replacement is in this region of Figure 2. It can be determined by comparing the amino acid sequences of hGH and hPL. h pt, (12-25) mutant has four amino acid substitutions, but what changes are there? These residues in the case of hPL(12-25) are 12.16.2 0 and 25.

Actual amino acid substitutions and other segment locations in the hPL(12-25) variant The conversion mutants are shown in Table Ⅰ. Each permutation is represented by a letter before and after the number. No. The letter 1 and number indicate the amino acid at the residue number of unmodified hGH.

The last letter indicates the amino acid substituted at that position. Therefore, Nl 2H In the hPL (12-25) mutant, asparagine 12 of hGll is histyl. This shows that it is substituted with gin.

Some of the actual permutations introduced therefore correspond to the corresponding segments in Figure 2. In this way, hPL (12-25) does not correspond to the substitution shown in If hPL (12-25) segment is replaced, 4 substitutions N12H, R1 It will include 6Q, L20A and F25L. However, the actual mutant is R1 6 and L20 are maintained, and the introduced substitutions are as shown in Table ■ There are only two of the four substitutions: N12H and F25L. ParenthG Other segment substitution mutants that retain one or more residues of H include regions A and E. and segment substitution mutants hPL (46-52) and pGH (1 67-181) are included.

Cloning each segment replacement human growth hormone mutant into a soluble hGH receptor In an in vitro system using displacement of (''I)hGH from the extracellular region of the Assay of segment substitution mutants relative to their somatogenic receptors Affinity was placed in place. Leung, o, w, et al. (1987 ) Nature 33G.

537, this truncated somatogenic receptor is younger than the mesenchymal receptor. Although the binding affinity is smaller (Kd-0,3nM), it shows similar activity. (Spencer, s, A, et al. (198B) J, Biol, Chem, 263.7862).

Residues 11-19°54-74 and 1, respectively, as detailed in the Examples. Selected segments A, C and F containing 64-191 are somatogenic receptors. It is the active domain of the hGH molecule that interacts with This means that within these areas Based on the observation of a Kd value that is more than 10 times lower for the hcH11 position, See Figure 4, this of course means that each amino acid residue within these active domains It does not mean that they are binding residues of somatogenic receptors, but rather that they The domain defines the amino acid sequence in which such active residues can be found.

Additionally, the positions of active domains A, C and F were determined. For example, mutant hPR L(12-33) to amino and carboxy terminal mutant hPRL(12-19) and hPRL (22-33). The results of this experiment indicate that the active domain of hGH Similarly, region F (pGH(167-18 The amino-terminal region of 1)) showed a significant decrease in binding affinity. most notable hPL with only two mutations E56D and R56M introduced for good effect (56-64) was found to be 30 times smaller than the binding.

Regions A, C, and F are very distant on the primary sequence of hGH, but the hormone The tertiary structure of the molecules makes them very close together.

See Figure 5, these active domains are located at the amino terminus of heliphus l (the active domain A), the loop from Cys-53 to the start of helix 2 (active domain C) and and the central region of helix 4 (active domain F).

Eight additional Mabs against hGH were assayed against segment-substituted hGH mutants. We created a map of hGH epitopes. In addition, the Mab is useful for hGH and hGH Used in competitive assays using mutants to inhibit hGH receptor binding to hGH The performance of each Mab was evaluated.

Combining these experimental results, it can be concluded that the substitution of the analogous segment to hGH is due to the original structure of the molecule. Evidence and observation that it does not significantly destroy! Activity is somatogenic receptor Direct effect on the interaction between the receptor and the receptor-substituted hGH mutant provides some evidence that it is. First, segment substitution mutants are Very important in disrupting binding to matogenic receptors or Mab. be selective. Second, somatogenic receptors and Mab have similar structures. I also recognize the structure. The receptor and at least four Mabs are located in the protein tertiary structure. Recognizes discontinuous epitopes of anti-inflammatory sensitivity. Third, circular dichroic spectra of all purified mutants. Tol is actually the same as wild type hGH. Fourth, pGH (164-190) is excluded. All mutants expressed essentially the same amount as the wild type.

In vivo proteolytic resistance has been used to screen for structural integrity.

Hecbt, M. H1+ et al. (1984) Proc.

Natl, ^cad, Sc1. [ISA 81.　5685　;Siritle (Shortle), D.

(1985) Genetics 110.539 segment substitution hGH mutations Changes in binding activity of the body do not necessarily indicate that the substituted residues in these mutants are directly somatogenic. Disruptive mutations that do not necessarily indicate contact with receptors are favored. This not only eliminates desirable interactions, but also induces undesirable interactions. For example hP The Nl 2R mutant in the RL (12-19) segment substitution hGH mutant is As In addition to changing the hydrogen bonding amide group of n12, the Arg substitution also changes the positive charge. Introducing larger side chains. Even more binding contacts are maintained between the analogs. As a result, some regions, if not all contacts, are replaced by segment hG. This can be investigated using the H mutant.

To identify specific active amino acids within active domains A, C, and F of FIG. We performed precise structural analysis of their active domains.

In this analysis, these three active domain residues were sequentially replaced with alanine. 6 Three single alanine mutants were created and each soluble hGH receptor (s hGH The binding constant for r) was determined.

Leung + D, II, etc. (198B) J, Biol, C hew, 263°7862° Based on this analysis, the amino acid residues listed in Table ■ are somatogenic receptors. constitutes the residues within the hGH molecule that are active for interaction with -. This is that The relative dissociation constants induced by alanine substitution of these residues compared to wild-type hGH This is based on the fact that it will more than quadruple. See Figure 7.

Preferred amino acid substitutions of these residues to create hGH variants are shown.

■Table hGH residue preferred amino acid substitution FIOG [! MARQSnlLIV P54 G[! MARQSnlLrV F55 GMPARQSDNKLH Chapter ■Table (continued) +58 T to GIuMP8RQS R64G [! MPAQSH, Niguchi N Q68 GEMPARSHKDN B171 N to GEMFARQSII 172 GEMFARQSH RoN E174 GMFARO3HDNKI.

Tl75 GEMFARQSvI SYWt to F176 GEM8R, mouth N to IVR178GEMF^QSH C182GEMPARQS V185 G [! Ml’A11OSITLY11 somatogenic receptor Table V shows other amino acid residues that are less active. Generally these residues The group shows a ~2-fold increase in relative Kd value when substituted with alanine.

Table V 14 N12 S55 B66 Q181P5 N14357 K70 R18 3L6 Li2 R59S71 G18737 R16S62 K168 R8R19N631179 hGH (showing greater affinity for Maddzienic receptors) The amino acid residues in are listed in Table 2.

■Table R2B65 3184 T3 Q69 B186 LIOI, 73 5188 R18R167R191 R64B174 A single-residue substitution hGH mutant, B174A, is surprisingly a somatogenic receptor. The dissociation constant for the protein is significantly (almost 5-fold) reduced. somatogenic In addition to showing binding affinity for the receptor, this mutant In multiple assays, it showed an increase in besomatogenic ability relative to hGH. This replacement and and other specific residue substitutions that showed a 1.4-fold or greater increase in somatogenic binding. Shown in the table.

■Table hGH variant R64K showing increased somatogenic binding E65 A L73 A E174 A, N, Q, S, G El 86 A 3188 A F], 91 A Other variants containing alanine substitutions not shown in Figure 7 are listed in Table 1. .

■Table □　Ka　(mM)　Kg(var)/Ki(wt)821＾　NU　- ni172^/F176^　201　543N47A　O,842,3 P48A NE - Q49 80, 36 1.0 750A 0.38 1.0 51A Q46＾　ME　- V173A NE - Note: NE - those that were not easily isolated in shake flasks (i.e. (i.e., approximately 5% or less of the wild-type expression level) Scanning amino acids for preliminary analysis of active amino acid residues for receptors The analysis is carried out by substituting various amino acids other than mutants were shown.

■Table □ Ka (nM) Kg (var) / 4 (wt) R77V O, 441, 3 L80D O,782,3 P176Y 3.2 8.6 Chapter ■ Table (continued) ! 174G O, 150, 43 [! 1740 ME E174HO,431,2 E174K 1.14 3.1 E174L 2.36 6.4 E174N 0.26 0.7 E174Q O,210,6 E174S O, 110, 3 E174V O,280,8 E174RNE 964K O, 210, 6 E65K NB [! 658 Ml! 17211 N [! 158L NE F25S N[! 026E　XE Q29S N! ! ! 300 NB R178K NE R1787Nil! R1780Nl! 1179M NE DI69N 3.6 10.5 Note: NE - those that were not expressed to easily isolated levels in shake flasks ( i.e., approximately 5% or less of the wild type expression level), in addition to the prepared hGH mutants, The table shows specific amino acid residues in hGH and the generation of variants with altered biological functions. The expected substituted amino acids are shown.

Table X 5 4 3 G[! MF 8R to ll mouth NF 44 GBMARO3YIILI VH18GEMFARQSKDNY E 6 5 GEFARQSIIDNKLL73 GEMFAROSIVY E l 8 6 GMPARO5HDNKS 1 8 8 GEMF AROI I mouth NKYF l 91 GEMARQSYIILIVF 97 GBMARO SYIIILIVA 9 8 GEMPRO 3 mouths NH N　9　9　GEMFARQS口KY 3 1 0 0 DNKYL to GBMF 8RO 1 0 1 GEMFAROS IVYV 102 GEMFAllOSITLY? 1Y] 03 GBMPA ROSIILIVG 104 EMf+＾RQSP R19GBMP＾QSIIKN口 Q2 2 KON to GEMFARS D26 GEMFAROSIIKN Q29 GEMFARSKKDN E 3 0 GMPARQSII NKLE 3 3 GMFARQSII N In another embodiment, the binding epitope of hGH to the prolactin receptor We have decided on the next step. h　G　H is also a growth hormone receptor and prolactin (PRL ) may also bind to receptors. These receptors as shown herein - compete with each other for binding to hG and H, indicating that their binding sites overlap. It shows that you are doing it. Scanning mutagenesis data are available at hPRL level. The hGH epitope for the receptor is located in the middle of helix 1 (residues Phe25 and /11sp26), loop region (including Ile58 and Arg64), ) and the middle of helix 4 (residues 168. to 172.EI74 and F 176). These residues are shown in the structural model map of hGH. Form a batch when making. This combined batch is described herein and B. C. Cunningham and J. A. Wells (I. (1989) 5ceince 244. 1081-1085) This overlaps with that determined for the hGH receptor published by It's not the same thing. The non-overlapping regions of these receptor binding sites for hGH By causing mutations, the selectivity of hGH can be adjusted by binding affinity to each receptor. 2000 times more or hPR to hGH receptor side without loss of tea It was shifted more than 20 times to the L receptor side. Similarly, mutations can occur in overlapping regions. As a result, binding to both receptors could be reduced by more than 500 times simultaneously. This way Receptor-selective mutants of eel hGH can be used in hGH species such as linear growth or lactation. Useful molecular probes linking various biological activities and specific receptor binding phenomena becomes.

In another embodiment, human growth hormone (hGH) receptor binding determinants are A synthetic analog was placed in human prolactin (hPRL). This specification and Kaninga Cunningham, B., C., and Ilellg. J, ^, ((1989) 5ceince 2nd Sat↓, 1081-1085) Published alanine scanning mutagenesis cloned from human liver We identified key residues in hGH that modulate its binding to hGH receptors.

Another mutation derived from hPRL was introduced into hGH and within the hGH receptor binding site. revealed that substitution of hPRL almost abolished the binding ability.

Thereafter, hPRL cDNA was cloned and expressed in E. coli.

Then, we sequentially introduced substitutions from hGH that seemed to be very important for receptor binding. After repeating one-site directed mutagenesis seven times, the association constant hP containing eight mutations that enhance the hGH receptor by more than 10,000 times. RL mutants were identified. This hPRL mutant binds only 6 times weaker than the wild type. However, the sequence identity with hGH was only 26%. These results indicate that hGH and hPRL, indicating the identity of the hGH receptor epitope. It is confirmed. Furthermore, these experiments generally involve the use of agonists or antagonists. It has been shown to be useful in designing hybrid hormones with new properties. It is easy to borrow receptor binding properties from near-green and functionally diverse hormones. It shows what is possible.

What follows is an example and is not intended to limit the scope of the invention.

Example 1 hGH mutagenesis and expression vectors for efficient mutagenesis Synthesis with 18 single restriction sites evenly distributed without changing the coding sequence Created the hGH gene. The synthetic hGH DNA sequences were each approximately 60 base pairs long. Ligation of seven synthetic NA cassettes with 10 bρ overlap with adjacent cassettes The 405 bp D indicated by BglII was assembled from N5il by I made an NA fragment.

This ligated fragment was purified on a polyacrylamide gel and eluted from it. Lucari phosphatase promoter and 5i11 signal sequence (Chi+7 (C Hang), C, N, et al. (1987) Gene 55゜189), Phage fl replication origin and b containing the replication origin and β-lactamase gene Recipient vector pB04 containing the region of pBR322 from p1205 to 4361 75 was similarly cut and cloned. This sequence was determined by dideoxy analysis. (Sanger, P. et al. (1977) Proc.

Natl, Acad, Sci, USA 74. 5463).

pBo475 was constructed as shown in FIG. Dral of length 475 bp, The f1-derived DNA from the fiber phage contained on the R5B1 fragment was p was cloned into the single Pvu11 site of BR322 to create plasmid p652. . Most of the tetracycline resistance genes contain p652 as Nhel and Narl. restriction cut and fill in its sticky ends using DNA polymerase and dNTPs, and ligation of the larger 3850 bp fragment itself. Plasmid pΔ652 was constructed by deletion. pΔ652 to EcoR1 , EcoRV, and the 3690 bp fragment was digested with alkaline phosphatase. ρhG containing the enzyme promoter, 5TII signal sequence and neutral hGH gene. 11411 (Chang, C. No. et al. (1987) Gene, 1300 bp EcoR1, EcoRV7 fragment derived from 55.189) Ligated. This construct was named pBO473. 11i, DNA was cloned into pBo473 in three ways. Vector pBO473 N5 i1. 240bp N5i1 cut with Gg1II and both synthesized from NA ．． Hind u [Fragment and 1170bpHjndll.

Bgll [ligated into fragment. This construct pBO475 is hG Mutagenesis of the hGH gene, which contains DNA encoding a neutral polypeptide of H. It has many new single restriction sites that allow for further manipulation. hGH a The entire DNA sequence of pB0475, including the amino acid sequence, is shown in FIG. pBO47 The single restriction site within the hGH sequence in 5 is derived from the analogue pGH, hPL and hPRL. allows the insertion of a mutagen cassette containing a DNA sequence encoding a similar segment of (Hells, J. 6 et al. (1985) Gene 3 Sat. 315), the hGH sequence was separately subjected to site-directed mutagenesis of the single-stranded vector pBO475. modification by a single restriction site followed by restriction selection for one of the single restriction sites (U. -ells, J, ^1 etc. (1986) Ph1los, Tran s, R, Sac, LondonSerA3ee, 415) Seventeen segment substitution hGH mutants listed in Table 1 were prepared. each wild hGll-hGH hybrid as described later. It was secreted into the periplasmic space of E. coli at levels exceeding 50%. hGH and hGH mutations The body is E. coli W3110. grown in low phosphate minimal medium. tonA strain (AT CC 27325) was expressed in (Chang, C, N, et al. (19B?)). ) Gene　55゜189).

hGH and hGH mutants were purified as follows. 4 to 200g cell pellet Double volume (800 ml) of 10 sM) Lith (pH 8.0) was added. Pipe this mixture The mixture was shaken at room temperature until the pellets were dissolved. Homogenize this mixture After enlarging and stirring in a cold room for 1 hour, the mixture was centrifuged at 7000 g for 15 minutes. On top of that After decanting the supernatant, ammonium sulfate was added to make it 45% saturated. (277 g/j) and stirred at room temperature for 1 hour.

After centrifugation at 11,000g for 30 minutes, the pellet was incubated with 40wj of 10mM) H8,0). This solution was added to 21 of 10-M Tris (pH 8,0). Dialyzed overnight, the sample was centrifuged or filtered through a 0.45 micron membrane. So After that, the sample was added to 100 liters of DEAR cellulose (Fast Flow, Pharmacia). Co., Ltd.) column. 10+sM (8 to IO times the column volume) (pH 8.0) ) The column was eluted using a gradient from zero to 300 mM NaCl. P.A. The hGH fluxin identified by GE was collected and diluted with lo-M Tris-H (1 (pH 8, 0> overnight. This sample was subjected to Centri-Prep 10 ultrafiltration. It was concentrated to approximately lag/mf.

Example 2 hGH and homologous recombinants of hGH The N-terminal region of various lengths of hGH was conjugated to the C-terminal region of porcine growth hormone. Random recombination of genes by serially combining random hybrid libraries containing Constructed by law. Gray, G, L, et al. (1986) J, Bacterol, the EcoR1 site of i6.635°pBO475 was Removed by EcoRI restriction treatment of Sumid and added with DNA polymerase and dNTPs. This plasmid was ligated after filling in the sticky ends by adding . Next, a new EcoR1 site was introduced just before the 3' end of the hGH gene. child This is the 345bp BglU of hGH-4R containing the EcoR1 site, and the EcoRV fragment. Safri fragment into a similar restriction lefter from the EcoR de pBO475 construct. This was done by rowing. Then the pGH gene (See Barb) burg) P, H, etc. were introduced immediately after the end. This is the EcoR of the construct mentioned earlier. I.

EcoRI containing pGH cDNA in the larger fragment of EcoRV digest , H3nd m (filling) fragment was introduced. this Plasmid pBO509 contains native hGH with a single EcoR1 site at the 3' end. hGH gene and the original pGH gene following it in the same direction. Due to the homology of the pGH gene, a part of pB0509 is derived from E. coli rec” MM. 294 (ATCC31446) caused in vivo recombination. and generates a hybrid hGH/pGH. These recombinants were collected Restrict the DNA with EcoRI to avoid recombination that would result in the loss of the EcoR1 site. The isolated plasmids were concentrated by linearization. 2 limit selections and large intestine After transformation into the bacteria rec' MM 294, almost all clones The loan became a hybrid hGH/pGH recombinant. 22 cubes Sequence analysis of the clone shows that the hGH/pGH hybrid has an amino-terminal hGH sequence and its Amino acid residue following +19. +29. +48. +94°+105. +123 and the pGH sequence starting at +164.

7 hGs with crossover points evenly distributed among hGH genes) (-pGH hybrids) obtained. However, the extreme carboxy terminal) 1 hybrid (hGH (1-1 Only 03)-pGH(164-191)) is available at sufficient levels for purification and analysis. Secreted by enterobacteria.

This hGH-pGH hybrid consists of three positions on the hydrophobic face of helix 4. (M170L, V173A and V180M). Therefore, Figure 2 Most of the sequence changes in the helical regions A, D, E, and F of the helical was designed to avoid mutation of residues on the hydrophobic face of . For example, the 71 Ipri mentioned above - Nodo hGH-pGH mutant is M170. V173. F176 and V180 Modified to maintain. Because these residues are located on the hydrophobic face of helix 4 Because it exists.

Example 3 Expression and purification of soluble human growth hormone receptor from E. coli Cloned DNA sequence encoding soluble human growth hormone receptor 5hGHr Leung + Lo W, et al. (1987) Nature 330 ．． 537) was subcloned into pBo 475 to create pJ1466 (11th (see Figure and Figure 12).

Vector pcls, 25HGHR (Leung, W. et al. (198 7) Nature 330. 537) was digested with Xbal and Kpnl, 1, Obp, which encodes the secretion signal and 246-codon extracellular region of hGH. The fragment was purified (Maniatis, T. et al. (198 2) Molecular Cloning, Cold Spring Harbor Laboratory, N. , Y, ). This fragment was ligated to similarly cut M13mp18. single-stranded DNA was purified for recombinants (Messing, J. .

(1983) Methods in Enzysology, 101! %20 pages ) 0-site directed mutagenesis (Carter, P., et al. (1986) Nucleic Ac1ds Res, 13. 4331) to Igset, Gigo Nucleotide 5' -A-AGT-GAT-GCA-TTT-TCT-GG-3 ' to introduce N5il at codon +l. The mutant sequence is a dideoxy sequence Analysis (Sanger, P. et al. Proc. Natl.

Acad, Sci, USA74. 5463). mutant two The double-stranded DNA was purified and cut with N5il and 5eal. hGH receptor 2 A 900 bp fragment containing the 46 codon extracellular region was isolated. pBo 4 75 was cut with N5il and EcoRV to create a 4.1 kb fragment (synthetic hG (lacking the H gene) was purified.

900 bρ fragment of the receptor and 4. lKb vector fragment ligation, and the recombinant clone (pJ1446) was confirmed by restriction map. did. This was transformed into E. coli KS303 using a straw. ch), 1st prize (1988) Proc, Natl, Acad, Sci, USA 85. 1576), then low phosphate medium (Chang +7) ), C, N, (1987) Gene 55° 189) at 30°C.

Receptor fragment proteins are analyzed using hGH affinity chromatography. Purified (Spencer, S., A., et al. (1988) J. Biol, Chew, 263°7862; Leung D, (1987) Nature 330°537). The sequence of pJ1446 The cloned receptor is shown in FIG. 12 along with its amino acid sequence.

Escherichia coli W3110. degP strain (Strauch, K, L (1988) PNAS USA Dan 5.1576) with ρJ1446 Fermentor Medium Low Phosphate Medium (Chang, C,N, (1987) Gene 55. 189) and grown at 30°C. 24 Preliminary data were obtained using 6 amino acid hGHr. Contains amino acids 1-238 A slightly shorter hGHr was used in Example 1 herein. About this receptor The results obtained were indistinguishable from the values obtained with the 246 amino acid hGHr.

A stop codon was placed after GIn238 to avoid carboxyl-terminal heterogeneity issues. The plasmid phGHr (1-238) was constructed. phGHr (1- From KS330 cultures containing phGHr (+-24 6) was produced in slightly higher yields and more uniformly than from cultures containing (data not shown), starting from cell pellets with a wet weight of 0.2 kg, 70-80% 20-40-g of high-purity bound protein was routinely isolated in yields of 1), N-terminal sequencing and mass spectrometry analysis Peptide mapping combined with It was confirmed that

Oligonucleotide site-directed mutagenesis of phGHr (1-246) template leotide (Carter + et al. (1986) Nuclei cAcids Res, 13. 4431-4443) PhGHr (1- 240°C241R) (in the formula, the asterisk stands for phGHr (1-246 ), underlined is a new single Mlu1 site, and CGT-T AG indicates the C241R mutation followed by a stop codon (Table XA).

Table X Amino- and carboxy-terminal sequences of hGH-binding protein constructs Hesei” 1 Tan / DNA / sil Table X (continued) Gin-^H Plasmid phGHr (1-238) was inserted into the Mlu1 site (Table X (A)) RuwIII! Selection (Wel Is et al. (1986) Phil, Tr. ans, R, Soc, Lond, ^, 317, 415-423) Site-directed mutagenesis of hGHr (1-240, C241R) tear plate Generated by. Simply put, oligonucleotides induces a translation stop codon after G I n 238 (CAA) librent) and the Mlul restriction site (underlined) was corrected. First trans by heteroduplex After recovering double-stranded DNA from Sufufusilan, it is re-transformed and the DNA sequence is - Increase the target phGHr (1-238) plasmid before sequencing. Ta.

Finally, the hGH-binding protein cloned into PhGHr (1-238) T51A mutation resulting in cDNA variants or cloning artifacts This difference was confirmed by DNA sequencing. Therefore A31T Reverter The sequence is the same as the reported sequence (Leung et al. (1987)). Nature (London) 330, 537-543°T at site 5I Purification and binding of proteins containing hr or Ala are indistinguishable (data not shown). Sazu). The characteristics of Ala51-binding protein mutants have been clarified, so selected for all analyses.

Characteristics of natural products isolated from human serum and recombinant hGH-binding proteins from E. coli In order to compare the isomers, the affinities of the wild type and various hGH mutants were measured.

Table X (B) K of hGH-binding proteins (nM) ± S, D.

wt　O,55±0.07　0.40+0.03　-　1.4158＾　21± 2 38±6 14±1 36±51.5R64A 12±1 22±4 11 +1 28±5 1.1E174A 0127±0.04 0.49±0.11 0.16±0.01 0.4±0.1 1.7P176A 71±7130± 2048±5120±20 1.5a, Kd value and its standard deviation (SD) are: Competitive binding analysis using wild-type hGH (wt) and many hGH mutants (Part 1) Figure 24).

b, hGH mutant and wild type hGH for each 1tGH binding protein ( The decrease in binding affinity calculated from the ratio of the dissociation constants of wt).

C1 Ratio of dissociation constants of two hGH binding proteins for a given hGH type.

Both proteins formed a complex with hGH in a specific stoichiometry (Figure 24), which As can be seen, the affinity 4- of wild-type hGH and its mutants has two bonds. The 11-change hGH-binding protein was almost the same among proteins (above, right column). It has a very high affinity compared to neutral proteins derived from human serum. child This reflects the purity and homogeneity of the recombinant protein. hGH binding Binding to four alanine mutants of hGH that destroy its ability to bind to proteins Both proteins have the same specificity as shown by changes in affinity. (Kd (mutant) / Kd (wt)), which extends to Tyr246. The affinity of hGH for binding proteins (Kd-0.36±0.08) M) is essentially the same as that ending after G1n23B (0,40±0 ．． 03nM), which means that the last eight residues of the molecule, including the seventh cysteine, indicates that it is not important for hGH binding.

Example 4 Receptor and monoclonal antibody binding assays Purified hGH or hGH mutants Binding of Example 3 to soluble hGH receptors (>95% purity) We conducted a test regarding. Laser decoding of combage-stained gel after 5DS-PAGE The concentration of purified hormone was determined by cytometer scanning. these values was in good agreement with the concentration determined by absorbance at 280n- (6□,.1%-〇, 93) . The dissociation constant (Kd) of binding to soluble hGH receptor at 25°C [” 'I) Calculated from Scatchard analysis for competitive displacement of hGH. this[" 'I) hGH is described by Spencer, S. A., et al. (1988), J, Bioches.

3.7862).

Different segment placements and residues were detected using enzyme-linked immunosorbent assay (ELISA). The binding of eight monoclonal antibodies to substituted hGH mutants was assayed. below Mab was used.

Mab name origin/method I MabA (*) 2　33.2　Hypurichiksha 3 Catt H-299-01 Mezzo 4”) Cusbiotech 4 72.3 Ha Iprichik company 5 [atj H-299-02 Mezzo 4” J with Kusuba Chikusha 6 Mab 653 'r Mini 17 ? MabD (*) 8 MabB (*) *Carbone, P, R, et al. (1985) J, Ian unol.

135.2609° Rabbit polyclonal antibody against hGH was affinity purified and 0.005 Final concentration of 2 μg/l in M sodium carbonate (pH 10), 24°C, 16-20 Coat the microplate (Nunc plate, Intermet, Denmark) at h. I started. The plate was washed in buffer B (50mM) with lysate [pH 7,5]. , 0.15M NaCff1, 2mM EDTA, 5+ag/5IBsA, 0. 0.05% Tween 20.0.02% Sodium Azide) 0.1agoml each h It was reacted with GH at 25°C for 2 hours. After washing the plate, incubate with buffer B at 150-0. The indicated Mab was serially diluted to 0.002 nM. 2 hours later play cells were washed and stained with horseradish peroxidase-conjugated anti-mouse antibody. It was tested. The obtained value is necessary to give half the maximum binding amount to each hGH variant. The concentration of each Mab required is shown.

Competitive displacement of hGH receptor from hGH by anti-h’GHMab is as follows. was measured. This assay was performed using anti-Gi (rabbit polyclonal antibody) as described above. This was done by immobilization of wild-type hGH in coated microplates. Receipt (fixed at 10 nM) and the designated anti-hGHMab (150-0°002 (diluted in nM range) into hGH-coated microplates at 25°C for 16-20 min. After incubation for 1 hour, unbound components were washed away. The amount of bound receptor is h Horseradish peroxidase that does not inhibit the binding of GH to its receptor Quantification was performed by adding an anti-receptor Mab conjugated to. The standardized replacement value is The receptor for half the Mabfi degree required to saturate hGII on the plate. Calculated from the ratio of 1 degree of Mab required to displace 50% of the protein. This value is It was used to compare the ability of Mab to replace the receptor.

Example 5 Soluble Somatogenic Somatogenic Receptor Binding Active Domain Example 3 Regarding binding to the monoclonal antibody described in Example 4, The 17 segment substitution hGH variants described in Examples 1 and 2 were assayed. Soma The results of the binding assay for togenic receptors are shown in Table 2.

As can be seen, segment substitutions that almost destroy binding ability are shown in Figures 4 and 4. These are the areas A, C, and F in Figure 5. Furthermore, these areas can be made even smaller. hG involved in binding to somatogenic receptors The location of the active domain of the H molecule was determined. Figure 4 shows the most important results from Table ■. This figure shows the replacement of the H sequence from the analogs hPRL, hPL and pGH. Bar graph showing the relative decrease in binding to soluble hGH receptor by be. 3 containing amino acid residues 12-19, 54-74 and 164-190, respectively. Three active domain regions A, C and F were identified. These areas are shown in Figure 5. Shown in a three-dimensional representation of an hGH molecule.

As can be seen, the three active domains A, C and F are the amino acid sequences of hGH. Defines the somatogenic link binding site for hGH, although it is discontinuous within the column. In the molecular structure, a continuous region is formed.

Example 6 hGH epitope mapping The binding of eight monoclonal antibodies to specific segment-substituted hGH variants was determined. Shown in Table XI.

Table XI wthGl(0,40,40,10,050,20,2G,080.1hPL( 12-25) 0.40.4 >75 >500.2 0.2 0.080.1 pGH(11-33) 0.4 > 1001.50.050.2 0.2 0. 080.1hPRL (12-33) 0.4 >100 >75 >500.2 0.2 0.0B 0.1hPRL (12-19) 0.4>12>75 >500.2 0.2 0.080.1hPRL (22-33) 0.40. 40.10.050.2 0.2 0.0B 0.1hpL (46-52) 0 ．． 40.40.10.050.2 0.2 θ, 400.1pGH (4B-52 ) 0.40.40.10.050.2 0.2 0.080.1 Table xl (continuation) hPL (56-64) 0.4 0.4 0.1 0.05 0.2 0.8 0.08 0.1pGH (57-73) 0.4 0.4 0.1 0.05> 200>200 0.08 0.1hPRL (54-74) 0.4 0.4 0.1 0.05 0.2 0.6 0.08 0.1hPRL (88-95 )>400 0.4 0.1 0.05 0.2 0.2 0.08 0.1 hPRL (97-104>>400>12 0.1 0,05 0.2 0. 2 0.08 0.1hPL (109-112)>12 0.4>75 15 0.2 0.2 0.08 0.1hPRL (111-129)>12 0. 4>75>50 0.2 0.2 0.08 0.1hPRL (126-1 36>0.4 0.4 0.1 0.05 0.2 0.2 0.08 0.1 pGII (164-190>0.4 0.4 0.5 0.03>25 12 ．． 5 0.20 0.4pGH (167-182) hGII (Δ32-46) 0.4 0.4 0.1 0.05 0.2 0 ．． 2>100>100N12A O, 40, 4>75>50 0.2 0. 2 0.0g 0.1C182^ 0.4 0.4 0.1. 0.05 2. 0 0.2 0.08 0.1pGH (each monoton except 167-190> mutant) The destruction of binding capacity for clonal antibodies is significant and highly selective. 13th Figure ~20 shows the location of the epitope for each Mab within the three-dimensional structure of hGH. I'm doing it. Figure 6 shows epitopes for binding sites for somatogenic receptors. It shows a loop.

For example, hPRL (88-95), hl'RL (97-104).

hPL (109-112) and hPRL, (111-129) mutants are Ma Other segment substituted hGHs that do not bind to bl but outside those regions are wild-type hGH It bound with similar efficiency to GH. Mab2゜3, 4. 5ii and the conclusion to σG This is a discontinuity in the primary sequence, but in the three-dimensional structure of the hormone, there are regions that are close to each other. (see Figure 6 and Figures 14-19). In contrast, Mabl, 7 and 8 are as shown in Figures 13, 19 and 20. was inhibited by mutations defined in contiguous sequences. In addition, a predetermined monochrome The region that inhibits binding to the null antibody is created by substituting a specific segment of the hGH8i region. subdivision into domains or variants with common substitutions already bound to a particular Mab. This was analyzed by variant analysis. For example, pGH(11-33) has a strong binding to Mab4. hPRL (12-33) had lost its binding ability. did Therefore, the disruptive mutation in the hPRL (12-33) mutant Residues not mutated in 1-33): N12. Li2. R16, D26 and R may be limited to 30. Furthermore, the hPRL (12-19) mutant lost its binding ability; Since hPRL(22-23) was not lost (see Figure 16), this is N. 12. Limited to L12 and R16. The N12H mutation causes pGH (11-3 3), this mutation explains the inhibition of binding to Mab4. obtain. This was tested with an alanine substitution at Asn-12. N12A residue substitution Binding of Mab3 or Mab4 to hGH mutants was reduced by more than 100-fold, but other Binding to Mab was not affected.

By using this cent hGHjij variant, most of the eight Mab The effects of all these Mabs even if binding is inhibited by a common set of mutations. Pitope can be elucidated. For example, hPRL(12-19) is associated with Mab2.3 and 4, but other mutants show that these Mabs recognize different structures. It was shown that

In particular, Mab2 and 3 are inhibited by pGH(11-33), but Mab4 is unaffected. It didn't resonate. Binding of Mab3 and 4 is inhibited by hPL(12-25) binding to Mab2 was not affected, thus the eight antibodies overlapped. However, they have epitopes that do not overlap completely. Mutations that disrupt binding are helical and loops and are always close in the hormone structure.

Taken together, most of the epitopes in the eight Mab sets are for this hormone. to cover. However, there are still regions where Mab does not bind. For example, 20 Three of the mutants were tested Mab (hPRL (2233), pGH (48 52) and hPRL (126-136)). I didn't.

There are significant differences between antibody epitopes and receptor binding sites. First, inhibitory mutations The punch defined by the difference is more about the receptor than about any Mab. The second difference is that the receptor is more sensitive to inhibitory displacement of the hormone than the Mab. It is more resistant.

This means that the maximum decrease in binding of any mutant to the receptor is approximately 70 times. However, almost all antibodies contain 1 as is the result of a single substitution such as N12A. There is at least one variant that causes a 1,000-fold decrease in binding.

Example 7 Competitive binding of Mab and 5hGHr Many mutants that disrupt receptor binding also disrupt binding of one or more Mab, thus , the ability of each of the eight Mabs to inhibit hGH receptor binding to hGH. was evaluated. The results of this test are shown in the XI Summer Table.

Table Xl1 1 0.4 > 150 > 375 2 0.4 0.8 2 3 0.1 150 1500 4 0.05 150 3000 5 0.2 0.2 1 6 0.2 0.2 1 7 0.08 0.4 5 8 0.1 >150° >1500 (*) Mab8 binding appears to slightly promote receptor binding to hGH. be.

+ Data from Table X regarding the binding of each Mab of hGH.

As can be seen from the table, Mab 5 and 6 are the most effective in inhibiting hGH receptor binding. are also efficient, indicating that these MAbs are in close contact with the determinants for the receptor. Determination of antigens present within repeating residues 54-74 and helix 4 By having a group (see FIGS. 5, 6, 17 and 18), Ma b2 is then a competitive antibody and shares a common inhibitory mutation with the receptor. (hPRL (12-19)), whereas Mab3 and 4 have Mab2 Although they are roughly 1000 times less competitive than It shares a recurrent inhibitory mutation with the protein.

See Figures 15 and 16, mutations in helix l that disrupt Mab 3 and 4. Residues that disrupt binding to Mab2 or receptors Table XIV (continued) tree) +58A 5.6 17.0 P59A　O,651,9 T60＾ Anal f’61＾　NB 562A　O,952,8 N63^ 1.12 3.3 R64＾　7.11　21.0 [1658 O,200,6 E66A 0.71 2.1 T67A NB (168A 18 5.2 069＾　0.3]　0.9 70＾　0.82　2.4 S71A 0.68 2.0 N72＾　NE L73A　O,240,70 [! 74A NE Table x ■ Amino acid scanning variant K of site 16'l-191 in hGH (nM ) Kd variant/Ka IITWT 0.34 1 R167＾　0.26　0.75 KI68A　O,371,1 D169A NU Table XrV (continued) M170A N11i Mouth 171A 2.4 7.1 172＾　4.6　14 V173A Nl! [+174A　O,0750,22 T175＾　NB T175S 5.9 16 P176＾　5.4　16 L177＾　NE　-+ R178A NE R178N], 4 4.2 1179＾　0.92　2.7 V180A 0.34 1.0 0181A 0.54 1.6 C182A 1.9 5.7 R183A　O,712,1 S184A　O,310,90 V185A 1.5 t 5 8186 80, 27 0.80 6187 80, 61 1.8 3188＾　0.24　0.7 C189＾　NU G190＾　Nl! R191A 0820 0.60 Alanine substitutions include residues 2-19 due to ambiguity in the amino-terminal residue position. (Abdel-Meguid, S., et al. (1999)) 87) Proc, Natl, ^cad, Sci, USA 84.643 4).

In fact, the most significant decrease in binding was in FIOA (6-fold) followed by helix l. With regard to binding, which occurred with an alanine substitution of residues 4-6 at the N-terminus (see Figure 21), Substantially larger effects (more than 20 times) were observed in the loop and carboxylate residues 54-74. A specific alanine substitution within the x-terminal sequence 167-191 occurred. some ara Binding of the nin mutant was increased 4.5-fold. The most dramatic example is the destructive alanine It was R174A, which is located in the middle of the mutations.

See FIGS. 4, 7 and 21.

The most destructive alanine substitutions are FIO to R64 and 0171 to v185. Form a batch of approximately 25 human squares on the hormone (see Figure 21). moreover These side chains appear to be oriented in the same direction on the molecule. For example, Herrick All alanine mutants (r)171A, which most affect binding with respect to 2A.

R174A, R176A, +179A, C182A and R1B3A) are It is limited to three and a half times of helix, and those side chains protrude from the same side of this helix. It is out (see Figure 21).

Based on this model, T175 and R17B are Since it occupies a central position in the sea urchin, it is expected to be involved in binding.

Tl75A mutant is not expressed in sufficient yield to assay in shake flasks However, it was a more conservative variant (T175S), thus Tl753 The mutant had a 16-fold reduction in receptor binding. Similarly, R178A is less expressed. Although not present, R178 was expressed in analyzable yields. R178N is The binding affinity decreased by more than 4 times.

The next disruptive mutant at the carboxy terminus was V185A. ■185Δ is Although outside of helix 4, this model allows for destructive projections within helix 4. It is expected that this will be in the same direction as the natural mutation. In contrast, Lanine mutations or alanine mutations therein pointing opposite to those mentioned above Mutations (R167A, 168A, V180A, C181A, 5184A, R18 6A, 5188A) generally have no or little effect on receptor binding. It has no effect.

Alanine mutants in helix 1 have only modest effects, but this also We conducted an analysis of The most bond-disrupting alanine substitution within this helix is of residues 6, 10 and 14, which are on the same side of the helix. More Both small disruptive alanine mutations (L9A, N12A and L15A) helical It exists on the opposite side of Kusu 1. Furthermore, this suggests that binding to hGH is Both anti-hGHMabs 3 and 4 bind to Asn-12, which does not compete with This is confirmed by the fact that See wxvi table.

Table xvi hGH and alanine modification for 8 anti-hGH monoclonal antibodies (Mab) combination of variants ab Hormone 1234 567 8 hGHO, 40, 40, 10,050, 20, 20, 0B 0. IPIG＾　0 ．． 4 0.4 0.1 0.05 0.2 0.2 0.08 0. lNl2^ 0.4 0.4 >75 >50 0.2 0.2 0.0B 0.1158 ＾　0.4　0.4　0.1　0.05　0.2　0.2　0.0B　0.1R 64＾　0.4　0.4　0.1　0.05　0.2　1.6　0.08　G, IC68＾　0.4　0.4　0.1　0.05　0.2　0.2　0.0B　 0.1 housing 168A O, 40, 40, 10,050, 20, 20, 080, 10 171^ 0.4 0.4 0.1 0.05 0.2 0.2 0.08 0 ．． lK172A　O,40,40,10,050,20,20,080,1!1 74＾　0.4　0.4　0.1　0.05　0.2　0.2　0.08　0. IF176A　O,40,40,10,050,20,20,0B　0. IC1 82^ 0.4 0.4 0.1 0.05 2.0 0.2 0.0B 0. IC182^　O, 40, 40, 10,050, 20, 20, 0B 0.154 The relative positions of the side chains within the -74 loop are such that they are of the side chains within helices 1 and 4. However, mutations in even-numbered residues result in greater binding than odd-numbered residues. There is a remarkable periodicity in the binding that is inhibited. This is especially true in this area. The first part fits well (54-59) and even-numbered residues project toward the receptor. This reflects the structure in which the odd-numbered residues are exposed and facing in opposite directions.

Example 9 Some evidence suggests that the structural integrity and binding energies of alanine-substituted hGH mutants Alanine substitutions that disrupt receptor binding do so by distorting the structure of the molecule. This shows that it is not rubbing. First, the eight Mabs have similar levels of hGH. Reacts with almost all alanine mutants that cause disruption of binding to the receptor.

See Table X■ above.

The exceptions are R64A and R64A, which selectively abolish binding to anti-hGHMabs 6 and 5, respectively. and C182A. These two Mabs have previously been shown to be somatotropic for binding to hGH. It was shown to compete with the genetic receptor. Further affects receptor binding We created two alanine mutants that did not. One is N which affects the binding of two Mab. 12^), the other does not affect any of the Mab (K168A), this data binding to the Mab or receptor is due to very local fluctuations in the mutant structure. It shows that it will be more destroyed. In addition, all hGH mutants have a circular dichroism pattern. The spectrum was virtually identical to wild type hGH.

Approximately 20% of the alanine variants (DIIA, T60A, R61A , T67A, N72A, R74A, D169A.

M170A, V173A, T175A, LL77A, and 178A.

C189A, G190A) are not secreted at sufficient levels to be isolated and separated. won. Genes encoding such variants are expressed in the same vector and Its expression is independent of specific alanine codons, so there are constant changes in expression levels. probably reflects differences in secretion levels and/or proteolysis of hGH variants. This is what I did. Some of the non-expressing alanine mutants in helix 4 are The hydrophobic side of the lix is shown in Figure 21, where the hydrophobic side is shown in white. (MI7OA, V173A and L177A). But some alanine placement The changes occur in helix 1 (L6A, L9A and FIOA) and helix 4 (F This is because it is accepted within the hydrophobic plane of 176A and V180A). It's not a general effect.

Furthermore, decreased expression of hGH mutants indicates that charged or neutral amino acids are alanine and I! exchange (DlIA, T60A, T67A, N72A, R 74A, D169A, T175A.

R178A). T175S and R178N that preserve the hydrogen bonding group at that site. Such mutations, if any, are expressed at lower levels than the wild type.

The nonexpressing C189A mutant destroys the carboxy-terminal sulfide and its phase The counterpart (CI82A) is also expressed at much lower levels than wild type. how many others The non-expressing alanine mutants (T60A, T61A and T67A) have a loop structure. Exists within the structure. Therefore, low level expression or non-expression results from many structural effects. However, it can be avoided by single structural or single functional substitutions.

Substitutions (158A, R64A) that showed no more than 10-fold effect on binding.

K172A, T175S1F176S, F176A) are involved in direct binding. It seems so. Naturally occurring (Farshed 87; Malivor) , R, et al. (1973) J, Mo1. Bio+.

225; Bryan, p., et al. (1986) Proc, Na tl.

^cad, Sci, USA-83, 3743; Hells, J, ^9 etc. 2222 i-Graph (Graf), t, etc. (1988) Pro c, Natl, ^cad.

Sci, USA 85. 4961) The strength of hydrogen bonds or salt bridges depends on the microenvironment. It is widely distributed in the range of 1 to 5 Kcal/sol. Regarding hGH The decrease in binding free energy due to K56. C68, DI71°K172 and R6 0.8.1.0.1.2.1.6 and 1.8 Kc for each alanine substitution of 4 al/mol (ΔΔG binding-+R, InKd (mutant)/Kd (wt) ), the energy of burying the hydrophobic side chain in the protein is the same as that in ethanol. tends to correspond to the free energy of transition (Estell, D , ^9 etc. (1986) 5science233.659; Nozaki ), Y, et al. (1980) 'Hydrophobic effect' (Willey edition, N, Y, pp. 4-21 ).

Therefore F175A, FIOA, R54A, 158A and V185A+7) The decrease in binding free energy is 1.6.1.0.0.9.1.7 and 0.9 respectively It was Kcal/mol. These values are from Phe+ Summer 1e or Vat Expected values of hydrophobic free energy for substitution with Ala, 2.0.2.4 and 2.0.2.4 respectively. and l, Q1 ((slightly smaller than al/sol, this analysis shows that T The effect of the 175S mutant (ΔΔG binding-1, 6 Kcal/mol) is Effect expected from loss of group (ΔΔG hydrophobicity = 0.7 Kcal/mol) The nature of the molecular contact between hGH and its somatogenic receptor is greater than Direct structural information is necessary for a thorough understanding. However, these alanine The binding energies of the mutants are the same as those for contacting residues in completely different systems. It shows that it is within the range of previous measurements taken. In fact, the best for receptor binding. Binding free energy of alanine substitution mutants excluding C182A, which is also destructive The total amount of energy is (-13.2Kcal/■ol) between hGH and its receptor. It is inferior to the loose binding free energy (-13Kcal/5ol).

Example 10 Reactivity of hGH mutant with anti-hGH polyclonal hGH mutant hPRL (2 2-23), E174A and hPRL (88-95>) in a rat body weight assay. Tested. The results of this test are shown in Figure 22, and it can be seen that hPRL ( Mutants other than 22-33) have a decreasing ability after about 14 days of growth. level of growth down is due to the expression of antibodies against various growth hormones that neutralize the biological effects do. The fact that the hPRL (22-33) mutant continued to grow indicates that it demonstrated that it does not have the same immunogenicity as native hGH or other variants used. Ru.

The reactivity of various hGH mutants to hGH was evaluated using human and polyclonal antibodies. Table XvII shows the results of comparison with serum from mice and mice.

Table XVII Binding site of serum anti-hGH antibodies to hGH variants Table XVII (continued) As can be seen from the table, mutants containing substitutions within the region of residues 22-33 differ from wild-type hGH. has substantially decreased binding activity with each human and mouse antiserum, or In some cases, it has no activity.

Mutant pGH5? Except for −73, mutants containing substitutions in other regions showed significant reactivity. Segment substitution mutants between residues 11 and 33 that do not show a decrease in somatodi These mutants are somatotropic because they retain the ability to bind to genetic receptors. Genesis maintains propulsion but has other properties, in this case anti-hGH polyclonal This shows the generation of mutants with altered reactivity to the antibody.

Example 1I Relationship between Kd and ability Kd (mutant)/Kd (wild type) ratio of specific hGH mutants and rat body weight test A semilog plot of the performance of these mutants in the test is shown in Figure 23, from which As can be seen, the existence of a linear relationship indicates binding affinity to somatogenic receptors. A decrease in - indicates a decrease in capacity.

As can be seen, hGH mutant E174A is associated with somatogenic receptors. It has a higher binding affinity than wild-type hGH. Also, this ability Approximately 12% larger than wild type hGH.

Furthermore, the mutant pPRL (94-104) has basically the same binding constant as the wild type. However, the capacity is about 2.7 times greater.

Example 12 The active domain of hGH for prolactin receptor binding Human Growth Hormone ( hGH) has linear growth, lactation, nitrogen retardation, diabetic and insulin-like effects. It exhibits many physiological effects, including macrophage activation. Battle, Ni, Chaura ( Chawla), J, S.

Parks and Mouth, Rudsan, ^nnu, 1lev.

Med, 34.519-547 (1983); L., G., P., Isaxo. Isaksson, et al. (1985), Annu, Rev, Phy siol, 47°483-499; c, Edwards (1988) Science 239, 769-771゜These effects are Each begins with the interaction of hGH with a specific cellular receptor. J, P, /1i HLIghs et al. (1985) Ainu, Rev, Physiol , 4 Doj, 469-482゜Creating a product that combines with hGH Leung gene liver hGH receptor (A, If, Leung) (1987) Nature (ondon) 330, 537-543) and Human prolactin (hPRL) receptor (J, M, Boutin (Bo (1988) Ce1153. ’69-77).

Receptor 1 spillover of hGH to hPRL receptors is associated with high levels of h Although GH-producing acromegaly patients have normal levels of hPRL, If hyperprolactinosis occurs (D clinical rll syndrome (J, E, 7 rad) Ki7 (Pradkin) et al. (1989) NewBngl, J, Med, 3 20. 640-644>, but there are other receptors that bind hGH. , these include placental lactogen (PL) tz receptor (M, free mark (Fr (eemark) et al. (1987) Bndocr inology) 120.186 5-1872), the binding site on hGH for these receptors is the same or which receptor (receptor group) is responsible for the pharmaceutical effect. Ruka didn't know. To solve these problems, hGH and hPRL receptors The protein binding site was located. The results obtained indicate that these receptor binding This shows that although the parts overlap, they are not the same. This is h Enables rational design of receptor-specific mutants of GH.

The hGH and hPRL receptors have extracellular homologs with 32% sequence homology. binding domain, a single transmembrane domain and a variety of sequences and lengths. There are several intracytoplasmic domains. The extracellular binding domain of hGH receptor is large. Expressed in enterobacteria and identical to those found in nature, such as soluble serum-binding proteins (S, A, Spencer et al. (1988) J , Biol, Che+*, 263. 7862-7867), similarly hPR The extracellular domain of the L receptor was expressed in E. coli and purified. hPRL The receptor fragment spans residues G1n1 to Thr 211 and has a single target. It ends just before the lance membrane main. it is a full length receptor has virtually the same high binding affinity and specificity as In this experiment The gene encoding the hPRL receptor used was derived from Montréal, Canada.

This DNA sequence was obtained from a human mammary gland cDNA library and Cross-species members of the Pter group identified with probes covering known conserved regions. Ta. For example, Davies J, A, et al. (1989) Endr. ocrinology 3.674-680; Edery et al. (1 989) Proc, Natl, ^cad, Sci, tlsA with high purity Scepter is very useful for rapid and accurate assay of binding affinity of hGH variants. It is a useful reagent.

hPRL and hGH receptors - Human hGH and hPRL receptors To determine whether the pitopes overlap or not, we use hPRL receptor fragment could replace the hGH receptor fragment derived from hG)I. (data not shown); in fact, the hPRL receptor fragment is hGH It competes for receptor binding sites with an apparent Kd value of 1 nM. This is hGH substantially the same affinity as measured by direct binding of the hPRL receptor to The results are the same (results not shown).

hPRL receptor using seven of the segment substitution hGH variants in Table 1. The relevant epitope on hGH was located. For this experiment, hGHΔ32-4 6 mutants were used. The method uses hPRLr as a receptor instead of hGHs. In addition to the above-mentioned determination of epitopes on hGH for the hGH receptor, This was done by the method used, namely the effect on binding of hGH variants.

The results for the 12 segment substitution hGH variants described above are shown in Table XVIII.

Table XVIII Homologue scan for the extracellular domain of hPRL receptor (hPRLr) Binding of hGH mutants prepared by mutagenesis; H analogue: Life according to the segment substituted from pGH, hPL, or hPRL. I named it. The exact description of the introduced mutations is a series of single mutations separated by commas. given by the body. Each single mutant has a one-letter code for wild-type residues followed by maturation. Its codon position and variant residues in hGH are indicated. hGH mutants first It was produced and purified as described in .

Binding to hPRLr was measured essentially as described for hGHr (Spen. Spencer, S, A, et al. (1988) J, []iol.

Chem, 263.7862-7867), but this method 1 is not suitable for hPRLr. Using an affinity-purified rabbit polyclonal antibody against the carrier-protein Then, the hPRLrv1 complex was precipitated using Gibco's BSA (crude). in general In addition, the standard deviation of the values was less than 20% of the reported values. against hGHr The relative decrease in binding affinity (K, (mutant)/to, (hGH)) is shown in Table 1. It was obtained from. Changes in receptor selectivity affect binding to hGHr. The affinity was calculated as the relative reduction ratio compared to hPRLr. WT = wild type.

111T hGH L 2.3 (1) (1) (1) (continued) wT hPRL L 7.6 3.3 > 100,000 - It can be seen from the table Sea urchin pGH(11-33) and pGH(57-73) are linked to hPRL receptors. pGH(48-52> had a large effect on the binding affinity. do not have. Unlike the hGH receptor, the hPRL receptor is similar to hPRL and hPL. but not pGH. As expected, the binding competing hormone h Virtually all substitutions from PRL or hPL did not affect binding. the only The exception is hPR, which has a >70-fold reduction in binding affinity to hPRL receptors. L (22-33). In this way, the hPRL receptor Mutations in the central region of Rix I and near the loop between residues 57 and 73 Very sensitive.

Homolog-scan data also revealed hPRL and hGH receptor epitopes. Several segment substitution mutants cause large changes in receptor binding selectivity. Letters (Table XVIII) that show that they are not identical because of their differences. For example, pGH (11-33) or to the binding caused by hpRL (22-33). The effect is approximately 1 more on hPRL receptors than on hGH receptors. 00 times bigger. On the other hand, hPL (56-64) and hPRL (54-74) have little effect on hPRL receptors, while they are 17-fold and The binding to the hGH receptor was 69 times weaker. Furthermore, these selective binding effects The effect (along with monoclonal antibody binding discussed earlier) is the receptor binding affinity. The decrease in the identity is local in the hGH mutant and may be due to global structural changes. It has been proven that there is no.

Alanine is a specific side chain in hGH that strongly regulates binding to hPRL receptors. Identified by canning mutagenesis and homologous substitution. Shown in Table x■ hGH mutants were prepared. hPRL substitutions, F25S and D26E are helical caused the greatest decrease in binding affinity at l (21 and 1, respectively). and 4.5 times). These residues protrude from the hydrophilic face of helix 1 and Other mutations in helix 1 (mainly R18A and and FIOA).

A loop region known to affect hGH receptor binding (54-68 ) and four residues in close proximity that do not affect hGH receptor binding. Two residues (Q49A and T50A) preceding the region were tested. most broken The disruptive mutants are l58A and R64A, which each have a binding affinity of decreased by 32 times and 6 times. The effects of the other four mutations are negligible.

Helix 1 and loop region (58-64>) are potent against hPRL receptor. The fact that it contains strong binding determinants is associated with helix 4. This is because This is because the lix is wedged between these two structures (Figure 25B). . In fact, the helix 4gl region between the disulfides bonded to 0165 to V185 Alanine scanning revealed strong binding determinants (Table XIX). most Most of the disruptive mutations are in four helical turns, R167 to R178. cloth and are present on the same hydrophilic side.

Table XIX hPRL or hGH receptor fragment (hPRLr or hGHr Binding of single mutants of hGH to ). Q2 created by site-directed mutagenesis Variants of hGH other than 2N, F25S, D26E, Q29S and R33 was prepared and purified as described above (Cunningham), B., C., and Zoller, M., J., and Sumi. Sm (1982) Nucleic Acids Res, 10. 6487-6499) o Receptor binding assay and variant names is shown in No. XVI11. Data on decreased binding affinity for hGHr Taken from Table ■.

ND indicates not measured.

Table XIX (continued) A functional topography was derived based on the location of hGH and hPRL receptors (second Figure 8), the highest epitope portion for the hPRL receptor (Figure 25B) is Mutations that exhibit a 2-fold or greater decrease in binding affinity are assigned. This standard Therefore, the epitope of the hPRL receptor is basically located in the helisox of FIO to Q29. 1 front, loops from F54 to Q6B and helisox from R167 to R178 4 limited to the hydrophilic side of On the other hand, the hGH receptor epitope (Figure 25A) From the amino terminal region to the front of hex 7x4 from 14 to M14, from F54 to 371 hex. Residues on the front face of helix 1 and the hydrophilic face of helix 4 from D171 to v185 Further mutational analysis will fill remaining gaps in hPRL epitopes, including This epitope overlaps with, but is identical to, the hGH receptor epitope. It is clear that this is not the case.

These data indicate that all of the binding determinants that recognize hG) (hGH and hPRL) Despite having 32% homology with the extracellular binding domain of the receptor, This shows that they are not the same.

Residues that cause large changes in receptor binding affinity may be due to indirect structural effects. So that's what we're doing. But most of these destructive effects have all been tested. A single mutant showed complete binding affinity for a group of eight hGH monoclonal antibodies. maintains the receptor activity and often leads to changes in receptor selectivity. It is thought that this is due to a local effect as it does not have a uniform effect on affinity. (see Table xIx and below).

Design of receptor-specific mutants of hGH Many single-hGH mutants are undergoes many changes in synthetic selectivity (Table XIX), most notably the hGH receptor. binding to the hPRL receptor led to a 4-fold increase in affinity for E174A decreases by more than 0 times. This is 120 times the receptor selectivity. It shows change. Another mutation (mainly R178N and 1179M) is hG H selectively binds to the hPRL receptor. In general, receptor specificity is Variants that also change are located in non-overlapping regions of the two receptor epitopes.

Even small mutations can occur if the free energy change in receptor binding is additive. It was believed that highly specific variants of hGH could be designed. Fact 2 The two most hGH receptor selective single mutants (K168A and E174A) Together, the double mutant has a 2300-fold selectivity for binding to the hGH receptor. (Table XX), and as pointed out earlier, the binding selectivity of hPL (56-64) The selectivity of hPL containing only two mutations B56D and R64M (56-6 4) (Table xi), these hGH mutants (K16 8A, , E174A or E56D, R64M) are substantially preferred receptors. , did not reduce selectivity for hGII or hPRL, respectively. Also both at the same time Binding to receptors can also be reduced.

Table XX to identify hGH and hPRL receptors (hGHr and hPRLr). Binding of engineered double mutants of hGH, hGH mutants amenable to site-directed mutagenesis to hGHr or hPRLr by the method described in Table xm. binding was assayed.

K, the standard deviation of the measured values was 1100% of the reported value. This was less than 20% of the reported value.

For example, the binding affinity for hGH and hPRL receptors individually can be increased. The total reduction of 172A and F176A is 550 and 1500 respectively. Produces 0x affinity destruction.

In all cases the binding free energy change (ΔΔG bond) is surprisingly additive. (Table XXI), the additive effects of mutations can be seen on enzyme-group interactions. (P, J, Carter 5167-5171), Protea Ze-Protease Inhibitor Interactions (M, Laskow) 1 protease inhibitors: medical and biochemical characteristics” (+ 983), N. Katunuwa VIE, Japan Science Society Cyater Edition, Tokyo, 55-68) and protein stability (D, Cyratol). (Shortle) et al., (1986) Protein, 1.81-89 ( 1986); M, H, Mecht, J, M, Starchban Sturtevant and R.T. Sawer Prot. ein 1.43-46) and published in these references. It is commonly observed that the mutated residues function independently and are in contact with each other, as in .

This suggests that the paired residues in many hGH variants function independently. It shows. Such additivity increases the desired receptor binding affinity and This creates a very predictable situation for creating hGH mutants with specificity and specificity. Ru.

Table XXI Regarding binding to hGH or hPRL receptor (hGHr or hPRLr) Additive effects of mutations in hGH. Binding free energy of mutants to wild-type hGH The energy change (ΔΔG bond) is ΔΔG bond = RT 1 n (CKD (mutant) / K e(hGH)) was calculated from the decrease in binding affinity. single also is the Ko (mutant)/of the multiple mutant hormone, and the (hGH) value is in Tables x■-Table xx Obtained from.

Similar to hGH, adrenergic receptors (R, J, Lefko Wisoc) witz) and M, G, Caron (198B) J, Biol, Chem, 1 engineering 1.4993-4996), ICF-Luceptor (M, ^, Cascieri et al. (1989) J.

Biol, Chew, 264.2199-2202), IL-2 receptor (11, J, Robb et al. (1984) J, Exp, Med, 1 60.1126-1146iR, J, Robb et al. (198B) Pr oc, Mail.

^cad, Sci, USA85, 5654-5658) and ANP receptor (D. Lowe and Goedel, private) There are many examples where more than one receptor or receptor subtype exists, such as Ru. In these cases, what links specific receptor functions to specific pharmaceutical effects is Difficult, but the use of receptor-specific hormone II (H-form) greatly facilitates this task. It can be simplified. For example, catechol J reamine cheek (beta-adrenergic receptor) elucidate the characteristics of receptor subtypes and link receptor function to physiological responses. (R, J, Lefkowitz, etc.) 1983)^nnu, Rev, Biochem, 52.159-186) .

Similarly, receptor-specific mutants of hGH are useful for the identification of other receptors for hGH and for hGH. Investigating the functions of hGH and hPRL receptors in the complex pharmacology of GH provide important tools for This study allows for the rational design of receptor-specific mutants. We present a systematic method for identifying receptor binding sites in hormones that are doing.

Example 13 Production of human prolactin that binds to human growth hormone Prolactin (PRL) contains growth hormone (GH), placental lactogen (PL) and prolifniline. It is a member of the homologous hormone group. Nicol, C, So, etc. (198 6) Endocrinol.

Rev, 7.169-203. Collectively this group of hormones promotes growth, differentiation , regulate a wide range of physiological effects such as electrolyte balance. Chituura (Chaw) la), R, K, et al. (1983) Ann, 1lev, Med.

↓↓, 519-547; Isaksson 0. G, P, etc. (1985) Ann, Rev. Physiol, Earthwork, 483-499. These pharmacological effects begin with binding to specific cellular receptors. For example hP RL binds to lactogenic receptors but not somatogenic receptors hGH does not bind to lactobacillus, and activates lactation but not bone growth. binds to both somatogenic and somatogenic receptors, and Activates both lactation and bone growth. Molecular basis of differences in receptor binding specificity is still not understood.

Cloning and expression of hPRL hPRL cDNA was extracted from the human pituitary cDNA library in λgtlO. (Huynh, T., V., et al. (1985)” DNA cloning Engineering Techniques Volume 1, D, M, Glover KLA (Author) Oxford IRL Press), 49-78) The 5' and and oligonucleotides corresponding to the 3′ end (Cooke) ,N,! , et al. (1981) J, Biol, Chew, 256.400 ? -4016) due to hyperdigestion (Maniatis) ),? , Mi (1982) “Moleki Error Cloning, Laboratory Manufacturer” manual (Cold Spring Harbor Laboratory, Cold Spring Harbor -1NY)) was cloned. A full-length cDNA clone was identified, and 720 bp BstIl from codon 22 to 55 bp downstream of the stop codon - old nd The nl fragment was subcloned into pJc118. Decode this array Proc (Sanger), P. et al. (1977) Proc, Na1 l, ^cad, Sci, USA 74.5463-5467) However, it was completely consistent with what had been reported previously (Coke ([:ooke), K, E, et al. (1981) J, Riot.

Ches, 256.4007-4016).

Intracellular expression vector pBO760 (Fig. 26) was constructed using standard methods (Maniaci Maniatis, T. (1982) Molecular Cloning Laboratory Laboratory Manual (Cold Spring Bar t<- Laboratory, Cold Spring Harbor-1NY).

Escherichia coli trp0-t-- derived from pHGH20'l-1 was used for hPRL gene transcription. (deBoer, H0^0 et al. (1982)) “Structure and function of tar” by Rodríguez, R, L and Chan. Chasberlin, M. J., ed. (Prager edition, New York) , 462-481). The hPRL coding sequence is a 47bρXbal-BstEII combination. NA cassette and hPRL cDNA derived?　20　bpBstE　l l-HlndII [contains fragment. p NA cassette The start codon is indicated by an asterisk. Phage “1ori” pBR322β-lactamase gene, pBR replication origin and pBR322β-lactamase gene are pBO 475 (derived from E. coli (MM294) containing Cunningham pB0760). 0.5L shake containing 100ml of M9 high case medium supplemented with μg/■l carbenicillin. Early logarithmic phase; A55@=0.1-0 ．． 3) (Miller, J. H. (1972) “Molecular Genetics Experiment” (Cold Spring Harbor Laboratory, Cold Spring Harbor Laboratory) bar-1NY). Add indole acrylic acid (final 50 μg/vbl) trp The promoter was induced. Cells were grown for an additional 6-8 h before harvesting by centrifugation. Thin Cell fractionation experiments showed that hPRL was almost exclusively present in the inclusion particles and that 5DS-PAG When analyzed with E, it was shown that it accounts for 2-5% of the total cell protein (data (not shown).

Purification and regeneration of hPRL. Encapsulated particles containing hPRL were treated with basic Tris (pH 8, 0), suspended in 1 gM EDTA (TE buffer) and finely divided by vigorous sonication. The cells were broken down. Insoluble substances collected with mb (10,000 x g for 15 minutes)25 ml of TE buffer. Add this suspension to 50% glycerin 0.2 hPRL was layered on a Little Cushion and centrifuged at 9000xg for 25 minutes. The included particles were pelletized. hPRL derived from this inclusion particle (purity 20%) Suspended in 5sf of TE buffer.

8NGnHC1 containing 0.3 g reduced glutathione (Sigma) in TE buffer pPRL was regenerated by dissolving the included particles in 156 ml of solution. 30 at room temperature After stirring briefly for a minute, the mixture was cooled to 0°C and 0.6 g of oxidized glutathione was added. diluted with cold TE buffer 844111 containing Leave this solution at 4°C overnight. Dialysis using 41 TE buffer T, changing the external solution 3 times in 24 hours after stirring thoroughly. I did it. Insoluble material was removed by centrifugation (10.000 x g for 20 minutes).

Furthermore, the regenerated and solubilized hPRL was added to 45% saturation (NIIJ with the addition of tsO4). and purification by precipitation by stirring at room temperature for 2.5 hours. This precipitate was centrifuged (12,0 OOxg for 30 minutes) and dissolved in 5 wj of TE buffer. After 30 minutes at room temperature After clarifying the solution (10,00 x glO minutes), filter it with a Millipore filter. (0.45 μm).

This solution was dialyzed against 0.5 liters of TE buffer overnight at 4°C. lastly This hPRL (purity 85%) was basically purified using the same method as described for the purification of hGH. until homogeneous (>95%) by FPLC using the same DEAE fast matrix. Purified (Cunningham) % B, C, et al. (1989 ) 5science, 24U, 1330-1335).

Mutagenesis and binding site-directed mutagenesis of hGH and hPRL variants (Zoller), A, Jo et al. (1982) Nucleic Ac 1ds Res, +10.64B? -6500) is a methylation repair defect in E. coli. strain Mut L (Ilrawer, B1 et al. (198 4) Ce113B, 879-887), enrichment of mutant clones Initial plasmid DNA obtained after transformation of a heteroduplex Each introduces a single restriction site so that the pool can be subjected to restriction purification or restriction selection. or by designing mutagenic oligonucleotides to exclude (llIells), J, ^1 et al. (1986) Ph11. Trans, R, Sac.

Lond, ^Lord, 415-423), all oligonucleotides are most 12 bp with exactly 77 matches 5' of the upstream mismatch and the most downstream mismatch. Designed to have 10 bP for an exact match on the 3' side of the match. of hGH In the case of mutagenesis, the hGH synthesis gene mentioned above contains many restriction sites. This was cloned into plasmid pBO475. Mutants of hGH are E. coli secreted into the pericellular space of the cells (Chang, C.M., et al. (1987)). Gene55, 189 196) Purified as previously described.

The Kd values for each congener are given herein and in Spencer S, ^.

(1988), J. Biol, Chew, 263.7862-786. (+xsB hG It was determined by competitive displacement of H. The previously mentioned hGH-binding protein (cloned human mer (Fuh), G, etc. 11 It was secreted from Escherichia coli and purified by the method described in 989), submitted by . Displacement curves were constructed in triplicate, and the standard deviation of Kd values was generally less than 20% of the reported value. Yes, the Kd value cannot exceed 50% of the reported value, except in cases where the Kd value is 10 μM or more. There wasn't.

Concentrations of hPRL and hPRL variants are extinction coefficient + S (0,1%, 280) - 0 , 9 (WeLIaufer), D, B.

(1962)^dv, In Prot, Chew, Sat. 1.303-390) Measured using Awes. This means that the mutant contains mutations in aromatic residues. If so, adjustments were made accordingly. Concentration values measured by absorbance are 5DS-PAG Laser densitometry using staining with combage blue for E and hGH The 0 circle dichroic spectrum that matched the measured value with an error of less than 10% was the Aviv Measurements were made using a Cary 60 Spectropolarimeter.

Find out which residues in hPRL are most disruptive to binding to the hGH receptor. (Figure 27), we first introduced many hPRL residues into hGH (Figure xxn). table).

Table XX11 Comparison of hPRL introduced into hGH and alanine substitution at site 58.64 in hGH. Single alanine substitutions at 176 and 178 significantly disrupt receptor binding; Substitution of hPRL residues at these sites had little effect. hPRL The greatest effect of the substitution is in helix 4 residues containing positions 176 and 178. There was. These data indicate that the residues in the helix 411 region of hPRL are It is shown that the best explanation is the lack of binding to the receptor.

Recombinant hPRL maintained native-like structure and functionality.

First, the near and far UV CD spectra (Figure 28) are the spectra of natural hPRL. It was the same as Le (Bewley,, Ding, ^.

(1979) 'Recent advances in hormone research' 351!, pp. 155-213, Demi-Press, N, Y, ), the far ultraviolet spectrum is the same as hGH, This indicates that there is an important difference in the average residue ellipticity in the 208n stomach and 224n stomach. This indicates the existence of a similar 4-helix bundle structure. These hormones The near-UV CD, which reflects the loss of aromatic residues and differences in the microenvironment, is markedly different. another lab In our study, recombinant hPRL showed sufficient immunological cross-reactivity in hPRL ELISA. (data not shown), and also proliferated rat lymphoid 11Nb2 cells. It was equivalent to hGH in causing 1980) J, Cl1n, Er+do.

Metab, S., 105B-1063); upon reduction, purified hPRL becomes 5DS -Also observed for hGH, which causes a significant decrease in mobility on PAGE), this This indicates the formation of disulfide bonds (Pollitt) , S. et al. (1983) J. Bacterol.

53.27-32), amino-terminal sequence analysis revealed that intracellularly expressed hPRL has an amino-terminal It was shown that methionine was maintained. However, like methionyl hGH, this This apparently does not affect its structure or function.

Binding of hPRL to hGH-binding proteins is reduced by more than 10' compared to hGH. (Table XX), which is below the detection limit of the binding assay.

Chapter XXI [Table Engineered residues in hPRL that allow binding to hGH-binding protein 1 Table XXI (continued) 1. hPRL mutants were generated, purified and analyzed as previously described. many mutations The body is represented by a series of single mutants separated by colons (Table xx■), with codon numbers The number is based on the hGH sequence (Figure 2).

2. The average standard error is calculated except when the Kd value exceeds 18M, where it falls by 50%. It was less than 20% of the reported value. This error occurs when the I[d value exceeds 10MM. It gets even bigger.

Combining three residues in helix 4 from hGH (H171D, N175T and Y176F) were introduced into hPRL. alanine scanning mutation Induction and hPRL substitution (Table Xx) indicates that these residues are It has been shown to be very important for binding to the target. This triple mutant of hPRL produces detectable binding to hGH-binding proteins, although it is 14,000 times weaker than hGH. I strained it. Furthermore, to create tetra-mutants, other important helix 4 residues (K178 R), this resulted in a level of binding that was only 660 times lower than in the wild type. hGH residue 1B4 to hPRL variant B, enhanced (variant B in Table xn+) Introduction of ~188 did not enhance binding to hGH-binding proteins. However, hPR E174A(7) introduction (Table xxm) gives the L-transformant C. E174A is hGH It has the same binding affinity to hGH-binding proteins as when incorporated into Furthermore, it increased by 3.5 times.

The loop region containing residues 54 to 74 has binding strength to the helix 4 region. analyzed. The loop region in hPRL was replaced with the hGH-derived sequence (hGH (Table xI1)). Complete substitution with 54-74) results in barely detectable impact on hGH-binding proteins. Give possible combinations. When this mutant is combined with mutant B, the binding affinity is increase substantially. However, this new mutant (B+hGH(54-74)) Binding affinity is almost 10-fold reduced compared to Mutant B alone. did Therefore, some hGH residues in the 54-74 loop replace hGH positions in helix 4. It was not compatible with the exchange.

We performed alanine scanning mutagenesis from the 54-74 loop of hGH. We selected seven residues that were shown to affect binding the most. R6 of hGH The 4A mutation reduces binding affinity more than 20-fold, but R64 of hGH mutant (hPRL substitution) has slightly more binding to hGH-binding proteins (factor X X11 table).

Therefore Lyss4 in hPRL remained unchanged. As a result hGH Six of the seven substitutions in which changes to alanine were most disruptive in hP This new mutant (B+H65F:556E:L581:E 56S:D68N:Q66B) is 50 times more potent than B+hGH (54-74) It binds, but the binding affinity is only 110 times smaller than that of wild-type hGH. That's it. However, this showed only a slight improvement over mutant B alone (6 ), as expected from the strong interaction previously observed in the loop region of hGH. It wasn't what. Therefore, we further disrupted the 6 mutations within the loop and H54 The combination of F:556E:L581 + mutant B binds 3 times more than mutant B alone. It turned out to be weak. Finally mutation E62S:D63N to variant C: Incorporation of Q66E (mutant D) has slightly lower binding affinity than hGH. This yields the IIfU body with the highest affinity, which is only 6 times lower. another single butt Natural mutations (854F, 356E, L581.A39P, N71S and L1791 ) increases the binding affinity of hPRL mutants to hGH-binding proteins. There was no such thing. The structure of the mutant was virtually determined by CD spectral analysis (Figure 28) or EL I Indistinguishable from native hPRL by SA activity (data not shown) Ta.

These studies use only functional information derived from site-directed mutagenesis experiments to This study demonstrated the possibility of restoring the binding properties of l[4GJ bodies. Alaninski of hGH Canning mutagenesis is an important aspect of regulating hGH binding to its receptor. provided a systematic analysis of the chains (Figure 27); this information is linked to the hGH receptor. Many residues in hPRL were highlighted (Fig. 29), which explains that However, further analysis showed that the alanine substitution in hGH was similar to the hPRL substitution in hGH. (Table xxn), as well as some hPR L substitutions can improve binding affinity, especially when larger side chains are present in hPRL. - were considerably more destructive than others. For example, conservative (only) hGH (bigger than that) F176YWt#I! is the binding affinity for the hGH receptor. a smaller R64 substitution causes an 8-fold decrease in binding affinity. showed a slight increase in nitty.

Thus, analysis of disruptive hPRL substitutions in hGH first focused on binding to hPRL. Figure 2 shows the introduction of a cluster of residues in helix 4 that carries out the identity. child Does this mean that no binding to the hGH receptor by wild-type hPRL is observed? is very important and yields binding affinities within the range of the assay used. Introducing several hGH substitutions simultaneously into hPRL (Kd 50 μM) is necessary.

By introducing functionally important residues into helix 4, hPRL can be easily detected. Introduced possible binding affinities. However, the loop region between 54-74 It turned out to be more difficult to create. The entire loop of hGH to hPRL The increase in coupling due to incorporation is smaller than expected, and the optimized helical When combined with S4 Mutant B, the binding was blocked and destructive. Our data is h The Δ4-74 loop structure of PRL is supported by other interactions in this protein. Which indicates that. This problem was resolved in stages. First, hPR in hGH Six loop residues from hGH identified as important for alanine substitutions accompanied by L substitutions. group was introduced into hPRL. Although this improved the situation some The combination of hGH mutations (limited to H54F, 356B and L581) is h It was destructive to PRL. These data indicate that some residues in the loop This shows that it is important to the structure and is more stable if left as is.

Many repeated cycles of mutagenesis result in tight binding of hPRL to the hGH receptor. It was necessary to unify the combination of residues that made it possible. This tactic has a mutation effect. It relies on the assumption that it is additive as observed in practice. For example E1 When the 74A mutation was added to hPRL variant C or hGH, binding decreased by 3 to 5. doubled. Furthermore, H54F, 356g, and L581 single against mutants The destructive effect of the mutant is (4,4 times) all three mutations added to mutant B. was approximately the same (3 times more) as the destruction caused by the combination of

Although the binding affinity of the mutant RI is only reduced by 6-fold V14N and H18 can be incorporated into mutants to further improve binding. There are several other residues such as 5V. In other words, alanine substitution in hGH There is a site that causes a 2-3 times decrease in binding (Figure 29); It helps with the planning process, but it is clearly not fundamental.

The cumulative effect of mutations is similar to protein evolution due to natural changes and cycles of selection. Connectivity can be converged in the same manner.

Traditional protein engineering experiments use high-resolution structural analysis to determine the natural We showed that it is possible to practically change the substrate specificity of mutant enzymes (Wells 0t ells) J, A4, etc. (1987) Proc, Natl, Acad, Sci, U S A 84.5167-5171i Wilkes (Illilk g), H, M, etc. (198B) 5science 242.1541-15 44), similarly others have binding properties due to the antigen-binding loop (Jones ), P, T, et al. (1986) Nature 321.522-525) and is a DNA receptor mouse (Wharton, R, P, (1985) NaLure 316.601-605) It was shown that it can be produced by replacing all units of . However, hG in hPRL Selective residue substitutions within the structural framework of hPRL are required to confer H receptor binding. shall be. Furthermore, CD spectral data indicate that the entire structure of the hPRL mutant is hG. Although it maintains similar binding properties to H, the structure of hPRL is very different. (It is similar, but not similar to hGH.

The fact that binding specificity for hGH receptors can be incorporated into hPRL suggests that somatic Confirms the functional importance of specific residues for genetic receptor binding.

These studies also show that hGH and hPRL have only a 23% concordance. Nevertheless, it provides irrefutable evidence for the structural relationship between the two. this is growth hormone This protein, including prolactin, prolifnilin or placental lactogen structures, Provides a rational way to approach novel receptor binding functions contained within the lumon group . Such /1 hybrids are associated with the pharmacological importance of receptor subtypes. Useful in distinguishing between receptor binding and activation. These analogs are chin New receptor specificities with more useful properties as antagonists or antagonists This will lead to the design of targeted hormones.

Example 14 Imparting binding properties of human growth hormone to human placental lactogen Human placental lactogen ( The binding affinity of hP L) to the hGH receptor is 3 higher than that of hGH. 0 times smaller (G, Bausann et al. (1986) J, Cl1n ,! ! ndocrino1. Metab, 62.134; ^ C, Herin Herington, et al. (1986) J.

Cl1n, Invest, 77.1817), previous mutation experiments The binding site of hGH to the protein is basically located near the amino terminus (residues 4-14). Two regions (residues 54-74 and 171- 185).

The entire sequence of hPL is 85% identical to hGH. Receptor binding epi on hGH Within the three regions that broadly constitute the tope, the hPL differs in only seven locations: Contains substitutions: P2Q, 14V.

N12H, R16Q, E56D, R64M, and 1179M.

(This nomenclature uses a one-letter code for residues in wild-type hGH, followed by residues in mature hGH. H site numbers and residues in hPL are indicated).

Single alanine substitutions of hGH were made at each of these seven sites.

Among these, four alanine substitutions, I4A, E56A, R64A.

and 1179A were found to cause a more than two-fold decrease in binding affinity. It was. In general, alanine substitutions are more binding than homologous substitutions derived from human prolactin. It also has a great effect. Therefore, of the hPL-derived substitutions introduced into hGH. While 1179M showed some effects, it was only 1.7 times more effective. It was.

However, R64A and R64M substitutions have identical and larger binding affinities. In addition, the hGH double mutant (E5611 1R64M) showed a further decrease of 30 times in total (Table 1); E56D and R64M basically act as receptor binding agents for hGH and hPL. Determine the difference in finity. Therefore, the double mutant D56E, M64R of hPL has substantially increased binding affinity to the hGH receptor. M179 Additional modifications such as 1 and V41 also enhance binding of hPL to hGH receptors. Ru.

Example 15 Effect of amino acid substitution at site 174 on binding to human growth hormone As shown previously, substitution of Glu174 by Ala (E174A) 0 parts that increase the affinity of growth hormone to its receptor by more than 4 times hGH substituted with other 12 residues to determine the optimal replacement residue at position 174 Mutants were created and their affinity with hGH-binding proteins was measured (Table XXIV) ).

Main Asn, G where side chain size determines binding affinity rather than charge It continues in the order of lu%olds, Lya, Leu, and Tyr.

Table XXIV Side chain Kd (nut) a, The mutation includes a Kpn1 site at site 178 cloned into pB0475. Mutant hGH gene was generated by site-directed mutagenesis ( Carter, P. et al. (1986) Nucl@tc Ac1 ds Res, 13.4431-4443>, oligonucleotide used for mutagenesis Creotide is It has an array represented by . Here NNN represents the new codon at position 174. Also, the asterisk indicates a mistake that excludes the Kpn1 site starting at codon 17B. It shows Tutsi. The mutated codons are as follows; Gln, CAG; AIL AAC; Ser+AGC; Lys, AAAHArg, AGG; H4s ,CAC;Gly.

GGG i Vat, GTG; Leu, CTG, following heteroduplex synthesis PCR was performed by restriction treatment with Kpnl to reduce the background of wild-type sequences. Enriched rasmidobule mutations. All mutant sequences are dideoxy sequences Confirmed by analysis (Sanger, F. et al. (1977) Proc. Natl, ^cad, Sci, U S A1 Sat, 5463-5467).

b, side chain puffing value is C, Chotia ((1984)^nn u.

Based on data from Rev. Bloches, 53.537).

c, %l separation constant depends on the hGH binding protein as mentioned earlier (+*s　1) Determined by competitive displacement of h　G　H. NF is a mutant hormone isolated and assayed. This indicates that it is expressed at a level that is too low for

Example 16 The hGH mutants shown in Table XXV were constructed. These phases for wild-type hGH Demonstrate interpersonal skills.

Table XXV Although preferred embodiments of the present invention have been described, it is possible to make various changes to these disclosed embodiments. It will be apparent to those skilled in the art that such modifications are within the scope of this invention. It is.

FIG.-3 Kd (mut) / Kd (wt) Engraving (no changes to the content) Engraving (no changes to the content) ;#):Invitation::h:e-old man? h　hPnL(97・1Il141 Strong bond ◆→−1 bond weak of Engraving (no changes to the content) Jyo 1!7 (no change in content) FIG.-II ・Neko× ・〉Ne ・(〉■× (〉■×　・ (■・ Nene■ Engraving (no changes to the content) FIG.-22 Engraving (no changes to the content) Capacity of GH analogs in rats 8 days after treatment 0.1 1.0 10.0 100.0 　1000. OKd mutant/Kd wild type FIG.-23 Engraving (no changes to the content) Binding determinants for hGHr Binding determinants for hPRLr 1 ori Engraving (no changes to the content) Wave length (nm) FIG, -28A Wave length (nm) FIG. -288 kuψ α” Minokuzu -Σ＞CD2 ~;≠2. -2C Comic View Procedural amendment (formality) %formula% 3. Person who makes corrections Relationship to the case: Applicant Name: Genentech Incorporated '5, date of amendment order: Self-issued international search report '11911110amaeelc711e,,,〒,,,q,q,i,,,.

Claims (1)

[Claims] 1. at least one that interacts with the first target substance in the amino acid sequence of the parent polypeptide. a) a first similar polypeptide segment derived from an analog to the parent polypeptide; substituting the first selected amino acid segment of the parent polypeptide with 2. Releasing the segment-replaced polypeptide live. wherein the parent polypeptide and the like Mimics have different interactions with the first target substance. b) contacting the first segment replacement polypeptide with the first target substance; measuring the interaction between a target substance and the segment-substituted polypeptide. c) repeating steps a) and b) with a second analogous polypeptide segment derived from an analog to the parent polypeptide to obtain at least one first segment comprising a similar amino acid segment different from the first segment replacement polypeptide; 2nd segment producing an agent-substituted polypeptide. d) comparing the difference in activity of the parent polypeptide and the first and second segment replacement polypeptides toward the first target substance; Check the location of the inn. A method comprising steps a) to d) above. 2. The unknown active domain is present in the primary amino acid sequence of the parent polypeptide. 2. The method of claim 1, wherein both segments include two discontinuous amino acid segments. 3. At least the first selected amino acid segment of the parent polypeptide is 2. The method of claim 1, comprising at least one amino acid residue present on the surface of the native structure of the repeptide. 4. 4. The method of claim 3, further comprising repeating steps a) and b) until substantially all amino acid residues on the surface of the parent polypeptide have been replaced by the analogous amino acid segment. . 5. 2. The method of claim 1, further comprising step a) or until about 15-100% of the amino acid sequence of said parent polypeptide is replaced with said similar amino acid segment. and b). 6. 2. The method of claim 1, further comprising step a) and until about 60-100% of the amino acid sequence of said parent polypeptide is replaced with said similar amino acid segment. and b). 7. 2. The method of claim 1, further comprising identifying a second unknown active domain that interacts with a second target substance of the parent polypeptide; A method comprising repeating steps a) to d) using the quality. 8. identifying at least one first active amino acid residue within the first active domain by the method of claim 1, and e) scanning for various first amino acid residues within the first active domain. the first residue position by substituting with a ning amino acid. producing a converted polypeptide. f) Bringing the first residue polypeptide into contact with the first target container and measuring the interaction between the target substance and the residue-substituted polypeptide. g) repeating steps e) and f) to at least is also reduced by replacing one second amino acid residue with a scanning amino acid. Producing at least one second residue substituted polypeptide. h) the first standard of the parent polypeptide and each of the first and second residue substituted polypeptides; and confirming the position of the active amino acid within the first active domain by comparing the difference in activity against the target substance. A method comprising steps e) to h). 9. 9. The method according to claim 8, further comprising repeating steps a) to h) using a second target substance to obtain a second active domain and at least one activity within the second active domain. 1. A method comprising identifying a sex amino acid residue. 10. 10. The method of claim 9, further comprising the step of substituting at least one of the active amino acid residues in the first active domain with various amino acids, the interaction with the first target substance being altered; the second target of the parent polypeptide A method of generating polypeptide variants that retain all interactions with the protein. 11. 11. The method of claim 10, further comprising at least one of said second active domains. also replace one of the active amino acid residues with various amino acids and replace the first and second targets. A method comprising producing a polypeptide variant that has an altered interaction with a target substance. 12. wherein the first and second active domains have at least one common active amino acid 10. The method of claim 9, wherein each of the first and second target substances has an acid residue, further substituting at least one of the common active amino acid residues with various amino acids. A method comprising producing a polypeptide variant that has an altered interaction with a substance. 13. 10. The method of claim 9, wherein the first and second active domains have at least one common active amino acid residue, and wherein at least one amino acid in the first active domain does not contain at least one common active amino acid residue. Polypeptides whose interaction with the first target substance has been changed by substituting residues with various amino acids A method comprising producing a peptide mutant. 14. A method of producing a growth hormone variant comprising replacing at least one active amino acid residue in a parent growth hormone with at least one different amino acid. A method that involves producing a growth hormone that has a different activity towards a particular labeling substance compared to the activity of the hormone. 15. A method of producing a growth hormone variant in which at least one active amino acid in the active domain of a parent growth hormone is replaced with at least one different amino acid, the activity of which differs toward a particular target substance as compared to that of the parent growth hormone. A method comprising producing an active growth hormone variant. 16. A method for identifying at least one active amino acid residue in a parent polypeptide, comprising: (a) scanning a first amino acid residue with residue number N in the parent polypeptide; (b) scanning each amino acid residue with residue numbers N+1 and N-1 for the first residue; (c) contacting each of the substituted polypeptides with a target substance and the target substance to produce N+1- and N-1-substituted polypeptides, respectively; (d) comparing the difference in activity of the parent polypeptide and the substituted polypeptide against the target substance; (e) determining the interaction between the target compound and the N+1 substituted polypeptide; When the difference in activity between the target substance and the N-1-substituted polypeptide is two times or more, the number of residues is increased, and when the difference in activity between the target substance and the N-1-substituted polypeptide is two times or more, the number of residues is decreased. Reverse steps (b) to (d). and (a) to (e) above. 17. Less than or equal to scanning amino acid substitutions in 4 contiguous residues In both cases, the four substituted polypeptides have an activity difference of less than twice that of the parent polypeptide. 17. The method of claim 16, wherein steps (b) to (d) are repeated until one is identified. 18. The parent polypeptide may be human growth hormone, human prolactin, alpha-interactive -feron, γ-interferon, tissue plasminogen activator, IGF-1, TGH-β1, EGF, CD-4, TNF, GMGSF, TGF, follicular stimulation Geki Hormone, Rejuvenating Hormone, Atreal Naturalistic Peptide 17. The method of claim 1, 8 or 16, wherein the method is selected from the group consisting of lactogen and placental lactogen. 19. The parent polypeptide is human growth hormone, human placental lactogen, and human protein. The method according to claim 18, which is lactin. 20. the parent polypeptide is human growth hormone and the analogue is human fetal growth hormone; selected from the group consisting of disc lactogen, protein growth hormone and human prolactin. The method according to claim 1. 21. Claim 8 or 8, wherein the scanning amino acid is an isosteric amino acid. is the method described in 16. 22. 17. The method according to claim 8 or 16, wherein the scanning amino acid is a neutral amino acid. 23. The neutral amino acids are selected from alanine, serine, glycine and cysteine. 23. The method of claim 22, wherein the method is selected from the group consisting of: 24. 24. The method according to claim 23, wherein the scanning amino acid is alanine. 25. 17. The method of claim 1, 8 or 16, wherein said activity is measured in an in vitro or in vivo assay. 26. the parent polypeptide is a hormone and the activity is a soluble hormone level 26. The method according to claim 25, wherein the method is determined by an in vitro assay using Scepter. 27. said hormone is human growth hormone, and said soluble hormone receptor 27. The method according to claim 26, wherein the target is ShGHLr. 28. said hormone is human growth hormone, and said soluble hormone receptor 27. The method of claim 26, wherein the target is ShPRLr. 29. 30. The method according to claim 29, wherein said activity indicates binding of said target substance to said parent polypeptide or catalysis of said target substance by said parent polypeptide. 30. 30. The method according to claim 29, wherein the activity between the target substance and the substituted polypeptide is increased compared to the parent polypeptide. 31. 30. The method according to claim 29, wherein the activity between the target substance and the substituted polypeptide is decreased compared to the parent polypeptide. 32. A growth molecule consisting of at least first, second and third parts in order from the N-terminus. wherein the first portion corresponds to at least a portion of the amino acid sequence of the parent growth hormone, the third portion corresponds to at least another portion of the amino acid sequence of the parent growth hormone, and the second portion corresponds to at least another portion of the amino acid sequence of the parent growth hormone. is the amino acid sequence analog of the natural analogue of the parent growth hormone. Growth hormone mutants corresponding to similar parts. 33. said parent growth hormone is hGH and said natural analogue is human placental hormone. 33. The growth hormone variant of claim 32, which is selected from the group consisting of lactogen, human prolactin, and porcine growth hormone. 34. The parent growth hormone is hGH, and the second portion is hPL(12-25), pGH(11-33), hPRL(12-33), hPRL(12-19), hPRL(22-33). , hPL(46-52), pGH(48-52), hPL(56-64), pGH(57-73), hPRL(54-74), hPRL(88-95), hPRL(97-104) , hPL(109-112), pGH(108-127), hPRL(111-129), hPRL(126-136), pGH(164-190), and pGH(167-181). 33. The growth hormone variant according to claim 32, which is selected. 35. The second part is hPRL (97-104), hPRL (54-74) and and hPL(56-64). Rumon mutant. 36. 35. The growth hormone variant of claim 34, wherein said second portion is hPRL(22-33). 37. The parent growth hormone is human growth hormone, and the second portion consists of a similar sequence of residues 11-33, 46-52, 54-74, 88-104, 108-136 and 164-190 of hGH. 33. The growth hormone variant of claim 32, comprising an amino acid sequence selected from the group. 38. It has an amino acid sequence that does not exist in nature and contains less human growth hormone. Both are human derivatives derived by substitution of one amino acid residue with various amino acids. In the long hormone variant, the amino acid residues are F10, F54, E56, 158, R6 4, Q68, D171, K172, E174, T175, F176, R178, C182, V185, 14, P5, L6, S7, R8, N12, M14, L15, R16, R19, S55, S57, P59, S62, N63, E66, K70, S71, K168, I179, C182, R183, G187, P2, T3, L10, H18, R64, E65, Q69, L73, R167, E174, S1 84, E186, S188, F191, H21, N47, P48, Q49, T5 0, S57, Q46, V173, R77, L80, F25, D26, Q29, E 30, D169, S43, F44 , F97, A98, N99, S100, L101, V102, Y103, G104, Q22, E33 and equivalents thereof. mutant. 39. 39. The growth hormone variant of claim 38, wherein said various amino acids are isosteric amino acids. 40. The amino acid residues are substituted from the human growth hormone amino acid residue group consisting of F10, F54, E56, I58, R64, Q68, D171, K172, E174, T175, F176, R178, C182, V185 and equivalents thereof. 39. The growth hormone variant according to claim 38, which has been discovered. 41. The above substitution is F10. GEMARQSYWLI and V, F54GEMAR QSYWLI and V, E56GMFARQSDNKL and V, 158GEM FARQSV and T, R64GEMFAQSH, KD and N, Q68GEM FARSHKD and N, D171GEMFARQSHK and N, K172G EMFARQSHD and N, E174GMFARQSHDNK and L, T1 75GEMFARQSV and I, F176GEMARQSYWLI and V , R178GEMFARQSHKD and N, C182GEMFARQ and S, 41. The growth hormone variant of claim 40, which is selected from the group consisting of V185GEMFARQSITLY and W. 42. 42. The growth hormone variant of claim 41, wherein said substitution is E174A. 43. The amino acid residues are I4, P5, L6, S7, R8, N12, M14, L15, R16, R19, S55, S57, P59, S62, N63, E66, K70, S71, K168, I179, C182, R183 , G187 and this 39. The growth hormone variant according to claim 38, which is selected from the group of amino acid residues of human growth hormone consisting of equivalents of the following. 44. The amino acid residues are P2, T3, L10, H18, R64, E65, Q69, L73, R167, E174, S184, E186, S188, F191 and and their equivalents. The growth hormone variant according to claim 38. 45. Human growth hormone amino acids in which the amino acid residues are H18, R64, E65, L73, E174, E186, S188, F191 and equivalents thereof. 39. The growth hormone variant according to claim 38, which is selected from the group of amino acid residues. 46. The substitution is selected from the group consisting of H18A, R64K, E65A, L73A, El74ANQS and G, E186A, S188A and F191A. The growth hormone variant according to claim 45. 47. 39. The growth hormone of claim 38, wherein said amino acid residue is selected from the group of human growth hormone amino acid residues consisting of H21, K172, F176, N47, P48, Q49, T50, S51, Q46, V173 and equivalents thereof. Hormone variant. 48. 48. The growth hormone variant according to claim 47, wherein the various amino acids are alanine. 49. The variant contains a double amino acid substitution consisting of K172A/F176A. The growth hormone variant according to claim 48. 50. The amino acid residues are S43, F44, H18, E65, L73, E186, S188, F191, F97, A98, N99, S100, L101, V102, Y103, G104, R19, Q22, D26, Q29, E30, E33 and and their equivalents. The growth hormone variant according to claim 38. 51. 51. The growth hormone variant of claim 50, wherein said amino acid residue is selected from the group consisting of F97, A98, N99, SlOO, L101, V102, Y103, G104 and equivalents thereof. 52. The various amino acids are F97GEMARQSYWLI and V, A98G EMFRQSDNH and K, N99GEMFARQSDK and Y, S100 GEMFARQHDNK and Y, L101GEMFARQSIV and Y, V 102GEMFARQSITLY and W, Y103GEMFA RQSWLI and V, G104EMFARQS and P. 53. 39. The growth hormone variant according to claim 38, wherein the amino acid residue is selected from the group consisting of Q22, F25, D26, Q29, E33, Q49, T50, R64, R167, K168, 158, F176, and R178. 54. Claim 53 wherein said substitution is selected from the group consisting of Q22N, F25S, D26E, Q29E, Q29S, E33K, Q49A, TsOA, R167A, K168A, I58L and A, R64 K and A, F176Y and A, R179K and N. Growth hormone variants described. 65. Claims: said variant comprises a double amino acid substitution selected from the group consisting of K168A:E174A, R178N:I179M and K172A:F176A. The growth hormone variant according to Box 38. 56. It has an amino acid sequence that does not exist in nature and contains less human growth hormone. Both are human growth hormone variants derived by substitution of one amino acid residue with various amino acids, such as F10, F25, D26, R167, K168, K172, B174, F176, 158, R64 and equivalents thereof. Karana A human growth hormone variant selected from a group of human growth hormone amino acid residues. 57. It has an amino acid sequence that does not exist in nature and contains less human growth hormone. Both are human growth hormone variants derived by substitution of one amino acid residue with various amino acids, the amino acids of human growth hormone consisting of F97, S100, L101, V102, Y103, T175 and equivalents thereof. A human growth hormone variant selected from the residue group. 58. 58. The growth hormone variant of claim 57, wherein said substitution is selected from the group consisting of F97A, S100A, L101A, V102A, Y103A and T175S. A different body. 59. Substitutions, insertions or insertions of amino acid residues in the amino acid sequence of human growth hormone a human growth hormone variant having the sequence containing one or more amino acid changes consisting of A human growth hormone variant that does not undergo conversion. 60. 60. The variant of claim 59, wherein said growth hormone is human growth hormone and said active domain comprises residues 2-33, 54-74 and 167-191. 61. The active domain comprises residues 6-14, 56-68 and 171-191. 61. The variant according to claim 60. 62. A growth hormone variant that contains one or more amino acid changes consisting of substitution, insertion, or deletion of amino acid residues in the amino acid sequence of human growth hormone, in which the active amino acid for the somatogenic receptor of the growth hormone is changed. do No growth hormone mutant. 63. the growth hormone is human growth hormone hGH, and the active amino acid 63. The variant of claim 62, wherein the acids include F10, F54, E56, 158, K64, Q68, D171, K172, E174, T175, F176, R178, C182 and V185. 64. The growth hormone is human growth hormone hGH, and the active amino acid 64. The variant of claim 63, wherein the acid comprises 14, P5, L6, S7, R8, S55, S57, P59, S62, N63, E66, K70, S71, K168, 1179, C182, R183 and G187. 65. It has an amino acid sequence that does not exist in nature and is derived by substituting at least one or more amino acid residues of human prolactin hormone with various amino acids. A human prolactin hormone variant in which the amino acid residues are H171, N175, Y176, K178, H54, S56, L58, A59, E62, D63, Q66, N71, L179, T55, K64, A67, M70, Q72, Human prolactin selected from the amino acid residue group of human prolactin consisting of K73 and D74 hormone mutants. 66. Claim 65, wherein said substitution comprises H171D:N175T:Y176F. human prolactin mutants. 67. 67. The human prolactin mutant according to claim 66, further comprising the substitution K178R. 68. The human prolactin mutant according to claim 67, further comprising analogues to the mutant. A variant containing a substitution with a similar amino acid sequence hGH (54-74). 69. The human prolactin mutant according to claim 67, further comprising analogues to the mutant. Variants containing substitutions with the similar amino acid sequence hGH (184-188). 70. 68. The human prolactin variant according to claim 67, further comprising the substitution H54F: S56E:L581:E62S:D63N:Q66E. 71. 68. The human prolactin variant according to claim 67, further comprising the substitution H54F: S56E:L581. 72. 68. The human prolactin mutant according to claim 67, further comprising the substitution E174A. 73. 73. The human prolactin mutant according to claim 72, further comprising the substitution E62S: D63N:Q66E. 74. 75. The human prolactin mutant according to claim 74, further comprising the substitution E54F. 75. 75. The human prolactin mutant according to claim 74, further comprising the substitution S56E. 76. 75. The human prolactin mutant according to claim 74, further comprising the substitution L58I. 77. 75. The human prolactin mutant according to claim 74, further comprising the substitution A59P. 78. 75. The human prolactin mutant according to claim 74, further comprising the substitution N71S. 79. 75. The human prolactin variant according to claim 74, further comprising the substitution L179I. 80. It has an amino acid sequence that does not exist in nature and contains at least one placental lactogen. human placental tissue induced by substitution of one amino acid residue with various amino acids. a human placental lactogen variant in which the amino acid residue is selected from the group consisting of Q2, V2, H12, Q16, D56, M64, M179 and equivalents thereof. mutant. 81. Early replacement is Q2P, V4I, H12N, Q16R, D56E, M64R and and M1791. A different body. 82. 81. The placental lactogen variant of claim 80, wherein said substitution is selected from the group consisting of V4A, D56A, M64A and M179A. 83. 81. The placental gland of claim 80, wherein said substitutions include D56E and M64R. togen mutant. 84. DNA sequences encoding the variants according to claims 32, 33, 38, 65 and 80. 85. An expression vector comprising the DNA sequence according to claim 84. 86. An expression host transformed with the expression vector according to claim 85.