RU2198180C2 - Analog of keratinocyte growth natural factor (versions), pharmaceutical composition, analog-encoding recombinant molecule of nucleic acid, expression vector, escherichia coli strain transformed by vector, method of analog preparing and method of stimulation of formation of nonfibroblastic epithelial cells - Google Patents

Abstract

FIELD: biotechnology, genetic and protein engineering. SUBSTANCE: invention relates to method of keratinocyte growth factor (KGF) analog preparing. Analog of KGF is prepared by modification of one or some amino acids in region of 32-55 of amino acid sequence SEQ ID NO:2. Cysteine residue in positions 32 and 46 is deleted or replaced with another amino acid. Expression vector comprising recombinant nucleic acid encoding analog of KGF is transformed in E. coli cells. Transformed strain of E. coli is cultured and KGF analog is obtained. KGF analog is used in pharmaceutical composition for stimulation of epithelial cells growth. Invention allows to prepare analogs of KGF eliciting enhanced stability in storage. Invention can be used for analog of KGF preparing. EFFECT: improved method of preparing, valuable properties of analog. 32 cl, 52 dwg, 5 tbl

Description

 The present invention relates to recombinant DNA technology and protein engineering. Specifically, recombinant DNA methodologies have been applied to obtain polypeptide analogs of keratinocyte growth factor (KGF), a strong growth mitogen of non-fibroblast epithelial cells, characterized in that the analogs have increased resistance compared to that of the KGF precursor.

BACKGROUND OF THE INVENTION
The complex process of tissue formation and repair is mediated by a number of protein factors, sometimes called soft tissue growth factors. These molecules are usually released by one type of cell and act to influence the proliferation of other cell types (Rubin et al. (1989), Proc. Natl. Acad. Sci. USA, 86: 802-806). Some soft tissue growth factors are secreted by certain types of cells and affect the proliferation, differentiation and / or maturation of sensitive cells during the development of multicellular organisms (Finch et al. (1989), Science, 24: 752-755). In addition to their role in the development of organisms, some of them are important for long-term health and the maintenance of more mature systems. For example, mammals have many systems in which a quick "circuit" of cells occurs. Such systems include the skin and the gastrointestinal tract, each of which consists of epithelial cells. This group of soft tissue growth factors includes the family of fibroblast growth factor proteins (FGF).

 To date, 8 members of the FGF family are known that have related primary structures: the main fibroblast growth factor, bFGF (Abraham et al. (1986), EMBO J., 5: 2523-2528); acidic fibroblast growth factor, aFGF (Jaye et al. (1986), Science, 233: 541-545); int-2 gene product, int-2 (Dickson and Peters (1987), Nature, 326: 833); hst / kFGF (Delli-Bovi et al. (1987), Cell, 50: 729-737 and Yoshida et al. (1987), Proc. Natl. Acad. Sci. USA, 84: 7305-7309); FGF-5 (Zhan et al. (1987), Mol. Cell. Biol, 8: 3487-3495); FGF-6 (Marics et al. (1989), Oncogene, 4: 335-340); keratinocyte growth factor (Finch et al. (1989), Science, 24: 752-755) and hisactophilin (Habazzettl et al. (1992), Nature, 359: 855-858).

 Among the FGF family of proteins, keratinocyte growth factor (KGF) is a unique proliferator of proliferation of non-fibroblast epithelial cells originating from mesenchymal tissues. The term “natural KGF” refers to a natural human (hKGF) or recombinant (rKGF) polypeptide (with or without signal sequence), as reflected by the amino acid sequence shown in SEQ ID NO: 2, or an allelic variant thereof. [Unless otherwise indicated, the numbering of amino acids in the molecules described below corresponds to that for the mature form of the natural molecule (ie minus the signal sequence), which is depicted by amino acids 32 to 194 in SEQ ID NO: 2].

 Natural KGF can be isolated from natural "human" sources (hKGF) or obtained using recombinant DNA (rKGF) (Finch et al. (1989), supra; Rubin et al. (1989), supra; Ron et al. (1993 ), The Journal of Biological Chemistry, 268 (4): 2984-2988 and Yan et al. (1991), In Vitro Cell. Dev. Biol., 27A: 437-438).

 Natural KGF is known to be relatively unstable in the aquatic environment and that it undergoes chemical and physical decomposition resulting in loss of biological activity due to processing and storage (Chen et al. (1994), Pharmaceutical Research, 11: 1582-1589). Natural KGF is also prone to aggregation at elevated temperatures and becomes inactive under acidic conditions (Rubin et al. (1989), Proc. Natl. Acad. Sci. USA, 86: 802-806). Aggregation of natural KGF in aqueous solution results in a deactivated protein. This is disadvantageous since such a loss of activity makes it impractical for storing aqueous formulations of natural KGF proteins for a long time or for introducing a protein for long periods. Moreover, the preparation of pharmaceutical formulations is particularly problematic since aggregated proteins are known to be immunogenic (Cleland et al. (1993), Crit. Rev. Therapeutic Drug Carrier Systems, 10: 307-377; Robbins et al. (1987) , Diabetes, 36: 838-845 and Pinckard et al. (1967), Clin. Exp. Immunol., 2: 331-340).

 Natural KGF contains five cysteine residues, namely amino acids 1, 15, 40, 102 and 106 (Finch et al. (1989), Science, 24: 752-755). Although the cysteine composition of natural KGF was reported, the role played by cysteine residues in activity (e.g., in particular biological activity) and tertiary structure (e.g., the tendency to form undesired inter- and intramolecular disulfide bonds) was not reported. Thus, in the previous technique there is no information, even if they are known, that cysteine residues are significant or are involved in the formation of undesired disulfide bonds, which makes the protein sensitive to aggregation and / or instability.

 To try to improve or otherwise change one or more characteristics of native KGF, protein engineering can be applied. Recombinant DNA technology has been applied to modify natural KGF sequences. Ron et al. (1993), J. Biol. Chem., 268 (4): 2984-2988 reported modified KGP polypeptides that deleted 3, 8, 27, 38, or 49 amino acids at the N-terminus. Polypeptides with deleted 3, 8 or 27 N-terminal residues retained the ability to bind to heparin; others are not. It was also reported that polypeptides with missing 3 and 8 residues are fully active, while a form with 27 residues removed was 10-20 times less active as a mitogen, and forms that lack 38 or 49 amino acids did not have mitogenic activity. The stability of the modified KGF polypeptides was not discussed and nothing was reported about it.

 PCT published patent application No. 90/08771, supra, also reported the production of a chimeric protein in which approximately the first 40 N-terminal amino acids of the mature form of natural KGF were linked to the C-terminal portion (approximately 140 amino acids) of aFGF. It was reported that the chimera targets keratinocytes like KGF, but it lacks sensitivity (predisposition) to heparin, which is typical for aFGF, but not KGF. The stability of the chimera was not discussed and not reported.

 Thus, no modified KGF molecule has been reported in the literature that has significantly higher stability than natural KGF. Moreover, the literature did not provide sufficient evidence or evidence for the successful production of KGF molecules with such desirable (given) characteristics.

 Typically, the effect on the biological activity of any substitution of amino acids in a protein varies depending on a number of factors, including the three-dimensional structure of the protein, and whether or not the modification of the receptor binding site of the primary protein sequence is relevant. Since neither the three-dimensional structure, nor the site of receptor binding of the primary structure of natural KGF have been published, knowledge of the previous technique does not allow a generalization about the effect of amino acid modifications on natural KGF based on the effect of amino acid modifications even on ordinary classes of proteins.

 The aim of the present invention is to obtain analogues of polypeptides and nucleic acid molecules encoding such analogues that exhibit increased stability (for example, at typical pH, thermal and / or other storage conditions) compared to natural KGF.

SUMMARY OF THE INVENTION
The present invention provides new, biologically active polypeptide analogs of KGF. For the purposes of this invention, the term “KGF” includes natural KGF and proteins having a peptide sequence substantially the same as a natural KGF peptide sequence that retains cysteine residues corresponding to Cys ¹ and Cys ^{15 of} natural KGF (Cys ³² and Cys ⁴⁶ SEQ ID NO: 2), and which retains all or part of the biological activity of natural KGF, in particular proliferation of non-fibroblast epithelial cells. When reference is made to FIGS. 4.7-24 and 37-50 and a specific amino acid position, the first member of the sequence is assigned a residue number of 0. By "characterized by a peptide sequence substantially the same as the natural KGF peptide sequence" is meant a peptide sequence, which is encoded by a DNA sequence capable of hybridizing with nucleotides 201 to 684 of SEQ ID NO: 1, preferably under stringent hybridization conditions.

 Determining the position of the corresponding amino acid between two amino acid sequences can be done by aligning (aligning and comparing) the two sequences in order to obtain the maximum number of residual matches, including the shift of the amino and / or carboxyl ends, introducing gaps, if necessary, and / or removing residues present in "candidate" as inserts. Database searches, sequencing, and manipulation can be performed using one of the well-known and commonly used sequence homology / identity analysis programs (for example, Pearson and Lipman (1988), Proc. Natl. Acad. Sci. USA, 85: 2444-2448; Altschul et al. (1990), J. Mot. Biol., 215: 403-410; Lipman and Pearson (1985), Science, 222: 1435 or Devereux et al. (1984), Nuc. Acids Res., 12: 387-395).

Stringent conditions in the context of hybridization are stringent (unifying) conditions regarding salt, temperature, organic solvents, and other parameters commonly controlled in hybridization reactions. Exemplary stringent hybridization conditions are hybridization at 4 • SSC at 62-67 ^° C followed by washing at 0.1 • SSC at 62-67 ^° C for about an hour. Or an example of stringent hybridization conditions is hybridization in 45-55% formamide, 4 • SSC at 40-45 ^{° C.} (See T. Maniatis et al. Molecular Cloning (A Laboratory Manual); Cold Spring Harbor Laboratory (1982), p. .387-389].

Thus, proteins include allelic variations, or deletion (s), substitution (s) or insertion (s) of amino acids, including fragments, chimeric or hybrid molecules of natural KGF. One example of KGF involves changing the charge of a polypeptide in which one or more amino acid residues 41-154 of natural KGF (preferably residues Arg ⁴¹ , Gln ⁴³ , Lys ⁵⁵ , Lys ⁹⁵ , Lys ¹²⁸ , Asn ¹³⁷ , Gln ¹³⁸ , Lys ¹³⁹ , Arg ¹⁴⁴ , Lys ¹⁴⁷ , Gln ¹⁵² , Lys ^153, or Thr ¹⁵⁴ ) are removed or replaced with a neutral residue or negatively charged residue selected to target a protein with a reduced positive charge (as disclosed in USSN 08 / 323,337, filed October 13, 1994 g), specifically including R (144) Q, KGF, having the substitution of arginine for glutamine in the floor Amino Acids 144 Natural KGF. Another example of KGF includes proteins obtained by substituting at least one amino acid with a higher loop potential of at least one amino acid in the loop forming region of Asn ¹¹⁵ -His ¹¹⁶ -Tyr ¹¹⁷ -Asn ¹¹⁸ -Thr ¹¹⁹ natural KGF (as disclosed USSN 08 / 323,473, filed October 13, 1994), specifically including H (116) G, KGF, having histidine to glycine substituted at position 116 of the amino acid of natural KGF. Another further example includes proteins with one or more amino acid substitutions, divisions or additions at 123-133 (amino acids 154-164 of SEQ ID NO: 2) of natural KGF; these proteins may exhibit agonistic or antagonistic activity. Surprisingly, it was discovered that when a KGF molecule (i.e., a precursor molecule) contains residues corresponding (as determined by the methods described above) to Cys ¹ and Cys ^{15 of} natural KGF (cysteine residues 32 and 46 of SEQ ID NO: 2), it is modified by replacing the corresponding cysteines, then the resulting KGF analog has increased stability compared to the precursor molecule. Preferably, in addition to increased resistance, the invention is directed to those analogues that fully exhibit biological activity (i.e., at least substantially the same receptor binding or affinity) compared to natural KGF.

 In another aspect of the invention, purified and isolated nucleic acid molecules encoding various biologically active polypeptide analogs of KGF are described. In one embodiment of the invention, such nucleic acids are DNA molecules cloned into biologically functional plasmid or virus vectors. In another embodiment of the invention, nucleic acid constructs can then be used to stably transform a prokaryotic or eukaryotic host cell. In yet another embodiment, the invention includes a process in which either a prokaryotic (preferably E. coli) or eukaryotic host cell stably transformed with a nucleic acid molecule is grown under conditions of an appropriate nutrient medium so that expression of the KGF analog is possible. After expression, the resulting recombinant polypeptide can be isolated and purified.

 An additional aspect according to this invention relates to pharmaceutical compositions containing a therapeutically effective amount of analog KGF and an acceptable pharmaceutical carrier. Such formulations will be useful in the treatment of patients suffering from diseases and lesions of the epithelium.

 In this regard, another aspect relates to methods of stimulating the growth of epithelial cells by administering to the patient a therapeutically effective amount of a KGF analog. In one embodiment of the invention, non-fibroblast epithelial cells are cells whose proliferation is stimulated. Such epithelial cells include various cells of the appendages, cells of the pancreas, liver cells, epithelium of the mucous membrane of the respiratory tract and gastrointestinal tract.

Brief Description of the Figures
Figures 1 depict the nucleotide (SEQ ID NO: 1) and amino acid (SEQ ID NO: 2) sequences of natural KGF (nucleotides encoding the mature form of natural KGF are shown by bases from 201 to 684 SEQ ID NO: 1, and the mature form of KGF is shown amino acid residues from 32 to 194 SEQ ID NO: 2).

 Figures 2A, 2B, and 2C depict plasmid maps of pCFM1156, pCFM1656, and pCFM3102, respectively.

 Figure 3 depicts the nucleotide (SEQ ID NO: 3) and amino acid (SEQ ID NO: 4) sequences of the RSH-KGF construct.

 Figure 4 shows the nucleotide (SEQ ID NO: 5) and amino acid (SEQ ID NO: 6) sequences of the construct contained in the plasmid with the KGF gene.

 Figure 5 shows chemically synthesized OLIGOs (oligonucleotides) (OLIGO # 6 to OLIGO # 11; SEQ ID NO: 12-17, respectively) used to replace the DNA sequence between the KpnI site and the EcoRI site (from amino acid position 46 to 85 SEQ ID NO: 6) in the construct contained in the KGF gene plasmid to produce the construct in the KGF gene plasmid (dsd).

 Figure 6 depicts the chemically synthesized OLIGO (oligonucleotides) (OLIGO # 12 to OLIGO # 24; SEQ ID NO: 18-30, respectively) used for the synthesis of KGF (codon optimized).

 Figure 7 depicts the nucleotide (SEQ ID NO: 31) and amino acid sequences (SEQ ID NO: 32) C (1.15) S, a KGF analogue with cysteine substitutions in series at amino acid positions 1 and 15 of natural KGF.

 Figure 8 depicts the nucleotide (SEQ ID NO: 33) and amino acid sequences (SEQ ID NO: 34) ΔN3 / C (15) S, an analogue of KGF with a deletion of the first three amino acids at the N-terminus and the replacement of cysteine in series at the amino acid position 15 of natural KGF.

 Figure 9 depicts the nucleotide (SEQ ID NO: 35) and amino acid (SEQ ID NO: 36) sequences of ΔN3 / C (15) -, a KGF analog having division of the first 3 amino acids at the N-terminus and division of cysteine at amino acid position 15 of the natural KGF.

 Figure 10 depicts the nucleotide (SEQ ID NO: 37) and amino acid (SEQ ID NO: 38) sequences of ΔN8 / C (15) S, a KGF analog having division of the first 8 amino acids at the N-terminus and the substitution of cysteine for serine at the amino acid position 15 natural KGF.

 Figure 11 depicts the nucleotide (SEQ ID NO: 39) and amino acid (SEQ ID NO: 40) sequences of ΔN8 / C (15) -, a KGF analog having division of the first 8 amino acids at the N-terminus and division of cysteine at amino acid position 15 of the natural KGF.

 Figure 12 depicts the nucleotide (SEQ ID NO: 41) and amino acid (SEQ ID NO: 42) sequences of ΔN15, a KGF analog having a deletion of the first 15 amino acids at the N-terminus of natural KGF.

 Figure 13 depicts the nucleotide (SEQ ID NO: 43) and amino acid (SEQ ID NO: 44) sequences of ΔN16, a KGF analog having a deletion of the first 16 amino acids at the N-terminus of natural KGF.

 Figure 14 depicts the nucleotide (SEQ ID NO: 45) and amino acid (SEQ ID NO: 46) sequences of ΔN17, a KGF analog having a deletion of the first 17 amino acids at the N-terminus of natural KGF.

 Figure 15 depicts the nucleotide (SEQ ID NO: 47) and amino acid (SEQ ID NO: 48) sequences of ΔN18, a KGF analog having a deletion of the first 18 amino acids at the N-terminus of natural KGF.

 Figure 16 depicts the nucleotide (SEQ ID NO: 49) and amino acid (SEQ ID NO: 50) sequences of ΔN19, a KGF analog having a deletion of the first 19 amino acids at the N-terminus of natural KGF.

 Figure 17 depicts the nucleotide (SEQ ID NO: 51) and amino acid (SEQ ID NO: 52) sequences of ΔN20, a KGF analog having a deletion of the first 20 amino acids at the N-terminus of natural KGF.

 Figure 18 depicts the nucleotide (SEQ ID NO: 53) and amino acid (SEQ ID NO: 54) sequences of ΔN21, a KGF analog having a deletion of the first 21 amino acids at the N-terminus of natural KGF.

 Figure 19 depicts the nucleotide (SEQ ID NO: 55) and amino acid (SEQ ID NO: 56) sequences of ΔN22, a KGF analog having a deletion of the first 22 amino acids at the N-terminus of natural KGF.

 Figure 20 depicts the nucleotide (SEQ ID NO: 57) and amino acid (SEQ ID NO: 58) sequences of ΔN23, a KGF analog having a deletion of the first 23 amino acids at the N-terminus of natural KGF.

 Figure 21 depicts the nucleotide (SEQ ID NO: 59) and amino acid (SEQ ID NO: 60) sequences of ΔN24, a KGF analog having a deletion of the first 24 amino acids at the N-terminus of natural KGF.

 Figure 22 depicts the nucleotide (SEQ ID NO: 61) and amino acid (SEQ ID NO: 62) sequences of C (1.15) S / R (144) E, an analogue of KGF having cysteine to serine substitutions at amino acid positions 1 and 15 and replacement of arginine with glutamic acid at position 144 of the amino acid of natural KGF.

 Figure 23 depicts the nucleotide (SEQ ID NO: 63) and amino acid (SEQ ID NO: 64) sequences of C (1.15) S / R (144) Q, a KGF analog having cysteine to serine substitutions at positions 1 and 15 of amino acids and replacement of arginine with glutamine at amino acid position 144 of natural KGF.

 Figure 24 depicts the nucleotide (SEQ ID NO: 65) and amino acid (SEQ ID NO: 66) sequences of ΔN23 / R (144) Q, a KGF analog having a deletion of the first 23 amino acids at the N-terminus and replacing arginine with glutamine at the amino acid position 144 natural KGF.

Figure 25 depicts the amount of remaining soluble protein during storage of natural KGF and C (1.15) S in 20 mM sodium phosphate, 0.15 M NaCl, pH 7.0 at 37 ^° C for 27 hours.

Figure 26 depicts the amount of remaining soluble protein during storage of natural KGF, ΔN15 and C (1.15) S in 50 mM NaPO ₄ , 0.15 M NaCl, pH 7.0 at 37 ^° C for 27 hours.

Figure 27 shows the amount of soluble protein of natural KGF, C (1.15) S, C (1.15) S / R (144) E and C (1.15) S / R (144) Q, as determined by size exclusion HPLC using particle size, depending on the thermostating time at 37 ^o C.

Figure 28 shows the calculated temperature (T _m ) as a function of the pH of natural KGF, C (1.15) S, C (1.15) S / R (144) Q, and C (1.15) S / R (144) E.

Figure 29 is a typical graph of the mitogenic activity of C (1.15) S determined by measuring the incorporation of [ ³ H] -thymidine during DNA synthesis and comparing it with a standard curve of natural KGF.

Figure 30 is a typical graph of mitogenic activity ΔN15 determined by measuring the incorporation of [ ³ H] -thymidine during DNA synthesis and comparing it with a standard curve of natural KGF.

Figure 31 depicts a typical graph of mitogenic activity ΔN23 determined by measuring the incorporation of [ ³ H] -thymidine during DNA synthesis and comparing it with a standard curve of natural KGF.

Figure 32 is a typical graph of the mitogenic activity of ΔN23 / R (144) Q determined by measuring the incorporation of [ ³ H] -thymidine during DNA synthesis and comparing it with a standard curve of natural KGF.

Figure 33 is a typical graph of the mitogenic activity of C (1.15) S / R (144) Q determined by measuring the incorporation of [ ³ H] -thymidine during DNA synthesis and comparing it with a standard curve of natural KGF.

Figure 34 depicts a typical graph of the mitogenic activity of C (1.15) S / R (144) E determined by measuring the incorporation of [ ³ H] -thymidine during DNA synthesis and comparing it with a standard curve of natural KGF.

 Figure 35 depicts the effect of natural KGF, KGF-a, ΔN15, C (1.15) S and ΔN23 on the chemical composition of serum.

 Figure 36 depicts the nucleotide (SEQ ID NO: 77) and amino acid (SEQ ID NO: 78) sequences of C (1,15,40) S, a KGF analog having cysteine to serine substitution at amino acid positions 1, 15 and 40 of natural KGF.

 Figure 37 depicts the nucleotide (SEQ ID NO: 79) and amino acid (SEQ ID NO: 80) sequences of C (1,15,102) S, a KGF analog having cysteine to serine substitution at amino acid positions 1, 15 and 102 of natural KGF.

 Figure 38 depicts the nucleotide (SEQ ID NO: 81) and amino acid (SEQ ID NO: 82) sequences of C (1,15,102,106) S, a KGF analog having cysteine to serine substitution at amino acid positions 1, 15, 102 and 106 of natural KGF.

 Figure 39 depicts the nucleotide (SEQ ID NO: 83) and amino acid (SEQ ID NO: 84) sequences of ΔN23 / N (137) E, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing asparagine with glutamic acid at position 137 amino acids of natural KGF.

 Figure 40 depicts the nucleotide (SEQ ID NO: 85) and amino acid (SEQ ID NO: 86) sequences of ΔN23 / K (139) E, a KGF analog having a deletion of the first 23 amino acids at the N-terminus and replacing lysine with glutamic acid at position 139 amino acids of natural KGF.

 Figure 41 depicts the nucleotide (SEQ ID NO: 87) and amino acid (SEQ ID NO: 88) sequences of ΔN23 / K (139) Q, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing lysine with glutamine at the 139 amino acid position natural KGF.

 Figure 42 depicts the nucleotide (SEQ ID NO: 89) and amino acid (SEQ ID NO: 90) sequences of ΔN23 / R (144) A, an analogue of KGF having a deletion of the first 23 amino acids at the N-terminus and replacing arginine with alanine at position 144 amino acids natural KGF.

 Figure 43 depicts the nucleotide (SEQ ID NO: 91) and amino acid (SEQ ID NO: 92) sequences of ΔN23 / R (144) E, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing arginine with glutamic acid at position 144 amino acids of natural KGF.

 Figure 44 depicts the nucleotide (SEQ ID NO: 93) and amino acid (SEQ ID NO: 94) sequences of ΔN23 / R (144) L, a KGF analog having a deletion of the first 23 amino acids at the N-terminus and replacing arginine with leucine at the amino acid position 144 natural KGF.

 Figure 45 depicts the nucleotide (SEQ ID NO: 95) and amino acid (SEQ ID NO: 96) sequences of ΔN23 / K (147) E, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing lysine with glutamic acid at position 147 amino acids of natural KGF.

 Figure 46 depicts the nucleotide (SEQ ID NO: 97) and amino acid (SEQ ID NO: 98) sequences of ΔN23 / K (147) Q, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing lysine with glutamine at the 147 amino acid position natural KGF.

 Figure 47 depicts the nucleotide (SEQ ID NO: 99) and amino acid (SEQ ID NO: 100) sequences of ΔN23 / K (153) E, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing lysine with glutamic acid at position 153 amino acids of natural KGF.

 Figure 48 depicts the nucleotide (SEQ ID NO: 101) and amino acid (SEQ ID NO: 102) sequences of ΔN23 / K (153) Q, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and replacing lysine with glutamine at the 153 amino acid position natural KGF.

 Figure 49 depicts the nucleotide (SEQ ID NO: 103) and amino acid (SEQ ID NO: 104) sequences of ΔN23 / Q (152) E / K (153) E, a KGF analogue having a deletion of the first 23 amino acids at the N-terminus and glutamine substitution glutamic acid at position 152 of the amino acid of natural KGF and lysine; glutamic acid at position 153 of the amino acid of natural KGF.

 Figure 50 depicts the effect of ΔN23 on streptozotocin-induced diabetes in Sprague-Dawley rats.

DETAILED DESCRIPTION OF THE INVENTION
In accordance with this invention, new analogues of KGF. It has now been found that four cysteine residues of natural KGF (Cys ¹ and Cys ¹⁵ , Cys ¹⁰² and Cys ¹⁰⁶ ) are involved in the formation of two disulfide bridges. Cys ^{40 is} not involved in the formation of an intramolecular disulfide bond. Therefore, KGF contains two small disulfide loops separated by nearly 90 amino acids. It is clear that determining which cysteine residues are involved in disulfide bridges allows one to determine which cysteine residues are free to form unwanted intermolecular crosslinking or intramolecular bonds that result in the protein acquiring an undesired tertiary structure (e.g., a conformation that reduces activity squirrel).

It is surprising that, as shown, the modification of KGF by deletions or substitutions of amino acid residues of cysteine residues corresponding to positions 1 and 15 of KGF (positions 32 and 46 of SEQ ID NO: 2) gives an analogue of KGF with significantly increased resistance (i.e., decreases problems caused by improper unwinding, the formation of intermolecular disulfide bonds and the aggregation of protein molecules). For example, KGF analogues usually, when purified, give a high yield of soluble, regular folded protein. Moreover, after purification, it is more stable with respect to pH, temperature, etc., compared with the stability of the precursor molecule. Although not theoretically explained, it is believed that Cys ¹ and Cys ¹⁵ KGF, in addition to the formation of an intramolecular disulfide bridge and N-terminal disulfide loop, under certain conditions also exist in the form of free cysteines, which are able to form intermolecular disulfide bridges, leading to protein instability and the formation of aggregated molecules. Moreover, it was found that a deletion of the N-terminal disulfide loop is not important for receptor binding or mitogenic activity.

As used herein, “KGF analogue” or “KGF polypeptide analogue” means any of the described natural and artificial polypeptides that differ in structure from KGF due to modification of the peptide sequence corresponding to 24 amino acids of the N-terminus of KGF (amino acids 32 to 55 SEQ ID NO: 2) in which Cys ¹ and Cys ¹⁵ KGF (amino acids 32 to 46 of SEQ ID NO: 2) are replaced or deleted. The method by which cysteine residues are replaced or removed is not important and includes, for example, the inclusion of polypeptide analogs or one or more deletions of amino acids and / or substitutions. Therefore, this invention represents a family of new proteins - keratinocyte growth factors. The ego family contains several groups of proteins.

 One group of KGF analogues includes molecules in which cysteine residues at positions 1 and 15 of KGF are replaced by amino acids, including those that are not found in natural proteins. A substitution strategy for analogs involves the use of site-directed mutagenesis (But et al. (1989), Gene, 77: 51-59; Innis et at. "PRC Protocols", Academic Press, Inc., San Diego, CA). [KGF analogues containing amino acid substitutions are denoted by the residue found in this position in the mature protein (minus the signal sequence) depicted in SEQ ID NO: 2, followed by the position of this amino acid in brackets and a new amino acid, For example, an analogue that includes a cysteine substitution on serine at amino acid position 15 KGF (position 46 of SEQ ID NO: 2), means "C (15) S"]. Preferably, cysteine is replaced with neutral amino acids such as glycine, valine, alanine, leucine, isoleucine, tyrosine, phenylalanine, histidine, tryptophan, serine, threonine and methionine, with serine, threonine and alanine being preferred due to their chemical similarities with cysteine. Example 1 details the production of C (1.15) S using partial gene synthesis (in combination with other recombinant methods) followed by recombinant expression in a stably transformed bacterial host.

Another group of KGF analogues includes molecules in which cysteine residues at positions 1 and 15 are removed from KGF. A different strategy, such as N-terminal truncation and site-specific deletions, or a combination thereof, can be used to obtain such KGF analogues. "N-terminal (terminal) truncation" refers to a modification of KGF in which 1 to 24 N-terminal amino acid residues of KGF (amino acids 32 to 55 of SEQ ID NO: 2), including Cys ¹ and Cys ¹⁵ , are removed. [KGF analogs with truncation of amino acids are indicated as the residue removed at this position of the mature protein (minus the signal sequence) depicted in SEQ ID NO: 2, starting from the division site, and the number of residues removed. For example, an KGF analog with an N-terminal truncation of 24 KGF residues (residues 1-55 of SEQ ID NO: 2) will be denoted by “ΔN24” analog]. Specifically, the ΔN23 group includes analogues of natural KGF in which one or more amino acid residues 41-154 (amino acids 72-185 SEQ ID NO: 2), specifically including amino acid residues 123-133 (amino acids 154-164 SEQ ID NO: 2) removed or replaced by a neutral residue and a negatively charged residue selected to act on a protein with a reduced positive charge. Preferred residues for modification are Arg ⁴¹ , Gln ⁴³ , Lys ⁵⁵ , Lys ⁹⁵ , Lys ¹²⁸ , Asn ¹³⁷ , Gln ¹³⁸ , Lys ¹³⁹ , Arg ¹⁴⁴ , Lys ¹⁴⁷ , Gln ¹⁵² , Lys ¹⁵³ or Thr ¹⁵⁴ , with more preferred being Gln ¹³⁸ , Lys ¹³⁹ , Arg ¹⁴⁴ , Lys ¹⁴⁷ , Gln ¹⁵² or Lys ¹⁵³ , and Arg ¹⁴⁴ is most preferred. This group also includes ΔN23 analogues of natural KGF with loop-forming modifications at the loop formation site of Asn ¹¹⁵ -His ¹¹⁶ -Tyr ¹¹⁷ -Asn ¹¹⁸ -Thr ¹¹⁹ (amino acids 146-150 SEQ ID NO: 2), preferably including modification with a change in charge of amino acid residue 116 , Gly substitution at position 116 is more preferred. Subsequently, this group ΔN23 of natural KGF analogues includes those with one or more amino acid substitutions, deletions or additions at site 123-133 (amino acids 154-164 of SEQ ID NO: 2) of natural KGF.

 Example 1 describes in detail the preparation of systematic N-terminal truncations carried out using partial gene synthesis in combination with other recombinant methods, followed by recombinant expression in a stably transformed bacterial host. Such illustrative examples of truncations of the N-terminus include ΔN15 to ΔN24. Moreover, Example 3 describes in detail the expression in culture cells of mammals of DNA encoding natural KGF and glycosylated isoforms with a heterogeneous N-terminus, purification of the preferred glycosylated isoform having N-terminal amino acid truncation 1-23 of the mature form of natural KGF.

In contrast, a site-specific deletion refers to a modification of KGF in which one or more amino acid residues (e.g., Cys ¹ or Cys ¹⁵ ) are deleted. When Cys ¹ or Cys ¹⁵ KGF has been specifically removed, the analog is shorter than KGF per amino acid. [KGF analogs having an amino acid deletion are denoted by the residue in this position in the mature protein (minus the signal sequence) depicted in SEQ ID NO: 2, followed by the position of this amino acid in brackets and the minus sign. For example, an analogue from this group having a cysteine division at position 15 of KGF (position 36 of SEQ ID NO: 2) is designated “C (15) -”]. Site-specific deletions can be carried out using site-directed mutagenesis, as described above.

Also included in this group are analogues in which Cys ¹ and Cys ¹⁵ KGF are removed by truncation or deletion. For example, a representative of the KGF analogs has a truncation of the first three residues at the N-terminus (Cys-Asn-Asp) of KGF or a truncation of the first eight amino acids (Cys-Asn-Asp-Met-Thr-Pro-Glu-Gln) of KGF together with a deletion of cysteine in position 15 of the amino acid KGF (these analogues are denoted by ΔN3 / C (15) - and ΔN8 / C (15) - respectively). Since the methionine residue is in the natural KGF at the fourth and fifth positions of the amino acid, no additional methionine codon is required to enable proper translation initiation.

 Another group includes molecules in which cysteine residues at positions 1 and 15 of KGF are substituted by truncation or substitution. For example, representatives of the KGF analogues are ΔN3 / C (15) S and ΔN8 / C (15) S. Such analogs have a truncation of the first three amino terminal KGF residues or a deletion of the first eight amino terminal KGF residues, together with the replacement of cysteine at amino acid position 15 with another amino acid, for example, serine.

 When KGF analogues are obtained biologically, i.e. they are cell expression products as opposed to solid state synthesis products derived from natural products by proteolysis or enzymes, etc., nucleic acids encoding such polypeptides will have one or more nucleotides different from the nucleotide sequence natural KGF. Such polynucleotides can be expressed and the resulting polypeptide can be purified by any of the recombinant DNA techniques known to those skilled in the art.

 DNA sequences encoding KGF analogues may, in whole or in part, include the inclusion of codons that are “preferred” for expression in selected “host cells” (eg, “colony for expression in E. coli”); the presence of restriction enzyme cleavage sites and the presence of additional initiation (initial), termination or intermediate nucleotide sequences (for example, an initiation methionine codon for expression in E. coli cells) to facilitate the construction of easily expressed vectors.

 The present invention also provides recombinant molecules or vectors for use in a method for expressing polypeptides. Such vectors can be DNA or RNA and can be ring (cyclic), linear, single-stranded or double-stranded by nature and can be natural or assembled from a number of components, natural or synthetic.

 Many examples of such expression vectors are known. The components of the vectors, for example replicons, marker genes, enhancers, promoters, and the like, can be obtained from natural sources or synthesized by known methods. In each case, the expression vectors useful in this invention must contain at least one expression control element operably associated with an inserted nucleic acid molecule encoding a KGF polypeptide analogue. This control element is responsible for regulating the expression of the polypeptide with the participation of the nucleic acid molecules of this invention. Used control elements include, for example, the lac system, trp system, operators and promoters from phage λ, the promoter of glycolytic yeast, the promoter of the gene for acid phosphatase of the yeast, α-mating factor of yeast and promoters derived from adenovirus, Epstein-Barr virus, polyoma, monkey virus, as well as those of various retroviruses. However, many other vectors and control elements suitable for prokaryotic and eukaryotic expression are known in the art and can be used in the practice of this invention.

 Examples of suitable cloning vectors of prokaryotes may include plasmids from E. coli (e.g. pBR322, col E1, pUC and the F factor), with preferred plasmids being pCFM1156 (ATCC 69702), pCFM1656 (ATCC 69576) and pCFM3102 (described in " Examples ", below). Other suitable expression vectors, numerous types of which are known in the practice of expression in mammalian, insect, yeast, fungal and bacterial cells, can also be used for this purpose. Transfection of these vectors into the corresponding host cell can cause the expression of polypeptides - analogues of KGF.

 The host microorganisms useful in this invention can be either prokaryotic or eukaryotic. Suitable prokaryotic hosts include various E. coli (e.g. FM5, HB101, DH5α, DH10 and MC1061), strains of Pseudomonas, Bacillus, Streptomyces, with E. coli being preferred. Suitable eukaryotic host cells include yeast and other fungi, insect cells, plant cells, and animal cells such as COS (e.g., COS-1 and COS-7) and CV-1 monkey cell lines, 3T3 lines derived from Swiss, Balb -c or NIH cells, HeLa and L-929 mouse cells, and CHO, BHK and HAC cells of the hamster. Depending on what acts as a host, recombinant polypeptides produced in accordance with this (with this invention) may be glycosylated by animals (mammals) or other eukaryotic carbohydrates or may not be glycosylated.

 The preferred method of preparation may vary depending on many factors and considerations; The optimal procedure for obtaining this situation will be obvious to an experienced specialist after a minimum number of experiments. The final expression product can then be purified to a state close to homogeneous, using methods known in the art. A typical method for producing prokaryotic cells involves the destruction of cell walls under high pressure or other methods: centrifugation or filtration to remove fragments of cells (cells), followed by ion exchange chromatography of the supernatant or filtrate, and finally chromatography on a hydrophobic medium. If the analog expresses in an insoluble form, another purification method involves first dissolving the inclusion bodies containing analogues, followed by ion exchange chromatography, then unfolding (unwinding) the protein, and finally chromatography on a hydrophobic carrier. Examples of purification techniques are disclosed in U.S.S.N. 08 / 323,339, filed October 13, 1994). Overall U.S.S.N. 08 / 323,339 discloses a method for purifying keratinocyte growth factors, which consists in: a) obtaining a solution containing KGF; b) binding of KGP from the solution according to (a) to a cation exchange resin; c) elution of KGF with a solution of an eluate with a cation exchange resin; d) or passing the eluting solution according to (c) through an appropriate column for molecular weight exclusion chromatography or during the chromatography of the eluate solution according to (c) on a hydrophobic medium; e) regeneration of KGF from a column for displacement chromatography by molecular weight or from a hydrophobic carrier.

Of course, analogues can be easily examined to evaluate their physical properties. The Examples section provides various well-known stability analysis methods, although the specific analysis used to test the analog is not necessary. Moreover, the level (magnitude) of biological activity (for example, receptor binding and / or affinity, mitogenic activity, cell proliferation activity and / or in vivo activity) can also be determined by various methods, some of which are presented in the Examples section. Many assay methods are well known and can be used to quickly examine KGF analogues to determine if they have acceptable biological activity or not. One such assay specifically tests KGF analogues for their ability to bind to the KGF receptor (KGFR) using competitive ¹²⁵ I-KGF binding (Bottaro et al. (1990), J. Biol. Chem., 265: 12767-12770; Ron et al . (1993), J. Biol. Chem., 268: 2984-2988). An alternative way to analyze the interactions of the KGFR / KGF analogue involves the use of methods such as (analysis) determining the actual (real) time of biospecific interaction (BIA) (Felder et al. (1993), Molecular and Cellular Biology, 13: 1449-1455). In addition, mitogenic analysis can be used to determine the ability of KGF analogues to stimulate DNA synthesis (Rubin et al. (1989), supra). Finally, cell proliferation assays can be used to determine the ability of KGF analogues to stimulate cell proliferation (Falco et al. (1988), Oncogene, 2: 573-578). Using any of the aforementioned assay systems, one can quickly examine KGF analogues for their biological activity.

 In a preferred embodiment, the invention relates to (relates to, describes) those KGF analogues that retain the full (i.e., at least substantially the same) in vitro and in vivo biological activity of natural KGF. Examples of KGF analogues with such properties determined using one or more of the above analyzes are C (1,15) S, ΔN3 / C (15) S, ΔN3 / C (15) -, ΔN8 / C (15) S, ΔN8 / C (15) -, ΔN15, ΔN16, ΔN17, ΔN18, ΔN19, ΔN20, ΔN21, ΔN22, ΔN23, ΔN24 or ΔN23 / R (144) Q.

 KGF analogs can be further modified so that they contain additional chemical residues, usually not included in the peptide. Such derivative residues (particles) may increase solubility, absorption, biological half-life, and the like of the KGF analogue. Or particles (residues) can eliminate or reduce unwanted side effects of protein and the like. Particles (residues) capable of mediating such effects are disclosed, for example, in REMINGTON'S PHARMACEUTICAL SCIENCES, 18th ed. , Mack Publishing Co., Easton, PA (1990). Covalent modifications can be introduced into the molecule by the reaction of targeted amino acid residues of the peptide with an organic derivative forming agent that is capable of reacting with selected side chains or terminal residues (i.e. Creighton (1983), PROTEINS: STRUCTURE AND MOLECULE PROPERTIES, WH Freeman and Co. , San Francisco, pp. 79-86). Polyethylene glycol ("PEG") is one of the chemical compounds used to formulate therapeutic protein products. The addition of polyethylene glycol residues has been shown to protect some proteins from proteolysis, Sada et al. (1991), J. Fermentation Bioengineering, 71: 137-139, and methods for attaching certain polyethylene glycol residues are available. See US Patent 4,179,337, Davis et al., "Non-Immunogenic Polypeptides", issued December 18, 1979; and US Patent 4,002,531, Royer, "Modifying enzymes with Polyethylene Glycol and Product Produced Thereby", issued January 11, 1977. For a review, see Abuchowski et al., in Enzymes as Drugs. (Holcerberg and Roberts (eds.) Pp. 367-383 (1981)). Various methods have been used to attach polyethylene glycol molecules to a protein. Typically, polyethylene glycol molecules bind to a protein using a reactive group located in the protein. Amino groups, such as those of lysine residues or at the N-terminus, are convenient (applicable) for such attachment. For example, Royer (US Pat. No. 4,002,531, supra) claims that reductive alkylation has been used to attach polyethylene glycol molecules to the enzyme. EP 0 539 167, published April 28, 1993, Wright, "Peg Imidates and Protein Derivates Thereof" states that peptides and organic compounds with free amino group (s) are modified with PEG imidates or corresponding water soluble organic polymers. Pat. USA 4,904,584, Shaw, issued February 27, 1990, relates the modification of a number of lysine residues in proteins by addition of polyethylene glycol molecules using reactive amino groups.

 In yet another embodiment, the invention describes a single administration of a unit: a medical formulation that can be safely administered parenterally or orally to treat the disease in warm-blooded animals (such as humans). Such medical formulations may be in the form of a lyophilized or otherwise dehydrated therapeutic or diagnostic agent, which may be reconstituted by the addition of a physiologically acceptable solvent. The solvent can be any medium, such as sterilized water, physiological saline, a solution of glucose or other aqueous carbohydrates (for example, polyols such as mannitol, xylitol or glycerin), which is capable of dissolving a dry composition that is compatible with the chosen route of administration and does not adversely affect active principle and on the used recovery stabilizers. In a specific embodiment, the present invention provides a kit for a single dose of a drug. The kit contains both a first pack of dry protein and a second pack of an aqueous formulation containing a reproduction stabilizer. As for the protein concentration in the solution, the volume of the solution in each package and the capacity of (all) packages (interrelated parameters that can be modified accordingly depending on the given concentration of the active principle in a unit of the final dose), they can vary in the interval well known to specialists .

KGF analogues of this invention can be used as therapeutic and diagnostic agents, as well as reagents for research. Thus, KGF analogs can be used for in vitro and / or in vivo diagnostic tests to quantify KGF in a tissue or organ sample or to detect and / or isolate cells that express KGFR (Bottaro et al. (1990), J. Biol. Chem., 265: 12767-12770; Ron et al. (1993), J. Biol. Chem., 268: 2984-2988). In the analysis of tissues or organs, the radioactivity resulting from the binding of ¹²⁵ I-KGF analog to KGFR is less than the value on the standard curve of binding of ¹²⁵ I-KGF analog due to the binding of unlabeled natural KGF to KGFR. Similarly, the ¹²⁵ I-KGF analogue can be used to determine the presence of KGFR in various types of cells.

 The present invention also contemplates the use of a KGF analog to produce antibodies against a peptide, these antibodies are also associated with natural KGF. In this embodiment of the invention, the antibodies are monoclonal or polyclonal in origin and are prepared using the KGF analogue. The resulting antibodies bind mainly with natural KGF, preferably when this protein is in its natural (biologically active) conformation. These antibodies can be used to determine or purify natural KGF.

Moreover, the present invention contemplates the use of KGF analogues for the discovery of high and low affinity KGF-binding molecules having therapeutic applications, for example, for efficient delivery of KGF or as an inhibitor of KGF activity. The therapeutic stability of KGF analogues is important for the identification of such binding molecules under physiological conditions (i.e., at 37 ^° C), since their affinity for KGF can be very temperature dependent and can be unpredictable based on the affinity at 4 ^° C.

 For in vivo use, KGF analogues may be formulated with additives. Such additives include buffers, carriers, stabilizers, fillers, preservatives, tonicity agents, antioxidants, and the like (e.g., viscosity correlating agents, or (dry) diluents). The choice of specific additives depends on the storage form (i.e., liquid or lyophilized) and the route of administration of the KGF analogue. Suitable formulations known in the art can be found at REMINGTON'S PHARMACEUTICAL SCIENCES (latest edition), Mack Publishing Company, Easton, PA.

 KGF analogs can be used in therapeutically effective amounts for tissues specifically characterized as having lesion or having a clinical deficiency of non-fibroblast epithelial cells. Areas in which KGF analogues can be used successfully include, but are not limited to: stimulation, proliferation, and differentiation of accessory structures such as hair follicles, sweat glands, sebaceous glands in patients with burns and other lesions of all or part of the layer; accelerated re-epithelialization of injuries caused by congenital bullous epidermolysis, which is a defect in the fusion (tight connection) of the epidermis with the dermis below, often leads to open painful water bubbles (blisters), which can cause a severe painful condition; preventing chemotherapy-induced baldness and treating male-pattern baldness or accelerated hair loss in men and women; treatment of stomach ulcers and duodenal ulcers; treatment of inflammatory diseases of the digestive tract, such as Crohn's disease (initially affecting the small intestine) and ulcerative colitis (initially affecting the large intestine); prevention or reduction of toxicity in the intestines during irradiation and chemotherapy due to treatment (for example, preliminary or subsequent) to protect or regenerate cells or both; stimulation of mucus in the gastrointestinal tract; stimulating the proliferation and differentiation of type II pneumocytes, which can help in the treatment or prevention of diseases such as hyaline membrane disease of newborns (i.e., respiratory distress syndrome in infants and bronchopulmonary dysplasia); stimulation of proliferation and differentiation of the epithelium of bronchioles and / or alveoli in acute and chronic (damage) lung disease or insufficiency due to respiratory failure (including high oxygen content), emphysema, the use of chemotherapeutic drugs affecting the lungs; trauma due to artificial respiration and others affecting mild circumstances; increased liver function to treat or prevent cirrhosis, sudden liver damage, damage caused by acute viral hepatitis and / or toxic liver damage; stimulation of the regeneration of corneal cells (corneal), for example, in the treatment of corneal lesions (abration); stimulation of the regeneration of epithelial cells in order to treat progressive gum disease; stimulation of regeneration of tympanic membrane epithelial cells in the treatment of middle ear disease and the treatment or prevention of cases of diabetes mellitus or as an aid in transplantation of islet cells (insulator).

 An effective amount of the KGF analogue may be administered to a patient in need of proliferation of non-fibroblast epithelial cells. An “effective amount” is the amount of KGF analogue that is required to achieve the desired response in a patient being treated, and is thus determined usually by the attending physician. Factors affecting the amount of KGF analog administered include the age and general condition of the patient, the disease being treated, etc. Typical doses range from 0.001 mg / kg body weight to 500 mg / kg body weight.

 The KGF analogue can be safely administered parenterally (for example, via IV, IT, IM, SC, or IP), orally or externally to warm-blooded animals (such as humans). KGF analogues can be used by single or multiple administration depending on the disease and condition of the patient. In some cases, the KGF analogue may be administered as an adjunct to other therapies and also with other pharmaceuticals.

 The following examples are included to more fully illustrate the present invention. It is understood that modifications may be made in the presented techniques without departing from the spirit of the present invention.

EXAMPLES
The standard methods for many of the procedures described in the following examples, or the corresponding alternative procedures are given in well-known textbooks on molecular biology, such as, for example, Molecular Cloning, Second Edition, Sambrook et al., Cold Spring Harbor Laboratory Press (1987) and Current Protocols in Molecular Biology, Ausabel et al., Green Publishing Associates / Wiley-Interscience, New York (1990).

Example 1: Obtaining DNA encoding KGF and KGF analogues
Cloning of the full-sized human KGF gene (encoding a polypeptide with the natural KGF sequence) was carried out using both the polymerase chain reaction (PCR) of RNA from an animal cell and PCR synthesized chemically (E. coli optimized by codon) oligonucleotides ("OLIGO" ) Both procedures are described below:
PCR amplification was performed using RNA isolated from cells that are known to produce a polypeptide. First, cells from the human fibroblast cell line AG1523A (obtained from the Human Genetic Mutant Cell Culture Repository Institute For Medical Research, Camden, New Jersey) were destroyed using guanidine thiocyanate, followed by extraction (according to the method of Chomyzinski et al. (1987), Anal. Biochem ., 172: 156). Using the standard reverse transcriptase protocol for total RNA, KGF cDNA was obtained. PCR (PCR # 1) amplification of the KGF gene was performed using KGF as the KGF cDNA template, and OLIGO # 1 and OLIGO # 2 primers, which encode DNA sequences directly at the 5 'and 3' ends of the KGF gene [Model 9600 Thermocycler (Perkin- Elmer Cetus, Norwalk, CT); 28 cycles; each cycle consists of denaturation for one minute at 94 ^° C, hybridization (annealing) for two minutes at 60 ^° C and elongation for three minutes at 72 ^° C]. A small aliquot of PCR # 1 product was then used as a template for a second PCR (PCR # 2) KGF amplification identical to the cycle conditions described above, except that hybridization was carried out at 50 ^° C. For expressing cloning of the KGF gene, loaded (inserted) PCR primers were used to create the corresponding restriction sites at both ends of the KGF gene. OLIGO # 3 and OLIGO # 4, respectively, were used to modify the KGF DNA of the KGF product from PCR # 2 to include MluI and BamHI restriction sites at the 5 'and 3' ends of the gene, respectively [PCR # 3; 30 cycles; each cycle consists of one minute at 94 ^o C (denaturation); two minutes at 60 ^o C for hybridization and three minutes at 72 ^o C for elongation]. This DNA was appropriately cut with MluI and BamHI, extracted with phenol and precipitated with ethanol. Then she was again suspended and
ligated (using T4 ligase) to plasmid pCFM1156 (Figure 2A), which contained the "RSH" signal sequence to create the RSH-KGF construct (Figure 3).

The ligation products were converted (according to the method of Hanahan (1983), J. Mol. Biol., 166: 557) to E. coli strain FM5 (ATCC: 53911) and placed in LB + kanamycin at 28 ^° C. Several transformants and grown in a small amount of liquid cultures containing 20 μg / ml kanamycin. Plasmid RSH-KGF was separated from the cells of each culture and DNA was sequenced. Due to the internal NdeI saiga in the KGF gene, it was not possible to directly clone the natural gene sequence into a given expression vector with the mentioned restriction sites NdeI and BamHI. This was accomplished by ligation of three parts. The RSH-KGF plasmid was cut with unique BsmI and SstI restriction sites, and a ~ 3 kpb DNA fragment (containing the 3 'end of the KGF gene) was isolated, followed by electrophoresis on a 1% agarose gel. PCR (PCR # 4) was performed as described for PCR # 3, except for replacing OLIGO # 3 with OLIGO # 5. The KGF DNA product was then cut with NdeI and BsmI and a 311 pb DNA fragment was isolated, followed by electrophoresis on a 4% agarose gel. The third ligation segment was a 1.8 kpb DNA fragment of pCFM1156, cut with NdeI and SstI followed by electrophoresis on a 1% agarose gel. After ligation (T4), transformation, (selection) selection using kanamycin and DNA sequencing as described above, a clone was selected containing the construct shown in Figure 4 and the plasmid designated KGF. Due to the internal ribosome binding site that produces the truncated product, the KGF DNA sequence between the unique KpnI and EcoRI sites was replaced by chemically synthesized OLIGOs (from OLIGO # 6 to OLIGO # 11) in order to minimize the use of the internal initiation site (Figure 5 ) (see sequence 1 at the end of the description).

 OLIGOs were phosphorylated with T4 polynucleotide kinase and then denatured by heating. Single-stranded (ss) OLIGO formed a ds DNA fragment with a slow decrease in temperature to room temperature. Then, T4 ligase was used to covalently bind the internal sticky ends and the whole ds OLIGO fragment to a KGF plasmid cut with KpnI and EcoRI. The new plasmid is designated KGF (dsd).

 The fully E. coli codon-optimized KGP gene was constructed using PCR amplification of chemically synthesized OLIGOs from # 12 to # 24 (see sequence 2).

OLIGOs # 12 to # 24 were designed so that the complete DNA sequence encoding the natural KGF was represented by OLIGO either the Watson strand or the Scream strand and PCR amplification would give the sequence of the desired double-stranded DNA (Figure 6) [PCR # 5 Model 9600 thermocycler (Perkin-Elmer Cetus); 21 cycles, each cycle consisting of denaturation at 94 ^° C for 31 seconds, hybridization at 50 ^° C for 31 seconds and elongation at 73 ^° C for 31 seconds; after 21 cycles, the PCR was completed after the final stage of elongation for 7 minutes]. After PCR amplification, the DNA fragment was cut with XbaI and BamHI and the 521 pb fragment was ligated into the expression plasmid pCFM1156, cut with the same enzymes. PCR # 5 used external primers (100 pmol / 100 μl r • n) OLIGO # 12 and OLIGO # 13 and 1 μl / 100 μl r • n KGF matrix obtained by ligation (using T4 ligase) from OLIGO # 14 to OLIGO # 19 (from OLIGO # 15 to OLIGO # 18 were phosphorylated with T4 polynucleotide kinase) using OLIGO # 20 to OLIGO # 24 as auxiliary oligonucleotides (Jayaraman et at. (1992), Biotechniques, 12: 392) for ligation. The KGF construct (codon optimized) was built.

 All of the KGF analogues described herein are composed in part of DNA sequences found in KGF (dsd), or KGF (codon optimized), or a combination thereof. The sequences are further modified by the introduction of DNA sequences at the corresponding restriction sites that encode the amino acids of a (specific) specific KGF analog obtained using one or more of the methods described above for synthesizing a DNA fragment. Any of the analogues can be obtained completely (in its entirety) using the above methods. However, as part of the overall OLIGO design, where applicable, E. coli optimized for the codon was used, although the presence of E. coli optimized for the codon in any of the studied genes, in part or in whole, slightly increased the protein yield that could be obtained from cultured bacterial cells. Figures 7 to 24 and 37 to 50 represent corresponding examples of nucleotide and amino acid sequence constructs of a particular KGF analogue: C (1.15) S (Figure 7); ΔN3 / C (15) S (Figure 8); ΔN3 / C (15) (Figure 9); ΔN8 / C (15) S (Figure 10); ΔN8 / C (15) (Figure 11); ΔN15 (Figure 12); ΔN16 (Figure 13); ΔN17 (Figure 14); ΔN18 (Figure 15); ΔN19 (Figure 16); ΔN20 (Figure 17); ΔN21 (Figure 18); ΔN22 (Figure 19); ΔN23 (Figure 20); ΔN24 (Figure 21); C (1.15) S / R (144) E (Figure 22); C (1.15) S / R (144) Q (Figure 23); DN23 / R (144) Q (Figure 24); C (1.15.40) S (Figure 36); C (1.15.102) S (Figure 37); C (1.15.102.106) S (Figure 38); DN23 / N (137) E (Figure 39); DN23 / K (139) E (Figure 40); DN23 / K (139) Q (Figure 41); DN23 / R (144) A (Figure 42); DN23 / R (144) E (Figure 43); DN23 / R (144) L (Figure 44); DN23 / K (147) E (Figure 45); DN23 / K (147) Q (Figure 46); DN23 / K (153) E (Figure 47); DN23 / K (153) Q (Figure 48) and DN23 / Q (152) / K (153) E (Figure 49). All KGF analog constructs described herein represented a confirmed DNA sequence.

-конца до промотора плазмидной репликации, Р_сорB, и до генов плазмидной репликации, изменения пар оснований, см. в последовательности 3.Example 2: Production in E. coli
A. Expression of KGF analogues
When cloning genes for KGF analogues, three different expression plasmids were used. They were pCFM1156 (ATCC 69702), pCFM1656 (ATCC 69576) and pCFM3102 (Figures 2A, 2B and 2C, respectively). Plasmid p3102 can be obtained from plasmid pCFM1156 by a series of site-directed major changes using PCR overlapping oligonucleotide mutagenesis. Starting from the BglII site (plasmid pCFM1656 bp # 180) directly 5

-end to the promoter of plasmid replication, P _CORB , and to the genes of plasmid replication, changes in base pairs, see sequence 3.

As can be seen, pCFM1156, pCFM1656, and pCFM3102 are very similar to each other and contain many identical restriction sites. Plasmids were selected on the basis of convenience, and the components of the DNA vectors can be easily changed in order to new designs. The host used for general (total) cloning is a strain FM5 E. coli (ATCC: 53911) and the transformations were carried out (according to the method Hanahan (1983), supra) or by electroelution using a Gene Pulser ^TM apparatus for transfection (BioRad Laboraties , Inc., Hercules, CA) in accordance with the manufacturer's instructions.

First, a small freshly grown inoculum of the given recombinant E. coli cells was initiated, containing the desired design on one of the three pCFM vectors, transferring 0.1 ml of the corresponding strain of the starting product frozen in glycerol into a 2 L vessel containing 500 ml of Luria broth. The culture was shaken (shaken) at 30 ^{° C.} for 16 hours, after which the culture was transferred to a 15 L fermenter containing an 8 L batch of sterile medium (Tsai, et al. (1987), J. Industrial Microbiol., 2: 181-187 )

Batch-fed fermentation was started with Feed # 1 as medium (Tsai, et al. (1987), supra). When the OD600 reached 35, expression of the desired KGF analog was induced by rapidly raising the temperature of the culture to 42 ^{° C.} in order to denature the CI (protein) repressor. Instead of make-up 1, they gave make-up 2, the initial rate of addition was 300 ml / hour. Make-up 2 contained 175 g / L trypticase peptone, 87.5 g / L yeast extract and 260 g / L glucose. After one hour at 42 ^{° C., the} temperature of the culture was reduced to 36 ^{° C.} and then this temperature was maintained for 6 hours.

Then the fermentation was stopped and the cells were collected by centrifugation in plastic containers placed inside 1 l of vessels in a centrifuge. Cells were planted by centrifugation at 400 rpm for 60 minutes, after which the supernatant was removed and the cell paste was frozen at -90 ^o C.

After the expression of various KGF analogues in E. coli, natural KGF, C (1.15) S, C (1.15) S / R (144) E, C (1.15) S / R (144) Q, ΔN15, ΔN23 and ΔN23 / R (144) Q proteins were purified using the following techniques. After fermentation with high cell density, the cell paste was suspended at 4 ^{° C.} in 0.2 M NaCl, 20 mM NaPO ₄ , pH 7.5 as a 10-20% solution (weight in volume) using a high shear mixer. The weighed cells were then lysed by passing the solution three times through a homogenizer (APV Gaulin, Inc., Everett, MA). The effluent homogenate was cooled to 4-8 ^{° C.} using an appropriate heat exchanger. The impurities were removed by centrifuging the lysate in a J-6B ^{TM centrifuge} (Beckman Instruments, Inc., Brea, CA) equipped with a JS 4.2 rotor at a speed of 4200 rpm for 30-60 minutes at 4 ^° C. The supernatants were carefully decanted and applied on a pre-prepared column of 450 ml (5 cm • 23 cm) with S-Sepharose Fast Flow ^{™ resin} (Pharmacia), balanced with 0.2 M NaCl, 20 mM NaPO ₄ , pH 7.5 at 4 ^° C. The column was then washed with five volumes ( 2250 ml) 0.4 M NaCl, 20 mM NaPO ₄ , pH 7.5 at 4 ^° C. The desired protein was eluted by washing with a 5 L column of 0.5 M NaCl, 20 mM NaPO ₄ , pH 7.5. Fractions of 50 ml were collected and A _{280 was} constantly monitored in the _effluent . Fractions in which the presence of eluted material was detected at A ₂₈₀ were then analyzed using SDS-PAGE on a 14% gel to confirm the presence of a given polypeptide.

Then the fractions containing the proteins of interest were pooled and diluted with an equal volume of distilled water. Then, the diluted sample was applied to a 450 ml (5 cm x 23 cm) pre-prepared column with S-Sepharose Fast Flow equilibrated with 0.4 M NaCl, 20 mM NaPO ₄ , pH 6.8 at 4 ^° C. The column was washed with 2250 ml 0.4 M NaCl, 20 mM NaPO ₄ , pH 6.8 and protein were eluted using 20 column volumes, in a linear concentration gradient from 0.4 M NaCl, 20 mM NaPO ₄ , pH 6.8 to 0.6 M NaCl, 20 mM NaPO ₄ , pH 6.8. Again, 50 ml fractions were collected with constant monitoring of A ₂₈₀ in the effluent. Protein-containing fractions (defined by 14% SDS-PAGE) were pooled, followed by concentration through a YM-10 membrane (molecular weight 10,000 cross-sectional membrane) in a 350 ml stirred vessel (Amicon, Inc., Mayberry, MA) to a volume of 30- 40 ml

The concentrate was then applied to a pre-made column of 1300 ml (4.4 x 85 cm) with Superdex-75 ^TM resin (Pharmacia), balanced with column buffer containing 1X PBS (physiological saline, buffered with phosphate, Dalbecco, "D-PBS", not containing calcium and magnesium) or 0.15 M NaCl, 20 mM NaPO ₄ , pH 7.0. After applying the sample to the column, the protein was eluted by gel filtration using a column buffer. Then, 10 ml fractions were collected and those containing the analogue (detected using 14% SDS-PAGE) were combined. Typically, the protein concentration was about 5-10 mg / ml in the final volume. All the above operations were carried out at 4-8 ^o C, unless otherwise indicated.

An alternative purification operation has been used to purify natural KGF, C (1.15) S and ΔN23. The procedure includes the following stages and, unless otherwise indicated, the temperature of the solutions and materials was maintained 23 ± 5 ^o C.

At the end of the bacterial fermentation step, the cell culture was cooled to 4-8 ^° C and the cells were harvested by centrifugation or the like. Based on the expected protein yield per unit weight of the cell paste and the required amount of purified protein, the corresponding amount of cell paste was suspended in five times (by weight) the amount of soft buffer solution 20 mM NaPO ₄ , 0.2 M NaCl, pH 7.5. Cells were dispersed to a homogeneous solution using a high shear mixer. The temperature of the dispersion of the cell paste was maintained at 4-8 ^o C during homogenization.

Cells were then lysed under pressure, for example, by passing a cell paste dispersion twice through a cell homogenizer with a suitable cross-sectional size. The homogenate was kept cooled at 5 ± 3 ^° C. To clarify the cell lysate, a volumetric filter device was used, equipped with a filter with a sufficient surface area balanced with an appropriate volume of 0.2 M NaCl, 20 mM NaPO ₄ , pH 7.5. Balancing and clarification (cleaning) was carried out at 5 ± 3 ^° C. Before cleaning, an appropriate amount of a suitable filtering accelerator was used to pre-coat the filter and mix thoroughly with the lysate, after which the lysate was clarified by passing the solution through a filtered apparatus. The filter was washed with 0.2 M NaCl, 20 mM NaPO ₄ , pH 7.5. The filtrate and washings were collected in a chilled vessel of a suitable capacity, while the temperature was everywhere kept below 10 ^o C.

After filtration, the lysate was passed through a pre-prepared column of SP-Sepharose Fast Flow with a ratio of at least 1 ml of resin per 2 g of cell paste. The column with SP-Sepharose Fast Flow was balanced with cold (5 ± 3 ^° С) 0.2 M NaCl, 20 mM NaPO ₄ , pH 7.5. The column temperature was maintained at a temperature below 10 ^o C. The filtered lysate (5 ± 3 ^o C) was then (loaded) applied onto an ion-exchange column, while the absorption of the eluate at 280 nm (A ₂₈₀ ) was constantly checked. After applying the sample, the column was washed with cold 0.2 M NaCl, 20 mm NaPO ₄ , pH 7.5, followed by washing with 0.3 M NaCl, 20 mm NaPO ₄ , pH 7.5 at 23 ± 5 ^o C.

To elute the target protein, a linear concentration gradient chromatography of 0.2-1 M NaCl, 20 mM NaPO ₄ , pH 7.5 was used. The entire product was collected in several fractions depending on A ₂₈₀ eluate. After elution, these fractions were combined and the volume was noticed.

In order to oxidize free sulfhydryl groups, an oxidation step was carried out. In the case of proteins with altered cysteine particles, compared with natural KGF, an oxidizing agent (for example, cystamindihydrochloride or another suitable oxidizing agent, for example cystine, oxidized glutathione or divalent copper) was added to a final concentration of 1-20 mM and the pH was adjusted to 7-9.5, in the case of the use of cystamindihydrochloride, preferably pH 9.0 ± 0.3. Oxidation was carried out at 10-30 ^o C for the appropriate time. In the case of natural KGF protein, oxidation was carried out by adding an appropriate amount of (NH ₄ ) ₂ SO ₄ , such as 1-2 M (NH ₄ ) ₂ SO ₄ , adjusting the pH to 7.5 ± 0.5 and maintaining the temperature at 23 ± 5 ^o C for the required period of time.

After oxidation, the pH of the solution was adjusted to between 6.5 and 9.5. If necessary, solid (NH ₄ ) ₂ SO _{4 was} added to the solution to a final concentration of 2 M. To remove particles, the solution was passed through appropriate clarification filters.

The filtered oxidized product was then subjected to hydrophobic support chromatography (HIC). The carrier in the HIC resin represented ^{Butyl-650M Toyopearl TM (Tosohaas,} Inc., Montgomeryville, PA). The solution containing the protein was applied to a column, which was previously balanced with 2 M (NH ₄ ) ₂ SO ₄ , 0.15 M NaCl, 20 mM NaPO ₄ , pH 7.0. After applying the sample, the column was washed with 2 M (NH ₄ ) ₂ SO ₄ , 0.15 M NaCl, 20 mM NaPO ₄ , pH 7.0. The desired protein was then eluted using chromatography in a decreasing linear concentration gradient of (NH ₄ ) ₂ SO ₄ from 2.0 M in 0.15 M NaCl, 20 mM NaPO ₄ , pH 7.0. When the target protein began to elute, as can be seen from the increase in A ₂₈₀ eluate, fractions were collected. An aliquot of each fraction was then analyzed by SDS-PAGE. Then the fractions containing the desired protein were combined, thoroughly mixed and the total volume, as well as the concentration of protein in it, was determined.

The combined HIC protein-containing eluate was concentrated and the elution buffer was replaced. Typically, proteins were concentrated to 5.0-10.0 mg / ml. Ultrafiltration was performed using an ultrafiltration system equipped with a PTGC Pellicon ™ cartridge ^system (Millipore, Inc., Bedford, MA) with a membrane with a suitable cut-off level.

After concentration, the sample was diafiltered against an appropriate buffer. What was retained (retentate) at the concentration stage was diafiltered against 0.15 M NaCl, 20 mM NaPO ₄ , pH 7.0 until the conductivity of the retentate was approximately 5% of the conductivity of the 0.15 M NaCl solution, 20 mM NaPO ₄ , pH 7.0.

In addition, to remove sediment and bacterial endotoxin that may be present, a concentrated, diafiltered, protein containing sample was passed through a 0.1 mm Posidyne® filter (Pall, Inc., Cortland, NY). After determining the concentration of protein in the solution and based on the desired concentration of the final mass of the product, the solution was diluted with 0.15 M NaCl, 20 mm sodium phosphate, pH 7.0 to a given final concentration. Then, final aseptic filtration was performed through a 0.22 mm filter and the resulting product mass was transferred to storage tanks with fire protection (at about 5 ^° C) for further use (for formulations).

B. Analysis
The analysis was carried out using E. coli - obtained natural KGF; C (1.15) S; C (1.15) S / R (144) Q; DN15; DN23 and DN23 / R (144) Q.

Toxicity
Toxicity studies of polypeptides have shown that they are not toxic.

Conformational stability
Polypeptides were compared by their storage stability, thermal unwinding transition temperature (T _m ), and stability over a wide pH range.

Storage stability
Figure 25 compares the natural KGF and C (1.15) S after storage in 2 mM NaPO ₄ , 0.15 NaCl, pH 7.0 at 37 ^° C for 27 hours. C (1.15) S contained a significantly higher soluble protein content compared to natural KGF.

Figure 26 compares the stability of natural KGF, ΔN15 (not shown) and C (1.15) S in PBS and in 50 mM NaPO ₄ , 0.15 NaCl, pH 7.0 after storage at 37 ^° C for 18 hours. The regeneration of soluble protein is significantly higher in saturated phosphate solution than in PBS. ΔN15, like C (1.15) S, showed significantly increased stability compared to natural KGF. ΔN15 and C (1.15) S also gave ~ 100% recovery (regeneration) in saturated phosphate, as expected from the results with natural KGF.

However, a single preliminary experiment was conducted comparing the storage stability of C (1,15,40) S and C (40) S (which is encoded with bases from 201 to 684 SEQ ID NO: 1, except that Ser ⁴⁰ is encoded by AGA) ; and C (1,15,102) S and C (102) S (which is encoded with bases from 201 to 684 of SEQ ID NO: 1, except that Ser ¹⁰² is encoded by AGA). Results (not shown) indicate low stability (i.e., less soluble protein after storage at 37 ^{° C.} ). However, the amount of soluble, correctly twisted C (1,15,40) S protein that was purified from the culture medium was greater than that of C (40) S; this suggests that C (1,15,40) S can, in fact, be more stable and that the results of the comparison example are unproven, unconvincing. The present invention preferably includes a KGF analog having substitutions other than Cys ⁴⁰ , Cys ¹⁰² , Cys ¹⁰⁶ , and more preferably excludes C (1,15,40) S, C (1,15,102) S and C (1, 15,102,106) S.

The ability of natural KGF, C (1,15) S, ΔN23, C (1,15) S / R (144) E, C (1,15) S / R (144) Q and ΔN23 / R (144) Q to prevent the formation of aggregated molecules at elevated temperatures has also been studied. Samples containing 0.5 mg / ml protein were prepared in D-PB8. Aliquots of 0.5 ml of each sample were placed in 1 ml 3-ml glass tubes. The tubes were closed with rubber stoppers and fixed on top of 13 mm with aluminum locking caps. Then these tubes were placed in a thermostat at 37 ^o C. At predetermined time intervals, the tubes were removed and analyzed the loss (decrease) of soluble protein. Visible sediments (particles) were removed by centrifugation of 250 ml of each sample through a 0.22 mm Spin-HF filter unit (Costar, Cambridge, MA). Soluble protein in the filtered solution was analyzed by size exclusion chromatography by HPLC. The amount of soluble protein was determined by integrating the HPLC peak area and plotting the result as a function of thermostating time at 37 ^° C. Results for natural KGF, C (1.15) S, C (1.15) S / R (144) E and C (1.15) S / R (144) Q and ΔN23 / R (144) Q are shown in Figure 26 (data for ΔN23 and ΔN23 / R (144) Q are not shown).

These kinetic curves were then used to estimate the half-life — the loss of half of the soluble, monomeric protein. Table 1 shows the half-life of soluble KGF for these proteins remaining after storage at 37 ^{° C.}

 As can be seen from Table 1 and Figure 27, natural KGF formed aggregated molecules most rapidly with a half-life of 0.6 days. C (1.15) S / R (144) Q, ΔN23 / R (144) Q and C (1.15) S / R (144) E showed a significant increase in half-life - loss of half solubility - to 13.3, 22.3 and 38 days respectively.

Thermal unwinding
Thermal unwinding was monitored by the circular dichroism (CD) method at 230 nm using a J-720 ^TM spectropolarimeter (Jasco, Inc., Easton, MD) equipped with a Peltier type RTS-343 temperature control system. Separate samples containing 0.1 mg / ml of the analyzed polypeptide in D-PBS (Life Technologies, Inc., Grand Island, NY) were prepared for CD analysis. About 2.5 ml of each sample was placed in a fluorescent cuvette (Hellma Cells, Inc., Jamaica, NY) from 10 mm long Suprasil ^TM rectangular quartz plates (Heraeus Quarzschmelze, GmbH, Hanau, Germany). The cuvette was then placed in a Peltier temperature control system in a spectropolarimeter. Thermal unwinding was carried out at a speed of 50 ^o C / hour. Changes in ellipticity were monitored at 230 nm to indicate deployment. T _{m of} each sample was evaluated by determining the temperature at which 50% of the protein molecules in the solution were unfolded (curled) (Biophysical Chemistry, Cantor and Schimmel (eds), WH Freeman and Co. San Francisco (1980). Defined T _m for each of three proteins are listed in Table 2.

As these results show, in C (1.15) S and ΔN15 T _{m is} increased by 1 ^o C compared with natural KGF. At ΔN23 T _m increased one more degree. However, replacing R144Q with C (1.15) S / R (144) Q or ΔN23 increases T _{m by} more than 6 ^° C and more than 7 ^° C compared to natural KGF. Moreover, C (1.15) S / R (144) E is more than 9 ^° C more stable than natural KGF.

pH
Acid stability C (1.15) S / R (144) Q and C (1.15) S / R (144) E were also compared with that of natural KGF, adjusting D-PBS to various pH values by adding concentrated HCl or NaOH. Approximately 2.35 ml of D-PBS with different pH values were mixed with 100 ml of a 2.45 mg / ml KGF protein solution in a quartz cuvette. These samples were thermally deployed at a speed of 50 ^o C / hour and controlled by CD at 230 nm. Figure 28 shows T _m as a function of pH for natural KGF, C (1.15) S, C (1.15) S / R (144) Q, and C (1.15) S / R (144) E. In the studied pH range, C (1.15) S, C (1.15) S / R (144) Q and C (1.15) S / R (144) E always had a higher T _m than natural KGF.

MEM), 50% приготовленного потребителем F12, 5 мг/мл трансферрина, 5 нг/мл селенита натрия, 0.0005% HSA и 0.005% Tween 20. Образцы KGF затем помещали в 96-луночные планшеты Falcon Primeria, засеянные клетками Balb/MK. В ходе синтеза ДНК измеряли [³H] -тимидин и преобразовывали в концентрацию (находили концентрацию) природного KGF на входе сравнением со стандартной кривой природного KGF. Результаты представлены на Фигурах от 29 до 34. Как видно из Фигур, анализируемые аналоги KGF, описанные на Фигурах от 28 до 33, имеют активность, сравнимую с природным KGF.In vitro biological activity
In vitro, mitogenic activity was determined in the same way as a function of protein concentration and half maximum concentrations by measuring the consumption of [ ³ H] -thymidine by Balb / MK cells (according to the methods of Rubin et al. (1989), supra). Typically, the concentration ratio of each of the KGF analogues was determined using in vitro biological analysis. Each KGF analog was then diluted and analyzed (for) biological activity using a Balb / MK mitogenic assay. First, the samples were diluted in a bioassay medium, consisting of 50% of the minimum support medium Eagle prepared by the consumer (

MEM), 50% consumer-prepared F12, 5 mg / ml transferrin, 5 ng / ml sodium selenite, 0.0005% HSA and 0.005% Tween 20. KGF samples were then placed in Falcon Primeria 96-well plates seeded with Balb / MK cells. During DNA synthesis, [ ³ H] -thymidine was measured and converted to the concentration (found concentration) of natural KGF at the input by comparison with the standard curve of natural KGF. The results are presented in Figures from 29 to 34. As can be seen from the Figures, the analyzed KGF analogues described in Figures from 28 to 33 have activity comparable to natural KGF.

Example 3: Production in Mammalian Cell Culture
This example describes the expression, isolation and characterization of two biologically active forms of recombinant KGF (rKGF) obtained in a mammalian expression system.

 The human KGF gene was isolated by PCR amplification of cDNA obtained from normal human dermis fibroblast cells (Clonetec, Inc., San Diego, CA). After obtaining cDNA using revertase, PCR (polymerase chain reaction) was used to amplify the KGF gene. For amplification of the gene from cDNA, OLIGO # 25 and OLIGO # 26 were used, and OLIGO # 27 and OLIGO # 28 were used to place the Hind III and Bgl II restriction sites at the ends of the fragment of the second PCR amplification, as shown in Figures 1a and 1b ( 4 cm sequence at the end of the text).

 After cloning and DNA sequence confirmation, the KGF gene DNA was used. Amplification was performed using OLIGO # 29 and OLIGO # 30 (5 cm sequence at the end of the text).

The OLIGO # 29 sense primer contained the XbaI site and the Kozak consensus translation sequence (5'-CCACC-3 ') of the upstream start codon, ATG. The antisense primer OLIGO # 30 included a SalI cloning site and an additional termination codon. After 18 cycles of PCR amplification (30 s of denaturation at 94 ^° C, 40 s of hybridization (firing) at 55 ^° C and 40 s of elongation at 73 ^° C), the product was hydrolytically digested with XbaI and SalI and ligated with a similarly hydrolyzed pDSRα2 DNA (by methods of Bourdrel et al. (1993), Protein Exp. and Purif., 4: 130-140 and Lu et al. (1992), Arch. Biochem. Biophys., 298: 150-158). This led to the plasmid KGF / pDSRα2, which placed the human KGF gene between the SV40 sequences of the early promoter and α-FSH polyadenylation. Two clones were selected and DNA sequencing confirmed the design of the desired vector.

Two micrograms of KGF / pDSRα2 DNA were then linearized using RvuI. Chinese hamster ovary cells (CHO), seeded a day earlier in culture dishes of 0.8 x 10 ⁶ cells / 60 mm, were transfected with treated DNA using the standard calcium phosphate precipitation method (Bourdrel et al., Supra). Two weeks later, individual colonies were selected and transferred to 24-well plates. Conditioned media are considered serum-free when the cells merge, and aliquots of them are analyzed by Western blotting using a polyclonal rabbit antiserum reagent against human E. coli expressed KGF.

Western blotting was performed by passing samples through 12.5% (w / v) SDS polyacrylamide gels, followed by electro-blotting for 400 hours at 400 mA onto nitrocellulose membranes using a semi-dry transfer apparatus (Hoefer Scientific Instruments, San Francisco, CA). 20 mM Tris, 150 mM glycine, 20% methanol served as transfer buffer. Nitrocellulose sheets were blocked by incubation with 10% normal goat serum in PBS. A rabbit antiserum obtained against E. coli-produced KGF was used as the primary antibody. For consumption, it was diluted 1/10 000 in 1% normal goat serum in PBS and incubated with blocked nitrocellulose sheets for 12 hours at room temperature, after which excess antibody was removed three times, washing in PBS for 30 minutes. Then, nitrocellulose membranes were thermostated in 100 ml of 1% normal goat serum in PBS containing Vectastain ^™ goat anti-rabbit IgG biotinylated (secondary antibody, Vector Labs. Burlingame, CA) for 30 minutes at room temperature. After three ten-minute washes in PBS, they were incubated for 30 minutes at room temperature in 100 ml of a 1% solution of normal goat serum containing streptavidin and biotinylated peroxidase, prepared in accordance with the manufacturer's recommendations (Vector Labs). After three washes in PBS, the KGF cross-reaction material was visualized by incubation in a mixture of 60 μl 30% (w / v) H ₂ O ₂ in 100 ml of PBS and 50 mg of 4-chloronaphthol in 20 ml of methanol. The reaction was stopped by pouring into water after 10 minutes.

 Blot analysis showed that a KGF-specific antibody is associated with three distinct protein bands, two of which are close in molecular weight to about 25-29 kDa, and one reaches a molecular weight of about 17 kDa compared to the expected molecular weight of about 18.8 for mature protein of 163 amino acids. In addition, several highly expressing clones secreting more than 2.0 mg of rKGF per liter, according to Western analysis, were selected and placed in rotating bottles (according to the method of Lu et al., Supra) in order to obtain large volumes of serum-free conditioned cleaning medium KGF cation exchange chromatography and gel filtration, as shown below.

KGF from 3 L of serum-free conditioned medium was purified by placing the medium directly on a cation exchange column (5 x 24 cm) filled with 450 ml of sulfoethyl S-Sepharose Fast Flow (Pharmacia), pre-equilibrated with 20 mM sodium phosphate, pH 7.5. After washing with five column volumes, 20 mM sodium phosphate, 0.2 M NaCl, pH 7.5, rKGF was eluted using 20 column volumes in a linear concentration gradient from 0.2 to 1.0 M NaCl in 20 mM sodium phosphate, pH 7.5. 50 ml fractions were collected with constant monitoring of A ₂₈₀ . KGF protein was detected by analyzing aliquots of each SDS-PAGE fraction. SDS-PAGE was performed in an electrophoretic system (Novex, San Diego, CA) using ready-made 14% trisglycine gels (according to the method of Laemmli (1970), Nature, 227: 680-685). Samples were mixed with a non-reducing sample SDS buffer without heating prior to application. Proteins were determined by staining with either Coomassie blue or silver. Two late eluting peaks apparently contained protein bands corresponding to 25-29 kDa and 17 kDa determined by western blot. Fractions containing each of these peaks were individually concentrated to a volume of less than 1 ml and subjected to gel filtration.

For gel filtration, columns with Superdex-75 ^TM resin (HR 10/30, Pharmacia), pre-equilibrated with PBS, pH 7.2, and calibrated according to the following well-known molecular weight standards (BioRad, San Francisco, CA): tyroglobulin (670 kDa) were used. ), g-globulin (158 kDa), ovalbulin (44 kDa), myoglobin (17 kDa) and vitamin B-12 (1.4 kDa). These purification steps give an approximately 2,000-fold purification of rKGF, with the inclusion of a specific material of 17 kDa and 30 kDa, as such, as determined by silver staining.

Using an example of a material with a higher molecular weight, rKGF was eluted as the main symmetric peak (A ₂₈₀ ), which was named KGF-a. By SDS-PAGE analysis of a smaller amount of this material, 3 μg / lane instead of 6 μg / lane, two bands were allowed with a molecular weight difference of 1-2 kDa. Using an example of a lower molecular weight material called KGF-b, gel filtration gave a protein preparation with the expected mobility. For both KGF-a and KGF-b, the overall yield after purification was approximately 30-40%.

The amino acid sequences of KGF-a and KGF-b were also analyzed. These analyzes were performed on an automated sequencer (Model 477A or 470A, Applied Biosystem, Inc., Foster City, CA) equipped with a 120A continuous PTH amino acid analyzer and Model 900A data acquisition system (according to the method of Lu et al. (1991), J. Biol. Chem., 266: 8102-8107). Edman sequencing showed in KGF-a the main N-terminal sequence of X ₁ -NDMTPEQMATNVX ₂ -X ₃ -S- (SEQ ID NO: 75). A minor sequence, starting from the third N-terminal amino acid, aspartic acid, was also present in 1.6% of all protein sequences. X ₁ , X ₂ and X ₃ were not assigned due to the lack of phenylthiohydantoinyl (PTH) amino acid signals during sequencing. The KGP N-terminal amino acid sequence predicted from the cDNA sequence indicates that X ₁ and X ₃ are Cys and X _{2 are} asparagine. The absence of X ₁ and X ₃ indicates that these cysteines can form disulfide bridges. Based on the sequence of matched N-linked glycosylation of Asn-X-Ser, the absence of the predicted Asn residue corresponding to X ₂ indicates that this is a potential N-linked glycosylation site.

Interestingly, the N-terminal sequencing of KGF-b showed the N-terminal amino acid sequence of SYDYMEGGDIRV- (SEQ ID NO: 76), which suggests that it was the N-terminal processed (truncated) form of KGF that was cleaved proteolytically by the Arg ²³ -Ser ²⁴ peptide bond .

 To further characterize purified KGF-a and KGF-b, the protein was exposed to glycosidases (neuraminidases, O-glycanases and / or N-glycanases) using known methods (Sasaki et al. (1987), J. Biol. Chem., 262: 12059-12076); Takeuchi et al. (1988), J. Biol. Chem., 263: 3657-3663); Zsebo at al. (1990), Cell, 63: 195-201). These data indicate that KGF-a contains N- and O-linked carbohydrates, although a lower molecular weight form of KGF-a may contain only N-linked sugar. Glycosidase treatment did not cause a decrease in the molecular weight of KGF-b, which indicates that the molecule is not glycosylated.

A glycosylated KGF-a sample is further characterized by mass spectroscopy of the endoproteinases of Glu-C peptides described above. Elucidation of the carbohydrate structure of a glycopeptide by effusion mass spectrometric analysis has been successfully applied to other proteins (Huddleston et al. (1993), Anal. Chem., 65: 877-884; Carr et al. (1993), Prot. Sci., 2 : 183-196). As confirmed by the isolation of the non-glycosylated polypeptide, Thr ²² appears to be partially glycosylated. A similar mass spectrometric analysis of Asn ¹⁴ suggested microheterogeneity during glycosylation with bi-, tri-, and tetraantenna structures with varying degrees of sialylation.

Table 3A summarizes the concentration of KGF that stimulates the [ ³ H] -thymidine incorporation of keratinocytes at half the maximum rate (according to the method of Rubin et al. (1989), supra). The interaction of the KGF receptor was studied using membrane preparations from the isolated KGF receptor prepared from keratinocytes of the epidermis of Balb / MK mice (as described by Massague (1983), J. Biol. Chem., 258: 13614-13620). Specifically, various forms of KGF were diluted with 50 mM Tris-HCl, pH 7.5, containing 0.2% bovine serum albumin so that the concentration range was from 0.8 ng to 100 ng per 50 μl. They were thermostated separately with a membrane preparation (75 ng / ml) and ¹²⁵ I-labeled E. coli-formed KGF (1.5 ng). Prescription binding and competition experiments were carried out at 4 ^{° C.} for 16 hours, after which the samples were removed, centrifuged and washed twice with the above diluted buffer to remove unbound and non-specifically bound labeled KGF. Samples were then counted for residual radioactivity. Competitive receptor binding curves between KGF samples and labeled KGF were constructed as a graph of the ratio (percentage) of non-competitiveness to the concentration of each KGF sample. Table 3B summarizes the concentration of KGF necessary to achieve 60% non-competitiveness against labeled E. coli-formed KGF, expressed in ng / ml.

As shown in Table 3A, KGF-a and KGF-b stimulate comparable [ ³ H] -thymidine incorporation, with a rate equal to half the maximum being stimulated by approximately 30 ng / ml of any analogue. Thus, shortening (truncation) does not reduce the biological activity of the molecule. However, two analogues have an activity of about 3 times lower than E. coli-formed full-sized KGF. As shown in Table 3B, non-competitive experiments using radioreceptor analysis showed that E. coli-formed KGF, KGF-a and KGF-b have similar receptor binding activity.

Example 4: In vivo biological analysis of KGF polypeptides produced in E. coli and mammalian cell culture
It has been shown that a single dose of KGF subcutaneously leads to a dose-dependent increase in serum cholesterol in mice within 24 hours. Natural KGF, KGF-a, C (1.15) S, ΔN15 and ΔN23 were also evaluated in terms of their ability to increase serum cholesterol depending on the dose.

 Each experimental group contained five female Balb / c mice (18-20 g) obtained from CRL. The protein was diluted with 0.9% saline so that the final injection volume was 100 μl / mouse. Each mouse was given a single subcutaneous injection in the following doses (Table 4).

 24 hours after injection, the mice were sacrificed and bled through a cardiac puncture. Blood samples were processed to determine serum cholesterol.

 As shown in Figure 34, it was found that each test KGF analogue increases serum cholesterol depending on the dose. Further, no apparent difference in bioactivity was found between any KGF analogs tested.

In vivo diabetes model
Chemically-induced (induced) diabetes mellitus patterns in various animal species have traditionally been used to study the disease and its treatment. Streptozotocin causes diabetes in mice, rats, hamsters, dogs, and monkeys, although studies on rats and mice are most commonly practiced. Jonod et al., Proc. Soc. Exp., Pio. Med. 126: 210-205 (1967); Rerup, Pharm. Pev. 22: 485-518 (1970); Rossini et al., PNAS 74: 2485-2489 (1977) and Ar'Rajab and Ahren, Pancreas 8: 50-57 (1977). In rats, doses of streptozotocin from 45 to 70 mg / kg as a single dose intravenously cause persistent disease. Doses below 45 mg / kg cause a (briefly) temporary painful condition that is reversible. Within one day after the administration of streptozotocin, a state of hyperglycemia is induced. The insulin content in the blood remains substantially unchanged compared to normal rats; however, the total content of insulin and C-peptide in the pancreas decreases sharply. Rats show the classic signs and symptoms of diabetes in humans: elevated blood glucose (hyperglycemia), glucose in the urine (glucosuria), increased thirst (polydipsia), increased urine output (polyuria), increased appetite (hyperphagia).

 The studies described in this discovery were conducted with a model of diabetes caused by streptozotocin in Sprague-Dawley rats. Male rats weighing 200-260 grams were used in research. Diabetes was caused by a single intravenous injection of streptozotocin in an amount of 50 mg of streptozotocin in sodium citrate buffer per 1 kg of body weight. Non-diabetic control rats received a single intravenous injection of sodium citrate buffer for control purposes. KGF was administered daily as a subcutaneous injection. The dose of KGF was from 3 to 5 mg / kg / day, depending on the experiment. In the first experiment, KGF therapy was started two days before the induction of diabetes and continued after the induction of diabetes, a total of eight injections. In the second and third experiments, KGF therapy was started a day after the induction of diabetes by streptozotocin. In the fourth experiment, a seven-day course of KGF therapy was started seven days after the administration of streptozotocin, and the animals continued it for another 12 weeks. In all experiments, except the fourth, blood glucose, urine glucose, and urine output were used as final indicators in the analysis. In addition, in some experiments, water intake, C-peptide content in urine, or total insulin and C-peptide content in the pancreas were determined. In the fourth experiment, the only end-point analyzed was blood glucose. Due to the fact that most of the insulin is removed (withdrawn) from the bloodstream by the liver, the determination of peripheral insulin concentrations reflects events of posthepatic metabolism (metabolism after the liver) rather than the secretion of insulin from the pancreas. Therefore, the definition of C-peptide is often used as a peripheral marker for insulin secretion. C-peptide is formed by processing proinsulin into insulin. Insulin and C-peptide are secreted by beta cells in equimolar ratios and only a small amount of C-peptide is extracted by the liver.

 This study examined the effects of KGF on streptozotocin-induced diabetes in Sprague-Dawley rats. On day 0, rat groups were exposed to either 45 or 50 mg / kg of streptozotocin (STZ). After that, the fasting blood glucose was monitored daily to determine the severity of an insulator lesion (islet cell lesion). On day 5, animals receiving SIZ were placed in one of two groups (20 / group) depending on the size (size) of hyperglycemia. The determining (separating) value was the blood glucose content of 300 mg / ml. A group of animals not receiving SIZ served as a control. On day 7, 10 animals from each group with hyperglycemia were given ΔN23 (3 mg / kg / day) or PBS as a subcutaneous injection and were given for a total of 7 days. Blood glucose was monitored daily, or every other day, or after a week; the data are presented in Figure 51. It should be noted that SIZ treated animals from both groups treated with ΔN23 showed significant decreases in blood glucose during the period of taking ΔN23. It is important that the average drop in blood glucose observed in SIZ-treated animals from the group with an initial blood glucose <300 mg / dl stabilized at about 150 mg / dl, while the drop in blood glucose observed in the group with its initial a value of <300 mg / dl was only (briefly) temporary. It should be noted that the scale is not linear with respect to days.

 Although the present invention has been described both generally and for presumptive embodiments, in light of the foregoing, it is understood that other variations and modifications may occur to those skilled in the art.

Claims (32)

 1. An analogue of natural keratinocyte growth factor corresponding to amino acids 32-194 of the sequence SEQ ID NO: 2, with or without a signal sequence, designated "KGF", characterized in that this analogue contains a KGF sequence with modifications of one or more amino acids 32-55 SEQ ID NO: 2, with the cysteine residue at position 32 of SEQ ID NO: 2 and the cysteine residue at position 46 of SEQ ID NO: 2 deleted or replaced with another amino acid.  2. The analogue according to claim 1, characterized in that the cysteine residue at position 32 of SEQ ID NO: 2 and the cysteine residue at position 46 of SEQ ID NO: 2 are deleted.  3. The analogue according to claim 1, characterized in that amino acids 32-55 of the sequence SEQ ID NO: 2 are deleted.  4. The analogue according to claim 1, characterized in that the cysteine residue at position 32 of the sequence SEQ ID NO: 2 is deleted, and the cysteine residue at position 46 of the sequence SEQ ID NO: 2 is replaced by another amino acid.  5. The analogue according to claim 1, characterized in that the amino acids up to the fourteen first N-terminal amino acids are deleted, and the cysteine residue at position 46 of the sequence SEQ ID NO: 2 is replaced by another amino acid.  6. The analogue according to claim 1, characterized in that the cysteine residue at position 32 of SEQ ID NO: 2 and the cysteine residue at position 46 of SEQ ID NO: 2 are replaced by another amino acid.  7. An analogue according to one of claims 4 to 6, characterized in that the cysteine residue is replaced by amino acid residues selected from alanine, leucine or serine residues.  8. The analogue according to claim 7, characterized in that the cysteine residue is replaced by a serine residue.  9. An analogue according to any one of claims 1 to 8, further comprising charge changing substitutions of one or more amino acids at positions 72 to 185 of the sequence SEQ ID NO: 2 with a neutral or negatively charged amino acid residue. 10. An analogue according to any one of claims 1 to 9, in which at least one amino acid residue in the loop forming amino acid sequence of Asn ¹⁴⁶ -Thr ¹⁵⁰ in the sequence of SEQ ID NO: 2 is replaced by a residue of another amino acid.  11. An analogue according to any one of paragraphs.9-10, characterized in that amino acids 32-54 of the sequence SEQ ID NO: 2 are deleted.  12. The analogue according to claim 1, characterized in that the analogue is selected from the group consisting of C (1.15) S (which corresponds to SEQ ID NO: 32, the first methionine residue being considered as “0” and is optional) ; C (1.15.40) S (which corresponds to SEQ ID NO: 78, with the first methionine residue being considered as “0” and optional); C (1.15.102) S (which corresponds to SEQ ID NO: 80, with the first methionine residue being considered as “0” and optional); C (1,15,102,106) S (which corresponds to SEQ ID NO: 82, with the first methionine residue being considered as “0” and optional); ΔN15 (which corresponds to SEQ ID NO: 42, with the first methionine residue being considered as “0” and optional); ΔN16 (which corresponds to SEQ ID NO: 44, with the first methionine residue being considered as “0” and optional); ΔN17 (which corresponds to SEQ ID NO: 46, with the first methionine residue being considered as “0” and optional); ΔN18 (which corresponds to SEQ ID NO: 42, with the first methionine residue being considered as “0” and optional); ΔN19 (which corresponds to SEQ ID NO: 50, with the first methionine residue being considered as “0” and optional); ΔN20 (which corresponds to SEQ ID NO: 52, with the first methionine residue being considered as “0” and optional); ΔN21 (which corresponds to SEQ ID NO: 54, with the first methionine residue being considered as “0” and optional); ΔN22 (which corresponds to SEQ ID NO: 56, with the first methionine residue being considered as “0” and optional); ΔN23 (which corresponds to SEQ ID NO: 58, while the first methionine residue is considered the residue "0" and is optional); ΔN24 (corresponding to SEQ ID NO: 60, with the first methionine residue being considered as “0” and optional); ΔN3 / C (15) S (which corresponds to SEQ ID NO: 34, with the first methionine residue being considered as “0” and optional); ΔN3 / C (15) (which corresponds to SEQ ID NO: 36, with the first methionine residue being considered as “0” and optional); ΔN8 / C (15) S (which corresponds to SEQ ID NO: 38, with the first methionine residue being considered as “0” and optional); ΔN8 / C (15) (which corresponds to SEQ ID NO: 40, with the first methionine residue being considered as “0” and optional); C (1.15) S / R (144) E (corresponding to SEQ ID NO: 62, the first methionine residue being considered as “0” and optional); C (1.15) S / R (144) Q (which corresponds to SEQ ID NO: 64, with the first methionine residue being considered as “0” and optional); ΔN23 / H (116) G (which corresponds to amino acids 54-194 of SEQ ID NO: 62), wherein the histidine residue at position 147 is replaced by a glycine residue; ΔN23 / N (137) E (which corresponds to SEQ ID NO: 84, while the first methionine residue is considered the residue "0" and is optional); ΔN23 / K (139) E (which corresponds to SEQ ID NO: 86, while the first methionine residue is considered the residue "0" and is optional); ΔN23 / K (139) Q (which corresponds to SEQ ID NO: 88, with the first methionine residue being considered as “0” and optional); ΔN23 / R (144) A (which corresponds to SEQ ID NO: 90, with the first methionine residue being considered as “0” and optional); ΔN23 / R (144) E (which corresponds to SEQ ID NO: 92, the first methionine residue being considered as “0” and is optional); ΔN23 / R (144) L (which corresponds to SEQ ID NO: 94, with the first methionine residue being considered as “0” and optional); ΔN23 / K (147) E (which corresponds to SEQ ID NO: 96, with the first methionine residue being considered as “0” and optional); ΔN23 / K (147) Q (which corresponds to SEQ ID NO: 98, with the first methionine residue being considered as “0” and optional); ΔN23 / N (153) E (which corresponds to SEQ ID NO: 100, with the first methionine residue being considered as “0” and optional); ΔN23 / NK (153) Q (which corresponds to SEQ ID NO: 102, wherein the first methionine residue is considered the residue "0" and is optional); ΔN23 / Q (152) E / K (153) E (which corresponds to SEQ ID NO: 104, with the first methionine residue being considered as "0" and optional) or ΔN23 / R (144) Q (which corresponds to SEQ ID NO : 66, with the first methionine residue being considered as “0” and optional).  13. The analogue according to claim 12, characterized in that it is ΔN16, which corresponds to SEQ ID NO: 44, while the first methionine residue is considered the residue "0" and is optional.  14. An analogue according to any one of claims 1 to 13, characterized in that it has a signal sequence or an amino terminal methionine residue.  15. The analogue of claim 14, wherein said signal sequence is represented by amino acids 1-31 of the sequence SEQ ID NO: 2.  16. An analogue according to any one of claims 1 to 15, characterized in that it is covalently linked to a chemical residue.  17. The analogue of claim 16, wherein said chemical residue is polyethylene glycol.  18. The analogue of the natural keratinocyte growth factor corresponding to amino acids 32-194 of the sequence SEQ ID NO: 2, with or without a signal sequence, is ΔN23, which corresponds to SEQ ID NO: 58, while the first methionine residue is considered the residue "0" and is optional and covalently linked to a chemical residue.  19. The analogue of claim 18, wherein said chemical residue is polyethylene glycol.  20. An analogue according to any one of claims 1 to 19, characterized in that it is lyophilized.  21. A pharmaceutical composition that stimulates the growth of epithelial cells, containing a therapeutically effective amount of an analogue of the natural keratinocyte growth factor according to any one of claims 1 to 20 and a pharmaceutically acceptable carrier.  22. The pharmaceutical composition according to item 21, designed to stimulate the formation of nephibroblast epithelial cells.  23. The pharmaceutical composition according to claim 22, wherein said non-fibroblastic epithelial cells are selected from adnexal structures, pancreatic cells, liver cells, mucosal epithelial cells of the respiratory and gastrointestinal tracts, corneal cells or inner ear epithelial cells.  24. The pharmaceutical composition according to item 21, intended for the prevention or treatment of a disease condition, which includes burns and other skin lesions throughout the thickness or partial; bullous epidermolysis; chemotherapy-induced baldness; male pattern baldness; progressive hair loss in men and women; ulcers of the stomach and duodenum; inflammatory bowel diseases such as Crohn's disease and ulcerative colitis; toxic intestinal lesions during radiotherapy and chemotherapy; hyaline membrane disease; acute or chronic lung damage; cirrhosis of the liver; transient liver failure; acute viral hepatitis; toxic liver damage; corneal abrasion; progressive gum disease; damage to the eardrum or diabetes.  25. Recombinant nucleic acid molecule encoding an analogue according to any one of claims 1 to 15.  26. An expression vector comprising a nucleic acid molecule according to claim 25, operably linked to regulatory elements.  27. The strain of the microorganism Esherichia coli FM5 (ATCC 53911), transformed with the vector according to p. 26, designed to obtain an analog of natural keratinocyte growth factor.  28. A method of producing an analog of natural keratinocyte growth factor, comprising growing a host cell transformed with a vector containing the nucleic acid of claim 25 under conditions that allow the expression of the encoded analog to be possible, followed by isolation of the resulting analog.  29. The method according to p, characterized in that the host cell is an E. coli cell.  30. The method according to p, characterized in that the host cell is a mammalian cell, preferably a Chinese hamster ovary cell.  31. A method of stimulating the formation of non-fibroblast epithelial cells, which consists in contacting such cells with an effective amount of an analogue of natural keratinocyte growth according to claim 1.  32. The method according to p, characterized in that said nephibroblast cells are selected from adnexal structures, pancreatic cells, liver cells, mucosal epithelial cells of the respiratory and gastrointestinal tracts, corneal cells or epithelial cells of the inner ear.

Images